From barry@zope.com  Tue Oct  1 23:41:21 2002
From: barry@zope.com (Barry A. Warsaw)
Date: Tue, 1 Oct 2002 18:41:21 -0400
Subject: [Mailman-Developers] Weirdness with Mailman and Apache 2.0.42?
Message-ID: <15770.9361.787672.183483@gargle.gargle.HOWL>


I just built out my new server, RH 7.3 (up-to-date), and Apache 2.0.42
with the current Mailman 2.1b-cvs.  I'm seeing a weird problem with
the "Bug in Mailman" page.  The page gets returned as Content-Type
text/plain even though it clearly prints "Content-Type: text/html"
before printing the standard error screen.

A little sluething with lynx shows that I'm seeing the MIME headers
get returned from the server like:

Server: Apache 2.0.42 (Unix)
 Content-type: text/html
^-------
notice this space, I believe this is what's messing up the browser's
calculation of the returned content type.

I don't understand what's going on.  With Apache 1.3.26 it all works
fine with the exact same Mailman code.  With Apache 2.0.42 (and .40) I
get the extra space.

Admittedly, I haven't dug into the Apache documentation much except to
browse through the CGI manual a bit.  Nothing jumped out at me as
being different b/w 2.0 and 1.3 for writing CGI scripts.

You folks may not have seen the problem if you never got the "We're
sorry" page.  Here's an easy way to provoke it:

- Edit $prefix/Mailman/Cgi/admin.py
- at the top of the file enter a line with "doink" on it

This'll provoke a Python SyntaxError which should be caught by
scripts/driver and printed in a nice format (if STEALTH_MODE = 0).

Can anybody verify this particular problem?  Does it ring any bells?
Is it a bug in Mailman, Python, or Apache (or somewhere else?).  I've
tried it with both lynx and Moz so I don't think it's a browser issue.

-Barry

From barry@python.org  Tue Oct  1 23:45:54 2002
From: barry@python.org (Barry A. Warsaw)
Date: Tue, 1 Oct 2002 18:45:54 -0400
Subject: [Mailman-Developers] Weirdness with Mailman and Apache 2.0.42?
References: <15770.9361.787672.183483@gargle.gargle.HOWL>
Message-ID: <15770.9634.144254.874343@gargle.gargle.HOWL>


Oh yeah, non error pages work fine, so that might be a clue <wink>.
-Barry

From barry@zope.com  Tue Oct  1 23:57:49 2002
From: barry@zope.com (Barry A. Warsaw)
Date: Tue, 1 Oct 2002 18:57:49 -0400
Subject: FW: [Mailman-Developers] Weirdness with Mailman and Apache
	2.0.42?
References: <15770.9361.787672.183483@gargle.gargle.HOWL>
	<B9BF750F.3E2A9%jon@apache.org>
Message-ID: <15770.10349.361721.606524@gargle.gargle.HOWL>


>>>>> "JS" == Jon Stevens <jon@apache.org> writes:

    JS> Not sure who to send this to, but it seems kind of
    JS> important...

Thanks for the forward Jon, but I just figured it out.  It is
definitely /not/ a problem with Apache.  Sorry for the false alarm.

-Barry

From chuqui@plaidworks.com  Wed Oct  2 04:18:10 2002
From: chuqui@plaidworks.com (Chuq Von Rospach)
Date: Tue, 1 Oct 2002 20:18:10 -0700
Subject: [Mailman-Developers] problem with 2.1b3+
Message-ID: <942EB3A0-D5B5-11D6-9A31-0003934516A8@plaidworks.com>


Ran into a problem with 2.1b3+

I configured my new system as "new.plaidworks.com" and did all my work 
there.

I've just put it into production and renamed it plaidworks.com. I've 
rebuilt everything. Unfortunately, I keep seeing mailman URLs showing 
up under the old "new.plaidworks.com" name.

Digging, i did a dumpdb of a list database. In it is:

     'web_page_url': 'http://new.plaidworks.com/mailman/',

uh, oh. I don't seem to be able to fix this with either config_list or 
the admin web pages. so bad things are happening.

Help?


-- 
Chuq Von Rospach, Architech
chuqui@plaidworks.com -- http://www.chuqui.com/

The first rule of holes: If you are in one, stop digging.


From dan.mick@sun.com  Wed Oct  2 04:45:05 2002
From: dan.mick@sun.com (Dan Mick)
Date: Tue, 01 Oct 2002 20:45:05 -0700
Subject: [Mailman-Developers] problem with 2.1b3+
References: <942EB3A0-D5B5-11D6-9A31-0003934516A8@plaidworks.com>
Message-ID: <3D9A6BC1.6000304@sun.com>

Chuq Von Rospach wrote:
> 
> Ran into a problem with 2.1b3+
> 
> I configured my new system as "new.plaidworks.com" and did all my work 
> there.
> 
> I've just put it into production and renamed it plaidworks.com. I've 
> rebuilt everything. Unfortunately, I keep seeing mailman URLs showing up 
> under the old "new.plaidworks.com" name.
> 
> Digging, i did a dumpdb of a list database. In it is:
> 
>     'web_page_url': 'http://new.plaidworks.com/mailman/',
> 
> uh, oh. I don't seem to be able to fix this with either config_list or 
> the admin web pages. so bad things are happening.
> 
> Help?

bin/fix_url.py



From chuqui@plaidworks.com  Wed Oct  2 05:00:22 2002
From: chuqui@plaidworks.com (Chuq Von Rospach)
Date: Tue, 1 Oct 2002 21:00:22 -0700
Subject: [Mailman-Developers] problem with 2.1b3+
In-Reply-To: <3D9A6BC1.6000304@sun.com>
Message-ID: <791EB577-D5BB-11D6-9A31-0003934516A8@plaidworks.com>

bingo. Thanks, dan. I looked for move_lists. Forgot it got redone...


On Tuesday, October 1, 2002, at 08:45  PM, Dan Mick wrote:

> Chuq Von Rospach wrote:
>> Ran into a problem with 2.1b3+
>> I configured my new system as "new.plaidworks.com" and did all my 
>> work there.
>> I've just put it into production and renamed it plaidworks.com. I've 
>> rebuilt everything. Unfortunately, I keep seeing mailman URLs showing 
>> up under the old "new.plaidworks.com" name.
>> Digging, i did a dumpdb of a list database. In it is:
>>     'web_page_url': 'http://new.plaidworks.com/mailman/',
>> uh, oh. I don't seem to be able to fix this with either config_list 
>> or the admin web pages. so bad things are happening.
>> Help?
>
> bin/fix_url.py
>
>
-- 
Chuq Von Rospach, Architech
chuqui@plaidworks.com -- http://www.chuqui.com/

He doesn't have ulcers, but he's a carrier.


From barry@python.org  Wed Oct  2 12:25:03 2002
From: barry@python.org (Barry A. Warsaw)
Date: Wed, 2 Oct 2002 07:25:03 -0400
Subject: [Mailman-Developers] problem with 2.1b3+
References: <942EB3A0-D5B5-11D6-9A31-0003934516A8@plaidworks.com>
Message-ID: <15770.55183.541944.33700@gargle.gargle.HOWL>


>>>>> "CVR" == Chuq Von Rospach <chuqui@plaidworks.com> writes:

    CVR> Ran into a problem with 2.1b3+

    CVR> I configured my new system as "new.plaidworks.com" and did
    CVR> all my work there.

    CVR> I've just put it into production and renamed it
    CVR> plaidworks.com. I've rebuilt everything. Unfortunately, I
    CVR> keep seeing mailman URLs showing up under the old
    CVR> "new.plaidworks.com" name.

    CVR> Digging, i did a dumpdb of a list database. In it is:

    CVR>      'web_page_url': 'http://new.plaidworks.com/mailman/',

    CVR> uh, oh. I don't seem to be able to fix this with either
    CVR> config_list or the admin web pages. so bad things are
    CVR> happening.

    CVR> Help?

web_page_url isn't configurable through the web in MM2.1 because too
many people messed up and made their lists unfunctional (screw it up
and you can't get to your list through the web anymore!).

You ought to be able to do it with config_list, so I'd be interested
in more information as to why that doesn't work.  But I'd prefer
fix_url.py instead for this task.

Once you've set up your mm_cfg.py with the new values, including
VIRTUAL_HOSTS, you can simply run

    % bin/withlist -l -r fix_url listname [options]

to switch listname's web_page_url and host_name attributes to their
new values.  Do

    % bin/fix_url.py

for docstring synopsis.

-Barry

From chuqui@plaidworks.com  Wed Oct  2 15:35:46 2002
From: chuqui@plaidworks.com (Chuq Von Rospach)
Date: Wed, 2 Oct 2002 07:35:46 -0700
Subject: [Mailman-Developers] problem with 2.1b3+
In-Reply-To: <15770.55183.541944.33700@gargle.gargle.HOWL>
Message-ID: <3D44977A-D614-11D6-9A31-0003934516A8@plaidworks.com>


On Wednesday, October 2, 2002, at 04:25  AM, Barry A. Warsaw wrote:

> You ought to be able to do it with config_list, so I'd be interested
> in more information as to why that doesn't work.

config_list didn't seem to have the value in it when I did a -o


-- 
Chuq Von Rospach, Architech
chuqui@plaidworks.com -- http://www.chuqui.com/

IMHO: Jargon. Acronym for In My Humble Opinion. Used to flag as an 
opinion
something that is clearly from context an opinion to everyone except the
mentally dense. Opinions flagged by IMHO are actually rarely humble. 
IMHO.
     (source: third unabridged dictionary of chuqui-isms).


From gargp@earthlink.net  Wed Oct  2 18:03:12 2002
From: gargp@earthlink.net (Pankaj K Garg)
Date: Wed, 2 Oct 2002 10:03:12 -0700
Subject: [Mailman-Developers] 
 RE: [Mailman-Users] Re: Re: One Button Direct Subscription
In-Reply-To: <FMELKIGJMCKDEONBJEDGMEGKEFAA.jimpop@rocketship.com>
Message-ID: <000901c26a35$98aa8c40$8d835f0f@home.org>

Yes, indeed this is the situation I am also in. I have an application,
somewhat like SourceForge, where we embed Mailman as the list manager.
The user is already logged in to the main application, and we know
the user's email address, etc. So, its not very nice to have:
(1) separate passwords for each mailing list, and (2) require the user
to type in their identity and password for subscription, options editing,
and un-subscribe requests.

I'm in the process of changing some of the Cgi scripts and making a
few changes to Mailman 2.0.9 to remove these aspects for an already
authenticated user.

If others are interested, I can put a patch together with the latest
Mailman code.

Regards

Pankaj
 

> -----Original Message-----
> From: mailman-users-admin@python.org
> [mailto:mailman-users-admin@python.org]On Behalf Of Jim Popovitch
> Sent: Wednesday, October 02, 2002 4:24 AM
> To: Will Yardley; mailman-users@python.org
> Subject: RE: [Mailman-Users] Re: Re: One Button Direct Subscription
> 
> 
> Hi Will,
> 
> That is not entirely true.  Sometimes you already have the person
> authenticated via another fashion (think: online banking 
> software), and all
> that you want to do is have the user subscribe for banking notices by
> pressing one button.
> 
> -Jim P.
> 
> > -----Original Message-----
> > From: Will Yardley
> > Sent: Wednesday, October 02, 2002 4:11 AM
> >
> > Well in any event, you should confirm the subscription; 
> otherwise anyone
> > can subscribe anyone else to your list.
> >
> > This is known as confirmed opt-in, and if you don't use a 
> system like
> > this, you're very likely to get booted by your ISP for spamming.
> >
> > --
> > Will Yardley
> > input: william < @ hq . newdream . net . >
> 
> 
> 
> 
> 
> 
> 
> 
> 
> 
> ------------------------------------------------------
> Mailman-Users mailing list
> Mailman-Users@python.org
> http://mail.python.org/mailman/listinfo/mailman-users
> Mailman FAQ: http://www.python.org/cgi-bin/faqw-mm.py
> Searchable Archives: 
http://www.mail-archive.com/mailman-users%40python.org/


From noreply@sourceforge.net  Thu Oct  3 07:14:50 2002
From: noreply@sourceforge.net (noreply@sourceforge.net)
Date: Wed, 02 Oct 2002 23:14:50 -0700
Subject: [Mailman-Developers] 
 [ mailman-Bugs-617930 ] MIME handling very dumb, breaks messages
Message-ID: <E17wzG6-0002a5-00@usw-sf-web4.sourceforge.net>

Bugs item #617930, was opened at 2002-10-03 00:14
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=100103&aid=617930&group_id=103

Category: mail delivery
Group: 2.0.x
Status: Open
Resolution: None
Priority: 5
Submitted By: Dylan Griffiths (inoshiro)
Assigned to: Nobody/Anonymous (nobody)
Summary: MIME handling very dumb, breaks messages

Initial Comment:
Mailman delivered a partially garbled message on a
mailing list today because it doesn't handle
non-text/plain messages.  The incoming message was:
Content-Transfer-Encoding: base64
Content-Type: text/plain;charset="utf-8"

So the content was something like this:
QW55b25lIG91dCB0aGVyZSB1c2luZyBNUlRHIHRvIGNvbGxlY3Qgcm91dGVyIGRhdGEgLCBJZiBz
byB3aGF0IGtpbmQgb2YgYmFja2VuZA0KYXJlIHlvdSB1c2luZyBJIGFtIHRyeWluZyB0byBjb25m
aWcgaXQgdG8gdXNlIHBvc3RncmVzICwgYW55IGhpbnRzIG9yIHRpcHMgYXMgdG8gdGhlIGJlc3Qg
d2F5IHRvIGdvIGFib3V0IHRoaXMNCiANCiANCiANCg==

To which Mailman blindly attached a text/plain
signature (rather than another mime section and marking
the body as mixed), which then decodes in any MIME
aware mailreader as a large blob of lower-ASCII crap:

I&#65533;$j&#65533;(&#65533;&#65533;&#65533;&#65533;&#65533;&#65533;&#65533;Lj)b&#65533;&#65533;&#65533;"&#65533;;
&#65533;&#65533;&#65533;Z&#65533;&#65533;&#1886;&#65533;&#65533;&#65533;&#65533;{&#65533;&#65533;&#65533;&#65533;z(&#65533;X&#65533;&#65533;&#1511;Z&#65533;&#65533;r&#65533;}:-&#65533;&#65533;ej^&#65533;&#65533;+
&#65533;&#65533;&#65533;+&#65533;&#65533;b&#65533;v&#65533;&#65533;&#65533;m&#65533;&#65533;b&#65533;&#1447;v&#731;&#65533;&#65533;&#65533;&#1576;&#65533;f&#65533;j&#65533;z{`&#65533;&#65533;zm&#65533;&#65533;&#65533;a&#65533;&#65533;&#65533;o&#65533;j)
fj&#65533;&#65533;b&#65533;&#65533;?slg=

Please adjust the behaviour to not append a signature
if it can't attach it in such a way that data is not
garbled.

----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=100103&aid=617930&group_id=103

From noreply@sourceforge.net  Thu Oct  3 12:43:27 2002
From: noreply@sourceforge.net (noreply@sourceforge.net)
Date: Thu, 03 Oct 2002 04:43:27 -0700
Subject: [Mailman-Developers] 
 [ mailman-Bugs-617930 ] MIME handling very dumb, breaks messages
Message-ID: <E17x4O7-0005UI-00@usw-sf-web4.sourceforge.net>

Bugs item #617930, was opened at 2002-10-03 02:14
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=100103&aid=617930&group_id=103

Category: mail delivery
Group: 2.0.x
>Status: Closed
>Resolution: Fixed
Priority: 5
Submitted By: Dylan Griffiths (inoshiro)
Assigned to: Nobody/Anonymous (nobody)
Summary: MIME handling very dumb, breaks messages

Initial Comment:
Mailman delivered a partially garbled message on a
mailing list today because it doesn't handle
non-text/plain messages.  The incoming message was:
Content-Transfer-Encoding: base64
Content-Type: text/plain;charset="utf-8"

So the content was something like this:
QW55b25lIG91dCB0aGVyZSB1c2luZyBNUlRHIHRvIGNvbGxlY3Qgcm91dGVyIGRhdGEgLCBJZiBz
byB3aGF0IGtpbmQgb2YgYmFja2VuZA0KYXJlIHlvdSB1c2luZyBJIGFtIHRyeWluZyB0byBjb25m
aWcgaXQgdG8gdXNlIHBvc3RncmVzICwgYW55IGhpbnRzIG9yIHRpcHMgYXMgdG8gdGhlIGJlc3Qg
d2F5IHRvIGdvIGFib3V0IHRoaXMNCiANCiANCiANCg==

To which Mailman blindly attached a text/plain
signature (rather than another mime section and marking
the body as mixed), which then decodes in any MIME
aware mailreader as a large blob of lower-ASCII crap:

I&#65533;$j&#65533;(&#65533;&#65533;&#65533;&#65533;&#65533;&#65533;&#65533;Lj)b&#65533;&#65533;&#65533;"&#65533;;
&#65533;&#65533;&#65533;Z&#65533;&#65533;&#1886;&#65533;&#65533;&#65533;&#65533;{&#65533;&#65533;&#65533;&#65533;z(&#65533;X&#65533;&#65533;&#1511;Z&#65533;&#65533;r&#65533;}:-&#65533;&#65533;ej^&#65533;&#65533;+
&#65533;&#65533;&#65533;+&#65533;&#65533;b&#65533;v&#65533;&#65533;&#65533;m&#65533;&#65533;b&#65533;&#1447;v&#731;&#65533;&#65533;&#65533;&#1576;&#65533;f&#65533;j&#65533;z{`&#65533;&#65533;zm&#65533;&#65533;&#65533;a&#65533;&#65533;&#65533;o&#65533;j)
fj&#65533;&#65533;b&#65533;&#65533;?slg=

Please adjust the behaviour to not append a signature
if it can't attach it in such a way that data is not
garbled.

----------------------------------------------------------------------

>Comment By: Barry A. Warsaw (bwarsaw)
Date: 2002-10-03 07:43

Message:
Logged In: YES 
user_id=12800

This won't be fixed in MM2.0, but it is fixed in MM2.1.

----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=100103&aid=617930&group_id=103

From chuqui@plaidworks.com  Thu Oct  3 15:17:14 2002
From: chuqui@plaidworks.com (Chuq Von Rospach)
Date: Thu, 3 Oct 2002 07:17:14 -0700
Subject: [Mailman-Developers] 2.1b3+ -- moderation checks out of order.
Message-ID: <D0BBAE03-D6DA-11D6-B994-0003934516A8@plaidworks.com>


Looks like the moderation checks are out of order. I ahve my system set 
up with a 12K limit on posts, and to auto-reject non-subscribers. If I 
get a piece of spam that's 15K, it gets held in the moderation queue, 
not rejected, because it checks the 12K limit before it checks the 
non-subscribe status....


-- 
Chuq Von Rospach, Architech
chuqui@plaidworks.com -- http://www.chuqui.com/

The Cliff's Notes Cliff's Notes on Hamlet:
     And they all died happily ever after


From wheakory@isu.edu  Thu Oct  3 19:59:12 2002
From: wheakory@isu.edu (Kory Wheatley)
Date: Thu, 03 Oct 2002 12:59:12 -0600
Subject: [Mailman-Developers] Edit a message then approve it
Message-ID: <3D9C9380.188031CB@isu.edu>

I have tried the below documentation for editing a held message then
approving it or sending it through, but it does not work.

Steps:

I forward the message to myself and click on the Reply-to, to bring it
up for editing
and then I edit the message (I'm using netscape messenger to do so). How
do I do this part of the documentation "
- Resend the message to the list using a Resent-To: header containing
the list posting address." There is no Resent-To: header option when you
compose a message in the Netscape Mail Client, so then I put the
Approval Header in the very first line of the body of the message
"Approved:password" which I use the site mailman password. I send the
message off  (which I use the mailman mailing list address), and the
message is sent to me with the "Approval:password" in the first line of
the message, the reason it was not held for approval is because I allow
my subscribed email to post without approval but the message did not
edit properly, because it shows the Approval:password in the first line
of the body. What I'm I doing wrong.



3.9. How do I edit a held message before approving it for the list?

Here's how to do this:

- Use the "forward this message" feature in the admindb page to forward
the message to yourself.

- At the same time, discard the original held message. You may want to
do this later, after you're sure the message was properly
forwarded to you.

- Edit the message in your mail reader. You should include a notice in
the message explaining that the list moderator has edited the
message. Please use proper netiquette!

- Resend the message to the list using a Resent-To: header containing
the list posting address. Also include an Approved: header
containing the list's admin password.

Viola! The edited message should be forwarded to the list membership.

==========

Alternatively, you can edit the actual held message found in
~mailman/data/... The message will be in a file called
heldmsg-<listname>-<number>.txt

You can use "grep" to find the message, and your favorite editor to
change it.

Once you have modified the message, feel free to approve it.

The web-interface shows you a portion of the message that has been
copied into a database called request.db. When you edit the
message via the web-interface you are actually editing the portion
captured in request.db and not the original message.

Approval of the message sends the actual held message from
~mailman/data/..

==========

Note for Mailman 2.1: the second suggestion about doesn't work so well
because by default, held messages are stored on disks in a
more efficient binary representation called a "pickle". Pickles aren't
normally editable. If you still want to use the second suggestion, you
might want to set HOLD_MESSAGES_AS_PICKLES in ListAdmin.py to 0, which
tells Mailman to store them using the older, less
efficient, plaintext representation instead of pickles.

I still don't like the second suggestion much because it is limited to
only those who have access to the system disk and command line.
Also, in MM2.1 the admindb interface doesn't fool you into thinking you
can edit the message, as the text areas are now read-only.

Edit this entry / Log info / Last changed on Thu Apr 18 23:31:06 2002 by
Barry Warsaw

--
#########################################
Kory Wheatley
Academic Computing Analyst Sr.
Phone 282-3874
#########################################
Everything must point to him.



From Dale@Newfield.org  Fri Oct  4 22:43:18 2002
From: Dale@Newfield.org (Dale Newfield)
Date: Fri, 4 Oct 2002 17:43:18 -0400 (EDT)
Subject: [Mailman-Developers] Disallowed characters in email address?
In-Reply-To: <15733.63405.157076.576891@anthem.wooz.org>
Message-ID: <Pine.OSX.4.44.0210041739380.10783-100000@core.newfield.org>


Just stumbled across this in Utils.py:
> # TBD: what other characters should be disallowed?
> _badchars = re.compile('[][()<>|;^,/]')

and thought I'd suggest that " and ' get added to that list...

I recently wound up with a list subscriber of the form "foo@bar.baz"
(*with* the quotes!) and had a more difficult time fixing it than you
might expect.

-Dale


From mrbill@mrbill.net  Sat Oct  5 21:31:39 2002
From: mrbill@mrbill.net (Bill Bradford)
Date: Sat, 5 Oct 2002 15:31:39 -0500
Subject: [Mailman-Developers] help!
Message-ID: <20021005203139.GL1223@mrbill.net>

Problems with a new MM 2.0.13 install, fresh compile of Python 2.2.1:

Moved lists to a new machine, new Python 2.2.1c1/Mailman 2.0.13 
install, and when I try to send mail to a list, I get this:

Oct  5 15:28:19 ohno.mrbill.net postfix/local[9535]: [ID 197553 mail.info] 80EFD499BC: to=<test@mrbill.net>, relay=local, delay=0, status=bounced (Command diedwith status 1: "/usr/local/mailman/mail/wrapper post test". Command output: Traceback (most recent call last):   File "/usr/local/mailman/scripts/post", line 33, in ?     from Mailman import MailList   File "/usr/local/mailman/Mailman/MailList.py", line 27, in ?     import errno ImportError: ld.so.1: /usr/local/bin/python: fatal: libgcc_s.so.1: open failed: No such file or directory )

Suggestions?  "make test" on Python looks fine, Mailman compiles/installs
cleanly.. I need to get these lists back up..

root@ohno:/var/adm> ls -al /usr/local/lib/libgcc*
lrwxrwxrwx    1 root     other          13 Oct  1 23:55 /usr/local/lib/libgcc_s.so -> libgcc_s.so.1
-rw-r--r--    1 bin      bin        802748 Aug 28 04:00 /usr/local/lib/libgcc_s.so.1

I cant tell which file it cant find/import.

Everything was compiled with GCC 3.2.

Bill

-- 
bill bradford
mrbill@mrbill.net
austin, texas

From mrbill@mrbill.net  Sat Oct  5 22:01:55 2002
From: mrbill@mrbill.net (Bill Bradford)
Date: Sat, 5 Oct 2002 16:01:55 -0500
Subject: [Mailman-Developers] help!
In-Reply-To: <20021005203139.GL1223@mrbill.net>
References: <20021005203139.GL1223@mrbill.net>
Message-ID: <20021005210155.GQ1223@mrbill.net>

On Sat, Oct 05, 2002 at 03:31:39PM -0500, Bill Bradford wrote:
> Oct  5 15:28:19 ohno.mrbill.net postfix/local[9535]: [ID 197553 mail.info] 80EFD499BC: to=<test@mrbill.net>, relay=local, delay=0, status=bounced (Command diedwith status 1: "/usr/local/mailman/mail/wrapper post test". Command output: Traceback (most recent call last):   File "/usr/local/mailman/scripts/post", line 33, in ?     from Mailman import MailList   File "/usr/local/mailman/Mailman/MailList.py", line 27, in ?     import errno ImportError: ld.so.1: /usr/local/bin/python: fatal: libgcc_s.so.1: open failed: No such file or directory )

Was a ld_library_path problem; fixed it with crle (solaris 9).

Bill

-- 
bill bradford
mrbill@mrbill.net
austin, texas

From barry@python.org  Sat Oct  5 23:11:44 2002
From: barry@python.org (Barry A. Warsaw)
Date: Sat, 5 Oct 2002 18:11:44 -0400
Subject: [Mailman-Developers] help!
References: <20021005203139.GL1223@mrbill.net>
	<20021005210155.GQ1223@mrbill.net>
Message-ID: <15775.25504.799774.229263@gargle.gargle.HOWL>


>>>>> "BB" == Bill Bradford <mrbill@mrbill.net> writes:

    BB> Was a ld_library_path problem; fixed it with crle (solaris 9).

Yep, there have been reports of problems building Python on Solaris 8
and 9 with gcc 3.1 and 3.2.  I think Python 2.2.2 (to be released next
week) might do better, but there may still be dynamic linker problems
lurking.

-Barry

From mjm@michaelmeltzer.com  Sun Oct  6 04:03:55 2002
From: mjm@michaelmeltzer.com (Michael Meltzer)
Date: Sat, 5 Oct 2002 23:03:55 -0400
Subject: [Mailman-Developers] 2.1b3+
Message-ID: <00c001c26ce5$02498b40$0b01a8c0@mjm2>

about 7 weeks ago in MimeDel.py it looks like the alternative collapse code was changed to "if mtype == 'multipart/alternative'"
Looks like it should be "if ctype == 'multipart/alternative'" with mtype the compare will always be false because mtype will be
multipart.

MJM


From noreply@sourceforge.net  Sun Oct  6 15:18:49 2002
From: noreply@sourceforge.net (noreply@sourceforge.net)
Date: Sun, 06 Oct 2002 07:18:49 -0700
Subject: [Mailman-Developers] 
 [ mailman-Patches-444884 ] Integration of Mailman & htdig for archi
Message-ID: <E17yCF7-0002OK-00@usw-sf-web4.sourceforge.net>

Patches item #444884, was opened at 2001-07-26 18:27
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=300103&aid=444884&group_id=103

Category: None
Group: None
Status: Open
Resolution: None
Priority: 7
Submitted By: Richard Barrett (ppsys)
Assigned to: Barry A. Warsaw (bwarsaw)
Summary: Integration of Mailman & htdig for archi

Initial Comment:
This patch is applicable to Mailman 2.0.6 release that 
has had search enhancement patch 444879 patch 
installed - if your Defaults.py has the 
ARCHIVE_INDEXING_ENABLE and ARCHIVE_INDEXING_DISABLE 
in it then you've got that patch.

It replaces earlier patches 401670 and 402423 and is 
mainly to correct some problems arising from fixes 
introduced into Mailman by bug fix releases since the 
402423 patch. 

This patch integrates htdig with Mailman and provides:

1. per list search facility with a search form on the 
list's TOC page.

2. maintenance of privacy of private archives which 
requires the user to establish their credentials via 
the normal private archive access before any access 
via htdig is allowed.
   
3. a common base URL for both public and private 
archive access via htsearch results so that htdig 
indices are unaffected by changingan archive from 
private to public and vice versa. All access to 
archives via htdig is controlled by a new wrapped cgi-
bin script called htdig.py.
   
4. a new cron activated script and extra crontab entry 
which runs htdig regularly to maintain the per list 
search indices.
   
5. automatic creation, deletion and maintenance of 
htdig configuration files and such. Beyond installing 
htdig and telling Mailman where it is via mm_cfg you 
do not have to do any other setup. Well not quite you 
do have to set up a single per installation symlink to 
allow htdig to find the automatically generated per 
list htdig configuration files.

You probably want to run this patch as follows:

cd <mailman 2.0.6 untarred and unzipped directory>
patch -p1 < <this patch file>  

----------------------------------------------------------------------

Comment By: Colin Mackinlay (cmackinlay)
Date: 2002-10-06 14:18

Message:
Logged In: YES 
user_id=624179

Just rebuilt MM as 2.1b3 with htdig.
Upgraded lists which had htdig before work fine
New lists give the obvious error:
  Unable to read word database file
  Did you run htmerge?
Running the cronjob doesn't fix as it used to, message is:
  Output from command /usr/bin/python -
S /usr/local/mailman/cron/nightly_htdig ..

Traceback (most recent call last):
  File "/usr/local/mailman/cron/nightly_htdig", line 153, in ?
    main()
  File "/usr/local/mailman/cron/nightly_htdig", line 118, in main
    file(rundig_run_file, 'w').close()
NameError: global name 'file' is not defined

The archive/htdig folder only contains the xx.conf file, but no 
db.xx files

If I copy in db.xx files from another list then the problem goes 
away (except I've now got an invalid set of references!)

Is this my elementary error or is it more sinister?!


----------------------------------------------------------------------

Comment By: Richard Barrett (ppsys)
Date: 2002-08-15 11:02

Message:
Logged In: YES 
user_id=75166

htdig-2.1b3-0.2.patch corrects a dumb syntax error in htdig-
2.1b3-0.1.patch which will typically show up as logged errors 
in the operation of the ArchRunner qrunner at line 721 of 
HyperArch.py

----------------------------------------------------------------------

Comment By: Richard Barrett (ppsys)
Date: 2002-08-14 16:51

Message:
Logged In: YES 
user_id=75166

htdig-2.1b3-0.1.patch is a revised version of the patch that is 
compatible with MM 2.1b3

----------------------------------------------------------------------

Comment By: Barry A. Warsaw (bwarsaw)
Date: 2002-08-08 16:33

Message:
Logged In: YES 
user_id=12800

I've sent Richard some comments off-line about this patch.

Meta comments: the 2.0.x patches can't be officially
supported, but I'm going to create an unofficial patches
page off the wiki for where the 2.0 patches can be migrated.

I think this patch set is too big for MM2.1, but if it's
cleaned up as per my private message, let's re-evaluate it
for MM2.2 (or 3.0).

----------------------------------------------------------------------

Comment By: Richard Barrett (ppsys)
Date: 2002-08-05 10:11

Message:
Logged In: YES 
user_id=75166

htdig-2.0.13-0.2.patch just adds a GPL notice to the patch


----------------------------------------------------------------------

Comment By: Richard Barrett (ppsys)
Date: 2002-08-01 16:35

Message:
Logged In: YES 
user_id=75166

htdig-2.1b2-0.1.patch is a revised version of the patch 
that is compatible with MM 2.1b2

----------------------------------------------------------------------

Comment By: Richard Barrett (ppsys)
Date: 2002-07-30 11:25

Message:
Logged In: YES 
user_id=75166

htdig-2.0.13-0.1.patch is purely cosmetic to get no mumble 
application to MM 2.0.13

----------------------------------------------------------------------

Comment By: Richard Barrett (ppsys)
Date: 2002-07-25 15:07

Message:
Logged In: YES 
user_id=75166

Do not use htdig-2.0.12-0.1.patch there is an error in it.
Use htdig-2.0.12-0.2.patch instead

----------------------------------------------------------------------

Comment By: Richard Barrett (ppsys)
Date: 2002-07-25 14:10

Message:
Logged In: YES 
user_id=75166

htdig-2.0.12-0.1.patch is a revised version of the patch that 
applies without complaint to MM 2.0.12.

It also add a facility for adding site wide htdig configuration 
attributes to all list specific htdig configuration files. 



----------------------------------------------------------------------

Comment By: Richard Barrett (ppsys)
Date: 2002-05-23 09:59

Message:
Logged In: YES 
user_id=75166

htdig-2.0.11-0.1.patch is a revised version of the patch that 
is compatible with MM 2.0.11

This version removes an incompatibility with Python 2.2 
which caused warning messages to be generated when any 
of the family cron/nightly_htdig scripts were run.

Some guidance on file access permissions for some htdig 
database files needed by rundig have been added to 
installation notes.

----------------------------------------------------------------------

Comment By: Richard Barrett (ppsys)
Date: 2002-04-19 10:59

Message:
Logged In: YES 
user_id=75166

htdig-2.0.10-0.1.patch is a revised version of the patch 
that is compatible with MM 2.0.10

----------------------------------------------------------------------

Comment By: Richard Barrett (ppsys)
Date: 2002-04-08 17:46

Message:
Logged In: YES 
user_id=75166

htdig-2.0.9-0.1.patch is a revised version of the patch 
that is compatible with MM 2.0.9

----------------------------------------------------------------------

Comment By: Richard Barrett (ppsys)
Date: 2002-03-06 16:22

Message:
Logged In: YES 
user_id=75166

htdig-2.1cvs-20020306.patch is a revised version of the patch that is compatible with the code published in 
mailman CVS on sourceforge as 12:30 GMT 6 Mar 2002

Known deficiency is that the non-English versions of files under $build/templates still contain text in English 
and need translations I cannot do. Also the necessary pygettext activity and subsequent translations in 
files under $build/messages remain to be done.

----------------------------------------------------------------------

Comment By: Richard Barrett (ppsys)
Date: 2001-12-17 16:56

Message:
Logged In: YES 
user_id=75166

htdig-2.1cvs-20011217.patch is a revised version of the 
patch that is compatible with the code published in mailman 
CVS on sourceforge as 11:50 GMT 17 Dec 2001

The only known deficiency is that the non-English versions 
of files under $build/templates still contain text in 
English and need translations I cannot do. Also the 
necessary pygettext activity and subsequent translations in 
files under $build/messages remain to be done.

----------------------------------------------------------------------

Comment By: Richard Barrett (ppsys)
Date: 2001-12-13 16:58

Message:
Logged In: YES 
user_id=75166

htdig-2.1a3-0.1.patch is a revised version of the patch that is compatible with the code published in 
mailman-2.1a3.tgz on sourceforge.

The only known deficiency is that the non-English versions of files under $build/templates still contain text 
in English and need translations I cannot do. Also the necessary pygettext activity and subsequent 
translations in files under $build/messages remain to be done. 

----------------------------------------------------------------------

Comment By: Richard Barrett (ppsys)
Date: 2001-11-28 17:33

Message:
Logged In: YES 
user_id=75166

The htdig-2.0.8-0.1.patch version of the patch resolves a problem that can arise with htdig indexing if the 
web_page_url for a list uses other than the http addressing (some folks want to use https). While specified 
as for MM 2.0.8 the revised patch should work OK with 2.0.7, 2.0.6 and probably back as far as 2.0.3. If 
you do not have the requirement for using other than http addressing in you lists web_page_urls it probably 
isn't worth the trouble of upgrading to this patch level.

----------------------------------------------------------------------

Comment By: Richard Barrett (ppsys)
Date: 2001-11-28 11:08

Message:
Logged In: YES 
user_id=75166

This patch should also apply without problems to MM 2.0.8

----------------------------------------------------------------------

Comment By: Richard Barrett (ppsys)
Date: 2001-11-27 12:00

Message:
Logged In: YES 
user_id=75166

This patch should also apply without problems to Mm 2.0.7

----------------------------------------------------------------------

Comment By: Richard Barrett (ppsys)
Date: 2001-11-09 11:54

Message:
Logged In: YES 
user_id=75166

The htdig-2.0.6-03.patch version of the patch makes some 
previously hard-coded things configurable and enhances the 
capability to run the htdig searches and indexing on a 
different machine to the one delivering Mailman and 
Mailman's web UI.

----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=300103&aid=444884&group_id=103

From darrell@grumblesmurf.net  Sun Oct  6 21:00:30 2002
From: darrell@grumblesmurf.net (Darrell Fuhriman)
Date: 06 Oct 2002 13:00:30 -0700
Subject: [Mailman-Developers] Disallowed characters in email address?
In-Reply-To: <Pine.OSX.4.44.0210041739380.10783-100000@core.newfield.org>
References: <Pine.OSX.4.44.0210041739380.10783-100000@core.newfield.org>
Message-ID: <tshhefz9w2p.fsf@startop.grumblesmurf.net>

Dale Newfield <Dale@Newfield.org> writes:

> Just stumbled across this in Utils.py:
> > # TBD: what other characters should be disallowed?
> > _badchars = re.compile('[][()<>|;^,/]')
> 

> and thought I'd suggest that " and ' get added to that list...

Nope, because both of those characters are valid e-mail address
components.  Not in the form you mention (i.e. the double quote
is not allowed in the domain part), but certainly in the local
part.

In fact this is a perfectly valid e-mail address:

"f@,'[& "@example.com

And the parser should be able to cope with them, or any other RFC
2822 compliant address.

Darrell

From barry@python.org  Mon Oct  7 21:38:09 2002
From: barry@python.org (Barry A. Warsaw)
Date: Mon, 7 Oct 2002 16:38:09 -0400
Subject: [Mailman-Developers] 2.1b3+
References: <00c001c26ce5$02498b40$0b01a8c0@mjm2>
Message-ID: <15777.61617.922059.524356@gargle.gargle.HOWL>


>>>>> "MM" == Michael Meltzer <mjm@michaelmeltzer.com> writes:

    MM> about 7 weeks ago in MimeDel.py it looks like the alternative
    MM> collapse code was changed to "if mtype ==
    MM> 'multipart/alternative'" Looks like it should be "if ctype ==
    MM> 'multipart/alternative'" with mtype the compare will always be
    MM> false because mtype will be multipart.

Clearly a typo, thanks.
-Barry

From ayanic01@cs.fiu.edu  Mon Oct  7 22:53:34 2002
From: ayanic01@cs.fiu.edu (Asher Yanich)
Date: Mon, 7 Oct 2002 17:53:34 -0400 (EDT)
Subject: [Mailman-Developers] information regarding request.db
Message-ID: <Pine.LNX.4.44.0210071749250.11180-100000@homer.cs.fiu.edu>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1


I am looking to write a tool to auto-flush held messages and I was
wondering what interface I could use to get into request.db.  I would like
to be able to intelligently look at the requests that are being held and
if they havent been moderated through within a certain period of time,
just clear them out.

So, I tried DB_File with perl and had no luck.

Thanks,
- -asher
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.0.7 (GNU/Linux)

iD8DBQE9ogJhAvdNfNYuUt8RAl7yAJ9pegCFOjze7w4VKdV+6mPaMiAqIwCeL7BV
7344/B1SqwJKY5LnP8axjow=
=kXYM
-----END PGP SIGNATURE-----


From sb.list@sb.org  Wed Oct  2 14:28:07 2002
From: sb.list@sb.org (Stonewall Ballard)
Date: Wed, 02 Oct 2002 09:28:07 -0400
Subject: [Mailman-Developers] problem with 2.1b3+
In-Reply-To: <15770.55183.541944.33700@gargle.gargle.HOWL>
Message-ID: <B9C06CA7.46BF2%sb.list@sb.org>

On 10/2/02 7:25 AM, "Barry A. Warsaw" <barry@python.org> wrote:

> web_page_url isn't configurable through the web in MM2.1 because too
> many people messed up and made their lists unfunctional (screw it up
> and you can't get to your list through the web anymore!).
> 
> You ought to be able to do it with config_list, so I'd be interested
> in more information as to why that doesn't work.  But I'd prefer
> fix_url.py instead for this task.
> 
> Once you've set up your mm_cfg.py with the new values, including
> VIRTUAL_HOSTS, you can simply run
> 
>   % bin/withlist -l -r fix_url listname [options]
> 
> to switch listname's web_page_url and host_name attributes to their
> new values.  Do
> 
>   % bin/fix_url.py
> 
> for docstring synopsis.
> 
> -Barry

That didn't work for me. I had to go in with "withlist" and manually set the
web_page_url for each list that I moved.

I created a bunch of lists under the domain 'mailman.stonetics.com', since I
had to create them before switching the A-record for
'lists.sudburysavoyards.org' over to the mailman server. I set the host_name
for each of these lists to 'lists.sudburysavoyards.org' on their admin
pages, then added these entries to my mm_cfg.py file, then ran fix_url, but
nothing happened:

add_virtualhost('lists.sb.org', 'lists.sb.org')
add_virtualhost('lists.sudburysavoyards.org', 'lists.sudburysavoyards.org')

They were still showing up on the web at 'mailman.stonetics.com'. Using
withlist manually fixed it, though.

I also noticed that the output file created by config_list has lots of
spurious line breaks about 2/3 of the way through. It looks unparsable.

 - Stoney


From barry@python.org  Mon Oct  7 23:06:56 2002
From: barry@python.org (Barry A. Warsaw)
Date: Mon, 7 Oct 2002 18:06:56 -0400
Subject: [Mailman-Developers] information regarding request.db
References: <Pine.LNX.4.44.0210071749250.11180-100000@homer.cs.fiu.edu>
Message-ID: <15778.1408.916755.720488@gargle.gargle.HOWL>


>>>>> "AY" == Asher Yanich <ayanic01@cs.fiu.edu> writes:

    AY> I am looking to write a tool to auto-flush held messages and I
    AY> was wondering what interface I could use to get into
    AY> request.db.  I would like to be able to intelligently look at
    AY> the requests that are being held and if they havent been
    AY> moderated through within a certain period of time, just clear
    AY> them out.

requests.db is a marshalled Python dictionary.

    AY> So, I tried DB_File with perl and had no luck.

What's Perl <wink>.

See this page for details:

http://www.python.org/doc/current/lib/module-marshal.html

or look at bin/dumpdb in the Mailman source code.

[Aside: request.db really ought to be a pickle not a marshal.  Oh
well.]

-Barry

From barry@python.org  Mon Oct  7 23:12:42 2002
From: barry@python.org (Barry A. Warsaw)
Date: Mon, 7 Oct 2002 18:12:42 -0400
Subject: [Mailman-Developers] message url in footer
References: <JHEOKEOOLIGLDHCMAHMOEEKBDPAA.marklists@mceahern.com>
Message-ID: <15778.1754.331928.16461@gargle.gargle.HOWL>


>>>>> "MM" == Mark McEahern <marklists@mceahern.com> writes:

    MM> It would be great if I could modify the footer of a mailing
    MM> list to insert the archive URL of the current message.

    MM> Is that easy/possible?

Nope because there's no standard API between Mailman and the archiver
to trade this information, and relying on something like the
Message-ID isn't quite reliable enough.  This is on the wish list
though (just now, officially).

    MM> I apologize if this is a FAQ.  I read the FAQ and the wish
    MM> list and couldn't find a reference to it.  I also did at least
    MM> a cursory google search and couldn't find anything obvious
    MM> related to this:

    MM>   http://www.google.com/search?q=mailman+footer

    MM>   http://www.google.com/search?q=mailman+footer+archive

    MM> Thanks,

    MM> // mark

    MM> p.s.  Suggestion: Put a "last updated" date on the wish list.

It's in the html comments, but I added an explicit RCS $Date$ key to
the file.

-Barry

From mjm@michaelmeltzer.com  Tue Oct  8 00:09:09 2002
From: mjm@michaelmeltzer.com (Michael Meltzer)
Date: Mon, 7 Oct 2002 19:09:09 -0400
Subject: [Mailman-Developers] large attachments replaced by links
References: <3D9334C4.60106@genesys.ro>
Message-ID: <00ea01c26e56$8b116110$34f820c0@ix1x1000>

I hear you, in my case is was a issue of small mailboxes(hotmail) and image attachments plus one to many people with virus. Saved
many headaches by stripping all attachments, adding a url text attachment in place of  the images and converting the html to text.

The good points:
1)average messages size dropped in half, mostly from the collapse 'multipart/alternative'.
2)not more bounce message from full mailboxes(ok, almost none :-)
3)no more headaches from dialup users
4)no more moderator message from "message size excited"
5)no more dueling virus checker messages
6)made the digest users happy.
7)got the list usage below the "radar screen" of line usage :-), no sharp peeks when passing images :-)

The bad points:
1)outlook 2000(not express) handles inline text attachments badly, they need to upgrade to outlook 2002
2)the private archive code does not handle the mime type's right, need to use a public archives(for now)
3)you have to watch out for what type's are url'ed, none of them should be server side scripts extensions
4)I think that quote-printable might have to be converted to 7bit, user bitch level is getting thier.

I have some "cheesy" code that does it, you would have to hardwire the code for the mime types and it will only work with the latest
CVS, if you want it you got it, Maybe you can talk Barry into adding it as a full feature of the system, I failed.

MJM

----- Original Message -----
From: "Silviu Marin-Caea" <silviu@genesys.ro>
To: <mailman-developers@python.org>
Sent: Thursday, September 26, 2002 12:24 PM
Subject: [Mailman-Developers] large attachments replaced by links


> Sorry if this is not a new ideea.
>
> We have some internal lists.  Users have the (bad) habbit of sending
> large attached files on these lists.
>
> What I'd like mailman to be able to do is to automatically take those
> attachments out of the posts, storing them to a directory I specify, and
> replace them with links to the stored files.
>
> Like this:
>
> "The file attached to this message can be found here
> http://bla-blaserver.net/attach/grossely_large_and_stupid_presentation.ppt"
>
> Configuring a http (or ftp) server to publish those attachments would be
> the job of the admin.
>
> Thank you.  Please CC to me the eventual replies.
>
> --
> Silviu Marin-Caea   Systems Engineer Linux/Unix   http://www.genesys.ro
> Phone +40723-267961
>
>
>
> _______________________________________________
> Mailman-Developers mailing list
> Mailman-Developers@python.org
> http://mail.python.org/mailman-21/listinfo/mailman-developers


From noreply@sourceforge.net  Tue Oct  8 03:06:58 2002
From: noreply@sourceforge.net (noreply@sourceforge.net)
Date: Mon, 07 Oct 2002 19:06:58 -0700
Subject: [Mailman-Developers] [ mailman-Bugs-620032 ] cmd_confirm: no lstrip
Message-ID: <E17yjly-0003pb-00@usw-sf-web3.sourceforge.net>

Bugs item #620032, was opened at 2002-10-08 02:06
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=100103&aid=620032&group_id=103

Category: (un)subscribing
Group: 2.1 beta
Status: Open
Resolution: None
Priority: 5
Submitted By: Ron Jarrell (jarrell)
Assigned to: Nobody/Anonymous (nobody)
Summary: cmd_confirm: no lstrip

Initial Comment:
Sep 23 18:45:41 2002 (574) Uncaught runner 
exception: 'NoneType' object has no attribute 'lstrip'
Sep 23 18:45:41 2002 (574) Traceback (most recent 
call last):
  
File "/home/mailman/Mailman/Queue/Runner.py", 
line 105, in _oneloop
    self._onefile(msg, msgdata)
  
File "/home/mailman/Mailman/Queue/Runner.py", 
line 154, in _onefile
    keepqueued = self._dispose(mlist, msg, msgdata)
  
File "/home/mailman/Mailman/Queue/CommandRu
nner.py", line 198, in _dispose
    res.process()
  
File "/home/mailman/Mailman/Queue/CommandRu
nner.py", line 93, in process
    stop = self.do_command(cmd, args)
  
File "/home/mailman/Mailman/Queue/CommandRu
nner.py", line 119, in do_command
    return handler.process(self, args)
  
File "/home/mailman/Mailman/Commands/cmd_co
nfirm.py", line 72, in process
    if line.lstrip() == match:
AttributeError: 'NoneType' object has no 
attribute 'lstrip'


----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=100103&aid=620032&group_id=103

From noreply@sourceforge.net  Tue Oct  8 03:07:39 2002
From: noreply@sourceforge.net (noreply@sourceforge.net)
Date: Mon, 07 Oct 2002 19:07:39 -0700
Subject: [Mailman-Developers] 
 [ mailman-Bugs-620033 ] gate_news should catch lost connection
Message-ID: <E17yjmd-0003qC-00@usw-sf-web3.sourceforge.net>

Bugs item #620033, was opened at 2002-10-08 02:07
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=100103&aid=620033&group_id=103

Category: nntp/news
Group: 2.1 beta
Status: Open
Resolution: None
Priority: 5
Submitted By: Ron Jarrell (jarrell)
Assigned to: Nobody/Anonymous (nobody)
Summary: gate_news should catch lost connection

Initial Comment:
Sep 20 10:51:35 2002 gate_news(9301): Traceback 
(most recent call last):
Sep 20 10:51:35 2002 gate_news(9301):   
File "/home/mailman/cron/gate_news", line 270, 
in ?
Sep 20 10:51:35 2002 gate_news(9301):      main()
Sep 20 10:51:35 2002 gate_news(9301):   
File "/home/mailman/cron/gate_news", line 250, in 
main
Sep 20 10:51:35 2002 gate_news(9301):      
process_lists(lock)
Sep 20 10:51:35 2002 gate_news(9301):   
File "/home/mailman/cron/gate_news", line 194, in 
process_lists
Sep 20 10:51:35 2002 gate_news(9301):      conn, 
first, last = open_newsgroup(mlist)
Sep 20 10:51:35 2002 gate_news(9301):   
File "/home/mailman/cron/gate_news", line 89, in 
open_newsgroup
Sep 20 10:51:35 2002 gate_news(9301):      
password=mm_cfg.NNTP_PASSWORD)
Sep 20 10:51:35 2002 gate_news(9301):   
File "/usr/local/lib/python2.2/nntplib.py", line 
117, in __init__
Sep 20 10:51:35 2002 gate_news(9301):      
self.welcome = self.getresp()
Sep 20 10:51:35 2002 gate_news(9301):   
File "/usr/local/lib/python2.2/nntplib.py", line 
203, in getresp
Sep 20 10:51:35 2002 gate_news(9301):      resp = 
self.getline()
Sep 20 10:51:35 2002 gate_news(9301):   
File "/usr/local/lib/python2.2/nntplib.py", line 
192, in getline
Sep 20 10:51:35 2002 gate_news(9301):      line = 
self.file.readline()
Sep 20 10:51:35 2002 gate_news(9301): IOError :  
[Errno 131] Connection reset by peer 


----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=100103&aid=620033&group_id=103

From noreply@sourceforge.net  Tue Oct  8 14:00:07 2002
From: noreply@sourceforge.net (noreply@sourceforge.net)
Date: Tue, 08 Oct 2002 06:00:07 -0700
Subject: [Mailman-Developers] 
 [ mailman-Bugs-620208 ] authinfo before mode reader needed
Message-ID: <E17yty3-0003Vk-00@usw-sf-web3.sourceforge.net>

Bugs item #620208, was opened at 2002-10-08 09:00
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=100103&aid=620208&group_id=103

Category: nntp/news
Group: 2.0.x
Status: Open
Resolution: None
Priority: 5
Submitted By: Scott Lipcon (slipcon)
Assigned to: Nobody/Anonymous (nobody)
Summary: authinfo before mode reader needed

Initial Comment:
My ISP just switched to giganews for NNTP service, and
my gate_news script stopped working - complaining about
authentication needed.  I tracked it down to the fact
that giganews' server gives a 480 authentication needed
error on the 'mode reader' command which mailman tries
to send before authentication.  

I solved the problem for myself by editing
Mailman/pythonlib/nntplib.py  and changing the NNTP
constructor.   I simply moved the block that sends the
authinfo commands in front of the block that sends the
'mode reader' command, and everything worked.   

I assume this is not the correct fix for this problem,
as other servers may not like the authinfo command
before the mode reader command.    Perhaps giganews'
server is broken, but it would be nice to work around
it.  Perhaps the authentication should be broken in to
its own subroutine, and have the code try
authentication whenever it gets a 480 authentication
required error.

I'm running mailman 2.0.13 on OpenBSD, with python
1.5.2, if it matters.  

Thanks,

Scott


----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=100103&aid=620208&group_id=103

From noreply@sourceforge.net  Tue Oct  8 16:54:15 2002
From: noreply@sourceforge.net (noreply@sourceforge.net)
Date: Tue, 08 Oct 2002 08:54:15 -0700
Subject: [Mailman-Developers] [ mailman-Patches-594771 ] i18n support for
	Archiver
Message-ID: <E17ywgZ-0002Vk-00@usw-sf-web4.sourceforge.net>

Patches item #594771, was opened at 2002-08-13 17:09
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=300103&aid=594771&group_id=103

Category: Pipermail
Group: Mailman 2.1
Status: Open
Resolution: None
>Priority: 7
Submitted By: Simone Piunno (pioppo)
>Assigned to: Barry A. Warsaw (bwarsaw)
Summary: i18n support for Archiver

Initial Comment:
I'm trying to provide changes in small chunks.
This 1st patch does the following:

- 7 new template files:

        archidxentry.html   
        archidxhead.html   
        archidxfoot.html     
        archliststart.html  
        archlistend.html    
        archtoc.html
        archtocentry.html

- _() marking for translatable strings

- a few small bugfixes (e.g. unwanted spaces,
   bin/arch not calling i18n.set_language, etc.)

TODO:

- show translated tags for archive volumes (e.g.
   "2002-August", "2002q3", "Week-of-Mon-XXXX")
   I have 2 alternatives: renaming files and dirs where
   the archives are built (cleaner, but screws up 
   everything in pre-existing archives) or build a 
   tag-to-description translator (better, I think)

- find a way to produce translated datetime strings
  without using locale.setlocale and time.strftime()
  (this is discouraged in Python manual)
  Here I really need some advice.


----------------------------------------------------------------------

Comment By: Barry A. Warsaw (bwarsaw)
Date: 2002-10-08 11:53

Message:
Logged In: YES 
user_id=12800

I'm looking at this patch now and am cleaning up the code. 
Is there a reason why one of the patches copied the
processUnixMailbox() method from pipermail.py to
HyperArch.py?  That doesn't appear necessary.

----------------------------------------------------------------------

Comment By: Tokio Kikuchi (tkikuchi)
Date: 2002-08-26 19:53

Message:
Logged In: YES 
user_id=67709

Thank you for updating.
Another comment:
I'v got error when I post a message to a list after applying
this patch. It is because a new variable Article._lang was
introduced and the older archive doesn't have this.  I had
to re-generate the whole archive. 
Another solution may be like.....
        if not self._lang:
            self._lang = self._mlist.preferred_language
        i18n.set_language(self._lang)



----------------------------------------------------------------------

Comment By: Simone Piunno (pioppo)
Date: 2002-08-26 15:12

Message:
Logged In: YES 
user_id=227443

I'm attaching the 3rd chunk modified as suggested by Tokio Kikuchi.
Maybe i18n_ctime is worth moving in Mailman/i18n.py ?


----------------------------------------------------------------------

Comment By: Tokio Kikuchi (tkikuchi)
Date: 2002-08-26 01:02

Message:
Logged In: YES 
user_id=67709

I have submitted a slight change to this patch. It is
uploaded in a separate patch ID.
Thank you Simone for exellent work!

----------------------------------------------------------------------

Comment By: Simone Piunno (pioppo)
Date: 2002-08-14 14:49

Message:
Logged In: YES 
user_id=227443

And finally, this is the 3rd piece of the puzzle: dates translation.
I had to reinvent the wheel, and it's an relatively expensive
hack (from a computational efficiency point of view) but it 
works without touching pipermail.py

This patch is incremental and must be applied after #2


----------------------------------------------------------------------

Comment By: Simone Piunno (pioppo)
Date: 2002-08-13 19:38

Message:
Logged In: YES 
user_id=227443

Here is the 2nd patch, providing the tag-to-description translator
for volume tags.  This patch is incremental and must be applied
after patch #¹

Now the last thing is dates.
I wouln'd want to reinvent the wheel....


----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=300103&aid=594771&group_id=103

From noreply@sourceforge.net  Tue Oct  8 16:53:35 2002
From: noreply@sourceforge.net (noreply@sourceforge.net)
Date: Tue, 08 Oct 2002 08:53:35 -0700
Subject: [Mailman-Developers] [ mailman-Patches-594771 ] i18n support for
	Archiver
Message-ID: <E17ywfv-0002Uz-00@usw-sf-web4.sourceforge.net>

Patches item #594771, was opened at 2002-08-13 17:09
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=300103&aid=594771&group_id=103

Category: Pipermail
Group: Mailman 2.1
Status: Open
Resolution: None
Priority: 5
Submitted By: Simone Piunno (pioppo)
Assigned to: Nobody/Anonymous (nobody)
Summary: i18n support for Archiver

Initial Comment:
I'm trying to provide changes in small chunks.
This 1st patch does the following:

- 7 new template files:

        archidxentry.html   
        archidxhead.html   
        archidxfoot.html     
        archliststart.html  
        archlistend.html    
        archtoc.html
        archtocentry.html

- _() marking for translatable strings

- a few small bugfixes (e.g. unwanted spaces,
   bin/arch not calling i18n.set_language, etc.)

TODO:

- show translated tags for archive volumes (e.g.
   "2002-August", "2002q3", "Week-of-Mon-XXXX")
   I have 2 alternatives: renaming files and dirs where
   the archives are built (cleaner, but screws up 
   everything in pre-existing archives) or build a 
   tag-to-description translator (better, I think)

- find a way to produce translated datetime strings
  without using locale.setlocale and time.strftime()
  (this is discouraged in Python manual)
  Here I really need some advice.


----------------------------------------------------------------------

>Comment By: Barry A. Warsaw (bwarsaw)
Date: 2002-10-08 11:53

Message:
Logged In: YES 
user_id=12800

I'm looking at this patch now and am cleaning up the code. 
Is there a reason why one of the patches copied the
processUnixMailbox() method from pipermail.py to
HyperArch.py?  That doesn't appear necessary.

----------------------------------------------------------------------

Comment By: Tokio Kikuchi (tkikuchi)
Date: 2002-08-26 19:53

Message:
Logged In: YES 
user_id=67709

Thank you for updating.
Another comment:
I'v got error when I post a message to a list after applying
this patch. It is because a new variable Article._lang was
introduced and the older archive doesn't have this.  I had
to re-generate the whole archive. 
Another solution may be like.....
        if not self._lang:
            self._lang = self._mlist.preferred_language
        i18n.set_language(self._lang)



----------------------------------------------------------------------

Comment By: Simone Piunno (pioppo)
Date: 2002-08-26 15:12

Message:
Logged In: YES 
user_id=227443

I'm attaching the 3rd chunk modified as suggested by Tokio Kikuchi.
Maybe i18n_ctime is worth moving in Mailman/i18n.py ?


----------------------------------------------------------------------

Comment By: Tokio Kikuchi (tkikuchi)
Date: 2002-08-26 01:02

Message:
Logged In: YES 
user_id=67709

I have submitted a slight change to this patch. It is
uploaded in a separate patch ID.
Thank you Simone for exellent work!

----------------------------------------------------------------------

Comment By: Simone Piunno (pioppo)
Date: 2002-08-14 14:49

Message:
Logged In: YES 
user_id=227443

And finally, this is the 3rd piece of the puzzle: dates translation.
I had to reinvent the wheel, and it's an relatively expensive
hack (from a computational efficiency point of view) but it 
works without touching pipermail.py

This patch is incremental and must be applied after #2


----------------------------------------------------------------------

Comment By: Simone Piunno (pioppo)
Date: 2002-08-13 19:38

Message:
Logged In: YES 
user_id=227443

Here is the 2nd patch, providing the tag-to-description translator
for volume tags.  This patch is incremental and must be applied
after patch #¹

Now the last thing is dates.
I wouln'd want to reinvent the wheel....


----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=300103&aid=594771&group_id=103

From noreply@sourceforge.net  Tue Oct  8 18:14:39 2002
From: noreply@sourceforge.net (noreply@sourceforge.net)
Date: Tue, 08 Oct 2002 10:14:39 -0700
Subject: [Mailman-Developers] [ mailman-Patches-594771 ] i18n support for
	Archiver
Message-ID: <E17yxwN-0003MR-00@usw-sf-web4.sourceforge.net>

Patches item #594771, was opened at 2002-08-13 17:09
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=300103&aid=594771&group_id=103

Category: Pipermail
Group: Mailman 2.1
>Status: Closed
>Resolution: Accepted
Priority: 7
Submitted By: Simone Piunno (pioppo)
Assigned to: Barry A. Warsaw (bwarsaw)
Summary: i18n support for Archiver

Initial Comment:
I'm trying to provide changes in small chunks.
This 1st patch does the following:

- 7 new template files:

        archidxentry.html   
        archidxhead.html   
        archidxfoot.html     
        archliststart.html  
        archlistend.html    
        archtoc.html
        archtocentry.html

- _() marking for translatable strings

- a few small bugfixes (e.g. unwanted spaces,
   bin/arch not calling i18n.set_language, etc.)

TODO:

- show translated tags for archive volumes (e.g.
   "2002-August", "2002q3", "Week-of-Mon-XXXX")
   I have 2 alternatives: renaming files and dirs where
   the archives are built (cleaner, but screws up 
   everything in pre-existing archives) or build a 
   tag-to-description translator (better, I think)

- find a way to produce translated datetime strings
  without using locale.setlocale and time.strftime()
  (this is discouraged in Python manual)
  Here I really need some advice.


----------------------------------------------------------------------

>Comment By: Barry A. Warsaw (bwarsaw)
Date: 2002-10-08 13:14

Message:
Logged In: YES 
user_id=12800

After removing the processUnixMailbox() in HyperArch.py,
this patch seems pretty good.  I'm going to go ahead and
commit it and we can fix any bugs that crop up later. 
(However, the simple things appear to work fine).

----------------------------------------------------------------------

Comment By: Barry A. Warsaw (bwarsaw)
Date: 2002-10-08 11:53

Message:
Logged In: YES 
user_id=12800

I'm looking at this patch now and am cleaning up the code. 
Is there a reason why one of the patches copied the
processUnixMailbox() method from pipermail.py to
HyperArch.py?  That doesn't appear necessary.

----------------------------------------------------------------------

Comment By: Tokio Kikuchi (tkikuchi)
Date: 2002-08-26 19:53

Message:
Logged In: YES 
user_id=67709

Thank you for updating.
Another comment:
I'v got error when I post a message to a list after applying
this patch. It is because a new variable Article._lang was
introduced and the older archive doesn't have this.  I had
to re-generate the whole archive. 
Another solution may be like.....
        if not self._lang:
            self._lang = self._mlist.preferred_language
        i18n.set_language(self._lang)



----------------------------------------------------------------------

Comment By: Simone Piunno (pioppo)
Date: 2002-08-26 15:12

Message:
Logged In: YES 
user_id=227443

I'm attaching the 3rd chunk modified as suggested by Tokio Kikuchi.
Maybe i18n_ctime is worth moving in Mailman/i18n.py ?


----------------------------------------------------------------------

Comment By: Tokio Kikuchi (tkikuchi)
Date: 2002-08-26 01:02

Message:
Logged In: YES 
user_id=67709

I have submitted a slight change to this patch. It is
uploaded in a separate patch ID.
Thank you Simone for exellent work!

----------------------------------------------------------------------

Comment By: Simone Piunno (pioppo)
Date: 2002-08-14 14:49

Message:
Logged In: YES 
user_id=227443

And finally, this is the 3rd piece of the puzzle: dates translation.
I had to reinvent the wheel, and it's an relatively expensive
hack (from a computational efficiency point of view) but it 
works without touching pipermail.py

This patch is incremental and must be applied after #2


----------------------------------------------------------------------

Comment By: Simone Piunno (pioppo)
Date: 2002-08-13 19:38

Message:
Logged In: YES 
user_id=227443

Here is the 2nd patch, providing the tag-to-description translator
for volume tags.  This patch is incremental and must be applied
after patch #¹

Now the last thing is dates.
I wouln'd want to reinvent the wheel....


----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=300103&aid=594771&group_id=103

From noreply@sourceforge.net  Tue Oct  8 20:05:17 2002
From: noreply@sourceforge.net (noreply@sourceforge.net)
Date: Tue, 08 Oct 2002 12:05:17 -0700
Subject: [Mailman-Developers] 
 [ mailman-Bugs-620208 ] authinfo before mode reader needed
Message-ID: <E17yzfR-00056C-00@usw-sf-web5.sourceforge.net>

Bugs item #620208, was opened at 2002-10-08 09:00
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=100103&aid=620208&group_id=103

Category: nntp/news
Group: 2.0.x
>Status: Closed
>Resolution: Out of Date
Priority: 5
Submitted By: Scott Lipcon (slipcon)
Assigned to: Nobody/Anonymous (nobody)
Summary: authinfo before mode reader needed

Initial Comment:
My ISP just switched to giganews for NNTP service, and
my gate_news script stopped working - complaining about
authentication needed.  I tracked it down to the fact
that giganews' server gives a 480 authentication needed
error on the 'mode reader' command which mailman tries
to send before authentication.  

I solved the problem for myself by editing
Mailman/pythonlib/nntplib.py  and changing the NNTP
constructor.   I simply moved the block that sends the
authinfo commands in front of the block that sends the
'mode reader' command, and everything worked.   

I assume this is not the correct fix for this problem,
as other servers may not like the authinfo command
before the mode reader command.    Perhaps giganews'
server is broken, but it would be nice to work around
it.  Perhaps the authentication should be broken in to
its own subroutine, and have the code try
authentication whenever it gets a 480 authentication
required error.

I'm running mailman 2.0.13 on OpenBSD, with python
1.5.2, if it matters.  

Thanks,

Scott


----------------------------------------------------------------------

>Comment By: Barry A. Warsaw (bwarsaw)
Date: 2002-10-08 15:05

Message:
Logged In: YES 
user_id=12800

Look for NNTP_USERNAME and NNTP_PASSWORD in the Defaults.py
file.

----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=100103&aid=620208&group_id=103

From noreply@sourceforge.net  Tue Oct  8 20:13:38 2002
From: noreply@sourceforge.net (noreply@sourceforge.net)
Date: Tue, 08 Oct 2002 12:13:38 -0700
Subject: [Mailman-Developers] 
 [ mailman-Bugs-620208 ] authinfo before mode reader needed
Message-ID: <E17yznW-0002dH-00@usw-sf-web3.sourceforge.net>

Bugs item #620208, was opened at 2002-10-08 09:00
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=100103&aid=620208&group_id=103

Category: nntp/news
Group: 2.0.x
>Status: Open
Resolution: Out of Date
Priority: 5
Submitted By: Scott Lipcon (slipcon)
Assigned to: Nobody/Anonymous (nobody)
Summary: authinfo before mode reader needed

Initial Comment:
My ISP just switched to giganews for NNTP service, and
my gate_news script stopped working - complaining about
authentication needed.  I tracked it down to the fact
that giganews' server gives a 480 authentication needed
error on the 'mode reader' command which mailman tries
to send before authentication.  

I solved the problem for myself by editing
Mailman/pythonlib/nntplib.py  and changing the NNTP
constructor.   I simply moved the block that sends the
authinfo commands in front of the block that sends the
'mode reader' command, and everything worked.   

I assume this is not the correct fix for this problem,
as other servers may not like the authinfo command
before the mode reader command.    Perhaps giganews'
server is broken, but it would be nice to work around
it.  Perhaps the authentication should be broken in to
its own subroutine, and have the code try
authentication whenever it gets a 480 authentication
required error.

I'm running mailman 2.0.13 on OpenBSD, with python
1.5.2, if it matters.  

Thanks,

Scott


----------------------------------------------------------------------

>Comment By: Scott Lipcon (slipcon)
Date: 2002-10-08 15:13

Message:
Logged In: YES 
user_id=91258

I think you misunderstood my bug report.  I have
NNTP_USERNAME and NNTP_PASSWORD set in mm_cfg.py, and that
was working for many months with my ISP's old new server.  I
believe they used supernews.    Last week, they switched to
using giganews, and my news gating stopped working.  

The code in nntplib.py attempts to issue a "mode reader"
command before doing any authentication.  Apparently
supernews' server dealt with that fine - perhaps 'mode
reader' is an unpriviledged command?    However, giganews'
server responds to the 'mode reader' command with a 480
authentication required error, which causes a stack trace in
gate_news.    

Like I said, I was able to change nntplib.py to do the
authentication BEFORE the mode reader command, and
everything works now.

If this still isn't clear, lets take this to email - I'm not
on any of the mailman lists, so email
slipcon@users.sourceforge.net if you need any more details
or logs.

Thanks,

Scott

----------------------------------------------------------------------

Comment By: Barry A. Warsaw (bwarsaw)
Date: 2002-10-08 15:05

Message:
Logged In: YES 
user_id=12800

Look for NNTP_USERNAME and NNTP_PASSWORD in the Defaults.py
file.

----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=100103&aid=620208&group_id=103

From noreply@sourceforge.net  Tue Oct  8 20:27:53 2002
From: noreply@sourceforge.net (noreply@sourceforge.net)
Date: Tue, 08 Oct 2002 12:27:53 -0700
Subject: [Mailman-Developers] 
 [ mailman-Bugs-620406 ] sync_members typo (patch included)
Message-ID: <E17z01J-0004Z9-00@usw-sf-web4.sourceforge.net>

Bugs item #620406, was opened at 2002-10-08 14:27
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=100103&aid=620406&group_id=103

Category: command line scripts
Group: None
Status: Open
Resolution: None
Priority: 5
Submitted By: Dave Habben (dhabben)
Assigned to: Nobody/Anonymous (nobody)
Summary: sync_members typo (patch included)

Initial Comment:
There is a typo in the sync_members script, attached is
a patch to correct it. This patch is against 2.0.13 but
the bug also appears in the CVS version of mailman. 


--- sync_members        2002-08-15 13:48:34.000000000 -0500
+++ sync_members        2002-10-08 14:23:42.000000000 -0500
@@ -48,7 +48,7 @@
         (or if no -d option given) they are added as
regular members.

     --notifyadmin[=<yes|no>]
-    --a[=<yes|no>]
+    -a[=<yes|no>]
         Specifies whether the admin should be notified
for each subscription
         or unsubscription.  If you're adding a lot of
addresses, you
         definitely want to turn this off!  With -a=yes
or -a, the admin is


----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=100103&aid=620406&group_id=103

From noreply@sourceforge.net  Tue Oct  8 20:29:56 2002
From: noreply@sourceforge.net (noreply@sourceforge.net)
Date: Tue, 08 Oct 2002 12:29:56 -0700
Subject: [Mailman-Developers] 
 [ mailman-Bugs-620208 ] authinfo before mode reader needed
Message-ID: <E17z03I-0005MR-00@usw-sf-web5.sourceforge.net>

Bugs item #620208, was opened at 2002-10-08 09:00
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=100103&aid=620208&group_id=103

Category: nntp/news
Group: 2.0.x
>Status: Closed
Resolution: Out of Date
Priority: 5
Submitted By: Scott Lipcon (slipcon)
Assigned to: Nobody/Anonymous (nobody)
Summary: authinfo before mode reader needed

Initial Comment:
My ISP just switched to giganews for NNTP service, and
my gate_news script stopped working - complaining about
authentication needed.  I tracked it down to the fact
that giganews' server gives a 480 authentication needed
error on the 'mode reader' command which mailman tries
to send before authentication.  

I solved the problem for myself by editing
Mailman/pythonlib/nntplib.py  and changing the NNTP
constructor.   I simply moved the block that sends the
authinfo commands in front of the block that sends the
'mode reader' command, and everything worked.   

I assume this is not the correct fix for this problem,
as other servers may not like the authinfo command
before the mode reader command.    Perhaps giganews'
server is broken, but it would be nice to work around
it.  Perhaps the authentication should be broken in to
its own subroutine, and have the code try
authentication whenever it gets a 480 authentication
required error.

I'm running mailman 2.0.13 on OpenBSD, with python
1.5.2, if it matters.  

Thanks,

Scott


----------------------------------------------------------------------

>Comment By: Barry A. Warsaw (bwarsaw)
Date: 2002-10-08 15:29

Message:
Logged In: YES 
user_id=12800

Ah sorry, I understand now.  I think upgrading to Python
2.1.3 or Python 2.2.1 should fix it for you.  The nntplib.py
as of Python 2.1 supports readermode both before and after
authorization.

----------------------------------------------------------------------

Comment By: Scott Lipcon (slipcon)
Date: 2002-10-08 15:13

Message:
Logged In: YES 
user_id=91258

I think you misunderstood my bug report.  I have
NNTP_USERNAME and NNTP_PASSWORD set in mm_cfg.py, and that
was working for many months with my ISP's old new server.  I
believe they used supernews.    Last week, they switched to
using giganews, and my news gating stopped working.  

The code in nntplib.py attempts to issue a "mode reader"
command before doing any authentication.  Apparently
supernews' server dealt with that fine - perhaps 'mode
reader' is an unpriviledged command?    However, giganews'
server responds to the 'mode reader' command with a 480
authentication required error, which causes a stack trace in
gate_news.    

Like I said, I was able to change nntplib.py to do the
authentication BEFORE the mode reader command, and
everything works now.

If this still isn't clear, lets take this to email - I'm not
on any of the mailman lists, so email
slipcon@users.sourceforge.net if you need any more details
or logs.

Thanks,

Scott

----------------------------------------------------------------------

Comment By: Barry A. Warsaw (bwarsaw)
Date: 2002-10-08 15:05

Message:
Logged In: YES 
user_id=12800

Look for NNTP_USERNAME and NNTP_PASSWORD in the Defaults.py
file.

----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=100103&aid=620208&group_id=103

From noreply@sourceforge.net  Tue Oct  8 20:57:32 2002
From: noreply@sourceforge.net (noreply@sourceforge.net)
Date: Tue, 08 Oct 2002 12:57:32 -0700
Subject: [Mailman-Developers] 
 [ mailman-Bugs-620406 ] sync_members typo (patch included)
Message-ID: <E17z0U0-0005I3-00@usw-sf-web2.sourceforge.net>

Bugs item #620406, was opened at 2002-10-08 15:27
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=100103&aid=620406&group_id=103

Category: command line scripts
>Group: 2.1 beta
>Status: Closed
>Resolution: Fixed
Priority: 5
Submitted By: Dave Habben (dhabben)
Assigned to: Nobody/Anonymous (nobody)
Summary: sync_members typo (patch included)

Initial Comment:
There is a typo in the sync_members script, attached is
a patch to correct it. This patch is against 2.0.13 but
the bug also appears in the CVS version of mailman. 


--- sync_members        2002-08-15 13:48:34.000000000 -0500
+++ sync_members        2002-10-08 14:23:42.000000000 -0500
@@ -48,7 +48,7 @@
         (or if no -d option given) they are added as
regular members.

     --notifyadmin[=<yes|no>]
-    --a[=<yes|no>]
+    -a[=<yes|no>]
         Specifies whether the admin should be notified
for each subscription
         or unsubscription.  If you're adding a lot of
addresses, you
         definitely want to turn this off!  With -a=yes
or -a, the admin is


----------------------------------------------------------------------

>Comment By: Barry A. Warsaw (bwarsaw)
Date: 2002-10-08 15:57

Message:
Logged In: YES 
user_id=12800

Thanks!  Applied for MM2.1

----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=100103&aid=620406&group_id=103

From mss@mawhrin.net  Tue Oct  8 23:09:11 2002
From: mss@mawhrin.net (Mikhail Sobolev)
Date: Tue, 8 Oct 2002 23:09:11 +0100
Subject: [Mailman-Developers] Re: [Mailman-checkins] mailman/bin arch,2.6,2.7
In-Reply-To: <E17yxwu-00046O-00@usw-pr-cvs1.sourceforge.net>
References: <E17yxwu-00046O-00@usw-pr-cvs1.sourceforge.net>
Message-ID: <20021008220911.GA3026@mawhrin.net>

On Tue, Oct 08, 2002 at 10:15:12AM -0700, barry@users.sourceforge.net wrote:
> Update of /cvsroot/mailman/mailman/bin
> In directory usw-pr-cvs1:/tmp/cvs-serv15760/bin
> 
> Modified Files:
> 	arch 
> Log Message:
> Integrating SF patch #594771, i18n'ified pipermail.
Can we hope that the mailman.pot file would be updated before release of
2.1b4?

--
Misha

From barry@python.org  Tue Oct  8 23:36:33 2002
From: barry@python.org (Barry A. Warsaw)
Date: Tue, 8 Oct 2002 18:36:33 -0400
Subject: [Mailman-Developers] Re: [Mailman-checkins] mailman/bin
	arch,2.6,2.7
References: <E17yxwu-00046O-00@usw-pr-cvs1.sourceforge.net>
	<20021008220911.GA3026@mawhrin.net>
Message-ID: <15779.24049.49790.268619@gargle.gargle.HOWL>


>>>>> "MS" == Mikhail Sobolev <mss@mawhrin.net> writes:

    MS> Can we hope that the mailman.pot file would be updated before
    MS> release of 2.1b4?

Yes.
-Barry

From noreply@sourceforge.net  Thu Oct 10 12:34:34 2002
From: noreply@sourceforge.net (noreply@sourceforge.net)
Date: Thu, 10 Oct 2002 04:34:34 -0700
Subject: [Mailman-Developers] 
 [ mailman-Patches-621257 ] MTA_ALIASES_STYLE = 'qmail-virtual'
Message-ID: <E17zbaM-0003dL-00@usw-sf-web3.sourceforge.net>

Patches item #621257, was opened at 2002-10-10 11:34
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=300103&aid=621257&group_id=103

Category: configure/install
Group: Mailman 2.1
Status: Open
Resolution: None
Priority: 5
Submitted By: John S. Denker (jsdenker)
Assigned to: Nobody/Anonymous (nobody)
Summary: MTA_ALIASES_STYLE = 'qmail-virtual'

Initial Comment:
I heartily endorse the recommendation in 
README.QMAIL to make a virtual mail server.

The patch (attached) improves support for this, causing 
bin/newlist to generate the correct commands for creating 
the aliases.

Included is the corresponding patch to the documentation.


----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=300103&aid=621257&group_id=103

From noreply@sourceforge.net  Thu Oct 10 23:06:39 2002
From: noreply@sourceforge.net (noreply@sourceforge.net)
Date: Thu, 10 Oct 2002 15:06:39 -0700
Subject: [Mailman-Developers] 
 [ mailman-Bugs-621597 ] list creation ignores DEFAULT_EMAIL_HOST
Message-ID: <E17zlS3-0003NR-00@usw-sf-web1.sourceforge.net>

Bugs item #621597, was opened at 2002-10-10 18:06
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=100103&aid=621597&group_id=103

Category: None
Group: 2.1 beta
Status: Open
Resolution: None
Priority: 5
Submitted By: Greg Ward (gward)
Assigned to: Nobody/Anonymous (nobody)
Summary: list creation ignores DEFAULT_EMAIL_HOST

Initial Comment:
If I set DEFAULT_EMAIL_HOST in mm_cfg.py, its value is 
ignored when creating a new list.  It's pretty easy to see 
why: MailList.InitVars uses DEFAULT_HOST_NAME.  But 
the two are only equal as long as mm_cfg.py is not used.

If DEFAULT_HOST_NAME and DEFAULT_EMAIL_HOST 
are really supposed to be the same, perhaps 
DEFAULT_HOST_NAME should not be set in Defaults.py -- 
rather, any code that looks at DEFAULT_EMAIL_HOST 
should check to see if DEFAULT_HOST_NAME happens to 
be defined in mm_cfg, and use it if so.  Or something like 
that.

I'm happy to supply a patch if someone agrees that this is a 
bug.

----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=100103&aid=621597&group_id=103

From mss@mawhrin.net  Thu Oct 10 23:10:12 2002
From: mss@mawhrin.net (Mikhail Sobolev)
Date: Thu, 10 Oct 2002 23:10:12 +0100
Subject: [Mailman-Developers] Re: [Mailman-checkins] mailman/templates/ru
 archidxentry.html,NONE,2.0 archidxfoot.html,NONE,2.0
 archidxhead.html,NONE,2.0 archlistend.html,NONE,2.0
 archliststart.html,NONE,2.0 archtocentry.html,NONE,2.0
 archtoc.html,NONE,2.0
In-Reply-To: <E17zl3s-0004Bz-00@usw-pr-cvs1.sourceforge.net>
References: <E17zl3s-0004Bz-00@usw-pr-cvs1.sourceforge.net>
Message-ID: <20021010221012.GA12092@mawhrin.net>

On Thu, Oct 10, 2002 at 02:41:40PM -0700, mss@users.sourceforge.net wrote:
> --- NEW FILE: archidxentry.html ---
> 
> --- NEW FILE: archlistend.html ---

I am not sure that that it's a really good idea to create that small
files...  However, I am not ready to submit a patch, so this is just to
bring the issue to the attention.

--
Misha

From noreply@sourceforge.net  Fri Oct 11 04:10:34 2002
From: noreply@sourceforge.net (noreply@sourceforge.net)
Date: Thu, 10 Oct 2002 20:10:34 -0700
Subject: [Mailman-Developers] 
 [ mailman-Bugs-621689 ] Subscribe requests dropped from admin db
Message-ID: <E17zqCA-0007wC-00@usw-sf-web5.sourceforge.net>

Bugs item #621689, was opened at 2002-10-10 20:10
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=100103&aid=621689&group_id=103

Category: (un)subscribing
Group: 2.1 beta
Status: Open
Resolution: None
Priority: 5
Submitted By: Mark Dadgar (mdadgar)
Assigned to: Nobody/Anonymous (nobody)
Summary: Subscribe requests dropped from admin db

Initial Comment:
Occasionally, subscribe requests (via the web page) to a list that 
requires moderator approval for subscriptions never make it to the 
admin db for that list.  They are listed in the subscribe log file, but 
never show up on the admin db page.

Log file entries look like this:

Oct 10 18:43:43 2002 (31168) eventmasters: pending   
XXXXXX.swfla.rr.com

But the admin db page for the list shows no outstanding issues.

I have verified that there was an actual subscription request 
behind this log entry.

----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=100103&aid=621689&group_id=103

From noreply@sourceforge.net  Fri Oct 11 05:29:43 2002
From: noreply@sourceforge.net (noreply@sourceforge.net)
Date: Thu, 10 Oct 2002 21:29:43 -0700
Subject: [Mailman-Developers] 
 [ mailman-Bugs-621689 ] Subscribe requests dropped from admin db
Message-ID: <E17zrQl-00026H-00@usw-sf-web2.sourceforge.net>

Bugs item #621689, was opened at 2002-10-10 20:10
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=100103&aid=621689&group_id=103

Category: (un)subscribing
Group: 2.1 beta
Status: Open
Resolution: None
Priority: 5
Submitted By: Mark Dadgar (mdadgar)
Assigned to: Nobody/Anonymous (nobody)
Summary: Subscribe requests dropped from admin db

Initial Comment:
Occasionally, subscribe requests (via the web page) to a list that 
requires moderator approval for subscriptions never make it to the 
admin db for that list.  They are listed in the subscribe log file, but 
never show up on the admin db page.

Log file entries look like this:

Oct 10 18:43:43 2002 (31168) eventmasters: pending   
XXXXXX.swfla.rr.com

But the admin db page for the list shows no outstanding issues.

I have verified that there was an actual subscription request 
behind this log entry.

----------------------------------------------------------------------

>Comment By: Mark Dadgar (mdadgar)
Date: 2002-10-10 21:29

Message:
Logged In: YES 
user_id=598228

I just verified, by cat'ing pending.pck, that the appropriate changes 
seem to be in there.  But the web interface never picks them up.

----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=100103&aid=621689&group_id=103

From marc_news@merlins.org  Sat Oct 12 03:44:56 2002
From: marc_news@merlins.org (Marc MERLIN)
Date: Fri, 11 Oct 2002 19:44:56 -0700
Subject: [Mailman-Developers] Re: [Exim] Overhauled 'Exim and Mailman' doc
In-Reply-To: <20021012014433.GA29438@cthulhu.gerg.ca>
References: <20021012014433.GA29438@cthulhu.gerg.ca>
Message-ID: <20021012024456.GA7671@merlins.org>

On Fri, Oct 11, 2002 at 09:44:33PM -0400, Greg Ward wrote:
> Hi all --
> 
> I've just spent the last couple of hours beating the README.EXIM
> document from Mailman's current CVS tree into shape.  (This is derived
> from Nigel Metheringham's "HOWTO - Using exim and Mailman together"
> document, which can be found at www.exim.org.)

Cool.

I've been meaning to add this:

----------------------------------------------------------------------------
If you do header and envelope callbacks, you can disable them for bounces to
mailman lists (it  is quite common for  internal hosts to bounce  with a non
reachable internal address). The  idea is that you typically  don't want non
bounceable Email, but  you'd better accept bounces to mailman  lists so that
you can unsubscribe the people who are bouncing.

The rcpt-to ACL should look something like this

  # Accept bounces to lists even if callbacks or other checks would fail
  warn     message	= X-WhitelistedRCPT-nohdrfromcallback: Yes
	   condition	= ${if and {{match{$local_part}{.*-bounces\+.*}} {exists {MAILMAN_HOME/lists/${sg{$local_part}{(.*)-bounces\+.*}{\$1}}}}}{yes}{no}} 

  accept  condition	= ${if and {{match{$local_part}{.*-bounces\+.*}} {exists {MAILMAN_HOME/lists/${sg{$local_part}{(.*)-bounces\+.*}{\$1}}}}}{yes}{no}} 

  # Now, do basic address checking, that we forgo if the receipient is in a
  # whilelist
  deny    hosts = !+localadds:!+host_disable_callback:*
          sender_domains = !+envdomain_disable_callback:*
	  local_parts	= !+noenvfromcallback
         !verify = sender/callout=90s/check_postmaster

and the data ACL, like this:
  deny    hosts = !+localadds:!+host_disable_callback:*
         !condition = $header_X-WhitelistedRCPT-nohdrfromcallback:
	 !condition = ${lookup{${domain:$header_from:}}lsearch{HDR_DOMAIN_DISABLE_CALLBACK} {yes}{no}}
	 !verify = header_sender/callout=90s/check_postmaster
	  # Thanks to the exim4.gooderror.diff patch, exim will return
	  # informative error messages. You can override that with a generic
	  # message though
	  #message = Header sender couldn't be verified
----------------------------------------------------------------------------

Marc
-- 
"A mouse is a device used to point at the xterm you want to type in" - A.S.R.
Microsoft is to operating systems & security ....
                                      .... what McDonalds is to gourmet cooking
Home page: http://marc.merlins.org/   |   Finger marc_f@merlins.org for PGP key

From bronto@csd-bes.net  Thu Oct 10 23:48:30 2002
From: bronto@csd-bes.net (bronto)
Date: Thu, 10 Oct 2002 15:48:30 -0700
Subject: [Mailman-Developers] Cron error messages
Message-ID: <a05111b0eb9cbb17782e6@[10.1.19.62]>

I've been getting some cron error messages, using v2.1b2 and then 
2.1b3.  I posted this recently in the regular mailman list, but got 
no help.  Maybe here?

Here's what I posted on September 23rd:

"I'm running Mailamn v2.1b2, and just got these errors sent to my email:

_________________________________________________________________________
Subject: Cron <mailman@g3> /usr/bin/python2 -S 
/usr/local/mailman/cron/gate_news

Traceback (most recent call last):
   File "/usr/local/mailman/cron/gate_news", line 44, in ?
     from Mailman import MailList
ValueError: bad marshal data

Subject: Cron <mailman@g3> /usr/bin/python2 -S 
/usr/local/mailman/cron/gate_news

Traceback (most recent call last):
   File "/usr/local/mailman/cron/gate_news", line 44, in ?
     from Mailman import MailList
   File "/usr/local/mailman/Mailman/MailList.py", line 65, in ?
     from Mailman.OldStyleMemberships import OldStyleMemberships
ValueError: bad marshal data

__________________________________________________________________________
They came five minutes apart.  The reference to "gate" suggests to me 
that it might have something to do with newsgroup syncing, but I'm 
not doing any of that.

Since posting that, I have discovered that one of my mailing lists 
has stopped functioning to some degree.  I had set up this particular 
list to be a simple personal archive of messages from another mailing 
list I subscribe to.  It doesn't send anything, so I hadn't noticed 
any problems.  While doing my mailman update to 2.1b3, I discovered 
that no messages have been archived since the 23rd when I got the 
above errors.  Snooping around the mailman directories, I see a ton 
of files (several hundred) in the qfiles/shunt directory, dated since 
9/23.  There are also a few dozen from 7/2, FWIW.  I'm betting this 
isn't a coincidence, but know nothing more than that.

Today at 12 noon, I received the following cron error:

_________________________________________
Traceback (most recent call last):
   File "/usr/local/mailman/cron/gate_news", line 44, in ?
     from Mailman import MailList
   File "/usr/local/mailman/Mailman/MailList.py", line 31, in ?
     import urllib
EOFError: EOF read where object expected
__________________________________________

Since that time, there has been a mailman process running taking up 
all available CPU cycles.  gtop shows that the CMD is 
/usr/bin/python2, FWIW.  There are a fair number of qrunner processes 
running, but my mailing lists are hardly active.

What's going on here?

Rob

From gward@mems-exchange.org  Sat Oct 12 02:44:33 2002
From: gward@mems-exchange.org (Greg Ward)
Date: Fri, 11 Oct 2002 21:44:33 -0400
Subject: [Mailman-Developers] Overhauled 'Exim and Mailman' doc
Message-ID: <20021012014433.GA29438@cthulhu.gerg.ca>

Hi all --

I've just spent the last couple of hours beating the README.EXIM
document from Mailman's current CVS tree into shape.  (This is derived
from Nigel Metheringham's "HOWTO - Using exim and Mailman together"
document, which can be found at www.exim.org.)

I reorganized the document, rewrote lots of text, reformatted just about
everything, simplified the Exim configuration code, and removed some
obsolete-looking stuff.  The document is now all about Mailman 2.1 on
Exim 3 or Exim 4; it doesn't explain how to have cohabiting MM 2.0 and
2.1 installations (although it should be possible -- I'm doing it on
mems-exchange.org with a config nearly identical to what's in the doc,
and Barry Warsaw is doing it on python.org with something vaguely
similar).

Exim-oids, please take a look at the hairy-scary routers/directors/
transports and make sure they look sane to you.  You could even test
them for me if you're feeling inclined; I have *not* actually tested
what I'm about to present.  Will do so sooner or later, I promise.  At
least with Exim 4 -- I don't use Exim 3 anywhere anymore.

Mailman-heads, I don't think there's much here for you, unless you're
also an Exim-oid, in which case see above.  Of course you should all
read this for clarity and grammar and so forth.  ;-)

Here 'tis:

------------------------------------------------------------------------
Using Exim and Mailman Together
===============================

[This is derived from Nigel Metheringham's "HOWTO - Using exim and
 Mailman together", which covers Mailman 2.0.x and Exim 3.  It
 has been updated to cover Mailman 2.1 and Exim 4.  The original
 document is here: http://www.exim.org/howto/mailman.html]


Mailman configuration
---------------------

There is no Mailman configuration needed other than the standard options
detailed in the Mailman install documentation.  The Exim configuration
is transparent to Mailman.  The uid/gid settings for Mailman must match
those in the config fragments given below.
   

Exim configuration
------------------

The Exim configuration is built so that a list created within Mailman
automatically appears to Exim without the need for defining any
additional aliases.

The drawback of this configuration is that it will work poorly on
systems supporting lists in several different mail domains.  While
Mailman handles virtual domains, it does not yet support having two
lists with the same name in different virtual domains, using the same
Mailman installation.  This will eventually change.  (But see below for
a variation on this scheme that should accomodate virtual domains
better.)

The configuration file excerpts below are for use in an already
functional Exim configuration, which accepts mail for the domain in
which the list resides.  If this domain is separate from the others
handled by your Exim configuration, then you'll need to:

  * add the list domain, "my.list.domain" to local_domains

  * add a "domains=my.list.domain" option to each of the directors
    (routers) for the list

  * (optional) exclude that domain from the other directors (routers)

[Note: the instructions in this document should work with either Exim 3
or Exim 4.  In Exim 3, you must have a 'local_domains' configuration
setting; in Exim 4, you most likely have a 'local_domains' domainlist.
If you don't, you probably know what you're doing and can adjust
accordingly.  Similarly, in Exim 4 the concept of "directors" has
disappeared -- there are only routers now.  So if you're using Exim 4,
whenever this document says "director", read "router".]

Whether you are using Exim 3 or Exim 4, you will need to add some macros
to the main section of your Exim config file.  You will also need to
define one new transport.  With Exim 3, you'll need to add two new
directors; with Exim 4, two new routers play the same role.

Finally, the configuration supplied here should allow co-habiting
Mailman 2.0 and 2.1 installations, with the proviso that you'll probably
want to use "mm21" in place of "mailman" -- e.g., MM21_HOME,
mm21_transport, etc.


Main configuration settings
~~~~~~~~~~~~~~~~~~~~~~~~~~~

First, you need to add some macros to the top of your Exim config file.
These just make the directors (routers) and transport below a bit
cleaner.  Obviously, you'll need to edit these based on how you
configured and installed Mailman.

  # Home dir for your Mailman installation -- aka Mailman's prefix
  # directory.
  MAILMAN_HOME=/usr/local/mailman
  MAILMAN_WRAP=MAILMAN_HOME/mail/mailman

  # User and group for Mailman, should match your --with-mail-gid
  # switch to Mailman's configure script.
  MAILMAN_UID=mailman
  MAILMAN_GID=mailman


Transport for Exim 3
~~~~~~~~~~~~~~~~~~~~

Add this to the transports section of your Exim config file,
i.e. somewhere between the first and second "end" line:

  mailman_transport:
    driver = pipe
    command = MAILMAN_WRAP \
              ${if def:local_part_suffix \
                   {${sg{$local_part_suffix}{-(\\w+)(\\+.*)?}{\$1}}}
                   {post}} \
              $local_part
    current_directory = MAILMAN_HOME
    home_directory = MAILMAN_HOME
    user = MAILMAN_UID
    group = MAILMAN_GID


Directors for Exim 3
~~~~~~~~~~~~~~~~~~~~

If you're using Exim 3, you'll need to add the following two directors
to your config file (directors go between the second and third "end"
lines).  Also, don't forget that order matters -- eg. you can make
Mailman lists take precedence over system aliases by putting these
directors in front of your aliasfile director, or vice-versa.

  # Director to handle a list's main posting address.  Should come first
  # in case there's a list named, eg. "mailman-owner".  Automatically
  # detects list existence by looking for lists/$local_part/config.pck
  # under MAILMAN_HOME.
  mailman_main_director:
    driver = smartuser
    require_files = MAILMAN_HOME/lists/$local_part/config.pck
    transport = mailman_transport

  # Handle all the other addresses related to a list: -bounces, -admin,
  # etc.  Same trick works to detect list existence thanks to the
  # 'suffix' option.
  mailman_director:
    driver = smartuser
    require_files = MAILMAN_HOME/lists/$local_part/config.pck
    suffix = -bounces : -bounces+* : \
             -confirm+* : -join : -leave : \
             -owner : -request : -admin
    transport = mailman_transport


Routers for Exim 4
~~~~~~~~~~~~~~~~~~

In Exim 4, there's no such thing as directors -- you instead need to add
two routers.  Also, the canonical order of the configuration file was
changed so routers come before transports, so the routers for Exim 4
come first here.  Put these two routers somewhere after the "begin
routers" line of your config file, and remember that order matters.

  mailman_main_router:
    driver = accept
    require_files = MAILMAN_HOME/lists/$local_part/config.pck
    transport = mailman_transport

  mailman_router:
    driver = accept
    require_files = MAILMAN_HOME/lists/$local_part/config.pck
    local_part_suffix = -bounces : -bounces+* : \
                        -confirm+* : -join : -leave : \
                        -owner : -request : -admin
    transport = mailman_transport


Transports for Exim 4
~~~~~~~~~~~~~~~~~~~~~

The transport for Exim 4 is the same as for Exim 3; just copy the
transport given above to somewhere under the "begin transports" line of
your Exim config file.


Notes
-----

Exim should be configured to allow reasonable volume -- e.g. don't
setting max_recipients down to a silly value -- and with normal degrees
of security -- specifically, be sure to allow relaying from 127.0.0.1,
but pretty much nothing else.  Parallel deliveries and other tweaks can
also be used if you like; experiment with your setup to see what works.
Delay warning messages should be switched off or configured to only
happen for non-list mail, unless you like receiving tons of mail when
some random host is down.


Problems
--------

  * Mailman will send as many MAIL FROM/RCPT TO as it needs. It may result
    in more than 10 or 100 messages sent in one connection, which will exceed
    the default value of Exim's smtp_accept_queue_per_connection
    This is bad because it will cause Exim to switch into queue mode and
    severely delay delivery of your list messages.
    The way to fix this is to set mailman's SMTP_MAX_SESSIONS_PER_CONNECTION
    (in ~mailman/Mailman/mm_cfg.py) to a smaller value than Exim's
    smtp_accept_queue_per_connection

  * Mailman should ignore Exim delay warning messages, even though
    Exim should never send this to list messages.  Mailman 2.1's
    general bounce detection and VERP support should greatly improve
    the bounce detector's hit rates.

  * List existence is checked on whether there is a config.pck file
    for a list.  If you delete lists by foul means, be aware of this.

  * If you are getting Exim or Mailman complaining about user ids
    when you send mail to a list, check that the MAILMAN_UID and
    MAILMAN_GID match those of Mailman itself (i.e. what were used
    in the configure script).  Also make sure you do not have
    aliases in the main alias file for the list.


Doing VERP with Exim and Mailman
--------------------------------

VERP will send one email, with a separate envelope sender (return path),
for each of your subscribers -- read the information in
~mailman/Mailman/Default.py for the options that start with VERP.  In a
nutshell, all you need to do to enable VERP with Exim is to add these
lines to ~mailman/Mailman/mm_cfg.py:

  VERP_PASSWORD_REMINDERS = 1
  VERP_PERSONALIZED_DELIVERIES = 1
  VERP_DELIVERY_INTERVAL = 1
  VERP_CONFIRMATIONS = 1

(The directors/routers above are smart enough to deal with VERP
bounces.)

       
Virtual Domains
---------------

One approach to handling virtual domains is to use a separate Mailman
installation for each virtual domain.  (Currently, this is the only way
to have lists with the same name in different virtual domains handled by
the same machine.)

In this case, the MAILMAN_HOME and MAILMAN_WRAP macros are useless --
you can remove them.  Change your directors (routers) to something like
this:

  require_files = /virtual/${domain}/mailman/lists/${lc:$local_part}/config.pck

and change your transport like this:

  command = /virtual/${domain}/mailman/mail/mailman \
            ${if def:local_part_suffix \
                 {${sg{$local_part_suffix}{-(\\w+)(\\+.*)?}{\$1}}}
                 {post}} \
              $local_part
  current_directory = /virtual/${domain}/mailman
  home_directory = /virtual/${domain}/mailman


List Verification
-----------------

This is how a set of address tests for the Exim lists look on a working
system for the quixote-users@mems-exchange.org list ("% " indicates the
Unix shell prompt):

  % exim -bt quixote-users
  quixote-users@mems-exchange.org
    router = mailman_main_router, transport = mailman_transport

  % exim -bt quixote-users-request
  quixote-users-request@mems-exchange.org
    router = mailman_router, transport = mailman_transport

  % exim -bt quixote-users-bounces
  quixote-users-bounces@mems-exchange.org
    router = mailman_router, transport = mailman_transport

  % exim -bt quixote-users-bounces+luser=example.com
  quixote-users-bounces+luser=example.com@mems-exchange.org
    router = mailman_router, transport = mailman_transport


Document History
----------------

Originally written by Nigel Metheringham <postmaster@exim.org>.
Updated by Marc Merlin <marc_soft@merlins.org> for Mailman 2.1, Exim 4.
Overhauled/reformatted/clarified/simplified by Greg Ward <gward@python.net>.
------------------------------------------------------------------------

-- 
Greg Ward - software developer                gward@mems-exchange.org
MEMS Exchange                            http://www.mems-exchange.org

From col.neeson@planetone-asia.com  Thu Oct 10 02:23:20 2002
From: col.neeson@planetone-asia.com (Colin Neeson)
Date: Thu, 10 Oct 2002 11:23:20 +1000
Subject: [Mailman-Developers] 2.1b3 gate_news problem Sparc/Linux 6.2
Message-ID: <B9CB13A8.88B%col.neeson@planetone-asia.com>


I posted this off to the users group, but nothing in response, so I thought
I would toss it here for comment.

I've got Mailman 2.0.13 in production, and currently am testing 2.1b3 on
Sparc/Linux 6.2.  I've had no problems with 2.1b3, until I try and use
news-mail (and vice versa) gateway.  I am not using this feature in 2.0.13,
so I am not sure whether it barfs there or not.  I'm using INN 2.2.2.

I'm not familiar with Python (clueless actually).  Could someone comment on
the following?

Traceback (most recent call last):
  File "/usr/local/mailman/cron/gate_news", line 270, in ?
    main()
  File "/usr/local/mailman/cron/gate_news", line 252, in main
    clearcache()
  File "/usr/local/mailman/cron/gate_news", line 107, in clearcache
    conn.quit()
  File "/usr/local/lib/python2.2/nntplib.py", line 552, in quit
    resp = self.shortcmd('QUIT')
  File "/usr/local/lib/python2.2/nntplib.py", line 247, in shortcmd
    self.putcmd(line)
  File "/usr/local/lib/python2.2/nntplib.py", line 187, in putcmd
    self.putline(line)
  File "/usr/local/lib/python2.2/nntplib.py", line 182, in putline
    self.sock.sendall(line)
socket.error: (32, 'Broken pipe')

The news server is up and running.  I can post and read news items from it,
so the socket.error is perplexing.

I would be very grateful for any assistance.

Many thanks,

-Colin.


From GustavoM.  Fri Oct 11 21:45:08 2002
From: GustavoM. (GustavoM.)
Date: Fri, 11 Oct 2002 20:45:08 +0000
Subject: [Mailman-Developers] problem with list member's vacation script
Message-ID: <E180C15-0001bz-00@real.local.evenreal.com>

Hi,

I manage a mailing list site with mailman, and I recently had the following 
problem:

Lets say I have a list called test@site.org with 3 members: m1@foo.org, 
m2@foo.org and m3@foo.org. m3 has its vacation script turned on. m1 sends an 
email to test@site.org, so m2,m3 and m1 itself receive the email. m3's 
vacation script replies to m1 and to test@site.org, so m1 gets 2 copies of 
the reply (one directly and one through test@site.org), m2 gets a copy and m1 
gets a copy of its own reply which in turn will be replied by its vacation 
script and so on. So my mail server ends up sending a lot of useless emails.
How can I stop this situation? 
I know I can reject or hold emails based on some header pattern matching, but 
I cannot do this on a case by case basis, not all vacation scripts behave the 
same, so this is not a solution for my problem.
I was thinking maybe there's a way to configure mailman so that if m1 sends 
an email to test@site.org mailman will not send a copy of the email back to 
m1 just because m1 is a list member.

Also, is there a way to limit the number of postings each member is allowed 
to sent per day?

Thanks,

-- 
Gustavo M.

From michael@adoptionmedia.com  Sat Oct 12 00:41:42 2002
From: michael@adoptionmedia.com (michael)
Date: Fri, 11 Oct 2002 16:41:42 -0700
Subject: [Mailman-Developers] displaying number of users on a website
Message-ID: <3DA761B6.3000807@adoptionmedia.com>

I'm trying to create a script that would display a dynamic number of 
subscribed users on my website. The website itself is written in php. 
Had anyone done this before? Thanks


From claw@kanga.nu  Sat Oct 12 07:14:32 2002
From: claw@kanga.nu (J C Lawrence)
Date: Fri, 11 Oct 2002 23:14:32 -0700
Subject: [Mailman-Developers] Overhauled 'Exim and Mailman' doc 
In-Reply-To: Message from Greg Ward <gward@mems-exchange.org> 
	of "Fri, 11 Oct 2002 21:44:33 EDT."
	<20021012014433.GA29438@cthulhu.gerg.ca> 
References: <20021012014433.GA29438@cthulhu.gerg.ca> 
Message-ID: <4162.1034403272@kanga.nu>

On Fri, 11 Oct 2002 21:44:33 -0400 
Greg Ward <gward@mems-exchange.org> wrote:

> Hi all -- I've just spent the last couple of hours beating the
> README.EXIM document from Mailman's current CVS tree into shape.
> (This is derived from Nigel Metheringham's "HOWTO - Using exim and
> Mailman together" document, which can be found at www.exim.org.)

Would you like to have the Exim/TMDA/Mailman HOW-TO as a rider on that,
or should we tuck it in as a separate README (Barry?).

-- 
J C Lawrence                
---------(*)                Satan, oscillate my metallic sonatas. 
claw@kanga.nu               He lived as a devil, eh?		  
http://www.kanga.nu/~claw/  Evil is a name of a foeman, as I live.


From pioppo@ferrara.linux.it  Fri Oct 11 19:55:42 2002
From: pioppo@ferrara.linux.it (Simone Piunno)
Date: Fri, 11 Oct 2002 20:55:42 +0200
Subject: [Mailman-Developers] Re: [Mailman-checkins] mailman/templates/ru
	archidxentry.html,NONE,2.0 archidxfoot.html,NONE,2.0 archidxhead.html,NONE,2.0
	archlistend.html,NONE,2.0 archliststart.html,NONE,2.0
	archtocentry.html,NONE,2.0 archtoc.html,NONE,2.0
In-Reply-To: <20021010221012.GA12092@mawhrin.net>
References: <E17zl3s-0004Bz-00@usw-pr-cvs1.sourceforge.net>
	<20021010221012.GA12092@mawhrin.net>
Message-ID: <20021011185542.GC4158@ferrara.linux.it>

On Thu, Oct 10, 2002 at 11:10:12PM +0100, Mikhail Sobolev wrote:

> > --- NEW FILE: archidxentry.html ---
> > 
> > --- NEW FILE: archlistend.html ---
> 
> I am not sure that that it's a really good idea to create that small
> files... 

where's the problem?
they're templates and admins are supposed to fill in to change the
look&feel on their own taste.

-- 
Adde parvum parvo magnus acervus erit.
Simone Piunno, FerraraLUG - http://members.ferrara.linux.it/pioppo

From noreply@sourceforge.net  Sat Oct 12 17:51:30 2002
From: noreply@sourceforge.net (noreply@sourceforge.net)
Date: Sat, 12 Oct 2002 09:51:30 -0700
Subject: [Mailman-Developers] 
 [ mailman-Patches-444884 ] Integration of Mailman & htdig for archi
Message-ID: <E180PUA-0005QT-00@usw-sf-web2.sourceforge.net>

Patches item #444884, was opened at 2001-07-26 18:27
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=300103&aid=444884&group_id=103

Category: None
Group: None
Status: Open
Resolution: None
Priority: 7
Submitted By: Richard Barrett (ppsys)
Assigned to: Barry A. Warsaw (bwarsaw)
Summary: Integration of Mailman & htdig for archi

Initial Comment:
This patch is applicable to Mailman 2.0.6 release that 
has had search enhancement patch 444879 patch 
installed - if your Defaults.py has the 
ARCHIVE_INDEXING_ENABLE and ARCHIVE_INDEXING_DISABLE 
in it then you've got that patch.

It replaces earlier patches 401670 and 402423 and is 
mainly to correct some problems arising from fixes 
introduced into Mailman by bug fix releases since the 
402423 patch. 

This patch integrates htdig with Mailman and provides:

1. per list search facility with a search form on the 
list's TOC page.

2. maintenance of privacy of private archives which 
requires the user to establish their credentials via 
the normal private archive access before any access 
via htdig is allowed.
   
3. a common base URL for both public and private 
archive access via htsearch results so that htdig 
indices are unaffected by changingan archive from 
private to public and vice versa. All access to 
archives via htdig is controlled by a new wrapped cgi-
bin script called htdig.py.
   
4. a new cron activated script and extra crontab entry 
which runs htdig regularly to maintain the per list 
search indices.
   
5. automatic creation, deletion and maintenance of 
htdig configuration files and such. Beyond installing 
htdig and telling Mailman where it is via mm_cfg you 
do not have to do any other setup. Well not quite you 
do have to set up a single per installation symlink to 
allow htdig to find the automatically generated per 
list htdig configuration files.

You probably want to run this patch as follows:

cd <mailman 2.0.6 untarred and unzipped directory>
patch -p1 < <this patch file>  

----------------------------------------------------------------------

Comment By: Colin Mackinlay (cmackinlay)
Date: 2002-10-12 16:51

Message:
Logged In: YES 
user_id=624179

Got a workaround!

The line referred to in the traceback:
 file(rundig_run_file, 'w').close()
is used to create a 'rundig_last_run' file of lenght 0 bytes
Creating this manually (or copying it) means the line isn't 
called and everything seems to work.

Either file() is not a valid function call or my python is broken - 
I'm not literate enough in python to know the answer though! 

----------------------------------------------------------------------

Comment By: Colin Mackinlay (cmackinlay)
Date: 2002-10-06 14:18

Message:
Logged In: YES 
user_id=624179

Just rebuilt MM as 2.1b3 with htdig.
Upgraded lists which had htdig before work fine
New lists give the obvious error:
  Unable to read word database file
  Did you run htmerge?
Running the cronjob doesn't fix as it used to, message is:
  Output from command /usr/bin/python -
S /usr/local/mailman/cron/nightly_htdig ..

Traceback (most recent call last):
  File "/usr/local/mailman/cron/nightly_htdig", line 153, in ?
    main()
  File "/usr/local/mailman/cron/nightly_htdig", line 118, in main
    file(rundig_run_file, 'w').close()
NameError: global name 'file' is not defined

The archive/htdig folder only contains the xx.conf file, but no 
db.xx files

If I copy in db.xx files from another list then the problem goes 
away (except I've now got an invalid set of references!)

Is this my elementary error or is it more sinister?!


----------------------------------------------------------------------

Comment By: Richard Barrett (ppsys)
Date: 2002-08-15 11:02

Message:
Logged In: YES 
user_id=75166

htdig-2.1b3-0.2.patch corrects a dumb syntax error in htdig-
2.1b3-0.1.patch which will typically show up as logged errors 
in the operation of the ArchRunner qrunner at line 721 of 
HyperArch.py

----------------------------------------------------------------------

Comment By: Richard Barrett (ppsys)
Date: 2002-08-14 16:51

Message:
Logged In: YES 
user_id=75166

htdig-2.1b3-0.1.patch is a revised version of the patch that is 
compatible with MM 2.1b3

----------------------------------------------------------------------

Comment By: Barry A. Warsaw (bwarsaw)
Date: 2002-08-08 16:33

Message:
Logged In: YES 
user_id=12800

I've sent Richard some comments off-line about this patch.

Meta comments: the 2.0.x patches can't be officially
supported, but I'm going to create an unofficial patches
page off the wiki for where the 2.0 patches can be migrated.

I think this patch set is too big for MM2.1, but if it's
cleaned up as per my private message, let's re-evaluate it
for MM2.2 (or 3.0).

----------------------------------------------------------------------

Comment By: Richard Barrett (ppsys)
Date: 2002-08-05 10:11

Message:
Logged In: YES 
user_id=75166

htdig-2.0.13-0.2.patch just adds a GPL notice to the patch


----------------------------------------------------------------------

Comment By: Richard Barrett (ppsys)
Date: 2002-08-01 16:35

Message:
Logged In: YES 
user_id=75166

htdig-2.1b2-0.1.patch is a revised version of the patch 
that is compatible with MM 2.1b2

----------------------------------------------------------------------

Comment By: Richard Barrett (ppsys)
Date: 2002-07-30 11:25

Message:
Logged In: YES 
user_id=75166

htdig-2.0.13-0.1.patch is purely cosmetic to get no mumble 
application to MM 2.0.13

----------------------------------------------------------------------

Comment By: Richard Barrett (ppsys)
Date: 2002-07-25 15:07

Message:
Logged In: YES 
user_id=75166

Do not use htdig-2.0.12-0.1.patch there is an error in it.
Use htdig-2.0.12-0.2.patch instead

----------------------------------------------------------------------

Comment By: Richard Barrett (ppsys)
Date: 2002-07-25 14:10

Message:
Logged In: YES 
user_id=75166

htdig-2.0.12-0.1.patch is a revised version of the patch that 
applies without complaint to MM 2.0.12.

It also add a facility for adding site wide htdig configuration 
attributes to all list specific htdig configuration files. 



----------------------------------------------------------------------

Comment By: Richard Barrett (ppsys)
Date: 2002-05-23 09:59

Message:
Logged In: YES 
user_id=75166

htdig-2.0.11-0.1.patch is a revised version of the patch that 
is compatible with MM 2.0.11

This version removes an incompatibility with Python 2.2 
which caused warning messages to be generated when any 
of the family cron/nightly_htdig scripts were run.

Some guidance on file access permissions for some htdig 
database files needed by rundig have been added to 
installation notes.

----------------------------------------------------------------------

Comment By: Richard Barrett (ppsys)
Date: 2002-04-19 10:59

Message:
Logged In: YES 
user_id=75166

htdig-2.0.10-0.1.patch is a revised version of the patch 
that is compatible with MM 2.0.10

----------------------------------------------------------------------

Comment By: Richard Barrett (ppsys)
Date: 2002-04-08 17:46

Message:
Logged In: YES 
user_id=75166

htdig-2.0.9-0.1.patch is a revised version of the patch 
that is compatible with MM 2.0.9

----------------------------------------------------------------------

Comment By: Richard Barrett (ppsys)
Date: 2002-03-06 16:22

Message:
Logged In: YES 
user_id=75166

htdig-2.1cvs-20020306.patch is a revised version of the patch that is compatible with the code published in 
mailman CVS on sourceforge as 12:30 GMT 6 Mar 2002

Known deficiency is that the non-English versions of files under $build/templates still contain text in English 
and need translations I cannot do. Also the necessary pygettext activity and subsequent translations in 
files under $build/messages remain to be done.

----------------------------------------------------------------------

Comment By: Richard Barrett (ppsys)
Date: 2001-12-17 16:56

Message:
Logged In: YES 
user_id=75166

htdig-2.1cvs-20011217.patch is a revised version of the 
patch that is compatible with the code published in mailman 
CVS on sourceforge as 11:50 GMT 17 Dec 2001

The only known deficiency is that the non-English versions 
of files under $build/templates still contain text in 
English and need translations I cannot do. Also the 
necessary pygettext activity and subsequent translations in 
files under $build/messages remain to be done.

----------------------------------------------------------------------

Comment By: Richard Barrett (ppsys)
Date: 2001-12-13 16:58

Message:
Logged In: YES 
user_id=75166

htdig-2.1a3-0.1.patch is a revised version of the patch that is compatible with the code published in 
mailman-2.1a3.tgz on sourceforge.

The only known deficiency is that the non-English versions of files under $build/templates still contain text 
in English and need translations I cannot do. Also the necessary pygettext activity and subsequent 
translations in files under $build/messages remain to be done. 

----------------------------------------------------------------------

Comment By: Richard Barrett (ppsys)
Date: 2001-11-28 17:33

Message:
Logged In: YES 
user_id=75166

The htdig-2.0.8-0.1.patch version of the patch resolves a problem that can arise with htdig indexing if the 
web_page_url for a list uses other than the http addressing (some folks want to use https). While specified 
as for MM 2.0.8 the revised patch should work OK with 2.0.7, 2.0.6 and probably back as far as 2.0.3. If 
you do not have the requirement for using other than http addressing in you lists web_page_urls it probably 
isn't worth the trouble of upgrading to this patch level.

----------------------------------------------------------------------

Comment By: Richard Barrett (ppsys)
Date: 2001-11-28 11:08

Message:
Logged In: YES 
user_id=75166

This patch should also apply without problems to MM 2.0.8

----------------------------------------------------------------------

Comment By: Richard Barrett (ppsys)
Date: 2001-11-27 12:00

Message:
Logged In: YES 
user_id=75166

This patch should also apply without problems to Mm 2.0.7

----------------------------------------------------------------------

Comment By: Richard Barrett (ppsys)
Date: 2001-11-09 11:54

Message:
Logged In: YES 
user_id=75166

The htdig-2.0.6-03.patch version of the patch makes some 
previously hard-coded things configurable and enhances the 
capability to run the htdig searches and indexing on a 
different machine to the one delivering Mailman and 
Mailman's web UI.

----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=300103&aid=444884&group_id=103

From chuqui@plaidworks.com  Sun Oct 13 04:43:08 2002
From: chuqui@plaidworks.com (Chuq Von Rospach)
Date: Sat, 12 Oct 2002 20:43:08 -0700
Subject: [Mailman-Developers] updated to CVS, and stuff broke....
Message-ID: <E360685D-DE5D-11D6-99E5-0003934516A8@plaidworks.com>

I updated my mailman to CVS tonight. a few things broke.

Looks like a couple of Defaults.py variables got added but weren't 
added to Defaults.py on the update.

# Default action to take on filtered messages.
# 0 = Discard, 1 = Reject, 2 = Forward, 3 = Preserve
DEFAULT_FILTER_ACTION = 0

was one, and

# Whether to allow list owners to preserve content filtered messages to 
a
# special queue on the disk.
OWNERS_CAN_PRESERVE_FILTERED_MESSAGES = 1

was, I believe, the other.

Then I went in and tried to turn on verping and personalization. When I 
turned on personalization, it broke:


# Set this variable to 1 to allow list owners to set the "personalized" 
flags
# on their mailing lists.  Turning these on tells Mailman to send 
separate
# email messages to each user instead of batching them together for 
delivery
# to the MTA.  This gives each member a more personalized message, but 
can
# have a heavy impact on the performance of your system.
OWNERS_CAN_ENABLE_PERSONALIZATION = 1

then turn it on in the non-digest page for the list.

The verping values are:

VERP_PASSWORD_REMINDERS = 1

# Another good opportunity is when regular delivery is personalized.  
Here
# again, we're already incurring the performance hit for addressing each
# individual recipient.  Set this to true to enable VERPs on all 
personalized
# regular deliveries (personalized digests aren't supported yet).
VERP_PERSONALIZED_DELIVERIES = 1

# And finally, we can VERP normal, non-personalized deliveries.  
However,
# because it can be a significant performance hit, we allow you to 
decide how
# often to VERP regular deliveries.  This is the interval, in number of
# messages, to do a VERP recipient address.  The same variable controls 
both
# regular and digest deliveries.  Set to 0 to disable occasional VERPs, 
set to
# 1 to VERP every delivery, or to some number > 1 for only occasional 
VERPs.
VERP_DELIVERY_INTERVAL = 10

# Set this to true to enable VERP-like (more user friendly) 
confirmations
VERP_CONFIRMATIONS = 1

Once I turn it on through the admin page, it stops working with:

plaidworks.com 110# Oct 12 20:36:07 2002 (21778) Uncaught runner 
exception: 'NoneType' object has no attribute 'encode'
Oct 12 20:36:07 2002 (21778) Traceback (most recent call last):
   File "/usr/local/mailman/Mailman/Queue/Runner.py", line 105, in 
_oneloop
     self._onefile(msg, msgdata)
   File "/usr/local/mailman/Mailman/Queue/Runner.py", line 154, in 
_onefile
     keepqueued = self._dispose(mlist, msg, msgdata)
   File "/usr/local/mailman/Mailman/Queue/OutgoingRunner.py", line 61, 
in _dispose
     self._func(mlist, msg, msgdata)
   File "/usr/local/mailman/Mailman/Handlers/SMTPDirect.py", line 135, 
in process
     msgdata['recips'] = chunk
   File "/usr/local/mailman/Mailman/Handlers/SMTPDirect.py", line 287, 
in verpdeliver
     # name is a byte string, convert it first to Unicode, given
AttributeError: 'NoneType' object has no attribute 'encode'

if I turn that back off, the lsit starts working again.





-- 
Chuq Von Rospach, Architech
chuqui@plaidworks.com -- http://www.plaidworks.com/chuqui/blog/

No! No! Dead girl, OFF the table! -- Shrek


From chuqui@plaidworks.com  Sun Oct 13 04:48:08 2002
From: chuqui@plaidworks.com (Chuq Von Rospach)
Date: Sat, 12 Oct 2002 20:48:08 -0700
Subject: [Mailman-Developers] HTML and text digests...
Message-ID: <969FACEA-DE5E-11D6-99E5-0003934516A8@plaidworks.com>


I've been playing with 2.1b3+ and inflicting it on my users. For the 
most part, it's been quite stable adn they seem to like it. the only 
real probelms have been self-inflicted, and Barry's gotten a good 
giggle over some of the things I did to break my stuff...

The one place I was getting complaints was HTML in the text digests.

I'd set things up to allow in multipart, and text/plain and text/html, 
plus image/{png,jpeg,gif}, just to see how folks used it and what they 
did with it.

That allowed HTML parts into the text digest, which doesn't deal with 
it well. that made people upset.

So I've turned on folding html to text via lynx, turned off the images, 
and we'll see.

But it seems to me that one way to allow that but protect the text 
digests was to use that lynx filtering, but only on the content sent to 
the text digest. Yes, I know, another @#$@%@#$ configuration option, 
but since the technology already exists....



-- 
Chuq Von Rospach, Architech
chuqui@plaidworks.com -- http://www.plaidworks.com/chuqui/blog/

Someday, we'll look back on this, laugh
nervously and change the subject.


From barry@python.org  Sun Oct 13 05:11:35 2002
From: barry@python.org (Barry A. Warsaw)
Date: Sun, 13 Oct 2002 00:11:35 -0400
Subject: [Mailman-Developers] Re: [Mailman-checkins] mailman/templates/ru
 archidxentry.html,NONE,2.0 archidxfoot.html,NONE,2.0
 archidxhead.html,NONE,2.0 archlistend.html,NONE,2.0
 archliststart.html,NONE,2.0 archtocentry.html,NONE,2.0
 archtoc.html,NONE,2.0
References: <E17zl3s-0004Bz-00@usw-pr-cvs1.sourceforge.net>
	<20021010221012.GA12092@mawhrin.net>
Message-ID: <15784.62071.201635.708438@gargle.gargle.HOWL>


>>>>> "MS" == Mikhail Sobolev <mss@mawhrin.net> writes:

    MS> On Thu, Oct 10, 2002 at 02:41:40PM -0700,
    MS> mss@users.sourceforge.net wrote:
    >> --- NEW FILE: archidxentry.html --- --- NEW FILE:
    >> archlistend.html ---

    MS> I am not sure that that it's a really good idea to create that
    MS> small files...  However, I am not ready to submit a patch, so
    MS> this is just to bring the issue to the attention.

These were necessary in order to i18n the pipermail pages.  CAn you
give more specifics about what the problem is?

-Barry

From marc_news@merlins.org  Sun Oct 13 06:28:20 2002
From: marc_news@merlins.org (Marc MERLIN)
Date: Sat, 12 Oct 2002 22:28:20 -0700
Subject: [Mailman-Developers] Re: [Exim] Overhauled 'Exim and Mailman' doc
In-Reply-To: <20021012184710.GB30217@cthulhu.gerg.ca>
References: <20021012014433.GA29438@cthulhu.gerg.ca>
	<20021012024456.GA7671@merlins.org> <20021012184710.GB30217@cthulhu.gerg.ca>
Message-ID: <20021013052820.GZ14183@merlins.org>

On Sat, Oct 12, 2002 at 02:47:10PM -0400, Greg Ward wrote:
> On 11 October 2002, Marc MERLIN said:
> > I've been meaning to add this:
> 
> Hmmm, good point.  Hadn't thought about it because SMTP callbacks seem a
> little *too* powerful to me.
> 
> I think it's more important to add something about disabling recipient
> verification for 127.0.0.1 first, but I wanted to get this text out for
> review first.

Just in case there was a misunderstanding here, the two are separate.
1) you want to disable from checks when the connection comes from
   localhost (as you said)
2) you also should skip callbacks if you are getting an Email going to a
   mailman bounce address (we might as well process a bounce coming from
   an unbouncable postmaster)

You may already have understood that, but I was just making sure :-)

Marc
-- 
"A mouse is a device used to point at the xterm you want to type in" - A.S.R.
Microsoft is to operating systems & security ....
                                      .... what McDonalds is to gourmet cooking 
Home page: http://marc.merlins.org/   |   Finger marc_f@merlins.org for PGP key

From tkikuchi@is.kochi-u.ac.jp  Sun Oct 13 07:05:02 2002
From: tkikuchi@is.kochi-u.ac.jp (Tokio Kikuchi)
Date: Sun, 13 Oct 2002 15:05:02 +0900
Subject: [Mailman-Developers] MimeDel i18n message undelivered
Message-ID: <3DA90D0E.2070008@is.kochi-u.ac.jp>

Hi,

While testing MimeDel with i18n Japanese configuration, message
delivery error occured when forwarding the filetered message to
the list administrator. Here is a quick patch:

--- /home/mailman/src/mailman/Mailman/Deliverer.py      Tue Oct  8 08:35:18 2002
+++ Deliverer.py        Sun Oct 13 14:18:06 2002
@@ -132,7 +132,7 @@
                         _charset=Utils.GetCharSet(self.preferred_language))
         attachment = MIMEMessage(msg)
         notice = Message.UserNotification(
-            recips, self.GetBouncesEmail(), subject)
+            recips, self.GetBouncesEmail(), subject, lang=self.preferred_language)
         notice.set_type('multipart/mixed')
         notice.attach(text)
         notice.attach(attachment)

Cheers,

-- 
Tokio Kikuchi, tkikuchi@ is.kochi-u.ac.jp
http://weather.is.kochi-u.ac.jp/


From mss@mawhrin.net  Sun Oct 13 20:33:38 2002
From: mss@mawhrin.net (Mikhail Sobolev)
Date: Sun, 13 Oct 2002 20:33:38 +0100
Subject: [Mailman-Developers] Re: [Mailman-checkins] mailman/templates/ru
	archidxentry.html,NONE,2.0 archidxfoot.html,NONE,2.0 archidxhead.html,NONE,2.0
	archlistend.html,NONE,2.0 archliststart.html,NONE,2.0
	archtocentry.html,NONE,2.0 archtoc.html,NONE,2.0
In-Reply-To: <15784.62071.201635.708438@gargle.gargle.HOWL>
References: <E17zl3s-0004Bz-00@usw-pr-cvs1.sourceforge.net>
	<20021010221012.GA12092@mawhrin.net>
	<15784.62071.201635.708438@gargle.gargle.HOWL>
Message-ID: <20021013193338.GA24768@mawhrin.net>

On Sun, Oct 13, 2002 at 12:11:35AM -0400, Barry A. Warsaw wrote:
> 
> >>>>> "MS" == Mikhail Sobolev <mss@mawhrin.net> writes:
> 
>     MS> On Thu, Oct 10, 2002 at 02:41:40PM -0700,
>     MS> mss@users.sourceforge.net wrote:
>     >> --- NEW FILE: archidxentry.html --- --- NEW FILE:
>     >> archlistend.html ---
> 
>     MS> I am not sure that that it's a really good idea to create that
>     MS> small files...  However, I am not ready to submit a patch, so
>     MS> this is just to bring the issue to the attention.
> 
> These were necessary in order to i18n the pipermail pages.  CAn you
> give more specifics about what the problem is?
It's not a _problem_.  It's just a feeling about a "right thing". :(

What I feel is not right that the resulting _file_ is assembled from
other _files_.  I believe, this is not how the rest of Mailman is done:
the "template" file _is_ the resulting file beside some substitions,
which are necessary to fill the some fields or other parts.

Again, it's not a problem.

--
Misha

From chuqui@plaidworks.com  Sun Oct 13 20:40:58 2002
From: chuqui@plaidworks.com (Chuq Von Rospach)
Date: Sun, 13 Oct 2002 12:40:58 -0700
Subject: [Mailman-Developers] more 2.1b3++ MIME funnies.
Message-ID: <B27BFB52-DEE3-11D6-99E5-0003934516A8@plaidworks.com>


Okay, is this me, or mailman?

I have content filtering on. My blacklist is empty. my whitelist is:

multipart/alternative
text/plain
text/html

I have convert text/html to plain text on.

My lynx setup is:

> # Command that is used to convert text/html parts into plain text.  
> This
> # should output results to standard output.  %(filename)s will contain 
> the
> # name of the temporary file that the program should operate on.
> HTML_TO_PLAIN_TEXT_COMMAND = '/usr/local/bin/lynx -dump %(filename)s'
>
> plaidworks.com 119# lynx -version
> Lynx Version 2.8.4rel.1 (17 Jul 2001)
> Built on darwin6.0 Aug 28 2002 22:10:30
>
> Copyrights held by the University of Kansas, CERN, and other 
> contributors.
> Distributed under the GNU General Public License.
> See http://lynx.browser.org/ and the online help for more information.

I'm finding that if someone sends a M/A message with a plain and an 
html part, it's getting processed, and I'm getting a message going out 
with two t/p parts, one the original, and one empty. Since the empty 
one is second, it gets displayed.

So something is whacked out. I'm not sure what.

What are people doing for their black/whitelists? am I just building 
this thing sideways?

What I'd like to be able to do (but can't) is allow M/A, t/plain and 
t/html, and use the lynx to defang HTML going into the text (but not 
MIME) digests. That seems the most reasonable set of compromises to 
allow styled mail. Unfortunately, mailman can't be set up that way...

so my backup plan was to allow text/plain, convert html to text if the 
text part didn't exist, and try to add features later. But I don't seem 
to be quite right yet. For what I'm trying, I *think* I should be 
whitelisting M/A and Text/Plain, but not text/HTMl. that screws people 
who only send text/html without a plain part, but they probably 
shouldn't do that anyway...

well, off to tweak the settings again....





-- 
Chuq Von Rospach, Architech
chuqui@plaidworks.com -- http://www.plaidworks.com/chuqui/blog/

The first rule of holes: If you are in one, stop digging.


From noreply@sourceforge.net  Mon Oct 14 12:50:42 2002
From: noreply@sourceforge.net (noreply@sourceforge.net)
Date: Mon, 14 Oct 2002 04:50:42 -0700
Subject: [Mailman-Developers] 
 [ mailman-Patches-444884 ] Integration of Mailman & htdig for archi
Message-ID: <E1813kA-0000he-00@usw-sf-web5.sourceforge.net>

Patches item #444884, was opened at 2001-07-26 18:27
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=300103&aid=444884&group_id=103

Category: None
Group: None
Status: Open
Resolution: None
Priority: 7
Submitted By: Richard Barrett (ppsys)
Assigned to: Barry A. Warsaw (bwarsaw)
Summary: Integration of Mailman & htdig for archi

Initial Comment:
This patch is applicable to Mailman 2.0.6 release that 
has had search enhancement patch 444879 patch 
installed - if your Defaults.py has the 
ARCHIVE_INDEXING_ENABLE and ARCHIVE_INDEXING_DISABLE 
in it then you've got that patch.

It replaces earlier patches 401670 and 402423 and is 
mainly to correct some problems arising from fixes 
introduced into Mailman by bug fix releases since the 
402423 patch. 

This patch integrates htdig with Mailman and provides:

1. per list search facility with a search form on the 
list's TOC page.

2. maintenance of privacy of private archives which 
requires the user to establish their credentials via 
the normal private archive access before any access 
via htdig is allowed.
   
3. a common base URL for both public and private 
archive access via htsearch results so that htdig 
indices are unaffected by changingan archive from 
private to public and vice versa. All access to 
archives via htdig is controlled by a new wrapped cgi-
bin script called htdig.py.
   
4. a new cron activated script and extra crontab entry 
which runs htdig regularly to maintain the per list 
search indices.
   
5. automatic creation, deletion and maintenance of 
htdig configuration files and such. Beyond installing 
htdig and telling Mailman where it is via mm_cfg you 
do not have to do any other setup. Well not quite you 
do have to set up a single per installation symlink to 
allow htdig to find the automatically generated per 
list htdig configuration files.

You probably want to run this patch as follows:

cd <mailman 2.0.6 untarred and unzipped directory>
patch -p1 < <this patch file>  

----------------------------------------------------------------------

>Comment By: Richard Barrett (ppsys)
Date: 2002-10-14 11:50

Message:
Logged In: YES 
user_id=75166

htdig-2.1b3-0.3.patch removes use of the file() function, used 
instead of the open() function, in three cron scripts added by 
the patch. Use of the file() function created an unnecessary 
dependency on Python 2.2

----------------------------------------------------------------------

Comment By: Colin Mackinlay (cmackinlay)
Date: 2002-10-12 16:51

Message:
Logged In: YES 
user_id=624179

Got a workaround!

The line referred to in the traceback:
 file(rundig_run_file, 'w').close()
is used to create a 'rundig_last_run' file of lenght 0 bytes
Creating this manually (or copying it) means the line isn't 
called and everything seems to work.

Either file() is not a valid function call or my python is broken - 
I'm not literate enough in python to know the answer though! 

----------------------------------------------------------------------

Comment By: Colin Mackinlay (cmackinlay)
Date: 2002-10-06 14:18

Message:
Logged In: YES 
user_id=624179

Just rebuilt MM as 2.1b3 with htdig.
Upgraded lists which had htdig before work fine
New lists give the obvious error:
  Unable to read word database file
  Did you run htmerge?
Running the cronjob doesn't fix as it used to, message is:
  Output from command /usr/bin/python -
S /usr/local/mailman/cron/nightly_htdig ..

Traceback (most recent call last):
  File "/usr/local/mailman/cron/nightly_htdig", line 153, in ?
    main()
  File "/usr/local/mailman/cron/nightly_htdig", line 118, in main
    file(rundig_run_file, 'w').close()
NameError: global name 'file' is not defined

The archive/htdig folder only contains the xx.conf file, but no 
db.xx files

If I copy in db.xx files from another list then the problem goes 
away (except I've now got an invalid set of references!)

Is this my elementary error or is it more sinister?!


----------------------------------------------------------------------

Comment By: Richard Barrett (ppsys)
Date: 2002-08-15 11:02

Message:
Logged In: YES 
user_id=75166

htdig-2.1b3-0.2.patch corrects a dumb syntax error in htdig-
2.1b3-0.1.patch which will typically show up as logged errors 
in the operation of the ArchRunner qrunner at line 721 of 
HyperArch.py

----------------------------------------------------------------------

Comment By: Richard Barrett (ppsys)
Date: 2002-08-14 16:51

Message:
Logged In: YES 
user_id=75166

htdig-2.1b3-0.1.patch is a revised version of the patch that is 
compatible with MM 2.1b3

----------------------------------------------------------------------

Comment By: Barry A. Warsaw (bwarsaw)
Date: 2002-08-08 16:33

Message:
Logged In: YES 
user_id=12800

I've sent Richard some comments off-line about this patch.

Meta comments: the 2.0.x patches can't be officially
supported, but I'm going to create an unofficial patches
page off the wiki for where the 2.0 patches can be migrated.

I think this patch set is too big for MM2.1, but if it's
cleaned up as per my private message, let's re-evaluate it
for MM2.2 (or 3.0).

----------------------------------------------------------------------

Comment By: Richard Barrett (ppsys)
Date: 2002-08-05 10:11

Message:
Logged In: YES 
user_id=75166

htdig-2.0.13-0.2.patch just adds a GPL notice to the patch


----------------------------------------------------------------------

Comment By: Richard Barrett (ppsys)
Date: 2002-08-01 16:35

Message:
Logged In: YES 
user_id=75166

htdig-2.1b2-0.1.patch is a revised version of the patch 
that is compatible with MM 2.1b2

----------------------------------------------------------------------

Comment By: Richard Barrett (ppsys)
Date: 2002-07-30 11:25

Message:
Logged In: YES 
user_id=75166

htdig-2.0.13-0.1.patch is purely cosmetic to get no mumble 
application to MM 2.0.13

----------------------------------------------------------------------

Comment By: Richard Barrett (ppsys)
Date: 2002-07-25 15:07

Message:
Logged In: YES 
user_id=75166

Do not use htdig-2.0.12-0.1.patch there is an error in it.
Use htdig-2.0.12-0.2.patch instead

----------------------------------------------------------------------

Comment By: Richard Barrett (ppsys)
Date: 2002-07-25 14:10

Message:
Logged In: YES 
user_id=75166

htdig-2.0.12-0.1.patch is a revised version of the patch that 
applies without complaint to MM 2.0.12.

It also add a facility for adding site wide htdig configuration 
attributes to all list specific htdig configuration files. 



----------------------------------------------------------------------

Comment By: Richard Barrett (ppsys)
Date: 2002-05-23 09:59

Message:
Logged In: YES 
user_id=75166

htdig-2.0.11-0.1.patch is a revised version of the patch that 
is compatible with MM 2.0.11

This version removes an incompatibility with Python 2.2 
which caused warning messages to be generated when any 
of the family cron/nightly_htdig scripts were run.

Some guidance on file access permissions for some htdig 
database files needed by rundig have been added to 
installation notes.

----------------------------------------------------------------------

Comment By: Richard Barrett (ppsys)
Date: 2002-04-19 10:59

Message:
Logged In: YES 
user_id=75166

htdig-2.0.10-0.1.patch is a revised version of the patch 
that is compatible with MM 2.0.10

----------------------------------------------------------------------

Comment By: Richard Barrett (ppsys)
Date: 2002-04-08 17:46

Message:
Logged In: YES 
user_id=75166

htdig-2.0.9-0.1.patch is a revised version of the patch 
that is compatible with MM 2.0.9

----------------------------------------------------------------------

Comment By: Richard Barrett (ppsys)
Date: 2002-03-06 16:22

Message:
Logged In: YES 
user_id=75166

htdig-2.1cvs-20020306.patch is a revised version of the patch that is compatible with the code published in 
mailman CVS on sourceforge as 12:30 GMT 6 Mar 2002

Known deficiency is that the non-English versions of files under $build/templates still contain text in English 
and need translations I cannot do. Also the necessary pygettext activity and subsequent translations in 
files under $build/messages remain to be done.

----------------------------------------------------------------------

Comment By: Richard Barrett (ppsys)
Date: 2001-12-17 16:56

Message:
Logged In: YES 
user_id=75166

htdig-2.1cvs-20011217.patch is a revised version of the 
patch that is compatible with the code published in mailman 
CVS on sourceforge as 11:50 GMT 17 Dec 2001

The only known deficiency is that the non-English versions 
of files under $build/templates still contain text in 
English and need translations I cannot do. Also the 
necessary pygettext activity and subsequent translations in 
files under $build/messages remain to be done.

----------------------------------------------------------------------

Comment By: Richard Barrett (ppsys)
Date: 2001-12-13 16:58

Message:
Logged In: YES 
user_id=75166

htdig-2.1a3-0.1.patch is a revised version of the patch that is compatible with the code published in 
mailman-2.1a3.tgz on sourceforge.

The only known deficiency is that the non-English versions of files under $build/templates still contain text 
in English and need translations I cannot do. Also the necessary pygettext activity and subsequent 
translations in files under $build/messages remain to be done. 

----------------------------------------------------------------------

Comment By: Richard Barrett (ppsys)
Date: 2001-11-28 17:33

Message:
Logged In: YES 
user_id=75166

The htdig-2.0.8-0.1.patch version of the patch resolves a problem that can arise with htdig indexing if the 
web_page_url for a list uses other than the http addressing (some folks want to use https). While specified 
as for MM 2.0.8 the revised patch should work OK with 2.0.7, 2.0.6 and probably back as far as 2.0.3. If 
you do not have the requirement for using other than http addressing in you lists web_page_urls it probably 
isn't worth the trouble of upgrading to this patch level.

----------------------------------------------------------------------

Comment By: Richard Barrett (ppsys)
Date: 2001-11-28 11:08

Message:
Logged In: YES 
user_id=75166

This patch should also apply without problems to MM 2.0.8

----------------------------------------------------------------------

Comment By: Richard Barrett (ppsys)
Date: 2001-11-27 12:00

Message:
Logged In: YES 
user_id=75166

This patch should also apply without problems to Mm 2.0.7

----------------------------------------------------------------------

Comment By: Richard Barrett (ppsys)
Date: 2001-11-09 11:54

Message:
Logged In: YES 
user_id=75166

The htdig-2.0.6-03.patch version of the patch makes some 
previously hard-coded things configurable and enhances the 
capability to run the htdig searches and indexing on a 
different machine to the one delivering Mailman and 
Mailman's web UI.

----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=300103&aid=444884&group_id=103

From mjm@michaelmeltzer.com  Mon Oct 14 17:00:36 2002
From: mjm@michaelmeltzer.com (Michael Meltzer)
Date: Mon, 14 Oct 2002 12:00:36 -0400
Subject: [Mailman-Developers] more 2.1b3++ MIME funnies.
References: <B27BFB52-DEE3-11D6-99E5-0003934516A8@plaidworks.com>
Message-ID: <00b101c2739a$d5744580$34f820c0@ix1x1000>

thier was a fix in Mimedel.py about 1 week ago with multipart/alternative that soulds pretty close to this bug,
How old is the CVS update you are using?

MJM
----- Original Message ----- 
From: "Chuq Von Rospach" <chuqui@plaidworks.com>
To: <mailman-developers@python.org>
Sent: Sunday, October 13, 2002 3:40 PM
Subject: [Mailman-Developers] more 2.1b3++ MIME funnies.


> 
> Okay, is this me, or mailman?
> 
> I have content filtering on. My blacklist is empty. my whitelist is:
> 
> multipart/alternative
> text/plain
> text/html
> 
> I have convert text/html to plain text on.
> 
> My lynx setup is:
> 
> > # Command that is used to convert text/html parts into plain text.  
> > This
> > # should output results to standard output.  %(filename)s will contain 
> > the
> > # name of the temporary file that the program should operate on.
> > HTML_TO_PLAIN_TEXT_COMMAND = '/usr/local/bin/lynx -dump %(filename)s'
> >
> > plaidworks.com 119# lynx -version
> > Lynx Version 2.8.4rel.1 (17 Jul 2001)
> > Built on darwin6.0 Aug 28 2002 22:10:30
> >
> > Copyrights held by the University of Kansas, CERN, and other 
> > contributors.
> > Distributed under the GNU General Public License.
> > See http://lynx.browser.org/ and the online help for more information.
> 
> I'm finding that if someone sends a M/A message with a plain and an 
> html part, it's getting processed, and I'm getting a message going out 
> with two t/p parts, one the original, and one empty. Since the empty 
> one is second, it gets displayed.
> 
> So something is whacked out. I'm not sure what.
> 
> What are people doing for their black/whitelists? am I just building 
> this thing sideways?
> 
> What I'd like to be able to do (but can't) is allow M/A, t/plain and 
> t/html, and use the lynx to defang HTML going into the text (but not 
> MIME) digests. That seems the most reasonable set of compromises to 
> allow styled mail. Unfortunately, mailman can't be set up that way...
> 
> so my backup plan was to allow text/plain, convert html to text if the 
> text part didn't exist, and try to add features later. But I don't seem 
> to be quite right yet. For what I'm trying, I *think* I should be 
> whitelisting M/A and Text/Plain, but not text/HTMl. that screws people 
> who only send text/html without a plain part, but they probably 
> shouldn't do that anyway...
> 
> well, off to tweak the settings again....
> 
> 
> 
> 
> 
> -- 
> Chuq Von Rospach, Architech
> chuqui@plaidworks.com -- http://www.plaidworks.com/chuqui/blog/
> 
> The first rule of holes: If you are in one, stop digging.
> 
> 
> _______________________________________________
> Mailman-Developers mailing list
> Mailman-Developers@python.org
> http://mail.python.org/mailman-21/listinfo/mailman-developers

From chuqui@plaidworks.com  Mon Oct 14 17:16:24 2002
From: chuqui@plaidworks.com (Chuq Von Rospach)
Date: Mon, 14 Oct 2002 09:16:24 -0700
Subject: [Mailman-Developers] more 2.1b3++ MIME funnies.
In-Reply-To: <00b101c2739a$d5744580$34f820c0@ix1x1000>
Message-ID: <490685F6-DF90-11D6-A426-0003934516A8@plaidworks.com>


On Monday, October 14, 2002, at 09:00  AM, Michael Meltzer wrote:

> thier was a fix in Mimedel.py about 1 week ago with 
> multipart/alternative that soulds pretty close to this bug,
> How old is the CVS update you are using?
>

8/12 -- saturday night...


-- 
Chuq Von Rospach, Architech
chuqui@plaidworks.com -- http://www.plaidworks.com/chuqui/blog/

The Cliff's Notes Cliff's Notes on Hamlet:
     And they all died happily ever after


-- 
Chuq Von Rospach, Architech
chuqui@plaidworks.com -- http://www.plaidworks.com/chuqui/blog/

Very funny, Scotty. Now beam my clothes down here, will you?


From chuqui@plaidworks.com  Mon Oct 14 17:16:24 2002
From: chuqui@plaidworks.com (Chuq Von Rospach)
Date: Mon, 14 Oct 2002 09:16:24 -0700
Subject: [Mailman-Developers] more 2.1b3++ MIME funnies.
In-Reply-To: <00b101c2739a$d5744580$34f820c0@ix1x1000>
Message-ID: <490685F6-DF90-11D6-A426-0003934516A8@plaidworks.com>


On Monday, October 14, 2002, at 09:00  AM, Michael Meltzer wrote:

> thier was a fix in Mimedel.py about 1 week ago with 
> multipart/alternative that soulds pretty close to this bug,
> How old is the CVS update you are using?
>

8/12 -- saturday night...


-- 
Chuq Von Rospach, Architech
chuqui@plaidworks.com -- http://www.plaidworks.com/chuqui/blog/

The Cliff's Notes Cliff's Notes on Hamlet:
     And they all died happily ever after


-- 
Chuq Von Rospach, Architech
chuqui@plaidworks.com -- http://www.plaidworks.com/chuqui/blog/

Very funny, Scotty. Now beam my clothes down here, will you?


From mjm@michaelmeltzer.com  Mon Oct 14 17:25:55 2002
From: mjm@michaelmeltzer.com (Michael Meltzer)
Date: Mon, 14 Oct 2002 12:25:55 -0400
Subject: [Mailman-Developers] more 2.1b3++ MIME funnies.
References: <490685F6-DF90-11D6-A426-0003934516A8@plaidworks.com>
Message-ID: <00ed01c2739e$5eff6ca0$34f820c0@ix1x1000>

unless that was a typo, that 2 months old, you might want to grab the latest and greatest.

The black list I use:
application
audio
video
model
image

White list:
text/plain
text/richtext
text/enriched
text/html
text/rfc822-headers
multipart/mixed
multipart/alternative
multipart/digest
multipart/parallel
multipart/related
multipart/report
multipart/message
multipart/signed
message/rfc822
message/partial
message/http


I saw you are playing around with images, I have a patch that will turn them into links on the fly. I have to line it up to the
latest but if you want it you got it.
Also I getting pretty good with that module, if the CVS does not do it, if you have a reproducible message I run it thought.

MJM

----- Original Message -----
From: "Chuq Von Rospach" <chuqui@plaidworks.com>
To: "Michael Meltzer" <mjm@michaelmeltzer.com>
Cc: <mailman-developers@python.org>
Sent: Monday, October 14, 2002 12:16 PM
Subject: Re: [Mailman-Developers] more 2.1b3++ MIME funnies.


>
> On Monday, October 14, 2002, at 09:00  AM, Michael Meltzer wrote:
>
> > thier was a fix in Mimedel.py about 1 week ago with
> > multipart/alternative that soulds pretty close to this bug,
> > How old is the CVS update you are using?
> >
>
> 8/12 -- saturday night...
>
>
> --
> Chuq Von Rospach, Architech
> chuqui@plaidworks.com -- http://www.plaidworks.com/chuqui/blog/
>
> The Cliff's Notes Cliff's Notes on Hamlet:
>      And they all died happily ever after
>
>
> --
> Chuq Von Rospach, Architech
> chuqui@plaidworks.com -- http://www.plaidworks.com/chuqui/blog/
>
> Very funny, Scotty. Now beam my clothes down here, will you?
>


From chuqui@plaidworks.com  Mon Oct 14 17:41:47 2002
From: chuqui@plaidworks.com (Chuq Von Rospach)
Date: Mon, 14 Oct 2002 09:41:47 -0700
Subject: [Mailman-Developers] more 2.1b3++ MIME funnies.
In-Reply-To: <00ed01c2739e$5eff6ca0$34f820c0@ix1x1000>
Message-ID: <D4808798-DF93-11D6-A426-0003934516A8@plaidworks.com>


On Monday, October 14, 2002, at 09:25  AM, Michael Meltzer wrote:

> unless that was a typo, that 2 months old, you might want to grab the 
> latest and greatest.

god, yes. I just wish it was a typo. I updated saturday night, then 
installed the update and started tweaking...

> The black list I use:
>
great. Thanks!

>
> I saw you are playing around with images,

I've turned that off for now. It (and html) cause too many problems to 
text digests, and that's more important right now. I'll whack on that 
later.

=--
Chuq Von Rospach, Architech
chuqui@plaidworks.com -- http://www.plaidworks.com/chuqui/blog/

Stress is when you wake up screaming and you realize you haven't fallen 
asleep yet.


From barry@python.org  Mon Oct 14 22:25:46 2002
From: barry@python.org (Barry A. Warsaw)
Date: Mon, 14 Oct 2002 17:25:46 -0400
Subject: [Mailman-Developers] more 2.1b3++ MIME funnies.
References: <B27BFB52-DEE3-11D6-99E5-0003934516A8@plaidworks.com>
Message-ID: <15787.13915.661937.113786@gargle.gargle.HOWL>


>>>>> "CVR" == Chuq Von Rospach <chuqui@plaidworks.com> writes:

    CVR> I'm finding that if someone sends a M/A message with a plain
    CVR> and an html part, it's getting processed, and I'm getting a
    CVR> message going out with two t/p parts, one the original, and
    CVR> one empty. Since the empty one is second, it gets displayed.

    CVR> So something is whacked out. I'm not sure what.

Odd.  I just tried this w/cvs.  No blacklist, whitelist as you
described, then I sent a m/a with a text/plain first and text/html
second.  In this configuration, I got a message with just the
text/plain part, as I'd expect.  Switch the two and I get the lynx'd
text/html part as expected.

Remember that one content filtering rule is that after filter/pass but
before html conversion, multipart/alternative parts are replaced with
their first non-empty subpart.

    CVR> What I'd like to be able to do (but can't) is allow M/A,
    CVR> t/plain and t/html, and use the lynx to defang HTML going
    CVR> into the text (but not MIME) digests. That seems the most
    CVR> reasonable set of compromises to allow styled
    CVR> mail. Unfortunately, mailman can't be set up that way...

Nope, because messages are processed for the two digests in the same
fell swoop.  I don't think it's ever come up before that you might
want different processing for the different digests, but that /does/
make sense.  ("Oh, TODO, please come here for a moment")

    CVR> so my backup plan was to allow text/plain, convert html to
    CVR> text if the text part didn't exist, and try to add features
    CVR> later. But I don't seem to be quite right yet. For what I'm
    CVR> trying, I *think* I should be whitelisting M/A and
    CVR> Text/Plain, but not text/HTMl. that screws people who only
    CVR> send text/html without a plain part, but they probably
    CVR> shouldn't do that anyway...

That's the setting that I think most people were talking about when
content filtering was added.

I think things are working properly, although the feature may not be
ideally useful yet.  I knew this one would be tough to make flexible
enough.

-Barry

From barry@python.org  Mon Oct 14 23:28:19 2002
From: barry@python.org (Barry A. Warsaw)
Date: Mon, 14 Oct 2002 18:28:19 -0400
Subject: [Mailman-Developers] Re: [Mailman-checkins] mailman/templates/ru
	archidxentry.html,NONE,2.0 archidxfoot.html,NONE,2.0
	archidxhead.html,NONE,2.0
	archlistend.html,NONE,2.0 archliststart.html,NONE,2.0
	archtocentry.html,NONE,2.0 archtoc.html,NONE,2.0
References: <E17zl3s-0004Bz-00@usw-pr-cvs1.sourceforge.net>
	<20021010221012.GA12092@mawhrin.net>
	<15784.62071.201635.708438@gargle.gargle.HOWL>
	<20021013193338.GA24768@mawhrin.net>
Message-ID: <15787.17667.206549.341275@gargle.gargle.HOWL>


>>>>> "MS" == Mikhail Sobolev <mss@mawhrin.net> writes:

    MS> It's not a _problem_.  It's just a feeling about a "right
    MS> thing". :(

    MS> What I feel is not right that the resulting _file_ is
    MS> assembled from other _files_.  I believe, this is not how the
    MS> rest of Mailman is done: the "template" file _is_ the
    MS> resulting file beside some substitions, which are necessary to
    MS> fill the some fields or other parts.

    MS> Again, it's not a problem.

Ah, I see what you're getting at.  Thanks.

I think we'll end up leaving it this way for MM2.1.

-Barry

From barry@python.org  Mon Oct 14 23:58:04 2002
From: barry@python.org (Barry A. Warsaw)
Date: Mon, 14 Oct 2002 18:58:04 -0400
Subject: [Mailman-Developers] MimeDel i18n message undelivered
References: <3DA90D0E.2070008@is.kochi-u.ac.jp>
Message-ID: <15787.19452.209740.892883@gargle.gargle.HOWL>


>>>>> "TK" == Tokio Kikuchi <tkikuchi@is.kochi-u.ac.jp> writes:

    TK> While testing MimeDel with i18n Japanese configuration,
    TK> message delivery error occured when forwarding the filetered
    TK> message to the list administrator. Here is a quick patch:

Thanks, applied.
-Barry

From barry@python.org  Tue Oct 15 00:56:47 2002
From: barry@python.org (Barry A. Warsaw)
Date: Mon, 14 Oct 2002 19:56:47 -0400
Subject: [Mailman-Developers] Overhauled 'Exim and Mailman' doc
References: <20021012014433.GA29438@cthulhu.gerg.ca>
Message-ID: <15787.22975.845040.875953@gargle.gargle.HOWL>


>>>>> "GW" == Greg Ward <gward@mems-exchange.org> writes:

    GW> I've just spent the last couple of hours beating the
    GW> README.EXIM document from Mailman's current CVS tree into
    GW> shape.  (This is derived from Nigel Metheringham's "HOWTO -
    GW> Using exim and Mailman together" document, which can be found
    GW> at www.exim.org.)

Good work Greg, thanks.

Speeaking of Mailman's README.EXIM, do we want to continue to
distribute a copy of the exim.org doc, or should we zap that and just
include a pointer to the canonical location on the web?  Which, I'm
assuming, will be http://www.exim.org;/howto/mailman.html and will
eventually be updated to include Greg's description.

-Barry

From barry@python.org  Tue Oct 15 00:58:14 2002
From: barry@python.org (Barry A. Warsaw)
Date: Mon, 14 Oct 2002 19:58:14 -0400
Subject: [Mailman-Developers] Overhauled 'Exim and Mailman' doc 
References: <20021012014433.GA29438@cthulhu.gerg.ca>
	<4162.1034403272@kanga.nu>
Message-ID: <15787.23062.767185.619766@gargle.gargle.HOWL>


>>>>> "JCL" == J C Lawrence <claw@kanga.nu> writes:

    JCL> Would you like to have the Exim/TMDA/Mailman HOW-TO as a
    JCL> rider on that, or should we tuck it in as a separate README
    JCL> (Barry?).

Is the document Exim specific, or can you write it to be somewhat
agnostic towards the mta?  If so, I'd be happy to add a README.TMDA,
or point to any documentation in the TMDA web pages.

-Barry

From marc_news@merlins.org  Tue Oct 15 01:05:55 2002
From: marc_news@merlins.org (Marc MERLIN)
Date: Mon, 14 Oct 2002 17:05:55 -0700
Subject: [Mailman-Developers] Overhauled 'Exim and Mailman' doc
In-Reply-To: <15787.22975.845040.875953@gargle.gargle.HOWL>
References: <20021012014433.GA29438@cthulhu.gerg.ca>
	<15787.22975.845040.875953@gargle.gargle.HOWL>
Message-ID: <20021015000555.GG2740@merlins.org>

On Mon, Oct 14, 2002 at 07:56:47PM -0400, Barry A. Warsaw wrote:
> Speeaking of Mailman's README.EXIM, do we want to continue to
> distribute a copy of the exim.org doc, or should we zap that and just
> include a pointer to the canonical location on the web?  Which, I'm
> assuming, will be http://www.exim.org;/howto/mailman.html and will
> eventually be updated to include Greg's description.

That'd be nice, but I think Nigel is a bit too busy, I sent him my new
version too, but he never had the time to post it on exim.org

Marc
-- 
"A mouse is a device used to point at the xterm you want to type in" - A.S.R.
Microsoft is to operating systems & security ....
                                      .... what McDonalds is to gourmet cooking
Home page: http://marc.merlins.org/   |   Finger marc_f@merlins.org for PGP key

From barry@python.org  Tue Oct 15 01:05:49 2002
From: barry@python.org (Barry A. Warsaw)
Date: Mon, 14 Oct 2002 20:05:49 -0400
Subject: [Mailman-Developers] HTML and text digests...
References: <969FACEA-DE5E-11D6-99E5-0003934516A8@plaidworks.com>
Message-ID: <15787.23517.239710.352242@gargle.gargle.HOWL>


>>>>> "CVR" == Chuq Von Rospach <chuqui@plaidworks.com> writes:

    CVR> But it seems to me that one way to allow that but protect the
    CVR> text digests was to use that lynx filtering, but only on the
    CVR> content sent to the text digest. Yes, I know, another
    CVR> @#$@%@#$ configuration option, but since the technology
    CVR> already exists....

But the architecture really doesn't, unfortunately.  Here's how
digests work:

For every message that makes it through the moderate-and-munge phase,
we simply append the message onto a Unix mailbox style folder.  Then,
when the cron job that actually sends out the digests fires up, it
slogs through that mbox and appends each message to two digests that
it's building up, one MIME and one plaintext.  The important thing is
that both digets are built at the same time, from the same source
mbox.  Makes for convenient and efficient code, but not for processing
the individual messages differently for each digests.

-Barry

From barry@python.org  Tue Oct 15 01:11:07 2002
From: barry@python.org (Barry A. Warsaw)
Date: Mon, 14 Oct 2002 20:11:07 -0400
Subject: [Mailman-Developers] updated to CVS, and stuff broke....
References: <E360685D-DE5D-11D6-99E5-0003934516A8@plaidworks.com>
Message-ID: <15787.23835.291844.48043@gargle.gargle.HOWL>


>>>>> "CVR" == Chuq Von Rospach <chuqui@plaidworks.com> writes:

    CVR> Looks like a couple of Defaults.py variables got added but
    CVR> weren't added to Defaults.py on the update.

You know that whenever you cvs up, you should always run
"./config.status" to update the generated files.  When new global
variables are added, they're always added to Defaults.py.in, and it's
only the re-running of configure (or config.status to retain your
previous options) that creates the Defaults.py file that Mailman
expects.

Another thing you always want to do when upgrading is to run
"bin/mailmanctl restart" after doing a "make install"

    CVR> Once I turn it on through the admin page, it stops working
    CVR> with:

    | plaidworks.com 110# Oct 12 20:36:07 2002 (21778) Uncaught runner 
    | exception: 'NoneType' object has no attribute 'encode'
    | Oct 12 20:36:07 2002 (21778) Traceback (most recent call last):
    |    File "/usr/local/mailman/Mailman/Queue/Runner.py", line 105, in 
    | _oneloop
    |      self._onefile(msg, msgdata)
    |    File "/usr/local/mailman/Mailman/Queue/Runner.py", line 154, in 
    | _onefile
    |      keepqueued = self._dispose(mlist, msg, msgdata)
    |    File "/usr/local/mailman/Mailman/Queue/OutgoingRunner.py", line 61, 
    | in _dispose
    |      self._func(mlist, msg, msgdata)
    |    File "/usr/local/mailman/Mailman/Handlers/SMTPDirect.py", line 135, 
    | in process
    |      msgdata['recips'] = chunk
    |    File "/usr/local/mailman/Mailman/Handlers/SMTPDirect.py", line 287, 
    | in verpdeliver
    |      # name is a byte string, convert it first to Unicode, given
    | AttributeError: 'NoneType' object has no attribute 'encode'

    CVR> if I turn that back off, the lsit starts working again.

This traceback is nonsensical.  First, it looks like line 287 is
pointing into a comment, and that should never happen.  So somehow
your source code is out of sync with what your daemon is running.
Do a cvs up, then follow the recipe above and see if that fixes your
problems.

-Barry

From barry@python.org  Tue Oct 15 01:18:28 2002
From: barry@python.org (Barry A. Warsaw)
Date: Mon, 14 Oct 2002 20:18:28 -0400
Subject: [Mailman-Developers] Overhauled 'Exim and Mailman' doc
References: <20021012014433.GA29438@cthulhu.gerg.ca>
	<15787.22975.845040.875953@gargle.gargle.HOWL>
	<20021015000555.GG2740@merlins.org>
Message-ID: <15787.24276.660576.335006@gargle.gargle.HOWL>


>>>>> "MM" == Marc MERLIN <marc_news@merlins.org> writes:

    MM> That'd be nice, but I think Nigel is a bit too busy, I sent
    MM> him my new version too, but he never had the time to post it
    MM> on exim.org

Ok.  Can you and Greg get together and agree on a new version for
Mailman's README.EXIM?  I'll install it in Mailman's cvs and then
Nigel will be free to update exim.org when he gets a chance.

-Barry

From gward@python.net  Sat Oct 12 19:47:10 2002
From: gward@python.net (Greg Ward)
Date: Sat, 12 Oct 2002 14:47:10 -0400
Subject: [Mailman-Developers] Re: [Exim] Overhauled 'Exim and Mailman' doc
In-Reply-To: <20021012024456.GA7671@merlins.org>
References: <20021012014433.GA29438@cthulhu.gerg.ca>
	<20021012024456.GA7671@merlins.org>
Message-ID: <20021012184710.GB30217@cthulhu.gerg.ca>

On 11 October 2002, Marc MERLIN said:
> I've been meaning to add this:

Hmmm, good point.  Hadn't thought about it because SMTP callbacks seem a
little *too* powerful to me.

I think it's more important to add something about disabling recipient
verification for 127.0.0.1 first, but I wanted to get this text out for
review first.

        Greg
-- 
Greg Ward <gward@python.net>                         http://www.gerg.ca/
I am deeply CONCERNED and I want something GOOD for BREAKFAST!

From marc_news@merlins.org  Tue Oct 15 01:27:14 2002
From: marc_news@merlins.org (Marc MERLIN)
Date: Mon, 14 Oct 2002 17:27:14 -0700
Subject: [Mailman-Developers] Overhauled 'Exim and Mailman' doc
In-Reply-To: <15787.24276.660576.335006@gargle.gargle.HOWL>
References: <20021012014433.GA29438@cthulhu.gerg.ca>
	<15787.22975.845040.875953@gargle.gargle.HOWL>
	<20021015000555.GG2740@merlins.org>
	<15787.24276.660576.335006@gargle.gargle.HOWL>
Message-ID: <20021015002714.GI2740@merlins.org>

On Mon, Oct 14, 2002 at 08:18:28PM -0400, Barry A. Warsaw wrote:
> 
> >>>>> "MM" == Marc MERLIN <marc_news@merlins.org> writes:
> 
>     MM> That'd be nice, but I think Nigel is a bit too busy, I sent
>     MM> him my new version too, but he never had the time to post it
>     MM> on exim.org
> 
> Ok.  Can you and Greg get together and agree on a new version for
> Mailman's README.EXIM?  I'll install it in Mailman's cvs and then
> Nigel will be free to update exim.org when he gets a chance.

I took the document from Nigel, modified it, and gave you my version for CVS
(and gave it to Nigel too)
Greg took my version from your CVS and further updated it, so he has the up
to date version, so feel free to take whatever he gives you.
(I gave him a few updates, but he's the maintainer of the day right now :-)

Marc
-- 
"A mouse is a device used to point at the xterm you want to type in" - A.S.R.
Microsoft is to operating systems & security ....
                                      .... what McDonalds is to gourmet cooking
Home page: http://marc.merlins.org/   |   Finger marc_f@merlins.org for PGP key

From barry@python.org  Tue Oct 15 01:27:40 2002
From: barry@python.org (Barry A. Warsaw)
Date: Mon, 14 Oct 2002 20:27:40 -0400
Subject: [Mailman-Developers] problem with list member's vacation script
References: <E180C15-0001bz-00@real.local.evenreal.com>
Message-ID: <15787.24828.611951.99407@gargle.gargle.HOWL>


>>>>> "GM" == Gustavo M <linux_news@nextphere.com> writes:

    GM> I manage a mailing list site with mailman, and I recently had
    GM> the following problem:

    GM> Lets say I have a list called test@site.org with 3 members:
    GM> m1@foo.org, m2@foo.org and m3@foo.org. m3 has its vacation
    GM> script turned on. m1 sends an email to test@site.org, so m2,m3
    GM> and m1 itself receive the email. m3's vacation script replies
    GM> to m1 and to test@site.org, so m1 gets 2 copies of the reply
    GM> (one directly and one through test@site.org), m2 gets a copy
    GM> and m1 gets a copy of its own reply which in turn will be
    GM> replied by its vacation script and so on. So my mail server
    GM> ends up sending a lot of useless emails.

And later...
    
    GM> How can I stop this situation? I know I can reject or hold
    GM> emails based on some header pattern matching, but I cannot do
    GM> this on a case by case basis, not all vacation scripts behave
    GM> the same,

And not all are written correctly, sadly.

Vacation programs should ignore all messages with a Precendence
header, with values bulk, list, or junk.  Correctly written vacation
programs will never auto-respond to Mailman messages because Mailman
always send its copies out with such a header.

Vacation programs should never respond to the list address either, nor
should they respond to the same address more than once in a certain
period of time, but broken vacation programs tend to be broken in
multiple ways.

There's no foolproof way to stop a broken vacation program short of
manually unsub'ing the user and sending them a nastygram.  I'm all in
favor of that approach.

That having been said, MM2.1 has some defenses against broken
replybots in general, of which vacation is just a common offender.
However, these defenses help primarily with avoiding reply storms for
Mailman auto-replies.

    GM> so this is not a solution for my problem.  I was thinking
    GM> maybe there's a way to configure mailman so that if m1 sends
    GM> an email to test@site.org mailman will not send a copy of the
    GM> email back to m1 just because m1 is a list member.

While this has nothing to do with broken vacation programs, users can
configure their subscriptions so that they do not receive copies of
their own postings through the list.

    GM> Also, is there a way to limit the number of postings each
    GM> member is allowed to sent per day?

Not currently.  It's on the TODO list for some post-2.1 release.

-Barry

From barry@python.org  Tue Oct 15 01:29:55 2002
From: barry@python.org (Barry A. Warsaw)
Date: Mon, 14 Oct 2002 20:29:55 -0400
Subject: [Mailman-Developers] Overhauled 'Exim and Mailman' doc
References: <20021012014433.GA29438@cthulhu.gerg.ca>
	<15787.22975.845040.875953@gargle.gargle.HOWL>
	<20021015000555.GG2740@merlins.org>
	<15787.24276.660576.335006@gargle.gargle.HOWL>
	<20021015002714.GI2740@merlins.org>
Message-ID: <15787.24963.347658.389683@gargle.gargle.HOWL>


>>>>> "MM" == Marc MERLIN <marc_news@merlins.org> writes:

    MM> I took the document from Nigel, modified it, and gave you my
    MM> version for CVS (and gave it to Nigel too) Greg took my
    MM> version from your CVS and further updated it, so he has the up
    MM> to date version, so feel free to take whatever he gives you.
    MM> (I gave him a few updates, but he's the maintainer of the day
    MM> right now :-)

Cool, thanks.  It's hard to keep track of everything.  ;}

Greg, whenever you're ready, send me an update!
-Barry

From chuqui@plaidworks.com  Tue Oct 15 02:48:16 2002
From: chuqui@plaidworks.com (Chuq Von Rospach)
Date: Mon, 14 Oct 2002 18:48:16 -0700
Subject: [Mailman-Developers] updated to CVS, and stuff broke....
In-Reply-To: <15787.23835.291844.48043@gargle.gargle.HOWL>
Message-ID: <2C268580-DFE0-11D6-8345-0003934516A8@plaidworks.com>


On Monday, October 14, 2002, at 05:11  PM, Barry A. Warsaw wrote:

> You know that whenever you cvs up, you should always run
> "./config.status" to update the generated files.

hmm. I thought I had. But you're right. I went and looked, and it's 
clear I didn't. oops.

> Another thing you always want to do when upgrading is to run
> "bin/mailmanctl restart" after doing a "make install"
>

tjat I did...

off to update CVS again.


-- 
Chuq Von Rospach, Architech, Apple IS&T E-mail systems
chuq@apple.com


From chuqui@plaidworks.com  Tue Oct 15 02:50:45 2002
From: chuqui@plaidworks.com (Chuq Von Rospach)
Date: Mon, 14 Oct 2002 18:50:45 -0700
Subject: [Mailman-Developers] HTML and text digests...
In-Reply-To: <15787.23517.239710.352242@gargle.gargle.HOWL>
Message-ID: <84F5B1DC-DFE0-11D6-8345-0003934516A8@plaidworks.com>


On Monday, October 14, 2002, at 05:05  PM, Barry A. Warsaw wrote:

> But the architecture really doesn't, unfortunately.

Yeah, I had that feeling. so we'll do without HTML for the time being. 
Honestly, not a huge loss, but I need functional text digests more. 
("you don't always get what you want....")





-- 
Chuq Von Rospach, Architech
chuqui@plaidworks.com -- http://www.plaidworks.com/chuqui/blog/

He doesn't have ulcers, but he's a carrier.


From barry@python.org  Tue Oct 15 02:52:45 2002
From: barry@python.org (Barry A. Warsaw)
Date: Mon, 14 Oct 2002 21:52:45 -0400
Subject: [Mailman-Developers] updated to CVS, and stuff broke....
References: <15787.23835.291844.48043@gargle.gargle.HOWL>
	<2C268580-DFE0-11D6-8345-0003934516A8@plaidworks.com>
Message-ID: <15787.29933.877496.973964@gargle.gargle.HOWL>


>>>>> "CVR" == Chuq Von Rospach <chuqui@plaidworks.com> writes:

    >> You know that whenever you cvs up, you should always run
    >> "./config.status" to update the generated files.

    CVR> hmm. I thought I had. But you're right. I went and looked,
    CVR> and it's clear I didn't. oops.

S'ok, I often forget myself.  Wanna see the bald spot where I scratch
my head too often? ;)

-Barry

From barry@python.org  Tue Oct 15 02:53:37 2002
From: barry@python.org (Barry A. Warsaw)
Date: Mon, 14 Oct 2002 21:53:37 -0400
Subject: [Mailman-Developers] HTML and text digests...
References: <15787.23517.239710.352242@gargle.gargle.HOWL>
	<84F5B1DC-DFE0-11D6-8345-0003934516A8@plaidworks.com>
Message-ID: <15787.29985.299934.233517@gargle.gargle.HOWL>


>>>>> "CVR" == Chuq Von Rospach <chuqui@plaidworks.com> writes:

    CVR> On Monday, October 14, 2002, at 05:05 PM, Barry A. Warsaw
    CVR> wrote:

    >> But the architecture really doesn't, unfortunately.

    CVR> Yeah, I had that feeling. so we'll do without HTML for the
    CVR> time being. Honestly, not a huge loss, but I need functional
    CVR> text digests more. ("you don't always get what you want....")

And if you can't be with the one you love... :)

-Barry

From chuqui@plaidworks.com  Tue Oct 15 02:56:10 2002
From: chuqui@plaidworks.com (Chuq Von Rospach)
Date: Mon, 14 Oct 2002 18:56:10 -0700
Subject: [Mailman-Developers] updated to CVS, and stuff broke....
In-Reply-To: <15787.29933.877496.973964@gargle.gargle.HOWL>
Message-ID: <4716AD46-DFE1-11D6-8345-0003934516A8@plaidworks.com>


On Monday, October 14, 2002, at 06:52  PM, Barry A. Warsaw wrote:

>

> S'ok, I often forget myself.  Wanna see the bald spot where I scratch
> my head too often? ;)
>

looks like mine...

Okay, updated to cvs. ran config.status. make, mailmanctl stop, make 
install, mailmanctl start, and went onto a list and turned on 
personalization.

sent a message, and...

*** error ***
Oct 14 18:54:32 2002 (2884) Uncaught runner exception: coercing to 
Unicode: need string or buffer, NoneType found
Oct 14 18:54:32 2002 (2884) Traceback (most recent call last):
   File "/usr/local/mailman/Mailman/Queue/Runner.py", line 105, in 
_oneloop
     self._onefile(msg, msgdata)
   File "/usr/local/mailman/Mailman/Queue/Runner.py", line 154, in 
_onefile
     keepqueued = self._dispose(mlist, msg, msgdata)
   File "/usr/local/mailman/Mailman/Queue/OutgoingRunner.py", line 61, 
in _dispose
     self._func(mlist, msg, msgdata)
   File "/usr/local/mailman/Mailman/Handlers/SMTPDirect.py", line 137, 
in process
     deliveryfunc(mlist, msg, msgdata, envsender, refused, conn)
   File "/usr/local/mailman/Mailman/Handlers/SMTPDirect.py", line 300, 
in verpdeliver
     'replace')
TypeError: coercing to Unicode: need string or buffer, NoneType found



-- 
Chuq Von Rospach, Architech
chuqui@plaidworks.com -- http://www.plaidworks.com/chuqui/blog/

Yes, I am an agent of Satan, but my duties
are largely ceremonial.


From barry@python.org  Tue Oct 15 02:58:25 2002
From: barry@python.org (Barry A. Warsaw)
Date: Mon, 14 Oct 2002 21:58:25 -0400
Subject: [Mailman-Developers] updated to CVS, and stuff broke....
References: <15787.29933.877496.973964@gargle.gargle.HOWL>
	<4716AD46-DFE1-11D6-8345-0003934516A8@plaidworks.com>
Message-ID: <15787.30273.727728.360263@gargle.gargle.HOWL>


Yikes.  Ok, simple bug.  Let me test a few things and then check in a
patch.  Should be should a few minutes.

-Barry

From barry@python.org  Tue Oct 15 03:10:08 2002
From: barry@python.org (Barry A. Warsaw)
Date: Mon, 14 Oct 2002 22:10:08 -0400
Subject: [Mailman-Developers] updated to CVS, and stuff broke....
References: <15787.29933.877496.973964@gargle.gargle.HOWL>
	<4716AD46-DFE1-11D6-8345-0003934516A8@plaidworks.com>
	<15787.30273.727728.360263@gargle.gargle.HOWL>
Message-ID: <15787.30976.510515.704359@gargle.gargle.HOWL>


>>>>> "BAW" == Barry A Warsaw <barry@python.org> writes:

    BAW> Yikes.  Ok, simple bug.  Let me test a few things and then
    BAW> check in a patch.  Should be should a few minutes.

Try it now.
-Barry

From chuqui@plaidworks.com  Tue Oct 15 03:56:13 2002
From: chuqui@plaidworks.com (Chuq Von Rospach)
Date: Mon, 14 Oct 2002 19:56:13 -0700
Subject: [Mailman-Developers] woo hoo!
Message-ID: <AA881E84-DFE9-11D6-8345-0003934516A8@plaidworks.com>


personalization works! Even better, it let me nail a bounce that was 
being forwarded through bigfoot to a place that was stripping all 
headers.

User `babsym98_thequeen@bigfoot.com' removed from list: ahl.

the bounce was coming back from:

Your message

   To:      babsym98_thequeen@bigfoot.com
   Subject: sorry, a quick test.
   Sent:    Tue, 15 Oct 2002 03:44:01 +0100

did not reach the following recipient(s):

sinclail@kilmarnock.ac.uk on Tue, 15 Oct 2002 03:49:02 +0100
     The recipient name is not recognized
         The MTS-ID of the original message is: c=gb;a= ;p=kilmarnock
colle;l=WEBSERVER02101502494NV0H90M
     MSEXCH:IMS:Kilmarnock College:MAIL:WEBSERVER 0 (000C05A6) Unknown
Recipient

You figure it out. Crap like this is why VERP was invented...

great stuff!

barry, this bug is still here, though:

*** subscribe ***
Oct 14 19:51:02 2002 (6753) ahl: deleted babsym98_thequeen@bigfoot.com

*** smtp ***
Oct 14 19:51:05 2002 (6128) 
<mailman.0.1034650262.6753.ahl@plaidworks.com> smtp for 1 recips, 
completed in 0.355 seconds
Oct 14 19:51:05 2002 (6128) 
<mailman.1.1034650262.6753.ahl@plaidworks.com> smtp for 1 recips, 
completed in 0.304 seconds
Oct 14 19:51:34 2002 (6128) 
<mailman.1.1034650262.6753.ahl@plaidworks.com> smtp for 1 recips, 
completed in 0.286 seconds

*** vette ***
Oct 14 19:51:33 2002 (6126) Precedence: bulk message ignored by: 
ahl-request@plaidworks.com

What's happening, Barry, is one part of Mailman is creating a message 
to let the admin know this guy was removed from the list, and then 
another part of Mailman is seeing the Precedence: Bulk header, and 
deleting it. so it's a message created and deleted within mailman 
without ever leaving the system, and the admin doesn't get notified. 
The message to the user confirming the unsubscription does work, since 
I see the bounce come back in later...



-- 
Chuq Von Rospach, Architech
chuqui@plaidworks.com -- http://www.plaidworks.com/chuqui/blog/

Very funny, Scotty. Now beam my clothes down here, will you?


From chuqui@plaidworks.com  Tue Oct 15 04:10:58 2002
From: chuqui@plaidworks.com (Chuq Von Rospach)
Date: Mon, 14 Oct 2002 20:10:58 -0700
Subject: [Mailman-Developers] hmm. One correction.
Message-ID: <B9E4ACFE-DFEB-11D6-8345-0003934516A8@plaidworks.com>

to my last. It looks like mailman is generating THREE messages to the 
unsubscribe now. One to the subscriber, one to the list admin and one 
-- where? I dunno. Or maybe I'm misreading the logs. but I definitely 
got the one to the list owner on this latest unsubscribe, which I 
didn't previous to this CVS update. but I'll have to see if I can 
figure out where that one deleted message is coming from or going to...


-- 
Chuq Von Rospach, Architech, Apple IS&T E-mail systems
chuq@apple.com


From chuqui@plaidworks.com  Tue Oct 15 04:18:45 2002
From: chuqui@plaidworks.com (Chuq Von Rospach)
Date: Mon, 14 Oct 2002 20:18:45 -0700
Subject: [Mailman-Developers] Even more on that log weirdie...
Message-ID: <D01BB111-DFEC-11D6-8345-0003934516A8@plaidworks.com>

Okay, here we go. I added an address to one of my lists, and then 
deleted it again. Here are the log entries.

plaidworks.com 99# remove_members --fromall xxx@chuqui.com
User `xxx@chuqui.com' removed from list: sharks.
plaidworks.com 100#
*** subscribe ***
Oct 14 20:16:13 2002 (7041) sharks: deleted xxx@chuqui.com

*** smtp ***
Oct 14 20:16:22 2002 (6128) 
<mailman.0.1034651773.7041.sharks@plaidworks.com> smtp for 1 recips, 
completed in 8.287 seconds
Oct 14 20:16:28 2002 (6128) 
<mailman.1.1034651773.7041.sharks@plaidworks.com> smtp for 1 recips, 
completed in 5.281 seconds

*** vette ***
Oct 14 20:16:35 2002 (6126) Precedence: bulk message ignored by: 
sharks-request@plaidworks.com

*** smtp ***
Oct 14 20:16:36 2002 (6128) 
<mailman.1.1034651773.7041.sharks@plaidworks.com> smtp for 1 recips, 
completed in 0.740 seconds


So it sends a message to the unsubscribed address, then it sends a 
message to the -request address, which mailman then happily throws out, 
and then it sends a message to -owner, which I got.

why is mailman sending the unsub to -request here? and throwing it out?


-- 
Chuq Von Rospach, Architech
chuqui@plaidworks.com -- http://www.plaidworks.com/chuqui/blog/

The Cliff's Notes Cliff's Notes on Hamlet:
     And they all died happily ever after


From barry@python.org  Tue Oct 15 05:29:15 2002
From: barry@python.org (Barry A. Warsaw)
Date: Tue, 15 Oct 2002 00:29:15 -0400
Subject: [Mailman-Developers] Even more on that log weirdie...
References: <D01BB111-DFEC-11D6-8345-0003934516A8@plaidworks.com>
Message-ID: <15787.39323.441755.51918@gargle.gargle.HOWL>


>>>>> "CVR" == Chuq Von Rospach <chuqui@plaidworks.com> writes:

    CVR> why is mailman sending the unsub to -request here? and
    CVR> throwing it out?

Oh this is twisted!  Note that this only happens when you have
admin_notify_mchanges enabled.

So, when you remove the address, the user gets a notification, which
is all well and good.  Also, the list-owner gets a notification
because of admin_notify_mchanges.  But note that this particular
message is sent to "listname-owner@dom.ain" -- so it gets to the
owners indirectly, not by using the owners explicitly as the
recipients of the message.

So, Mailman happily sends out the list-owner message, which your mta
dutifully returns to Mailman's incoming queue.  Now this message
correctly has a Precendence: bulk header on it, and Mailman now goes
through the normal processing for an -owner addressed message.  At
this point it has no idea it generated the message internally and fed
it back to itself via a loop through the MTA.

Now the logs/vette message is coming from Replybot.py and it's
actually both lying and misleading <wink>.  It's lying about the
"ignored by" address -- it just says the -request address no matter
where the message came from.  It's also misleading because although
the log message ways "Precedence: bulk message ignored by:" the only
thing that really happens is that the replybot won't auto-reply to the
message.  All the other normal processing of the message occurs, so
eventually it ends in the outgoing queue and sent on to the list
owners.

Fneh.  Two fixes are in order i believe.  1) admin_notify_mchanges
should use the owners addresses explicitly in the recipients list so
you don't need a roundtrip through the mta to get this message
delivered, and 2) the logs/vette message needs to be improved.

Thanks for the reproducible recipe Chuq.  I'll work out a fix and
check it in before I head to sleep.

-Barry

From barry@python.org  Tue Oct 15 05:33:26 2002
From: barry@python.org (Barry A. Warsaw)
Date: Tue, 15 Oct 2002 00:33:26 -0400
Subject: [Mailman-Developers] Even more on that log weirdie...
References: <D01BB111-DFEC-11D6-8345-0003934516A8@plaidworks.com>
	<15787.39323.441755.51918@gargle.gargle.HOWL>
Message-ID: <15787.39574.361767.739902@gargle.gargle.HOWL>


>>>>> "BAW" == Barry A Warsaw <barry@python.org> writes:

    BAW> 2) the logs/vette message needs to be improved.

On second though, I'm inclined to just remove the log message.  It was
mostly just helpful for debugging.

pep-282-ly y'rs,
-Barry

From claw@kanga.nu  Tue Oct 15 05:39:13 2002
From: claw@kanga.nu (J C Lawrence)
Date: Mon, 14 Oct 2002 21:39:13 -0700
Subject: [Mailman-Developers] Overhauled 'Exim and Mailman' doc 
In-Reply-To: Message from barry@python.org (Barry A. Warsaw) 
	of "Mon, 14 Oct 2002 19:58:14 EDT."
	<15787.23062.767185.619766@gargle.gargle.HOWL> 
References: <20021012014433.GA29438@cthulhu.gerg.ca>
	<4162.1034403272@kanga.nu>  <15787.23062.767185.619766@gargle.gargle.HOWL> 
Message-ID: <23556.1034656753@kanga.nu>

On Mon, 14 Oct 2002 19:58:14 -0400 
Barry A Warsaw <barry@python.org> wrote:

>>>>>> "JCL" == J C Lawrence <claw@kanga.nu> writes:

>> Would you like to have the Exim/TMDA/Mailman HOW-TO as a rider on
>> that, or should we tuck it in as a separate README (Barry?).

> Is the document Exim specific, or can you write it to be somewhat
> agnostic towards the mta?  

It is absurdly MTA specific.  There's really not much choice about that
-- TMDA really needs to get into the mail system at a level just above
acceptance and just below delivery, and that level is quite MTA
specific.  Achieving the same ends under Postfix requires bleeding edge
features of Postfix (we;ll, bleeding edge when I did the HOW-TO, I've
not checked since then) and resorting to sub-domains and other gross
perturbations of a mail system.  I've not looked at what would be needed
for Sendmail.  Jason M of TMDA fame at one point said he'd write a QMail
based HOW-TO, but I haven't noticed that go by yet.

> If so, I'd be happy to add a README.TMDA, or point to any
> documentation in the TMDA web pages.

There's a FAQ entry:

  http://www.python.org/cgi-bin/faqw-mm.py?req=show&file=faq06.007.htp

It needs updating for Exim4 and newer revs of TMDA, but I'm a mite short
of time for that.

-- 
J C Lawrence                
---------(*)                Satan, oscillate my metallic sonatas. 
claw@kanga.nu               He lived as a devil, eh?		  
http://www.kanga.nu/~claw/  Evil is a name of a foeman, as I live.


From chuqui@plaidworks.com  Tue Oct 15 07:28:48 2002
From: chuqui@plaidworks.com (Chuq Von Rospach)
Date: Mon, 14 Oct 2002 23:28:48 -0700
Subject: [Mailman-Developers] here we go again... (are these headers really
	right?)
Message-ID: <5D172B58-E007-11D6-9047-0003934516A8@plaidworks.com>


Okay, looking at the headers of a personalized message. is this really 
right?

Here's the problem. From points to the author. To points to me. In this 
case, they're both the same, but...

If you reply -- it goes to the author.

If you reply-to-all -- it goes to the author, because nowhere in the To 
or CC lines is the mail list mentioned.

I realize we're against co-ercing reply-to, but I think this has gotten 
a bit extreme.

Shouldn't the list be in the CC: line?



From: Chuq Von Rospach <chuqui@plaidworks.com>
Date: Mon Oct 14, 2002  9:48:14  PM US/Pacific
To: chuqui@plaidworks.com
Subject: a quick explanation of what's been going on with the list 
server....
Return-Path: <sharks-bounces+chuqui=plaidworks.com@plaidworks.com>
Received: from www.plaidworks.com (localhost [127.0.0.1]) by 
plaidworks.com (8.12.2/8.12.2) with ESMTP id g9F4mi1F008176 for 
<chuqui@plaidworks.com>; Mon, 14 Oct 2002 21:48:45 -0700 (PDT)
Received: from plaidworks.com (dsl081-078-186.sfo1.dsl.speakeasy.net 
[64.81.78.186]) by plaidworks.com (8.12.2/8.12.2) with ESMTP id 
g9F4mGhh008141 for <sharks@plaidworks.com>; Mon, 14 Oct 2002 21:48:16 
-0700 (PDT)
Mime-Version: 1.0 (Apple Message framework v546)
Content-Type: text/plain; charset=US-ASCII; format=flowed
Content-Transfer-Encoding: 7bit
Message-Id: <5087C41E-DFF9-11D6-8345-0003934516A8@plaidworks.com>
X-Mailer: Apple Mail (2.546)
X-Beenthere: sharks@plaidworks.com
X-Mailman-Version: 2.1b3+
Precedence: list
List-Unsubscribe: <http://www.plaidworks.com/mailman/listinfo/sharks>, 
<mailto:sharks-request@plaidworks.com?subject=unsubscribe>
List-Id: A mailing list for fans of the San Jose Sharks 
<sharks.plaidworks.com>
List-Post: <mailto:sharks@plaidworks.com>
List-Help: <mailto:sharks-request@plaidworks.com?subject=help>
List-Subscribe: <http://www.plaidworks.com/mailman/listinfo/sharks>, 
<mailto:sharks-request@plaidworks.com?subject=subscribe>
Sender: sharks-bounces+chuqui=plaidworks.com@plaidworks.com
Errors-To: sharks-bounces+chuqui=plaidworks.com@plaidworks.com


-- 
Chuq Von Rospach, Architech
chuqui@plaidworks.com -- http://www.plaidworks.com/chuqui/blog/

No! No! Dead girl, OFF the table! -- Shrek


From jwblist@olympus.net  Tue Oct 15 07:42:29 2002
From: jwblist@olympus.net (John W Baxter)
Date: Mon, 14 Oct 2002 23:42:29 -0700
Subject: [Mailman-Developers] more 2.1b3++ MIME funnies.
In-Reply-To: <15787.13915.661937.113786@gargle.gargle.HOWL>
References: <B27BFB52-DEE3-11D6-99E5-0003934516A8@plaidworks.com>
 <15787.13915.661937.113786@gargle.gargle.HOWL>
Message-ID: <p05200a01b9d16858f7ed@[192.168.1.50]>

At 17:25 -0400 10/14/2002, Barry A. Warsaw wrote:
>    CVR> What I'd like to be able to do (but can't) is allow M/A,
>    CVR> t/plain and t/html, and use the lynx to defang HTML going
>    CVR> into the text (but not MIME) digests. That seems the most
>    CVR> reasonable set of compromises to allow styled
>    CVR> mail. Unfortunately, mailman can't be set up that way...
>
>Nope, because messages are processed for the two digests in the same
>fell swoop.  I don't think it's ever come up before that you might
>want different processing for the different digests, but that /does/
>make sense.  ("Oh, TODO, please come here for a moment")

And the archive represents a history of...
   which version of the list...
     non-digest, text digest, MIME digest???

Probably non-digest--except for a digest-only list--but proving what was
sent by relying on the archive becomes difficult.

I do *not* think the above musings imply "don't do it."

  --John
-- 
John Baxter   jwblist@olympus.net      Port Ludlow, WA, USA

From chuqui@plaidworks.com  Tue Oct 15 07:51:08 2002
From: chuqui@plaidworks.com (Chuq Von Rospach)
Date: Mon, 14 Oct 2002 23:51:08 -0700
Subject: [Mailman-Developers] more 2.1b3++ MIME funnies.
In-Reply-To: <p05200a01b9d16858f7ed@[192.168.1.50]>
Message-ID: <7B8EFBE1-E00A-11D6-9047-0003934516A8@plaidworks.com>


>> Nope, because messages are processed for the two digests in the same
>> fell swoop.  I don't think it's ever come up before that you might
>> want different processing for the different digests, but that /does/
>> make sense.  ("Oh, TODO, please come here for a moment")

Yup. they really serve different purposes. One's designed for MIME, one 
isn't. So (now, long after it makes sense to think of it for 2.1....) 
making them separate processes now seems to make sense. but not for 
2.1. If someone absolutely positively has to have full MIME with all 
that stuff, they can force everyone to MIME digests, or do away with 
digests. to me, it's what I call a "nice to have", not a "gotta have".

> And the archive represents a history of...
>    which version of the list...
>      non-digest, text digest, MIME digest???

to me, archives always ought to represent to non-digest. the digest 
versions of things are convenient alternatives, but more or less by 
definition altered versions. Now, a GOOD archive would store all of 
that, but allow you the same general parsing restriction setup as 
mailman now does (or wil).

> I do *not* think the above musings imply "don't do it."
>

Hey, nobody said that Mailman 2.1 was "it", and we could all go off and 
stop working on it.... (giggle)


-- 
Chuq Von Rospach, Architech
chuqui@plaidworks.com -- http://www.plaidworks.com/chuqui/blog/

Stress is when you wake up screaming and you realize you haven't fallen 
asleep yet.


From claw@kanga.nu  Tue Oct 15 08:29:23 2002
From: claw@kanga.nu (J C Lawrence)
Date: Tue, 15 Oct 2002 00:29:23 -0700
Subject: [Mailman-Developers] here we go again... (are these headers
	really right?) 
In-Reply-To: Message from Chuq Von Rospach <chuqui@plaidworks.com> 
	of "Mon, 14 Oct 2002 23:28:48 PDT."
	<5D172B58-E007-11D6-9047-0003934516A8@plaidworks.com> 
References: <5D172B58-E007-11D6-9047-0003934516A8@plaidworks.com> 
Message-ID: <27038.1034666963@kanga.nu>

On Mon, 14 Oct 2002 23:28:48 -0700 
Chuq Von Rospach <chuqui@plaidworks.com> wrote:

> Here's the problem. From points to the author. To points to me. In
> this case, they're both the same, but...

Translation:

  A post which was BCC'ed to a list.

> I realize we're against co-ercing reply-to, but I think this has
> gotten a bit extreme.

> Shouldn't the list be in the CC: line?

Arguable.  It really depends on your view of BCCs and lists, and whether
you consider them a Good and Useful thing, or an annoyance of the
devil.  cf followup-to in netnews.  

  The principle of minimum invasion (ie lists should dick with headers
  as little as possible) says that BCC'ed posts should stay BCC'ed.

  The principle or minimum surprise likely says (there's debate) that
  some version of a reply (and RFC 2369 probably doesn't count) should
  enclude the list.

You get to pick.

-- 
J C Lawrence                
---------(*)                Satan, oscillate my metallic sonatas. 
claw@kanga.nu               He lived as a devil, eh?		  
http://www.kanga.nu/~claw/  Evil is a name of a foeman, as I live.


From Nigel.Metheringham@dev.InTechnology.co.uk  Tue Oct 15 09:37:12 2002
From: Nigel.Metheringham@dev.InTechnology.co.uk (Nigel Metheringham)
Date: 15 Oct 2002 09:37:12 +0100
Subject: [Exim] Re: [Mailman-Developers] Overhauled 'Exim and Mailman'
	doc
In-Reply-To: <20021015002714.GI2740@merlins.org>
References: <20021012014433.GA29438@cthulhu.gerg.ca>
	<15787.22975.845040.875953@gargle.gargle.HOWL>
	<20021015000555.GG2740@merlins.org>
	<15787.24276.660576.335006@gargle.gargle.HOWL> 
	<20021015002714.GI2740@merlins.org>
Message-ID: <1034671034.1147.1.camel@gaspode.localnet>

On Tue, 2002-10-15 at 01:27, Marc MERLIN wrote:
> >     MM> That'd be nice, but I think Nigel is a bit too busy, I sent
> >     MM> him my new version too, but he never had the time to post it
> >     MM> on exim.org

Tell me about it :-)


> Greg took my version from your CVS and further updated it, so he has the up
> to date version, so feel free to take whatever he gives you.
> (I gave him a few updates, but he's the maintainer of the day right now :-)

I'll do a batch of exim web site updates in the next couple of days. 
Get your submissions in now...

	Nigel.
-- 
[ Nigel Metheringham           Nigel.Metheringham@InTechnology.co.uk ]
[ - Comments in this message are my own and not ITO opinion/policy - ]


From barry@python.org  Tue Oct 15 16:11:46 2002
From: barry@python.org (Barry A. Warsaw)
Date: Tue, 15 Oct 2002 11:11:46 -0400
Subject: [Mailman-Developers] Overhauled 'Exim and Mailman' doc 
References: <20021012014433.GA29438@cthulhu.gerg.ca>
	<4162.1034403272@kanga.nu>
	<15787.23062.767185.619766@gargle.gargle.HOWL>
	<23556.1034656753@kanga.nu>
Message-ID: <15788.12338.645244.564716@gargle.gargle.HOWL>


>>>>> "JCL" == J C Lawrence <claw@kanga.nu> writes:

    >> Is the document Exim specific, or can you write it to be
    >> somewhat agnostic towards the mta?

    JCL> It is absurdly MTA specific.  There's really not much choice
    JCL> about that -- TMDA really needs to get into the mail system
    JCL> at a level just above acceptance and just below delivery, and
    JCL> that level is quite MTA specific.  Achieving the same ends
    JCL> under Postfix requires bleeding edge features of Postfix
    JCL> (we;ll, bleeding edge when I did the HOW-TO, I've not checked
    JCL> since then) and resorting to sub-domains and other gross
    JCL> perturbations of a mail system.  I've not looked at what
    JCL> would be needed for Sendmail.  Jason M of TMDA fame at one
    JCL> point said he'd write a QMail based HOW-TO, but I haven't
    JCL> noticed that go by yet.

Gotcha.

    >> If so, I'd be happy to add a README.TMDA, or point to any
    >> documentation in the TMDA web pages.

    JCL> There's a FAQ entry:

    JCL> http://www.python.org/cgi-bin/faqw-mm.py?req=show&file=faq06.007.htp

    JCL> It needs updating for Exim4 and newer revs of TMDA, but I'm a
    JCL> mite short of time for that.

The FAQ entry is certainly extensive!  If you do get it updated and
can distill it out of html, I'd be happy to add a README if you want.

-Barry

From barry@python.org  Tue Oct 15 16:16:32 2002
From: barry@python.org (Barry A. Warsaw)
Date: Tue, 15 Oct 2002 11:16:32 -0400
Subject: [Mailman-Developers] more 2.1b3++ MIME funnies.
References: <B27BFB52-DEE3-11D6-99E5-0003934516A8@plaidworks.com>
	<15787.13915.661937.113786@gargle.gargle.HOWL>
	<p05200a01b9d16858f7ed@[192.168.1.50]>
Message-ID: <15788.12624.851539.770446@gargle.gargle.HOWL>


>>>>> "JWB" == John W Baxter <jwblist@olympus.net> writes:

    | And the archive represents a history of...
    |    which version of the list...
    |      non-digest, text digest, MIME digest???

    JWB> Probably non-digest--except for a digest-only list--but
    JWB> proving what was sent by relying on the archive becomes
    JWB> difficult.

For the most part, what your non-digest members see is what the
archive sees.  So that means it's post content filtering and other
moderate-and-munge operations.

But the archives job is not to provide proof of what was posted to the
list.  For that you should hook into your mail server to archive
postings before they hit Mailman.

    JWB> I do *not* think the above musings imply "don't do it."

I'm not sure what I shouldn't be doing that I already isain't <wink>.

-Barry

From wheakory@isu.edu  Tue Oct 15 16:18:16 2002
From: wheakory@isu.edu (Kory Wheatley)
Date: Tue, 15 Oct 2002 09:18:16 -0600
Subject: [Mailman-Developers] Mailman editing messages
Message-ID: <3DAC31B8.73A42708@isu.edu>

I noticed in mailman 2.1.b3 that you can't edit a message from
"mailman/data" anymore and then approve the message with the edited
changes, (through the command line). Will there be a work around that
you
will be able to do this, because currently has I've talked about I'm not

able to use the web interface to edit a message through Netscape Mailer
(which really is not considered a web interface, but you know what I'm
talking about) so I currently do it through the command line in
/mailman/data with mailman 2.0.13. When Mailman 2.1 comes out I will not

be able to use either approach, so I hope that maybe these issues have
been expressed.

I don't mean to be critical, because I've used Mailman since its first
existence with our 170 production Universities Mailing lists and I like
it.
.

--
#########################################
Kory Wheatley
Academic Computing Analyst Sr.
Phone 282-3874
#########################################
Everything must point to him.



From claw@kanga.nu  Tue Oct 15 16:32:38 2002
From: claw@kanga.nu (J C Lawrence)
Date: Tue, 15 Oct 2002 08:32:38 -0700
Subject: [Mailman-Developers] Overhauled 'Exim and Mailman' doc 
In-Reply-To: Message from barry@python.org (Barry A. Warsaw) 
	of "Tue, 15 Oct 2002 11:11:46 EDT."
	<15788.12338.645244.564716@gargle.gargle.HOWL> 
References: <20021012014433.GA29438@cthulhu.gerg.ca>
	<4162.1034403272@kanga.nu> <15787.23062.767185.619766@gargle.gargle.HOWL>
	<23556.1034656753@kanga.nu>  <15788.12338.645244.564716@gargle.gargle.HOWL> 
Message-ID: <2255.1034695958@kanga.nu>

On Tue, 15 Oct 2002 11:11:46 -0400 
Barry A Warsaw <barry@python.org> wrote:

> The FAQ entry is certainly extensive!  If you do get it updated and
> can distill it out of html, I'd be happy to add a README if you want.

Uhh, the FAQ is plain text.  Its just the FAQ software which is
HTML-ising (kinda structured text style).

-- 
J C Lawrence                
---------(*)                Satan, oscillate my metallic sonatas. 
claw@kanga.nu               He lived as a devil, eh?		  
http://www.kanga.nu/~claw/  Evil is a name of a foeman, as I live.


From barry@python.org  Tue Oct 15 16:35:32 2002
From: barry@python.org (Barry A. Warsaw)
Date: Tue, 15 Oct 2002 11:35:32 -0400
Subject: [Mailman-Developers] more 2.1b3++ MIME funnies.
References: <p05200a01b9d16858f7ed@[192.168.1.50]>
	<7B8EFBE1-E00A-11D6-9047-0003934516A8@plaidworks.com>
Message-ID: <15788.13764.73262.310234@gargle.gargle.HOWL>


>>>>> "CVR" == Chuq Von Rospach <chuqui@plaidworks.com> writes:

    CVR> Yup. they really serve different purposes. One's designed for
    CVR> MIME, one isn't. So (now, long after it makes sense to think
    CVR> of it for 2.1....) making them separate processes now seems
    CVR> to make sense. but not for 2.1. If someone absolutely
    CVR> positively has to have full MIME with all that stuff, they
    CVR> can force everyone to MIME digests, or do away with
    CVR> digests. to me, it's what I call a "nice to have", not a
    CVR> "gotta have".

I do think some adjustments to the pipeline idea is in order, along
with splitting the two digest formats into two different message
sinks.  The trick will be doing that efficiently, with approachable
configuration management.

    >> And the archive represents a history of...  which version of
    >> the list...  non-digest, text digest, MIME digest???

    CVR> to me, archives always ought to represent to non-digest. the
    CVR> digest versions of things are convenient alternatives, but
    CVR> more or less by definition altered versions. Now, a GOOD
    CVR> archive would store all of that, but allow you the same
    CVR> general parsing restriction setup as mailman now does (or
    CVR> wil).

Remember that there's also the raw(er) mbox archive.  Right now
they're tied, but it could make sense to split those as well.

    >> I do *not* think the above musings imply "don't do it."
    >>

    CVR> Hey, nobody said that Mailman 2.1 was "it", and we could all
    CVR> go off and stop working on it.... (giggle)

Aw shoot.  Honey, cancel my vacation.
-Barry

From chuqui@plaidworks.com  Tue Oct 15 16:41:36 2002
From: chuqui@plaidworks.com (Chuq Von Rospach)
Date: Tue, 15 Oct 2002 08:41:36 -0700
Subject: [Mailman-Developers] here we go again... (are these headers
	really right?) 
In-Reply-To: <27038.1034666963@kanga.nu>
Message-ID: <96FB3D38-E054-11D6-9047-0003934516A8@plaidworks.com>


On Tuesday, October 15, 2002, at 12:29  AM, J C Lawrence wrote:

> On Mon, 14 Oct 2002 23:28:48 -0700
> Chuq Von Rospach <chuqui@plaidworks.com> wrote:
>
>> Here's the problem. From points to the author. To points to me. In
>> this case, they're both the same, but...
>
> Translation:
>
>   A post which was BCC'ed to a list.

Except that's not true.

I sent it To: my mailing list. The personalization removed that, and 
replaced with with a To: to the individual on the maling list. Leaving 
no linkage left back to the list for reply or reply-all to find.

In non-personalized mail, to To: address becomes the list. In 
personalized, that linkage is nuked. That effectively is treating ALL 
mail to the list as if it were BCCed to the list.

So reply is dead (I won't call it broken) if personalization is turned 
on. I don't believe that's what my users are going to expect.

I know, I know, I'm calling for more configuration stuff, but... Is 
there some way we can turn on personalization while leaving to/from/cc 
in the old style setup? Just allow return-path and header/footer tweak 
type stuff?

I just feel like if people start using personalization like this, it 
won't be what they expect. the removal of the linkage back to the list 
bothers me. If someone chooses to break the linkage with Bcc, great. If 
not...

god, can we go back and shoot the guys who mis-designed reply-to? 
List-post: isn't quite right here (or widely supported yet), reply-to 
doesn't do the "right" thing, either.

I guess another question would be this -- should the list regularlize 
how headers are written to set it up so reply goes back to the author 
and reply-all to the list? isn't that (barring a coerced reply to) how 
most users are expecting things now?

>> Shouldn't the list be in the CC: line?
>
> Arguable.  It really depends on your view of BCCs and lists, and 
> whether
> you consider them a Good and Useful thing, or an annoyance of the
> devil.  cf followup-to in netnews.

would be true, if I'd Bcc:ed the list. I explicitly didn't.

I guess I'm arguing that if I turn personalization on, it shouldn't 
hcange how the systme handles to/from/cc -- and it does. And that's 
going to cause unhappiness.


-- 
Chuq Von Rospach, Architech
chuqui@plaidworks.com -- http://www.plaidworks.com/chuqui/blog/

IMHO: Jargon. Acronym for In My Humble Opinion. Used to flag as an 
opinion
something that is clearly from context an opinion to everyone except the
mentally dense. Opinions flagged by IMHO are actually rarely humble. 
IMHO.
     (source: third unabridged dictionary of chuqui-isms).


From chuqui@plaidworks.com  Tue Oct 15 16:44:03 2002
From: chuqui@plaidworks.com (Chuq Von Rospach)
Date: Tue, 15 Oct 2002 08:44:03 -0700
Subject: [Mailman-Developers] more 2.1b3++ MIME funnies.
In-Reply-To: <15788.13764.73262.310234@gargle.gargle.HOWL>
Message-ID: <EE3461DA-E054-11D6-9047-0003934516A8@plaidworks.com>


On Tuesday, October 15, 2002, at 08:35  AM, Barry A. Warsaw wrote:

> I do think some adjustments to the pipeline idea is in order, along
> with splitting the two digest formats into two different message
> sinks.  The trick will be doing that efficiently, with approachable
> configuration management.

don't disagree, but IMHO, unless it's a trivial job, I wouldn't hold up 
2.1 for it.




-- 
Chuq Von Rospach, Architech
chuqui@plaidworks.com -- http://www.plaidworks.com/chuqui/blog/

Someday, we'll look back on this, laugh
nervously and change the subject.


From barry@python.org  Tue Oct 15 16:53:15 2002
From: barry@python.org (Barry A. Warsaw)
Date: Tue, 15 Oct 2002 11:53:15 -0400
Subject: [Mailman-Developers] Overhauled 'Exim and Mailman' doc 
References: <20021012014433.GA29438@cthulhu.gerg.ca>
	<4162.1034403272@kanga.nu>
	<15787.23062.767185.619766@gargle.gargle.HOWL>
	<23556.1034656753@kanga.nu>
	<15788.12338.645244.564716@gargle.gargle.HOWL>
	<2255.1034695958@kanga.nu>
Message-ID: <15788.14827.344878.52545@gargle.gargle.HOWL>


>>>>> "JCL" == J C Lawrence <claw@kanga.nu> writes:

    JCL> Uhh, the FAQ is plain text.  Its just the FAQ software which
    JCL> is HTML-ising (kinda structured text style).

Still, is the FAQ entry enough, or does it make sense to extract that
into a README?

-Barry

From barry@python.org  Tue Oct 15 16:56:09 2002
From: barry@python.org (Barry A. Warsaw)
Date: Tue, 15 Oct 2002 11:56:09 -0400
Subject: [Mailman-Developers] Mailman editing messages
References: <3DAC31B8.73A42708@isu.edu>
Message-ID: <15788.15001.511675.885092@gargle.gargle.HOWL>


>>>>> "KW" == Kory Wheatley <wheakory@isu.edu> writes:

    KW> I noticed in mailman 2.1.b3 that you can't edit a message from
    KW> "mailman/data" anymore and then approve the message with the
    KW> edited changes, (through the command line). Will there be a
    KW> work around that you will be able to do this, because
    KW> currently has I've talked about I'm not

    KW> able to use the web interface to edit a message through
    KW> Netscape Mailer (which really is not considered a web
    KW> interface, but you know what I'm talking about) so I currently
    KW> do it through the command line in /mailman/data with mailman
    KW> 2.0.13. When Mailman 2.1 comes out I will not

    KW> be able to use either approach, so I hope that maybe these
    KW> issues have been expressed.

    KW> I don't mean to be critical, because I've used Mailman since
    KW> its first existence with our 170 production Universities
    KW> Mailing lists and I like it.  .

Set HOLD_MESSAGES_AS_PICKLES=0 in your mm_cfg.py file and any messages
waiting for approval will be stored as plaintext instead of Python
pickles.  It's less efficient, but who cares?

Just be sure that when you edit the held message file, you save it in
a form that the email Parser's lax parsing mode can grok.  IOW,
conform to the RFCs and you'll be cool.

-Barry

From barry@python.org  Tue Oct 15 16:57:43 2002
From: barry@python.org (Barry A. Warsaw)
Date: Tue, 15 Oct 2002 11:57:43 -0400
Subject: [Mailman-Developers] more 2.1b3++ MIME funnies.
References: <15788.13764.73262.310234@gargle.gargle.HOWL>
	<EE3461DA-E054-11D6-9047-0003934516A8@plaidworks.com>
Message-ID: <15788.15095.495406.840214@gargle.gargle.HOWL>


>>>>> "CVR" == Chuq Von Rospach <chuqui@plaidworks.com> writes:

    >> I do think some adjustments to the pipeline idea is in order,
    >> along with splitting the two digest formats into two different
    >> message sinks.  The trick will be doing that efficiently, with
    >> approachable configuration management.

    CVR> don't disagree, but IMHO, unless it's a trivial job, I
    CVR> wouldn't hold up 2.1 for it.

Oh sorry, that's a given!  If 2.1 final doesn't come out by the end of
the year, I'm going to start farming soybeans.

-Barry

From claw@kanga.nu  Tue Oct 15 17:46:10 2002
From: claw@kanga.nu (J C Lawrence)
Date: Tue, 15 Oct 2002 09:46:10 -0700
Subject: [Mailman-Developers] Overhauled 'Exim and Mailman' doc 
In-Reply-To: Message from barry@python.org (Barry A. Warsaw) 
	of "Tue, 15 Oct 2002 11:53:15 EDT."
	<15788.14827.344878.52545@gargle.gargle.HOWL> 
References: <20021012014433.GA29438@cthulhu.gerg.ca>
	<4162.1034403272@kanga.nu> <15787.23062.767185.619766@gargle.gargle.HOWL>
	<23556.1034656753@kanga.nu> <15788.12338.645244.564716@gargle.gargle.HOWL>
	<2255.1034695958@kanga.nu>  <15788.14827.344878.52545@gargle.gargle.HOWL> 
Message-ID: <3905.1034700370@kanga.nu>

On Tue, 15 Oct 2002 11:53:15 -0400 
Barry A Warsaw <barry@python.org> wrote:

> Still, is the FAQ entry enough, or does it make sense to extract that
> into a README?

Given how few seem to see/read the FAQ, I'd say extracting it is
worthwhile.

-- 
J C Lawrence                
---------(*)                Satan, oscillate my metallic sonatas. 
claw@kanga.nu               He lived as a devil, eh?		  
http://www.kanga.nu/~claw/  Evil is a name of a foeman, as I live.


From barry@python.org  Tue Oct 15 17:54:49 2002
From: barry@python.org (Barry A. Warsaw)
Date: Tue, 15 Oct 2002 12:54:49 -0400
Subject: [Mailman-Developers] here we go again... (are these headers
	really right?) 
References: <27038.1034666963@kanga.nu>
	<96FB3D38-E054-11D6-9047-0003934516A8@plaidworks.com>
Message-ID: <15788.18521.851601.663482@gargle.gargle.HOWL>


>>>>> "CVR" == Chuq Von Rospach <chuqui@plaidworks.com> writes:

    >> Here's the problem. From points to the author. To points to
    >> me. In this case, they're both the same, but...
    >> Translation: A post which was BCC'ed to a list.

    CVR> Except that's not true.

Nor intended.  When personalizing, I intended that the To field gets
the address of the individual, but I didn't mean to erase all traces
of the list or make it impossible to reply to the list.

So our options are these:

1. Stick the list posting address back in the To header and probably
   not include the recipient's address in a mail header.

2. Leave the recip's address in To, but add a CC back to the list

3. Munge Reply-To so that the list's address gets in there somehow

Did you realize you can already do #3?  Just set reply_goes_to_list to
point to "This List".  I don't like #1 because I want a friendlier To
header for personalized messages.  That's not a strong reason, so I
could be persuaded.

So I'm inclined to add #2, but only if reply_goes_to_list isn't "This
List", since I don't think you need the list address in there twice.

    CVR> god, can we go back and shoot the guys who mis-designed
    CVR> reply-to? List-post: isn't quite right here (or widely
    CVR> supported yet), reply-to doesn't do the "right" thing,
    CVR> either.

Damn, Guido took the keys to his time machine, but even if he hadn't
we'd still probably only have 20-50% buyin my mua's.

    CVR> I guess another question would be this -- should the list
    CVR> regularlize how headers are written to set it up so reply
    CVR> goes back to the author and reply-all to the list? isn't that
    CVR> (barring a coerced reply to) how most users are expecting
    CVR> things now?

How's that different than what we do now (modulo one of the fixes
above)?

-Barry

From chuqui@plaidworks.com  Tue Oct 15 18:05:27 2002
From: chuqui@plaidworks.com (Chuq Von Rospach)
Date: Tue, 15 Oct 2002 10:05:27 -0700
Subject: [Mailman-Developers] here we go again... (are these headers
	really right?) 
In-Reply-To: <15788.18521.851601.663482@gargle.gargle.HOWL>
Message-ID: <4D9937A1-E060-11D6-9047-0003934516A8@plaidworks.com>


On Tuesday, October 15, 2002, at 09:54  AM, Barry A. Warsaw wrote:

> 1. Stick the list posting address back in the To header and probably
>    not include the recipient's address in a mail header.

my first preference.

>
> 2. Leave the recip's address in To, but add a CC back to the list


my second choice.

Both of these re-enable what I think is "right" here -- reply to 
author, reply-all to author and list.


>
> 3. Munge Reply-To so that the list's address gets in there somehow

I don't like this -- because it removes the reply/reply-to dichotomy 
above.

> So I'm inclined to add #2, but only if reply_goes_to_list isn't "This
> List", since I don't think you need the list address in there twice.
>

that seems to me best compromise. If you don't coerce reply-to, stick 
the list in CC so the replies work right. If you DO coerce reply-to, 
then it takes precedence and leave it alone.

> How's that different than what we do now (modulo one of the fixes
> above)?
>

It's not...


-- 
Chuq Von Rospach, Architech
chuqui@plaidworks.com -- http://www.plaidworks.com/chuqui/blog/

No! No! Dead girl, OFF the table! -- Shrek


From chuqui@plaidworks.com  Tue Oct 15 18:08:33 2002
From: chuqui@plaidworks.com (Chuq Von Rospach)
Date: Tue, 15 Oct 2002 10:08:33 -0700
Subject: [Mailman-Developers] here we go again... (are these headers
	really right?) 
In-Reply-To: <4D9937A1-E060-11D6-9047-0003934516A8@plaidworks.com>
Message-ID: <BC2EE70D-E060-11D6-9047-0003934516A8@plaidworks.com>

Oh, by the way. I don't remember seeing Barry note this, but...

mailman-developers is now running under Mailman 2.1b3+.

and I'm seeing amazingly good performance on return messages. To the 
point where the first few times, I thought I'd replied to myself, not 
the list.

(clap clap clap clap clap clap)


-- 
Chuq Von Rospach, Architech
chuqui@plaidworks.com -- http://www.plaidworks.com/chuqui/blog/

The Cliff's Notes Cliff's Notes on Hamlet:
     And they all died happily ever after


From barry@python.org  Tue Oct 15 18:30:51 2002
From: barry@python.org (Barry A. Warsaw)
Date: Tue, 15 Oct 2002 13:30:51 -0400
Subject: [Mailman-Developers] here we go again... (are these headers
	really right?) 
References: <15788.18521.851601.663482@gargle.gargle.HOWL>
	<4D9937A1-E060-11D6-9047-0003934516A8@plaidworks.com>
Message-ID: <15788.20683.541273.960731@gargle.gargle.HOWL>


>>>>> "CVR" == Chuq Von Rospach <chuqui@plaidworks.com> writes:

    >> 2. Leave the recip's address in To, but add a CC back to the
    >> list

    CVR> my second choice.

Done.

-Barry

From barry@python.org  Tue Oct 15 18:33:12 2002
From: barry@python.org (Barry A. Warsaw)
Date: Tue, 15 Oct 2002 13:33:12 -0400
Subject: [Mailman-Developers] here we go again... (are these headers
	really right?) 
References: <4D9937A1-E060-11D6-9047-0003934516A8@plaidworks.com>
	<BC2EE70D-E060-11D6-9047-0003934516A8@plaidworks.com>
Message-ID: <15788.20824.59653.818109@gargle.gargle.HOWL>


>>>>> "CVR" == Chuq Von Rospach <chuqui@plaidworks.com> writes:

    CVR> Oh, by the way. I don't remember seeing Barry note this,
    CVR> but...

    CVR> mailman-developers is now running under Mailman 2.1b3+.

I thought I sent around a message about it, but maybe I forgot.  Good
to see you noticed! :)

    CVR> and I'm seeing amazingly good performance on return
    CVR> messages. To the point where the first few times, I thought
    CVR> I'd replied to myself, not the list.

    CVR> (clap clap clap clap clap clap)

Thanks!  The real interesting thing will be to see how performance
holds up with dozens (or more) of lists running on the same
installation.

BTW, I'll be converting the other mailman-* lists over when I put out
b4, hopefully by the end of this week.

are-we-having-fun-yet-ly y'rs,
-Barry

From mjm@michaelmeltzer.com  Tue Oct 15 18:28:27 2002
From: mjm@michaelmeltzer.com (Michael Meltzer)
Date: Tue, 15 Oct 2002 13:28:27 -0400
Subject: [Mailman-Developers] attachments to links :-)
Message-ID: <001301c27470$45b898c0$34f820c0@ix1x1000>

Barry you seem in a good mood today. Figured no good deed should so unpunished :-) I try to pitch it again :-)

Any chance to get attachments to links in 2.1b4, very good stuff :-)

I can not offer a soybean farm but I always like growing mushroom better, They do best in old mines, completely in the dark and you
only concern is how high and fresh  the sh-t is :-)

MJM


From barry@python.org  Tue Oct 15 19:01:05 2002
From: barry@python.org (Barry A. Warsaw)
Date: Tue, 15 Oct 2002 14:01:05 -0400
Subject: [Mailman-Developers] attachments to links :-)
References: <001301c27470$45b898c0$34f820c0@ix1x1000>
Message-ID: <15788.22497.311441.176991@gargle.gargle.HOWL>


>>>>> "MM" == Michael Meltzer <mjm@michaelmeltzer.com> writes:

    MM> Barry you seem in a good mood today. Figured no good deed
    MM> should so unpunished :-) I try to pitch it again :-)

I love the chutzpah!  We try to have the same attitude in the Python
world too. :)

    MM> Any chance to get attachments to links in 2.1b4, very good
    MM> stuff :-)

Remind me again, is there a SF patch waiting to be applied?  Not that
I'm promising to add it...

    MM> I can not offer a soybean farm but I always like growing
    MM> mushroom better, They do best in old mines, completely in the
    MM> dark and you only concern is how high and fresh the sh-t is
    MM> :-)

My patent is pending on hallucinogenic soybeans.  A little wasabi
coating and dude, duuuudde!  Taste, good for you, and hours of
entertainment in each bag(gie).  I have a very special XEmacs
font-lock color scheme just for when the test batches roll out of the
oven.

-Barry

From noreply@sourceforge.net  Tue Oct 15 19:19:32 2002
From: noreply@sourceforge.net (noreply@sourceforge.net)
Date: Tue, 15 Oct 2002 11:19:32 -0700
Subject: [Mailman-Developers] 
 [ mailman-Bugs-620033 ] gate_news should catch lost connection
Message-ID: <E181WI0-0000tI-00@usw-sf-web2.sourceforge.net>

Bugs item #620033, was opened at 2002-10-07 22:07
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=100103&aid=620033&group_id=103

Category: nntp/news
Group: 2.1 beta
Status: Open
Resolution: None
Priority: 5
Submitted By: Ron Jarrell (jarrell)
Assigned to: Nobody/Anonymous (nobody)
Summary: gate_news should catch lost connection

Initial Comment:
Sep 20 10:51:35 2002 gate_news(9301): Traceback 
(most recent call last):
Sep 20 10:51:35 2002 gate_news(9301):   
File "/home/mailman/cron/gate_news", line 270, 
in ?
Sep 20 10:51:35 2002 gate_news(9301):      main()
Sep 20 10:51:35 2002 gate_news(9301):   
File "/home/mailman/cron/gate_news", line 250, in 
main
Sep 20 10:51:35 2002 gate_news(9301):      
process_lists(lock)
Sep 20 10:51:35 2002 gate_news(9301):   
File "/home/mailman/cron/gate_news", line 194, in 
process_lists
Sep 20 10:51:35 2002 gate_news(9301):      conn, 
first, last = open_newsgroup(mlist)
Sep 20 10:51:35 2002 gate_news(9301):   
File "/home/mailman/cron/gate_news", line 89, in 
open_newsgroup
Sep 20 10:51:35 2002 gate_news(9301):      
password=mm_cfg.NNTP_PASSWORD)
Sep 20 10:51:35 2002 gate_news(9301):   
File "/usr/local/lib/python2.2/nntplib.py", line 
117, in __init__
Sep 20 10:51:35 2002 gate_news(9301):      
self.welcome = self.getresp()
Sep 20 10:51:35 2002 gate_news(9301):   
File "/usr/local/lib/python2.2/nntplib.py", line 
203, in getresp
Sep 20 10:51:35 2002 gate_news(9301):      resp = 
self.getline()
Sep 20 10:51:35 2002 gate_news(9301):   
File "/usr/local/lib/python2.2/nntplib.py", line 
192, in getline
Sep 20 10:51:35 2002 gate_news(9301):      line = 
self.file.readline()
Sep 20 10:51:35 2002 gate_news(9301): IOError :  
[Errno 131] Connection reset by peer 


----------------------------------------------------------------------

>Comment By: Barry A. Warsaw (bwarsaw)
Date: 2002-10-15 14:19

Message:
Logged In: YES 
user_id=12800

I didn't realize that nntplib.NNTP() can raise an IOError --
I thought it would have been a socket.error.  Oh well,
here's an untested patch.  I think I'll check this in just
because I don't see how it could break anything.

----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=100103&aid=620033&group_id=103

From noreply@sourceforge.net  Tue Oct 15 20:00:22 2002
From: noreply@sourceforge.net (noreply@sourceforge.net)
Date: Tue, 15 Oct 2002 12:00:22 -0700
Subject: [Mailman-Developers] 
 [ mailman-Bugs-620033 ] gate_news should catch lost connection
Message-ID: <E181WvW-00033A-00@usw-sf-web1.sourceforge.net>

Bugs item #620033, was opened at 2002-10-07 22:07
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=100103&aid=620033&group_id=103

Category: nntp/news
Group: 2.1 beta
>Status: Closed
>Resolution: Accepted
Priority: 5
Submitted By: Ron Jarrell (jarrell)
Assigned to: Nobody/Anonymous (nobody)
Summary: gate_news should catch lost connection

Initial Comment:
Sep 20 10:51:35 2002 gate_news(9301): Traceback 
(most recent call last):
Sep 20 10:51:35 2002 gate_news(9301):   
File "/home/mailman/cron/gate_news", line 270, 
in ?
Sep 20 10:51:35 2002 gate_news(9301):      main()
Sep 20 10:51:35 2002 gate_news(9301):   
File "/home/mailman/cron/gate_news", line 250, in 
main
Sep 20 10:51:35 2002 gate_news(9301):      
process_lists(lock)
Sep 20 10:51:35 2002 gate_news(9301):   
File "/home/mailman/cron/gate_news", line 194, in 
process_lists
Sep 20 10:51:35 2002 gate_news(9301):      conn, 
first, last = open_newsgroup(mlist)
Sep 20 10:51:35 2002 gate_news(9301):   
File "/home/mailman/cron/gate_news", line 89, in 
open_newsgroup
Sep 20 10:51:35 2002 gate_news(9301):      
password=mm_cfg.NNTP_PASSWORD)
Sep 20 10:51:35 2002 gate_news(9301):   
File "/usr/local/lib/python2.2/nntplib.py", line 
117, in __init__
Sep 20 10:51:35 2002 gate_news(9301):      
self.welcome = self.getresp()
Sep 20 10:51:35 2002 gate_news(9301):   
File "/usr/local/lib/python2.2/nntplib.py", line 
203, in getresp
Sep 20 10:51:35 2002 gate_news(9301):      resp = 
self.getline()
Sep 20 10:51:35 2002 gate_news(9301):   
File "/usr/local/lib/python2.2/nntplib.py", line 
192, in getline
Sep 20 10:51:35 2002 gate_news(9301):      line = 
self.file.readline()
Sep 20 10:51:35 2002 gate_news(9301): IOError :  
[Errno 131] Connection reset by peer 


----------------------------------------------------------------------

>Comment By: Barry A. Warsaw (bwarsaw)
Date: 2002-10-15 15:00

Message:
Logged In: YES 
user_id=12800

I'm applying this patch

----------------------------------------------------------------------

Comment By: Barry A. Warsaw (bwarsaw)
Date: 2002-10-15 14:19

Message:
Logged In: YES 
user_id=12800

I didn't realize that nntplib.NNTP() can raise an IOError --
I thought it would have been a socket.error.  Oh well,
here's an untested patch.  I think I'll check this in just
because I don't see how it could break anything.

----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=100103&aid=620033&group_id=103

From mjm@michaelmeltzer.com  Tue Oct 15 19:57:42 2002
From: mjm@michaelmeltzer.com (Michael Meltzer)
Date: Tue, 15 Oct 2002 14:57:42 -0400
Subject: [Mailman-Developers] attachments to links :-)
References: <001301c27470$45b898c0$34f820c0@ix1x1000>
	<15788.22497.311441.176991@gargle.gargle.HOWL>
Message-ID: <005c01c2747c$bd8f3140$34f820c0@ix1x1000>

chutzpah works for me, Thank You Thank You, My NewYorker is coming out :-)

I just took a shot at patching against the CVS, you been a busy camper :-) dispose is a nice touch and stops "where did it go
questions" . I need tonight to line my stuff up, I gave you a quick rundown below, forgive me, python is not my first langue :-) I
am pretty sure I can have a patch for MimeDel.py and Private.py, the template's I am clue less :-), I understand no promises, just
wanted to get back to you, see if their is interest, Or find out my lame code it not up to standards :-)

MJM


But in English:

For MimeDel.py
1)
from Mailman.Handlers.Scrubber import save_attachment
from time import strftime
2)
in filter_parts add the call to make_attachment, by it changing the mtype/ctype it stays out of the way of the filtering code, also
need to insert it near the top of the code to catch a single message of these mime types.

def make_attachment(mlist, subpart):
!      #should be set from mlist, work in progress
!      #BTW this will act real stupid with mulipart, it need the real object not the house keeping
!     attach_filter = ['image/bmp', 'image/jpeg', 'image/tiff', 'image/gif', 'image/png', 'image/pjpeg', 'image/x-png',
'image/x-wmf']
!     ctype = subpart.get_type('text/plain')
!     mtype = subpart.get_main_type('text')
!     if ctype in attach_filter or mtype in attach_filter:
!         cctype = subpart.get_type()
!         #size is off, just could not stand to call decode to correct, might just take off 20% and be done
!         size = len(subpart.get_payload())
!         desc = subpart.get('content-description', (_('not available')))
!         filename = subpart.get_filename(_('not available'))
!         url = save_attachment(mlist, subpart, strftime("attch/%Y%m/%d"))
!         del subpart['content-type']
!         del subpart['content-transfer-encoding']
!         del subpart['content-disposition']
!         del subpart['content-description']
!         subpart.add_header('Content-Type', 'text/plain', charset='us-ascii')
!         subpart.add_header('Content-Transfer-Encoding', '7bit')
!         subpart.add_header('Content-disposition', 'inline')
!         subpart.set_payload(_("""\
! Name: %(filename)s Type: %(cctype)s Size: %(size)d bytes Desc: %(desc)s
! Url: %(url)s
! """))
!         return 1
!     else:
!         return 0
!
!
3)
attach_filter should be on mlist and part of the filter_content template

4)their is a problem in private.py(I think this will show for some of the scrubber functionality too), it dose not return the mime
type to the browser for the extended types, might be a good idea to confirm it with the types from attach_filter for good security.



----- Original Message -----
From: "Barry A. Warsaw" <barry@python.org>
To: "Michael Meltzer" <mjm@michaelmeltzer.com>
Cc: <mailman-developers@python.org>
Sent: Tuesday, October 15, 2002 2:01 PM
Subject: Re: [Mailman-Developers] attachments to links :-)


>
> >>>>> "MM" == Michael Meltzer <mjm@michaelmeltzer.com> writes:
>
>     MM> Barry you seem in a good mood today. Figured no good deed
>     MM> should so unpunished :-) I try to pitch it again :-)
>
> I love the chutzpah!  We try to have the same attitude in the Python
> world too. :)
>
>     MM> Any chance to get attachments to links in 2.1b4, very good
>     MM> stuff :-)
>
> Remind me again, is there a SF patch waiting to be applied?  Not that
> I'm promising to add it...
>
>     MM> I can not offer a soybean farm but I always like growing
>     MM> mushroom better, They do best in old mines, completely in the
>     MM> dark and you only concern is how high and fresh the sh-t is
>     MM> :-)
>
> My patent is pending on hallucinogenic soybeans.  A little wasabi
> coating and dude, duuuudde!  Taste, good for you, and hours of
> entertainment in each bag(gie).  I have a very special XEmacs
> font-lock color scheme just for when the test batches roll out of the
> oven.
>
> -Barry


From barry@python.org  Tue Oct 15 21:42:08 2002
From: barry@python.org (Barry A. Warsaw)
Date: Tue, 15 Oct 2002 16:42:08 -0400
Subject: [Mailman-Developers] Cron error messages
References: <a05111b0eb9cbb17782e6@[10.1.19.62]>
Message-ID: <15788.32160.73629.424590@gargle.gargle.HOWL>


>>>>> "bronto" ==   <bronto@csd-bes.net> writes:

    | Traceback (most recent call last):
    |    File "/usr/local/mailman/cron/gate_news", line 44, in ?
    |      from Mailman import MailList
    | ValueError: bad marshal data

Because you're getting this on an import, I have to conclude that your
.pyc files are corrupt.  I've no idea why that could be (maybe you ran
out of disk space once, or have other h/w problems).

One thing you can do is to temporarily disable cron, the web
interface, and qrunner, then cd to /usr/local/mailman and run this:

% find . -name \*.pyc -print | xargs rm

Then cd back to your source directory and do a "make install".  This
will regenerate all your pyc files.  Watch carefully for any errors,
especially during the "compileall" phase.

    bronto> Since posting that, I have discovered that one of my
    bronto> mailing lists has stopped functioning to some degree.  I
    bronto> had set up this particular list to be a simple personal
    bronto> archive of messages from another mailing list I subscribe
    bronto> to.  It doesn't send anything, so I hadn't noticed any
    bronto> problems.  While doing my mailman update to 2.1b3, I
    bronto> discovered that no messages have been archived since the
    bronto> 23rd when I got the above errors.  Snooping around the
    bronto> mailman directories, I see a ton of files (several
    bronto> hundred) in the qfiles/shunt directory, dated since 9/23.
    bronto> There are also a few dozen from 7/2, FWIW.  I'm betting
    bronto> this isn't a coincidence, but know nothing more than that.

    bronto> Today at 12 noon, I received the following cron error:

    | _________________________________________
    | Traceback (most recent call last):
    |    File "/usr/local/mailman/cron/gate_news", line 44, in ?
    |      from Mailman import MailList
    |    File "/usr/local/mailman/Mailman/MailList.py", line 31, in ?
    |      import urllib
    | EOFError: EOF read where object expected
    | __________________________________________

Again, that's happening on an import, but now it's while importing a
standard Python module.  That might mean that your Python installation
is corrupt in a similar way.

    bronto> Since that time, there has been a mailman process running
    bronto> taking up all available CPU cycles.  gtop shows that the
    bronto> CMD is /usr/bin/python2, FWIW.  There are a fair number of
    bronto> qrunner processes running, but my mailing lists are hardly
    bronto> active.

    bronto> What's going on here?

I don't know if those other problems are related, but you first have
to make sure those import problems go away before you can track down
any other issues.

-Barry

From barry@python.org  Tue Oct 15 21:45:38 2002
From: barry@python.org (Barry A. Warsaw)
Date: Tue, 15 Oct 2002 16:45:38 -0400
Subject: [Mailman-Developers] 2.1b3 gate_news problem Sparc/Linux 6.2
References: <B9CB13A8.88B%col.neeson@planetone-asia.com>
Message-ID: <15788.32370.294487.156001@gargle.gargle.HOWL>


>>>>> "CN" == Colin Neeson <col.neeson@planetone-asia.com> writes:

    CN> socket.error: (32, 'Broken pipe')

    CN> The news server is up and running.  I can post and read news
    CN> items from it, so the socket.error is perplexing.

Very perlexing, especially since you get this error on the QUIT.  You
don't say what version of Python you're using, although I can guess
from the traceback it's 2.2.something.  First, upgrade to Python
2.2.2, which was just released yesterday.  That may or may not help.

If you're still having the problems, try some tests manually with the
nntplib module.  See

    http://www.python.org/doc/current/lib/module-nntplib.html

for details and examples.  If you can connect to your server, read
some data, grab n.g. articles and such, without error, then we'll have
to take a different tack.

-Barry

From gward@python.net  Tue Oct 15 22:43:54 2002
From: gward@python.net (Greg Ward)
Date: Tue, 15 Oct 2002 17:43:54 -0400
Subject: [Mailman-Developers] Re: [Exim] Overhauled 'Exim and Mailman' doc
In-Reply-To: <20021012014433.GA29438@cthulhu.gerg.ca>
References: <20021012014433.GA29438@cthulhu.gerg.ca>
Message-ID: <20021015214354.GB4223@cthulhu.gerg.ca>

[follow up on exim-users please -- let's try to keep this on one list]

On 11 October 2002, I said:
> I've just spent the last couple of hours beating the README.EXIM
> document from Mailman's current CVS tree into shape.  (This is derived
> from Nigel Metheringham's "HOWTO - Using exim and Mailman together"
> document, which can be found at www.exim.org.)

OK, here's the latest version.  I wrote about receiver verification, and
SMTP callback based on Marc's suggestion.  Rewrote his text and
streamlined the config a bit to improve readability -- hope it still
works.

I also actually tested my suggested transport, and the version here is
known to work with Exim 4.  ;-)

------------------------------------------------------------------------
Using Exim and Mailman Together
===============================

[This is derived from Nigel Metheringham's "HOWTO - Using exim and
 Mailman together", which covers Mailman 2.0.x and Exim 3.  It
 has been updated to cover Mailman 2.1 and Exim 4.  The original
 document is here: http://www.exim.org/howto/mailman.html]


Mailman configuration
---------------------

There is no Mailman configuration needed other than the standard options
detailed in the Mailman install documentation.  The Exim configuration
is transparent to Mailman.  The uid/gid settings for Mailman must match
those in the config fragments given below.
   

Exim configuration
------------------

The Exim configuration is built so that a list created within Mailman
automatically appears to Exim without the need for defining any
additional aliases.

The drawback of this configuration is that it will work poorly on
systems supporting lists in several different mail domains.  While
Mailman handles virtual domains, it does not yet support having two
lists with the same name in different virtual domains, using the same
Mailman installation.  This will eventually change.  (But see below for
a variation on this scheme that should accomodate virtual domains
better.)

The configuration file excerpts below are for use in an already
functional Exim configuration, which accepts mail for the domain in
which the list resides.  If this domain is separate from the others
handled by your Exim configuration, then you'll need to:

  * add the list domain, "my.list.domain" to local_domains

  * add a "domains=my.list.domain" option to each of the directors
    (routers) for the list

  * (optional) exclude that domain from the other directors (routers)

[Note: the instructions in this document should work with either Exim 3
or Exim 4.  In Exim 3, you must have a 'local_domains' configuration
setting; in Exim 4, you most likely have a 'local_domains' domainlist.
If you don't, you probably know what you're doing and can adjust
accordingly.  Similarly, in Exim 4 the concept of "directors" has
disappeared -- there are only routers now.  So if you're using Exim 4,
whenever this document says "director", read "router".]

Whether you are using Exim 3 or Exim 4, you will need to add some macros
to the main section of your Exim config file.  You will also need to
define one new transport.  With Exim 3, you'll need to add two new
directors; with Exim 4, two new routers play the same role.

Finally, the configuration supplied here should allow co-habiting
Mailman 2.0 and 2.1 installations, with the proviso that you'll probably
want to use "mm21" in place of "mailman" -- e.g., MM21_HOME,
mm21_transport, etc.


Main configuration settings
~~~~~~~~~~~~~~~~~~~~~~~~~~~

First, you need to add some macros to the top of your Exim config file.
These just make the directors (routers) and transport below a bit
cleaner.  Obviously, you'll need to edit these based on how you
configured and installed Mailman.

  # Home dir for your Mailman installation -- aka Mailman's prefix
  # directory.
  MAILMAN_HOME=/usr/local/mailman
  MAILMAN_WRAP=MAILMAN_HOME/mail/mailman

  # User and group for Mailman, should match your --with-mail-gid
  # switch to Mailman's configure script.
  MAILMAN_UID=mailman
  MAILMAN_GID=mailman


Transport for Exim 3
~~~~~~~~~~~~~~~~~~~~

Add this to the transports section of your Exim config file,
i.e. somewhere between the first and second "end" line:

  mailman_transport:
    driver = pipe
    command = MAILMAN_WRAP \
              '${if def:local_part_suffix \
                    {${sg{$local_part_suffix}{-(\\w+)(\\+.*)?}{\$1}}} \
                    {post}}' \
              $local_part
    current_directory = MAILMAN_HOME
    home_directory = MAILMAN_HOME
    user = MAILMAN_UID
    group = MAILMAN_GID

(XXX this is untested by me under Exim 3!  Can someone using Exim 3
please let me know if it works?)


Directors for Exim 3
~~~~~~~~~~~~~~~~~~~~

If you're using Exim 3, you'll need to add the following two directors
to your config file (directors go between the second and third "end"
lines).  Also, don't forget that order matters -- eg. you can make
Mailman lists take precedence over system aliases by putting these
directors in front of your aliasfile director, or vice-versa.

  # Director to handle a list's main posting address.  Should come first
  # in case there's a list named, eg. "mailman-owner".  Automatically
  # detects list existence by looking for lists/$local_part/config.pck
  # under MAILMAN_HOME.
  mailman_main_director:
    driver = smartuser
    require_files = MAILMAN_HOME/lists/$local_part/config.pck
    transport = mailman_transport

  # Handle all the other addresses related to a list: -bounces, -admin,
  # etc.  Same trick works to detect list existence thanks to the
  # 'suffix' option.
  mailman_director:
    driver = smartuser
    require_files = MAILMAN_HOME/lists/$local_part/config.pck
    suffix = -bounces : -bounces+* : \
             -confirm+* : -join : -leave : \
             -owner : -request : -admin
    transport = mailman_transport


Routers for Exim 4
~~~~~~~~~~~~~~~~~~

In Exim 4, there's no such thing as directors -- you instead need to add
two routers.  Also, the canonical order of the configuration file was
changed so routers come before transports, so the routers for Exim 4
come first here.  Put these two routers somewhere after the "begin
routers" line of your config file, and remember that order matters.

  mailman_main_router:
    driver = accept
    require_files = MAILMAN_HOME/lists/$local_part/config.pck
    transport = mailman_transport

  mailman_router:
    driver = accept
    require_files = MAILMAN_HOME/lists/$local_part/config.pck
    local_part_suffix = -bounces : -bounces+* : \
                        -confirm+* : -join : -leave : \
                        -owner : -request : -admin
    transport = mailman_transport


Transports for Exim 4
~~~~~~~~~~~~~~~~~~~~~

The transport for Exim 4 is the same as for Exim 3; just copy the
transport given above to somewhere under the "begin transports" line of
your Exim config file.


Notes
-----

Exim should be configured to allow reasonable volume -- e.g. don't set
max_recipients down to a silly value -- and with normal degrees of
security -- specifically, be sure to allow relaying from 127.0.0.1, but
pretty much nothing else.  Parallel deliveries and other tweaks can also
be used if you like; experiment with your setup to see what works.
Delay warning messages should be switched off or configured to only
happen for non-list mail, unless you like receiving tons of mail when
some random host is down.


Problems
--------

  * Mailman will send as many MAIL FROM/RCPT TO as it needs. It may result
    in more than 10 or 100 messages sent in one connection, which will exceed
    the default value of Exim's smtp_accept_queue_per_connection
    This is bad because it will cause Exim to switch into queue mode and
    severely delay delivery of your list messages.
    The way to fix this is to set mailman's SMTP_MAX_SESSIONS_PER_CONNECTION
    (in ~mailman/Mailman/mm_cfg.py) to a smaller value than Exim's
    smtp_accept_queue_per_connection

  * Mailman should ignore Exim delay warning messages, even though
    Exim should never send this to list messages.  Mailman 2.1's
    general bounce detection and VERP support should greatly improve
    the bounce detector's hit rates.

  * List existence is determined by the existence of a config.pck file
    for a list.  If you delete lists by foul means, be aware of this.

  * If you are getting Exim or Mailman complaining about user ids
    when you send mail to a list, check that the MAILMAN_UID and
    MAILMAN_GID match those of Mailman itself (i.e. what were used
    in the configure script).  Also make sure you do not have
    aliases in the main alias file for the list.


Receiver Verification
---------------------

Exim's receiver verification feature is very useful -- it lets Exim
reject unrouteable addresses at SMTP time.  However, this is most useful
for externally-originating mail that is addresses to mail in one of your
local domains.  For Mailman list traffic, mail originates on your
server, and is addressed to random external domains that are not under
your control.  Furthermore, each message is addressed to many recipients
-- up to 500 if you use Mailman's default configuration, and don't tweak
SMTP_MAX_RCPTS.

Doing receiver verification on Mailman list traffic is a recipe for
trouble.  In particular, Exim will attempt to route every recipient
addresses in outgoing Mailman list posts.  Even though this requires
nothing more than a few DNS lookups for each address, it can still
introduce significant delays.  Therefore, you should disable recipient
verification for Mailman traffic.

Under Exim 3, put this in your main configuration section:

  receiver_verify_hosts = !127.0.0.1

Under Exim 4, this is probably already taken care of for you by the
default recipient verification ACL statement (in the "RCPT TO" ACL):

  accept  domains       = +local_domains
          endpass
          message       = unknown user
          verify        = recipient

which only does recipient verification on addresses in your domain.
(That's not exactly the same as doing recipient verification only on
messages coming from non-127.0.0.1 hosts, but it should do the trick for
Mailman.)


SMTP Callback
-------------

Exim's SMTP callback feature is an even more powerful way to detect
bogus sender addresses than normal sender verification.  Unfortunately,
lots of servers send bounce messages with a bogus address in the header,
and there are plenty that sound bounces with bogus envelope senders
(even though they're supposed to just use an emtpy envelope sender for
bounces).

In order to ensure that Mailman can disable/remove bouncing addresses,
you generally want to receive bounces for Mailman lists, even if those
bounces are themselves not bouncable.  Thus, you might want to disable
SMTP callback on bounce messages.

If you do header and envelope callbacks, you can disable them for bounces to
mailman lists (it  is quite common for  internal hosts to bounce  with a non
reachable internal address). The  idea is that you typically  don't want non
bounceable Email, but  you'd better accept bounces to mailman  lists so that
you can unsubscribe the people who are bouncing.

With Exim 4, you can accomplish this using something like the following
in your "RCPT TO" ACL:

  # Accept bounces to lists even if callbacks or other checks would fail
  warn     message	= X-WhitelistedRCPT-nohdrfromcallback: Yes
	   condition	= \
           ${if and {{match{$local_part}{(.*)-bounces\+.*}}
                     {exists {MAILMAN_HOME/lists/$1/config.pck}}} \
                {yes}{no}}
                {yes}{no}}

  accept   condition	= \
           ${if and {{match{$local_part}{(.*)-bounces\+.*}}
                     {exists {MAILMAN_HOME/lists/$1/config.pck}}} \
                {yes}{no}}
                {yes}{no}}

  # Now, check sender address with SMTP callback.
  deny   !verify = sender/callout=90s/check_postmaster

If you also do SMTP callbacks on header addresses, you'll want something
like this in your "DATA" ACL:

  deny   !condition = $header_X-WhitelistedRCPT-nohdrfromcallback:
	 !verify = header_sender/callout=90s/check_postmaster

[XXX all this stuff is completely untested by me!  -Greg]


Doing VERP with Exim and Mailman
--------------------------------

VERP will send one email, with a separate envelope sender (return path),
for each of your subscribers -- read the information in
~mailman/Mailman/Default.py for the options that start with VERP.  In a
nutshell, all you need to do to enable VERP with Exim is to add these
lines to ~mailman/Mailman/mm_cfg.py:

  VERP_PASSWORD_REMINDERS = 1
  VERP_PERSONALIZED_DELIVERIES = 1
  VERP_DELIVERY_INTERVAL = 1
  VERP_CONFIRMATIONS = 1

(The directors/routers above are smart enough to deal with VERP
bounces.)

       
Virtual Domains
---------------

One approach to handling virtual domains is to use a separate Mailman
installation for each virtual domain.  (Currently, this is the only way
to have lists with the same name in different virtual domains handled by
the same machine.)

In this case, the MAILMAN_HOME and MAILMAN_WRAP macros are useless --
you can remove them.  Change your directors (routers) to something like
this:

  require_files = /virtual/${domain}/mailman/lists/${lc:$local_part}/config.pck

and change your transport like this:

  command = /virtual/${domain}/mailman/mail/mailman \
            ${if def:local_part_suffix \
                 {${sg{$local_part_suffix}{-(\\w+)(\\+.*)?}{\$1}}}
                 {post}} \
              $local_part
  current_directory = /virtual/${domain}/mailman
  home_directory = /virtual/${domain}/mailman


List Verification
-----------------

This is how a set of address tests for the Exim lists look on a working
system.  The list in question is quixote-users@mems-exchange.org, and
these commands were run on the mems-exchange.org mail server ("% "
indicates the Unix shell prompt):

  % exim -bt quixote-users
  quixote-users@mems-exchange.org
    router = mailman_main_router, transport = mailman_transport

  % exim -bt quixote-users-request
  quixote-users-request@mems-exchange.org
    router = mailman_router, transport = mailman_transport

  % exim -bt quixote-users-bounces
  quixote-users-bounces@mems-exchange.org
    router = mailman_router, transport = mailman_transport

  % exim -bt quixote-users-bounces+luser=example.com
  quixote-users-bounces+luser=example.com@mems-exchange.org
    router = mailman_router, transport = mailman_transport

If your "exim -bt" output looks something like this, that's a start: at
least it means Exim will pass the right messages to the right Mailman
commands.  It by no means guarantees that your Exim/Mailman installation
is functioning perfectly, though!


Document History
----------------

Originally written by Nigel Metheringham <postmaster@exim.org>.
Updated by Marc Merlin <marc_soft@merlins.org> for Mailman 2.1, Exim 4.
Overhauled/reformatted/clarified/simplified by Greg Ward <gward@python.net>.
------------------------------------------------------------------------

-- 
Greg Ward <gward@python.net>                         http://www.gerg.ca/
Never put off till tomorrow what you can put off till the day after tomorrow.

From noreply@sourceforge.net  Tue Oct 15 22:52:08 2002
From: noreply@sourceforge.net (noreply@sourceforge.net)
Date: Tue, 15 Oct 2002 14:52:08 -0700
Subject: [Mailman-Developers] 
 [ mailman-Patches-623783 ] Option to discard non-member postings
Message-ID: <E181Zbk-00007B-00@usw-sf-web5.sourceforge.net>

Patches item #623783, was opened at 2002-10-15 14:52
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=300103&aid=623783&group_id=103

Category: list administration
Group: Mailman 2.0.x
Status: Open
Resolution: None
Priority: 5
Submitted By: Alan Batie (abatie)
Assigned to: Nobody/Anonymous (nobody)
Summary: Option to discard non-member postings

Initial Comment:
Spam has become such an epidemic that any even
semi-public list must restrict postings to members
only, but then it's up to the administrator to go in
and manually delete it all, which is a real pain
through the web interface in mailman.  This patch adds
an option to just throw all non-member postings away.

----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=300103&aid=623783&group_id=103

From barry@python.org  Tue Oct 15 23:14:54 2002
From: barry@python.org (Barry A. Warsaw)
Date: Tue, 15 Oct 2002 18:14:54 -0400
Subject: [Mailman-Developers] Re: [Exim] Overhauled 'Exim and Mailman' doc
References: <20021012014433.GA29438@cthulhu.gerg.ca>
	<20021015214354.GB4223@cthulhu.gerg.ca>
Message-ID: <15788.37726.767628.525739@gargle.gargle.HOWL>


Thanks Greg.  I've installed this as README.EXIM (with a ispell pass
;).   Also, in MM2.1b4, MAILMAN_UID and MAILMAN_GID are replaced with
MAILMAN_USER and MAILMAN_GROUP, so I changed these in the file.  You
might want to double check against cvs.

Thanks!
-Barry

From bronto@csd-bes.net  Wed Oct 16 02:18:46 2002
From: bronto@csd-bes.net (bronto)
Date: Tue, 15 Oct 2002 18:18:46 -0700
Subject: [Mailman-Developers] Cron error messages
In-Reply-To: <15788.32160.73629.424590@gargle.gargle.HOWL>
References: <a05111b0eb9cbb17782e6@[10.1.19.62]>
 <15788.32160.73629.424590@gargle.gargle.HOWL>
Message-ID: <a05111b00b9d26d45bbac@[10.1.19.62]>

Thanks for the reply, and yes, there was a hard disk failure a couple 
weeks prior to this.  Perhaps corrupted python/mailman was the first 
effect, that didn't show up until later.

Since the original "bad marshal data" problem, I've upgraded to 
v2.1b3, so unless make install does not replace existing files I 
think my mailman installation should be OK now (?).  I will 
re-install python2 to make sure it's OK as well.

The remaining question is that of all the files in the shunt 
directory.  These must represent missing posts from the mailing list. 
Can these be recovered and processed?  Is it simply a matter of 
moving them into a different directory?  To me, "shunt" sounds like 
"move these here because I can't deal with them".

Thanks again

Rob



>  >>>>> "bronto" ==   <bronto@csd-bes.net> writes:
>
>     | Traceback (most recent call last):
>     |    File "/usr/local/mailman/cron/gate_news", line 44, in ?
>     |      from Mailman import MailList
>     | ValueError: bad marshal data
>
>Because you're getting this on an import, I have to conclude that your
>.pyc files are corrupt.  I've no idea why that could be (maybe you ran
>out of disk space once, or have other h/w problems).
>
>One thing you can do is to temporarily disable cron, the web
>interface, and qrunner, then cd to /usr/local/mailman and run this:
>
>% find . -name \*.pyc -print | xargs rm
>
>Then cd back to your source directory and do a "make install".  This
>will regenerate all your pyc files.  Watch carefully for any errors,
>especially during the "compileall" phase.
>
>     bronto> Since posting that, I have discovered that one of my
>     bronto> mailing lists has stopped functioning to some degree.  I
>     bronto> had set up this particular list to be a simple personal
>     bronto> archive of messages from another mailing list I subscribe
>     bronto> to.  It doesn't send anything, so I hadn't noticed any
>     bronto> problems.  While doing my mailman update to 2.1b3, I
>     bronto> discovered that no messages have been archived since the
>     bronto> 23rd when I got the above errors.  Snooping around the
>     bronto> mailman directories, I see a ton of files (several
>     bronto> hundred) in the qfiles/shunt directory, dated since 9/23.
>     bronto> There are also a few dozen from 7/2, FWIW.  I'm betting
>     bronto> this isn't a coincidence, but know nothing more than that.
>
>     bronto> Today at 12 noon, I received the following cron error:
>
>     | _________________________________________
>     | Traceback (most recent call last):
>     |    File "/usr/local/mailman/cron/gate_news", line 44, in ?
>     |      from Mailman import MailList
>     |    File "/usr/local/mailman/Mailman/MailList.py", line 31, in ?
>     |      import urllib
>     | EOFError: EOF read where object expected
>     | __________________________________________
>
>Again, that's happening on an import, but now it's while importing a
>standard Python module.  That might mean that your Python installation
>is corrupt in a similar way.
>
>     bronto> Since that time, there has been a mailman process running
>     bronto> taking up all available CPU cycles.  gtop shows that the
>     bronto> CMD is /usr/bin/python2, FWIW.  There are a fair number of
>     bronto> qrunner processes running, but my mailing lists are hardly
>     bronto> active.
>
>     bronto> What's going on here?
>
>I don't know if those other problems are related, but you first have
>to make sure those import problems go away before you can track down
>any other issues.
>
>-Barry


From barry@python.org  Wed Oct 16 03:13:19 2002
From: barry@python.org (Barry A. Warsaw)
Date: Tue, 15 Oct 2002 22:13:19 -0400
Subject: [Mailman-Developers] Cron error messages
References: <a05111b0eb9cbb17782e6@[10.1.19.62]>
	<15788.32160.73629.424590@gargle.gargle.HOWL>
	<a05111b00b9d26d45bbac@[10.1.19.62]>
Message-ID: <15788.52031.724617.30867@gargle.gargle.HOWL>


>>>>> "bronto" ==   <bronto@csd-bes.net> writes:

    bronto> Since the original "bad marshal data" problem, I've
    bronto> upgraded to v2.1b3, so unless make install does not
    bronto> replace existing files I think my mailman installation
    bronto> should be OK now (?).  I will re-install python2 to make
    bronto> sure it's OK as well.

You should be ok, since the installation process forces a recompile of
all the pyc files.

    bronto> The remaining question is that of all the files in the
    bronto> shunt directory.  These must represent missing posts from
    bronto> the mailing list. Can these be recovered and processed?
    bronto> Is it simply a matter of moving them into a different
    bronto> directory?  To me, "shunt" sounds like "move these here
    bronto> because I can't deal with them".

Correct!  It shunt is there as a failsafe, so when an uncaught
exception occurs during the processing of a message, at least you
don't lose it -- there's a chance that the bug will be fixed and you
can "unshunt" the message.

Actually, there's a handy script, bin/unshunt that moves all the
messages from qfiles/shunt to the proper directory.  I run that
several times a day while I'm developing new code. :)

Just run bin/unshunt and watch your log files.

-Barry

From chuqui@plaidworks.com  Wed Oct 16 05:22:27 2002
From: chuqui@plaidworks.com (Chuq Von Rospach)
Date: Tue, 15 Oct 2002 21:22:27 -0700
Subject: [Mailman-Developers] oops. One more buglet.
Message-ID: <E11FB554-E0BE-11D6-9047-0003934516A8@plaidworks.com>


okay, updated to CVS. Updated. turned on personalization and...

it all works! (woo hoo!)

but there seems to be one more glitch. Posted to a list, got the 
response. Did a reply-all ot that message, and sent it back to the 
list. and I got...

From: Chuq Von Rospach <chuqui@plaidworks.com>
Date: Tue Oct 15, 2002  9:11:40  PM US/Pacific
To: chuqui@plaidworks.com
Cc: "Yes, we have a xxxxx!" <xxxx@plaidworks.com>
Cc: "Yes, we have a xxxxx!" <xxxx@plaidworks.com>
Subject: Re: yes! another test!

Yup. Two CC: lines to the list.

So, how should we handle this? If a CC: to the list already exists, 
don't add another? Should we have two CC: lines anyway? (is that legal 
by the RFC?). Merge al CC:s into one line? How will mailman handle an 
existing CC?

god, so many questions....



-- 
Chuq Von Rospach, Architech
chuqui@plaidworks.com -- http://www.plaidworks.com/chuqui/blog/

Yes, I am an agent of Satan, but my duties
are largely ceremonial.


From barry@python.org  Wed Oct 16 05:47:41 2002
From: barry@python.org (Barry A. Warsaw)
Date: Wed, 16 Oct 2002 00:47:41 -0400
Subject: [Mailman-Developers] oops. One more buglet.
References: <E11FB554-E0BE-11D6-9047-0003934516A8@plaidworks.com>
Message-ID: <15788.61293.814067.914641@gargle.gargle.HOWL>


>>>>> "CVR" == Chuq Von Rospach <chuqui@plaidworks.com> writes:

    CVR> Yup. Two CC: lines to the list.

Oh dang.

    CVR> So, how should we handle this? If a CC: to the list already
    CVR> exists, don't add another? Should we have two CC: lines
    CVR> anyway? (is that legal by the RFC?). Merge al CC:s into one
    CVR> line? How will mailman handle an existing CC?

RFC 2822 says zero or one CC header, so we just need to merge and zap
duplicates.  It's a bit of code, but not insurmountabe.  I'll check in
a fix.

Thanks,
-Barry

From noreply@sourceforge.net  Wed Oct 16 07:30:14 2002
From: noreply@sourceforge.net (noreply@sourceforge.net)
Date: Tue, 15 Oct 2002 23:30:14 -0700
Subject: [Mailman-Developers] 
 [ mailman-Bugs-609692 ] Non-member hold reason is wrong
Message-ID: <E181hh8-000226-00@usw-sf-web5.sourceforge.net>

Bugs item #609692, was opened at 2002-09-15 22:07
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=100103&aid=609692&group_id=103

Category: mail delivery
Group: 2.1 beta
Status: Open
Resolution: None
Priority: 5
Submitted By: Les Niles (lniles)
Assigned to: Nobody/Anonymous (nobody)
Summary: Non-member hold reason is wrong

Initial Comment:
Maybe I'm missing something, but it would seem 
that Handlers.Moderate.process() should call 
hold.hold_for_approval() with a fourth arg of 
Hold.NonMemberPost instead of Hold.ModeratedPost 
at line 85.  The latter causes confusion for the typical 
case of a member (person) posting from other than 
her subscribed email address, since she gets back a 
message saying that the post was to a moderated 
list.  And the default reject message for the 
moderator, something about the posting being 
deemed inappropriate, is not appropriate. :)


----------------------------------------------------------------------

Comment By: Tokio Kikuchi (tkikuchi)
Date: 2002-10-16 06:30

Message:
Logged In: YES 
user_id=67709

I am sending a patch to raise attention: sorry for the
inconvenience with the folded lines but only two lines. (TK)

--- /home/mailman/src/mailman/Mailman/Handlers/Moderate.py 
    Wed Jul 10 04:29
:53 2002
+++ Moderate.py Wed Oct 16 15:25:26 2002
@@ -82,7 +82,7 @@
     if matches_p(sender, mlist.accept_these_nonmembers):
         return
     if matches_p(sender, mlist.hold_these_nonmembers):
-        Hold.hold_for_approval(mlist, msg, msgdata,
Hold.ModeratedPost)
+        Hold.hold_for_approval(mlist, msg, msgdata,
Hold.NonMemberPost)
         # No return
     if matches_p(sender, mlist.reject_these_nonmembers):
         do_reject(mlist)
@@ -98,7 +98,7 @@
         # Accept
         return
     elif mlist.generic_nonmember_action == 1:
-        Hold.hold_for_approval(mlist, msg, msgdata,
Hold.ModeratedPost)
+        Hold.hold_for_approval(mlist, msg, msgdata,
Hold.NonMemberPost)
     elif mlist.generic_nonmember_action == 2:
         do_reject(mlist)
     elif mlist.generic_nonmember_action == 3:


----------------------------------------------------------------------

Comment By: Les Niles (lniles)
Date: 2002-09-18 16:56

Message:
Logged In: YES 
user_id=92931

Oops -- Hold.NonMemberPost should replace 
Hold.ModeratedPost at line 101, not line 85.  (This is 
in Sept 3 2002 CVS.)


----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=100103&aid=609692&group_id=103

From mjm@michaelmeltzer.com  Wed Oct 16 09:23:47 2002
From: mjm@michaelmeltzer.com (Michael Meltzer)
Date: Wed, 16 Oct 2002 04:23:47 -0400
Subject: [Mailman-Developers] attachments to links :-)
References: 
	<001301c27470$45b898c0$34f820c0@ix1x1000><15788.22497.311441.176991@gargle.gargle.HOWL>
	<005c01c2747c$bd8f3140$34f820c0@ix1x1000>
Message-ID: <02c601c274ed$594cc5f0$0b01a8c0@mjm2>

Had some heartburn trying to code if mm_cfg.attach_filter does not exist leave it alone, never quit got it, but The code is tested
and working.

MJM


in mm_cfg.py
attach_filter = ['image/bmp', 'image/jpeg', 'image/tiff', 'image/gif', 'image/png', 'image/pjpeg', 'image/x-png', 'image/x-wmf']

Index: MimeDel.py
===================================================================
RCS file: /cvsroot/mailman/mailman/Mailman/Handlers/MimeDel.py,v
retrieving revision 2.5
diff -u -r2.5 MimeDel.py
--- MimeDel.py 8 Oct 2002 00:02:00 -0000 2.5
+++ MimeDel.py 16 Oct 2002 08:16:28 -0000
@@ -35,6 +35,8 @@
 from Mailman.Queue.sbcache import get_switchboard
 from Mailman.Logging.Syslog import syslog
 from Mailman.Version import VERSION
+from Mailman.Handlers.Scrubber import save_attachment
+from time import strftime
 from Mailman.i18n import _


@@ -45,6 +47,8 @@
         return
     if msgdata.get('isdigest'):
         return
+
+    make_attachment(mlist, msg)
     # We also don't care about our own digests or plaintext
     ctype = msg.get_content_type()
     mtype = msg.get_content_maintype()
@@ -64,7 +68,7 @@
     if msg.is_multipart():
         # Recursively filter out any subparts that match the filter list
         prelen = len(msg.get_payload())
-        filter_parts(msg, filtertypes, passtypes)
+        filter_parts(mlist, msg, filtertypes, passtypes)
         # If the outer message is now an empty multipart (and it wasn't
         # before!) then, again it gets discarded.
         postlen = len(msg.get_payload())
@@ -121,7 +125,7 @@



-def filter_parts(msg, filtertypes, passtypes):
+def filter_parts(mlist, msg, filtertypes, passtypes):
     # Look at all the message's subparts, and recursively filter
     if not msg.is_multipart():
         return 1
@@ -129,9 +133,12 @@
     prelen = len(payload)
     newpayload = []
     for subpart in payload:
-        keep = filter_parts(subpart, filtertypes, passtypes)
+        keep = filter_parts(mlist, subpart, filtertypes, passtypes)
         if not keep:
             continue
+        if make_attachment(mlist, subpart):
+            newpayload.append(subpart)
+            continue
         ctype = subpart.get_content_type()
         mtype = subpart.get_content_maintype()
         if ctype in filtertypes or mtype in filtertypes:
@@ -218,3 +225,33 @@
         badq.enqueue(msg, msgdata)
     # Most cases also discard the message
     raise Errors.DiscardMessage
+
+
+
+def make_attachment(mlist, subpart):
+     #should be set from mlist, work in progress
+     #BTW this will act real stupid with mulipart, it need the real object not the house keeping
+    if not mm_cfg.attach_filter:
+        return 0
+    ctype = subpart.get_content_type()
+    mtype = subpart.get_content_maintype()
+    if ctype in mm_cfg.attach_filter or mtype in mm_cfg.attach_filter:
+        #size is off, just could not stand to call decode to correct, might just take off 20% and be done
+        size = len(subpart.get_payload())
+        desc = subpart.get('content-description', (_('not available')))
+        filename = subpart.get_filename(_('not available'))
+        url = save_attachment(mlist, subpart, strftime("attch/%Y%m/%d"))
+        del subpart['content-type']
+        del subpart['content-transfer-encoding']
+        del subpart['content-disposition']
+        del subpart['content-description']
+        subpart.add_header('Content-Type', 'text/plain', charset='us-ascii')
+        subpart.add_header('Content-Transfer-Encoding', '7bit')
+        subpart.add_header('Content-disposition', 'inline')
+        subpart.set_payload(_("""\
+Name: %(filename)s Type: %(ctype)s Size: %(size)d bytes Desc: %(desc)s
+Url: %(url)s
+"""))
+        return 1
+    else:
+        return 0

Index: private.py
===================================================================
RCS file: /cvsroot/mailman/mailman/Mailman/Cgi/private.py,v
retrieving revision 2.16
diff -u -r2.16 private.py
--- private.py 22 May 2002 03:00:19 -0000 2.16
+++ private.py 16 Oct 2002 08:18:10 -0000
@@ -20,6 +20,7 @@
 import sys
 import os
 import cgi
+import mimetypes

 from Mailman import mm_cfg
 from Mailman import Utils
@@ -46,6 +47,8 @@
 def content_type(path):
     if path[-3:] == '.gz':
         path = path[:-3]
+    if mm_cfg.attach_filter and mimetypes.guess_type(os.path.splitext(path)[1])[0] in mm_cfg.attach_filter:
+        return mimetypes.guess_type(os.path.splitext(path)[1])[0]
     if path[-4:] == '.txt':
         return 'text/plain'
     return 'text/html'





From gward@python.net  Wed Oct 16 15:03:09 2002
From: gward@python.net (Greg Ward)
Date: Wed, 16 Oct 2002 10:03:09 -0400
Subject: [Mailman-Developers] Latest tweaks to README.EXIM
Message-ID: <20021016140309.GA4992@cthulhu.gerg.ca>

Barry -- Nigel M. suggested a couple of tweaks to README.EXIM on
exim-users; here's the patch.  Not tested yet.

--- README.EXIM	15 Oct 2002 22:14:16 -0000	2.5
+++ README.EXIM	16 Oct 2002 14:01:23 -0000
@@ -26,10 +26,10 @@
 The drawback of this configuration is that it will work poorly on
 systems supporting lists in several different mail domains.  While
 Mailman handles virtual domains, it does not yet support having two
-lists with the same name in different virtual domains, using the same
-Mailman installation.  This will eventually change.  (But see below for
-a variation on this scheme that should accommodate virtual domains
-better.)
+distinct lists with the same name in different virtual domains, using
+the same Mailman installation.  This will eventually change.  (But see
+below for a variation on this scheme that should accommodate virtual
+domains better.)
 
 The configuration file excerpts below are for use in an already
 functional Exim configuration, which accepts mail for the domain in
@@ -38,10 +38,10 @@
 
   * add the list domain, "my.list.domain" to local_domains
 
-  * add a "domains=my.list.domain" option to each of the directors
-    (routers) for the list
+  * add a "domains=my.list.domain" option to the director
+    (router) for the list
 
-  * (optional) exclude that domain from the other directors (routers)
+  * (optional) exclude that domain from your other directors (routers)
 
 [Note: the instructions in this document should work with either Exim 3
 or Exim 4.  In Exim 3, you must have a 'local_domains' configuration
@@ -53,8 +53,8 @@
 
 Whether you are using Exim 3 or Exim 4, you will need to add some macros
 to the main section of your Exim config file.  You will also need to
-define one new transport.  With Exim 3, you'll need to add two new
-directors; with Exim 4, two new routers play the same role.
+define one new transport.  With Exim 3, you'll need to add a new
+director; with Exim 4, a new router plays the same role.
 
 Finally, the configuration supplied here should allow co-habiting
 Mailman 2.0 and 2.1 installations, with the proviso that you'll probably
@@ -66,7 +66,7 @@
 ~~~~~~~~~~~~~~~~~~~~~~~~~~~
 
 First, you need to add some macros to the top of your Exim config file.
-These just make the directors (routers) and transport below a bit
+These just make the director (router) and transport below a bit
 cleaner.  Obviously, you'll need to edit these based on how you
 configured and installed Mailman.
 
@@ -103,53 +103,41 @@
 please let me know if it works?)
 
 
-Directors for Exim 3
-~~~~~~~~~~~~~~~~~~~~
+Director for Exim 3
+~~~~~~~~~~~~~~~~~~~
 
-If you're using Exim 3, you'll need to add the following two directors
-to your config file (directors go between the second and third "end"
+If you're using Exim 3, you'll need to add the following director to
+your config file (directors go between the second and third "end"
 lines).  Also, don't forget that order matters -- e.g. you can make
-Mailman lists take precedence over system aliases by putting these
-directors in front of your aliasfile director, or vice-versa.
-
-  # Director to handle a list's main posting address.  Should come first
-  # in case there's a list named, e.g. "mailman-owner".  Automatically
-  # detects list existence by looking for lists/$local_part/config.pck
-  # under MAILMAN_HOME.
-  mailman_main_director:
-    driver = smartuser
-    require_files = MAILMAN_HOME/lists/$local_part/config.pck
-    transport = mailman_transport
+Mailman lists take precedence over system aliases by putting this
+director in front of your aliasfile director, or vice-versa.
 
-  # Handle all the other addresses related to a list: -bounces, -admin,
-  # etc.  Same trick works to detect list existence thanks to the
-  # 'suffix' option.
+  # Handle all addresses related to a list 'foo': the posting address.
+  # Automatically detects list existence by looking
+  # for lists/$local_part/config.pck under MAILMAN_HOME.
   mailman_director:
     driver = smartuser
     require_files = MAILMAN_HOME/lists/$local_part/config.pck
+    suffix_optional
     suffix = -bounces : -bounces+* : \
              -confirm+* : -join : -leave : \
              -owner : -request : -admin
     transport = mailman_transport
 
 
-Routers for Exim 4
-~~~~~~~~~~~~~~~~~~
+Router for Exim 4
+~~~~~~~~~~~~~~~~~
 
-In Exim 4, there's no such thing as directors -- you instead need to add
-two routers.  Also, the canonical order of the configuration file was
-changed so routers come before transports, so the routers for Exim 4
-come first here.  Put these two routers somewhere after the "begin
-routers" line of your config file, and remember that order matters.
-
-  mailman_main_router:
-    driver = accept
-    require_files = MAILMAN_HOME/lists/$local_part/config.pck
-    transport = mailman_transport
+In Exim 4, there's no such thing as directors -- you need to add a new
+router instead.  Also, the canonical order of the configuration file was
+changed so routers come before transports, so the router for Exim 4
+comes first here.  Put this router somewhere after the "begin routers"
+line of your config file, and remember that order matters.
 
   mailman_router:
     driver = accept
     require_files = MAILMAN_HOME/lists/$local_part/config.pck
+    local_part_suffix_optional
     local_part_suffix = -bounces : -bounces+* : \
                         -confirm+* : -join : -leave : \
                         -owner : -request : -admin
@@ -306,8 +294,7 @@
   VERP_DELIVERY_INTERVAL = 1
   VERP_CONFIRMATIONS = 1
 
-(The directors/routers above are smart enough to deal with VERP
-bounces.)
+(The director (router) above is smart enough to deal with VERP bounces.)
 
        
 Virtual Domains
@@ -319,7 +306,7 @@
 the same machine.)
 
 In this case, the MAILMAN_HOME and MAILMAN_WRAP macros are useless --
-you can remove them.  Change your directors (routers) to something like
+you can remove them.  Change your director (router) to something like
 this:
 
   require_files = /virtual/${domain}/mailman/lists/${lc:$local_part}/config.pck



-- 
Greg Ward <gward@python.net>                         http://www.gerg.ca/
And now for something completely different.

From noreply@sourceforge.net  Wed Oct 16 15:05:35 2002
From: noreply@sourceforge.net (noreply@sourceforge.net)
Date: Wed, 16 Oct 2002 07:05:35 -0700
Subject: [Mailman-Developers] [ mailman-Patches-624112 ] mail password
	reminder bug
Message-ID: <E181onn-0000to-00@usw-sf-web1.sourceforge.net>

Patches item #624112, was opened at 2002-10-16 14:05
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=300103&aid=624112&group_id=103

Category: mail delivery
Group: Mailman 2.0.x
Status: Open
Resolution: None
Priority: 5
Submitted By: Marcin Deranek (marcin_deranek)
Assigned to: Nobody/Anonymous (nobody)
Summary: mail password reminder bug

Initial Comment:
cron/mailpasswds script sends improper headers (Sender,
Errors-To, X-BeenThere, Return-Path). It uses headers
of first publicly available mailing list.
My patch uses MAILMAN_OWNER variable from config file
as sender of password reminder. Patch was created for
version 2.0.13.


----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=300103&aid=624112&group_id=103

From stoney@sb.org  Wed Oct 16 18:38:38 2002
From: stoney@sb.org (Stonewall Ballard)
Date: Wed, 16 Oct 2002 13:38:38 -0400
Subject: [Mailman-Developers] Bounce with verp on not recognized
Message-ID: <B9D31C5E.48F0C%stoney@sb.org>

I just updated to the latest CVS version, turned on personalization and
subscribed a bunch of people. I got an "Uncaught Bounce Notification" back,
but as far as I can tell, there was no personalization on the reply-to
address going out.

I put this in my mm_cfg.py:

VERP_PASSWORD_REMINDERS = 1
VERP_PERSONALIZED_DELIVERIES = 1
VERP_CONFIRMATIONS = 1
OWNERS_CAN_ENABLE_PERSONALIZATION = 1

It doesn't appear from the MTA logs that the envelope sender was munged to
contain the verp extension. I'm not certain of this, not having the original
outgoing message, but I believe that the MTA (Communigate Pro) will not do
more than reject a message when it has a problem with the envelope from
address.

Should I have expected that welcome messages would contain verp'd reply-to's
or is this likely to be a misconfiguration somewhere on my part?

BTW, this is a _wonderful_ feature.

 - Stoney

-- 
Stonewall Ballard 
stoney@sb.org           http://www.sb.org/stoney/



From stoney@sb.org  Wed Oct 16 20:36:28 2002
From: stoney@sb.org (Stonewall Ballard)
Date: Wed, 16 Oct 2002 15:36:28 -0400
Subject: [Mailman-Developers] VERP'd reply-to not being used by bounce
	detector
Message-ID: <B9D337FC.48F97%stoney@sb.org>

I'm using Mailman 2.1b+, from a CVS build earlier today.

I sent a batch of messages out, and got a dozen bounces. Each one was
forwarded to me as "uncaught", but the reply-to had the verp'd address in
it.

An example bounce to: address is
reannounce-bounces+jron=rcn.net@lists.firstparish.org

Is the bounce detector simply not coded to handle these yet?

This also indicates that Welcome messages should have the VERP extension,
but don't (see my earlier message).

 - Stoney

-- 
Stonewall Ballard 
stoney@sb.org           http://www.sb.org/stoney/



From chuqui@plaidworks.com  Wed Oct 16 22:11:15 2002
From: chuqui@plaidworks.com (Chuq Von Rospach)
Date: Wed, 16 Oct 2002 14:11:15 -0700
Subject: [Mailman-Developers] encoding the subject line..
Message-ID: <CE7F5E09-E14B-11D6-9047-0003934516A8@plaidworks.com>


Barry, why is mailman encoding the subject line when it doesn't have to?

In-Reply-To: <4.2.0.58.20021016132636.00a99290@mail.claritycoaching.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Subject: =?iso-8859-1?q?10/19_game_moved_to_1pm=2E=2E=2E?=
X-BeenThere: sharks@plaidworks.com
X-Mailman-Version: 2.1b3+
Precedence: list


-- 
Chuq Von Rospach, Architech
chuqui@plaidworks.com -- http://www.plaidworks.com/chuqui/blog/

He doesn't have ulcers, but he's a carrier.


From che@debian.org  Wed Oct 16 22:59:47 2002
From: che@debian.org (Ben Gertzfield)
Date: Wed, 16 Oct 2002 14:59:47 -0700
Subject: [Mailman-Developers] encoding the subject line..
References: <CE7F5E09-E14B-11D6-9047-0003934516A8@plaidworks.com>
Message-ID: <3DADE153.5000808@debian.org>

Chuq Von Rospach wrote:

> Barry, why is mailman encoding the subject line when it doesn't have to?
>
> Subject: =?iso-8859-1?q?10/19_game_moved_to_1pm=2E=2E=2E?=


This is a message that came from outside?  How are you sure that it 
wasn't encoded in the first place?  

Mailman doesn't (as far as I know) ever ADD charset information to a 
message, so if it didn't come in as iso-8859-1, and if this message was 
not internally generated from a (non-English) Mailman list, I don't see 
how the iso-8859-1 charset information got attached.

Ben





From chuqui@plaidworks.com  Wed Oct 16 23:22:25 2002
From: chuqui@plaidworks.com (Chuq Von Rospach)
Date: Wed, 16 Oct 2002 15:22:25 -0700
Subject: [Mailman-Developers] encoding the subject line..
In-Reply-To: <3DADE153.5000808@debian.org>
Message-ID: <BF978DF8-E155-11D6-9047-0003934516A8@plaidworks.com>


On Wednesday, October 16, 2002, at 02:59  PM, Ben Gertzfield wrote:

> This is a message that came from outside?  How are you sure that it 
> wasn't encoded in the first place?


as far as I know, it wasn't.

went and checked. here's the subject for that message as it is in the 
mbox format file I use to archive all incoming mail (makes debugging 
MUCH easier)

Subject: Re: 10/19 game moved to 1pm...

his stuff is NOT encoded coming in the subject line. his MIME is:

Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"; format=flowed

not 8859. So Mailman is making that change to the subject line. I 
believe it started when I turned on personalization last night after I 
did the latest CVS update with Barry's fixes. Don't know if it did it 
before, but it wasn't encoding this way yesterday.

-- 
Chuq Von Rospach, Architech
chuqui@plaidworks.com -- http://www.plaidworks.com/chuqui/blog/

IMHO: Jargon. Acronym for In My Humble Opinion. Used to flag as an 
opinion
something that is clearly from context an opinion to everyone except the
mentally dense. Opinions flagged by IMHO are actually rarely humble. 
IMHO.
     (source: third unabridged dictionary of chuqui-isms).


From chuqui@plaidworks.com  Wed Oct 16 23:25:03 2002
From: chuqui@plaidworks.com (Chuq Von Rospach)
Date: Wed, 16 Oct 2002 15:25:03 -0700
Subject: [Mailman-Developers] encoding the subject line..
In-Reply-To: <3DADE153.5000808@debian.org>
Message-ID: <1DE5E538-E156-11D6-9047-0003934516A8@plaidworks.com>


FYI, I just sent a test.

Subject: =?iso-8859-1?q?test=2C_one_two?=


*I* definitely didn't encode in 8859 going out, but it came back that 
way. the body is still in us-ascii, so this is something specific to 
the subject line, and it seems to be turned on when I turn on 
personalization.


-- 
Chuq Von Rospach, Architech
chuqui@plaidworks.com -- http://www.plaidworks.com/chuqui/blog/

He doesn't have ulcers, but he's a carrier.


From dagilbert@raytheon.com  Wed Oct 16 23:43:33 2002
From: dagilbert@raytheon.com (David A Gilbert)
Date: Wed, 16 Oct 2002 15:43:33 -0700
Subject: [Mailman-Developers] MHonArc integration with Mailman
Message-ID: <OFBA5AFBFF.84A521E1-ON88256C54.007B0D5E@mck.us.ray.com>

Hi all,

Does anyone know if there is a patch, a HOW-TO, or something that explains
how to integrate MHonArc with Mailman? I've checked the archives of the
users and developers lists as well as the FAQs at http://www.list.org/ and
http://www.python.org/cgi-bin/faqw-mm.py. At the latter FAQ , I found one
called:
4.4. How can I use an external archiver with Mailman like MHonArc? that
referenced a HOW-TO in the mailman-developers list archives but I could not
find the referenced HOW-TO or anything like it in the archives.

Any help much appreciated, thanks.

Regards,
David



From barry@python.org  Thu Oct 17 00:00:43 2002
From: barry@python.org (Barry A. Warsaw)
Date: Wed, 16 Oct 2002 19:00:43 -0400
Subject: [Mailman-Developers] encoding the subject line..
References: <3DADE153.5000808@debian.org>
	<BF978DF8-E155-11D6-9047-0003934516A8@plaidworks.com>
Message-ID: <15789.61339.346705.414151@gargle.gargle.HOWL>


>>>>> "CVR" == Chuq Von Rospach <chuqui@plaidworks.com> writes:

    CVR> not 8859. So Mailman is making that change to the subject
    CVR> line. I believe it started when I turned on personalization
    CVR> last night after I did the latest CVS update with Barry's
    CVR> fixes. Don't know if it did it before, but it wasn't encoding
    CVR> this way yesterday.

I can't reproduce this by sending a message with a subject line

    Subject: 10/19 game moved to 1pm...

to an English list, where the user has English as their language.  The
Subject comes through unencoded.

What's your subject_prefix set to?  Actually, do this:

% bin/config_list -o - mylist | grep subject_prefix
subject_prefix = '[My List] '

-Barry

From chuqui@plaidworks.com  Thu Oct 17 00:09:21 2002
From: chuqui@plaidworks.com (Chuq Von Rospach)
Date: Wed, 16 Oct 2002 16:09:21 -0700
Subject: [Mailman-Developers] encoding the subject line..
In-Reply-To: <15789.61339.346705.414151@gargle.gargle.HOWL>
Message-ID: <4DC2F634-E15C-11D6-9047-0003934516A8@plaidworks.com>


On Wednesday, October 16, 2002, at 04:00  PM, Barry A. Warsaw wrote:

> bin/config_list -o - mylist | grep subject_prefix

subject_prefix = ''


It's happening on all of my lists, FWIW. And I don't think it's 
sendmail, because the subject isn't encoded when I receive it, but it 
seems to be encoded leaving mailman.

weird.

-- 
Chuq Von Rospach, Architech
chuqui@plaidworks.com -- http://www.plaidworks.com/chuqui/blog/

Someday, we'll look back on this, laugh
nervously and change the subject.


From chuqui@plaidworks.com  Thu Oct 17 00:10:37 2002
From: chuqui@plaidworks.com (Chuq Von Rospach)
Date: Wed, 16 Oct 2002 16:10:37 -0700
Subject: [Mailman-Developers] encoding the subject line..
In-Reply-To: <15789.61339.346705.414151@gargle.gargle.HOWL>
Message-ID: <7B3B365C-E15C-11D6-9047-0003934516A8@plaidworks.com>


On Wednesday, October 16, 2002, at 04:00  PM, Barry A. Warsaw wrote:
>

> to an English list, where the user has English as their language.  The
> Subject comes through unencoded.
>

Stupid comment: are you sure? Because I didn't know this until a user 
using an explicitly non-MIME mailer wrote to complain, and I started 
looking at email in "raw source" mode. All I see is the subject line 
looking just fine, because my client interpreted and cleaned up the 
encoding for me...


-- 
Chuq Von Rospach, Architech
chuqui@plaidworks.com -- http://www.plaidworks.com/chuqui/blog/

No! No! Dead girl, OFF the table! -- Shrek


From sb.list@sb.org  Thu Oct 17 00:20:45 2002
From: sb.list@sb.org (Stonewall Ballard)
Date: Wed, 16 Oct 2002 19:20:45 -0400
Subject: [Mailman-Developers] encoding the subject line..
In-Reply-To: <7B3B365C-E15C-11D6-9047-0003934516A8@plaidworks.com>
Message-ID: <B9D36C8D.49052%sb.list@sb.org>

On 10/16/02 7:10 PM, "Chuq Von Rospach" <chuqui@plaidworks.com> wrote:

> Stupid comment: are you sure? Because I didn't know this until a user
> using an explicitly non-MIME mailer wrote to complain, and I started
> looking at email in "raw source" mode. All I see is the subject line
> looking just fine, because my client interpreted and cleaned up the
> encoding for me...

I've had a couple mailings go out today after I converted to personalized,
and I'm not seeing this (in raw mode).

For example:

  Subject: [REAnnounce] City Year - 10/26/02
  Subject: [Activism] FYI- Forum Oct 20 on African Debt Relief

 - Stoney


From chuqui@plaidworks.com  Thu Oct 17 00:26:39 2002
From: chuqui@plaidworks.com (Chuq Von Rospach)
Date: Wed, 16 Oct 2002 16:26:39 -0700
Subject: [Mailman-Developers] encoding the subject line..
In-Reply-To: <B9D36C8D.49052%sb.list@sb.org>
Message-ID: <B88A367D-E15E-11D6-9047-0003934516A8@plaidworks.com>


On Wednesday, October 16, 2002, at 04:20  PM, Stonewall Ballard wrote:

> I've had a couple mailings go out today after I converted to 
> personalized,
> and I'm not seeing this (in raw mode).
>

weird. Barry, if you want my full mm_cfi and a config_list to see how 
I'm set up, let me know.



-- 
Chuq Von Rospach, Architech
chuqui@plaidworks.com -- http://www.plaidworks.com/chuqui/blog/

IMHO: Jargon. Acronym for In My Humble Opinion. Used to flag as an 
opinion
something that is clearly from context an opinion to everyone except the
mentally dense. Opinions flagged by IMHO are actually rarely humble. 
IMHO.
     (source: third unabridged dictionary of chuqui-isms).


From tkikuchi@is.kochi-u.ac.jp  Thu Oct 17 01:07:50 2002
From: tkikuchi@is.kochi-u.ac.jp (Tokio Kikuchi)
Date: Thu, 17 Oct 2002 09:07:50 +0900
Subject: [Mailman-Developers] encoding the subject line..
References: <CE7F5E09-E14B-11D6-9047-0003934516A8@plaidworks.com>
Message-ID: <3DADFF56.9050907@is.kochi-u.ac.jp>


Chuq Von Rospach wrote:
> 
> Barry, why is mailman encoding the subject line when it doesn't have to?

Why? because the recipient prefers it.

The member has set his preferred language as non-english.
You will see this on a non-personalized list if it's language
has set non-english.

The subject is re-generated within mailman because it is munged,
while the charset (content-type header) are kept as it was.

May be email should not encode the header if the string doesn't
contain non-ascii charcter.

> 
> In-Reply-To: <4.2.0.58.20021016132636.00a99290@mail.claritycoaching.com>
> MIME-Version: 1.0
> Content-Type: text/plain; charset=us-ascii
> Subject: =?iso-8859-1?q?10/19_game_moved_to_1pm=2E=2E=2E?=
> X-BeenThere: sharks@plaidworks.com
> X-Mailman-Version: 2.1b3+
> Precedence: list
> 
> 


-- 
Tokio Kikuchi, tkikuchi@ is.kochi-u.ac.jp
http://weather.is.kochi-u.ac.jp/


From che@debian.org  Thu Oct 17 01:14:53 2002
From: che@debian.org (Ben Gertzfield)
Date: Wed, 16 Oct 2002 17:14:53 -0700
Subject: [Mailman-Developers] encoding the subject line..
References: <CE7F5E09-E14B-11D6-9047-0003934516A8@plaidworks.com>
	<3DADFF56.9050907@is.kochi-u.ac.jp>
Message-ID: <3DAE00FD.4080805@debian.org>

Tokio Kikuchi wrote:

> May be email should not encode the header if the string doesn't
> contain non-ascii charcter.


What happens to iso-2022-jp strings, then? :)  They have no non-ascii 
characters. (Here we go again..)

Ben



From chuqui@plaidworks.com  Thu Oct 17 01:17:08 2002
From: chuqui@plaidworks.com (Chuq Von Rospach)
Date: Wed, 16 Oct 2002 17:17:08 -0700
Subject: [Mailman-Developers] encoding the subject line..
In-Reply-To: <3DADFF56.9050907@is.kochi-u.ac.jp>
Message-ID: <C609779C-E165-11D6-9047-0003934516A8@plaidworks.com>


On Wednesday, October 16, 2002, at 05:07  PM, Tokio Kikuchi wrote:

>
> Chuq Von Rospach wrote:
>> Barry, why is mailman encoding the subject line when it doesn't have 
>> to?
>
> Why? because the recipient prefers it.
>
> The member has set his preferred language as non-english.
> You will see this on a non-personalized list if it's language
> has set non-english.

Except I'm seeing it and my preferred language is US English. I just 
went and double-checked. And the user that reported it to me is, too.

So that might be how it's supposed to work, but it doesn't seem quite 
right on my site.

-- 
Chuq Von Rospach, Architech
chuqui@plaidworks.com -- http://www.plaidworks.com/chuqui/blog/

The first rule of holes: If you are in one, stop digging.


From tkikuchi@is.kochi-u.ac.jp  Thu Oct 17 01:20:48 2002
From: tkikuchi@is.kochi-u.ac.jp (Tokio Kikuchi)
Date: Thu, 17 Oct 2002 09:20:48 +0900
Subject: [Mailman-Developers] encoding the subject line..
References: <CE7F5E09-E14B-11D6-9047-0003934516A8@plaidworks.com>
	<3DADFF56.9050907@is.kochi-u.ac.jp> <3DAE00FD.4080805@debian.org>
Message-ID: <3DAE0260.7070702@is.kochi-u.ac.jp>

Ben Gertzfield wrote:
> Tokio Kikuchi wrote:
> 
>> May be email should not encode the header if the string doesn't
>> contain non-ascii charcter.
> 
> 
> 
> What happens to iso-2022-jp strings, then? :)  They have no non-ascii 
> characters. (Here we go again..)

OK. non-ascii-printable characters. ESC is not printable, I guess.;-)



-- 
Tokio Kikuchi, tkikuchi@ is.kochi-u.ac.jp
http://weather.is.kochi-u.ac.jp/


From tkikuchi@is.kochi-u.ac.jp  Thu Oct 17 01:45:39 2002
From: tkikuchi@is.kochi-u.ac.jp (Tokio Kikuchi)
Date: Thu, 17 Oct 2002 09:45:39 +0900
Subject: [Mailman-Developers] encoding the subject line..
References: <C609779C-E165-11D6-9047-0003934516A8@plaidworks.com>
Message-ID: <3DAE0833.5050005@is.kochi-u.ac.jp>


>>> Barry, why is mailman encoding the subject line when it doesn't have to?
>>
>>
>> Why? because the recipient prefers it.
>>
>> The member has set his preferred language as non-english.

Some time ago, in CookHeader.py, english is determined to use
'iso-8859-1' because many like to use 8bit character in
us-ascii declared mail.

Again, I think email module should be revised.
I had a little experiment

in ~/pythonlib (email version is 2.4.3)

Python 2.1.3 (#1, Sep 19 2002, 17:00:05)
[GCC 2.95.2 19991024 (release)] on freebsd4
Type "copyright", "credits" or "license" for more information.
 >>> from email.Header import Header
 >>> a = Header('abc','iso-8859-1',70,'Subject')
 >>> print a
=?iso-8859-1?q?abc?=
 >>>

Or, is this version problem? I'm using python 2.1.3 (for Zope's sake).

-- 
Tokio Kikuchi, tkikuchi@ is.kochi-u.ac.jp
http://weather.is.kochi-u.ac.jp/


From philb@philb.us  Thu Oct 17 07:48:20 2002
From: philb@philb.us (Phil Barnett)
Date: Thu, 17 Oct 2002 02:48:20 -0400
Subject: [Mailman-Developers] qrunner
Message-ID: <200210170248.20057.philb@philb.us>


What's the downside of qrunner not running each minute?

Would every 10 minutes be less cpu intensive and still generally effectiv=
e?

What am I trading off?


From claw@kanga.nu  Thu Oct 17 10:09:39 2002
From: claw@kanga.nu (J C Lawrence)
Date: Thu, 17 Oct 2002 02:09:39 -0700
Subject: [Mailman-Developers] MHonArc integration with Mailman 
In-Reply-To: Message from "David A Gilbert" <dagilbert@raytheon.com> 
	of "Wed, 16 Oct 2002 15:43:33 PDT."
	<OFBA5AFBFF.84A521E1-ON88256C54.007B0D5E@mck.us.ray.com> 
References: <OFBA5AFBFF.84A521E1-ON88256C54.007B0D5E@mck.us.ray.com> 
Message-ID: <5280.1034845779@kanga.nu>

On Wed, 16 Oct 2002 15:43:33 -0700 
David A Gilbert <David> wrote:

> Does anyone know if there is a patch, a HOW-TO, or something that
> explains how to integrate MHonArc with Mailman?

Loosely, there are two ways to go about it:

  1) To plug MHonArc into Mailman as an external archiver.

  2) To subscribe an archive address to each list and then have mail
  sent to that address archived as appropriate.

I do the latter.  An archiving address is subscribed to each list I want
to archive (whether I run that list locally or it remotely hosted).
Mail received by that address runs thru a procmail filter which stashes
it off in per list folders which a cronjob then periodically runs
MHonArc against.

Full details on this setup are in the FAQ.

-- 
J C Lawrence                
---------(*)                Satan, oscillate my metallic sonatas. 
claw@kanga.nu               He lived as a devil, eh?		  
http://www.kanga.nu/~claw/  Evil is a name of a foeman, as I live.


From barry@python.org  Thu Oct 17 15:47:56 2002
From: barry@python.org (Barry A. Warsaw)
Date: Thu, 17 Oct 2002 10:47:56 -0400
Subject: [Mailman-Developers] Re: Latest tweaks to README.EXIM
References: <20021016140309.GA4992@cthulhu.gerg.ca>
Message-ID: <15790.52636.726738.639461@yyz.zope.com>


Thanks, applied.
-Barry

From barry@python.org  Thu Oct 17 16:52:41 2002
From: barry@python.org (Barry A. Warsaw)
Date: Thu, 17 Oct 2002 11:52:41 -0400
Subject: [Mailman-Developers] encoding the subject line..
References: <15789.61339.346705.414151@gargle.gargle.HOWL>
	<7B3B365C-E15C-11D6-9047-0003934516A8@plaidworks.com>
Message-ID: <15790.56521.561501.859301@gargle.gargle.HOWL>


>>>>> "CVR" == Chuq Von Rospach <chuqui@plaidworks.com> writes:

    >> to an English list, where the user has English as their
    >> language.  The Subject comes through unencoded.

    CVR> Stupid comment: are you sure?

Positive.  VM has really excellent MIME support.  When an encoded
(header or body) message comes in, it displayes it to me in the
friendliest format, and the buffer is labeled "Inbox Presentation".
But I can hit `D' repeatedly to toggle through various
representations, including "give it to me, raw" which shows me the
undecoded raw message as it's sitting in my mail folder.
    
-Barry

From barry@python.org  Thu Oct 17 16:54:34 2002
From: barry@python.org (Barry A. Warsaw)
Date: Thu, 17 Oct 2002 11:54:34 -0400
Subject: [Mailman-Developers] encoding the subject line..
References: <B9D36C8D.49052%sb.list@sb.org>
	<B88A367D-E15E-11D6-9047-0003934516A8@plaidworks.com>
Message-ID: <15790.56634.232428.511629@gargle.gargle.HOWL>


>>>>> "CVR" == Chuq Von Rospach <chuqui@plaidworks.com> writes:

    >> I've had a couple mailings go out today after I converted to
    >> personalized, and I'm not seeing this (in raw mode).

    CVR> weird. Barry, if you want my full mm_cfi and a config_list to
    CVR> see how I'm set up, let me know.

That's probably the next step.  Send me the config.db file and the
mm_cfg.py file for the list.

-Barry

From barry@python.org  Thu Oct 17 16:58:10 2002
From: barry@python.org (Barry A. Warsaw)
Date: Thu, 17 Oct 2002 11:58:10 -0400
Subject: [Mailman-Developers] encoding the subject line..
References: <CE7F5E09-E14B-11D6-9047-0003934516A8@plaidworks.com>
	<3DADFF56.9050907@is.kochi-u.ac.jp>
Message-ID: <15790.56850.677692.968336@gargle.gargle.HOWL>


>>>>> "TK" == Tokio Kikuchi <tkikuchi@is.kochi-u.ac.jp> writes:

    >> Barry, why is mailman encoding the subject line when it doesn't
    >> have to?

    TK> Why? because the recipient prefers it.

Actually, the Subject header is never encoded specific to the user, so
personalization /should/ have nothing to do with it.

The Subject is munged in prefix_subject() in CookHeaders.py.  When
personalization is on, only the user's name is encoded in
verpdelivery() in SMTPDirect.py.  So the Subject is modified based on
what the list's configuration only.  We don't care what the end user's
language is for purposes of the Subject header.

-Barry

From barry@python.org  Thu Oct 17 17:41:10 2002
From: barry@python.org (Barry A. Warsaw)
Date: Thu, 17 Oct 2002 12:41:10 -0400
Subject: [Mailman-Developers] encoding the subject line..
References: <C609779C-E165-11D6-9047-0003934516A8@plaidworks.com>
	<3DAE0833.5050005@is.kochi-u.ac.jp>
Message-ID: <15790.59430.206821.692862@gargle.gargle.HOWL>


>>>>> "TK" == Tokio Kikuchi <tkikuchi@is.kochi-u.ac.jp> writes:

    >> Barry, why is mailman encoding the subject line when it doesn't
    >> have to?
    >> Why? because the recipient prefers it.  The member has set his
    >> preferred language as non-english.

    TK> Some time ago, in CookHeader.py, english is determined to use
    TK> 'iso-8859-1' because many like to use 8bit character in
    TK> us-ascii declared mail.

But that's only because Header already tries us-ascii first, so it
wouldn't make sense to include us-ascii twice in the "try to find the
best charset" loop.

    TK> Again, I think email module should be revised.
    TK> I had a little experiment

    TK> in ~/pythonlib (email version is 2.4.3)

    TK> Python 2.1.3 (#1, Sep 19 2002, 17:00:05) [GCC 2.95.2 19991024
    TK> (release)] on freebsd4 Type "copyright", "credits" or
    TK> "license" for more information.
    >> from email.Header import Header a =
    >> Header('abc','iso-8859-1',70,'Subject') print a
    TK> =?iso-8859-1?q?abc?=
    >>

    TK> Or, is this version problem? I'm using python 2.1.3 (for
    TK> Zope's sake).

No, but you're not exactly reproducing what CookHeader.py does!  In
our case, the first argument will always be a unicode string, in which
case the charset argument is just a hint.  So this is closer to what
we really do:

>>> h = Header(u'abc', 'iso-8859-1', 70, 'Subject')
>>> print h
abc

See, no encoding!

Now, I just realized that I think we're not doing the string ->
unicode conversion correctly, in CookHeaders.py, prefix_subject().  So
I'll check in a fix for that, but for me, it makes no difference for
reproducing Chuq's problem.

-Barry

From chuqui@plaidworks.com  Thu Oct 17 17:57:27 2002
From: chuqui@plaidworks.com (Chuq Von Rospach)
Date: Thu, 17 Oct 2002 09:57:27 -0700
Subject: [Mailman-Developers] encoding the subject line..
In-Reply-To: <15790.59430.206821.692862@gargle.gargle.HOWL>
Message-ID: <8413E943-E1F1-11D6-99E3-0003934516A8@plaidworks.com>


On Thursday, October 17, 2002, at 09:41  AM, Barry A. Warsaw wrote:
>
> Now, I just realized that I think we're not doing the string ->
> unicode conversion correctly, in CookHeaders.py, prefix_subject().  So
> I'll check in a fix for that, but for me, it makes no difference for
> reproducing Chuq's problem.
>

I updated to CVS and tried it, and I'm still encoded.

Subject: =?iso-8859-1?q?quick_test?=

by the way, in case it might matter:

plaidworks.com 14# python -v
# /usr/lib/python2.2/site.pyc matches /usr/lib/python2.2/site.py
import site # precompiled from /usr/lib/python2.2/site.pyc
# /usr/lib/python2.2/os.pyc matches /usr/lib/python2.2/os.py
import os # precompiled from /usr/lib/python2.2/os.pyc
import posix # builtin
# /usr/lib/python2.2/posixpath.pyc matches 
/usr/lib/python2.2/posixpath.py
import posixpath # precompiled from /usr/lib/python2.2/posixpath.pyc
# /usr/lib/python2.2/stat.pyc matches /usr/lib/python2.2/stat.py
import stat # precompiled from /usr/lib/python2.2/stat.pyc
# /usr/lib/python2.2/UserDict.pyc matches /usr/lib/python2.2/UserDict.py
import UserDict # precompiled from /usr/lib/python2.2/UserDict.pyc
Python 2.2 (#1, 07/14/02, 23:25:09)
[GCC Apple cpp-precomp 6.14] on darwin
Type "help", "copyright", "credits" or "license" for more information.





-- 
Chuq Von Rospach, Architech
chuqui@plaidworks.com -- http://www.plaidworks.com/chuqui/blog/

Stress is when you wake up screaming and you realize you haven't fallen 
asleep yet.


From barry@python.org  Thu Oct 17 18:29:57 2002
From: barry@python.org (Barry A. Warsaw)
Date: Thu, 17 Oct 2002 13:29:57 -0400
Subject: [Mailman-Developers] encoding the subject line..
References: <15790.59430.206821.692862@gargle.gargle.HOWL>
	<8413E943-E1F1-11D6-99E3-0003934516A8@plaidworks.com>
Message-ID: <15790.62357.57583.142130@gargle.gargle.HOWL>


Dang, even with your setting I still get the header unencoded.

I don't have Python 2.2[.0] or OSX around at the moment to see if
that's an issue, so I'd like you to try two things.  Save the attached
file and then do this, assuming you installed Mailman in the default
location.

% PYTHONPATH=/usr/local/mailman/pythonlib python test.py
email version: 2.4.3
email path: /usr/local/mailman/pythonlib/email/__init__.pyc
quick test
% python test.py
email version: 2.4.3
email path: /usr/local/lib/python2.2/email/__init__.pyc
quick test

-Barry

From barry@python.org  Thu Oct 17 19:10:35 2002
From: barry@python.org (Barry A. Warsaw)
Date: Thu, 17 Oct 2002 14:10:35 -0400
Subject: [Mailman-Developers] Is anyone gatewaying to a moderated newsgroup?
Message-ID: <15790.64795.122440.828483@gargle.gargle.HOWL>


I'd like to know if anybody else uses Mailman to gateway a mailing
list to a moderated newsgroup.  On python.org we gateway
python-announce-list to comp.lang.python.announce, but we have to use
SF patch #401270 to do so.

Under prodding by the FSF, I've thought about this issue some more,
and it's something I have to deal with before we can move
python-announce-list over to the MM2.1 code base.  I had a brainstorm
about how to make this much easier to support, but I'd like to bounce
this off of someone else for some sanity checking.

This seems like an esoteric subject, so I don't want to bother the
lists if no one else is doing this.  If there's interest, I'll follow
up with a message to just mailman-developers.

Thanks,
-Barry

From chuqui@plaidworks.com  Thu Oct 17 19:48:18 2002
From: chuqui@plaidworks.com (Chuq Von Rospach)
Date: Thu, 17 Oct 2002 11:48:18 -0700
Subject: [Mailman-Developers] my encoding problem: fixed!
Message-ID: <00A8B67D-E201-11D6-99E3-0003934516A8@plaidworks.com>

	
Barry, genius that he is, tracked down my encoding problem.

My mail libs were downrev, at 2.3 instead of 2.4.3

For some reason they weren't being upgraded by CVS, when they should 
have been. We still dont' know why, but we need to see if it was "just 
me", or if others are having this problem, too.

If you've been doing CVS updates, can you check src/mailman/misc and 
see what release of the email directory you have? If it's NOT 2.4.3, 
you have a similar problem and we need to know if anyone else is being 
affected, so we can find out if we have to figure out a fix.

chuq

-- 
Chuq Von Rospach, Architech
chuqui@plaidworks.com -- http://www.plaidworks.com/chuqui/blog/

IMHO: Jargon. Acronym for In My Humble Opinion. Used to flag as an 
opinion
something that is clearly from context an opinion to everyone except the
mentally dense. Opinions flagged by IMHO are actually rarely humble. 
IMHO.
     (source: third unabridged dictionary of chuqui-isms).


From sb.list@sb.org  Thu Oct 17 20:14:24 2002
From: sb.list@sb.org (Stonewall Ballard)
Date: Thu, 17 Oct 2002 15:14:24 -0400
Subject: [Mailman-Developers] my encoding problem: fixed!
In-Reply-To: <00A8B67D-E201-11D6-99E3-0003934516A8@plaidworks.com>
Message-ID: <B9D48450.49285%sb.list@sb.org>

On 10/17/02 2:48 PM, "Chuq Von Rospach" <chuqui@plaidworks.com> wrote:

> If you've been doing CVS updates, can you check src/mailman/misc and
> see what release of the email directory you have? If it's NOT 2.4.3,
> you have a similar problem and we need to know if anyone else is being
> affected, so we can find out if we have to figure out a fix.

I have "email-2.4.3.tar.gz" in my misc directory. Is that the file you mean?

 - Stoney


From chuqui@plaidworks.com  Thu Oct 17 20:18:24 2002
From: chuqui@plaidworks.com (Chuq Von Rospach)
Date: Thu, 17 Oct 2002 12:18:24 -0700
Subject: [Mailman-Developers] my encoding problem: fixed!
In-Reply-To: <B9D48450.49285%sb.list@sb.org>
Message-ID: <35136B74-E205-11D6-99E3-0003934516A8@plaidworks.com>


On Thursday, October 17, 2002, at 12:14  PM, Stonewall Ballard wrote:

> I have "email-2.4.3.tar.gz" in my misc directory. Is that the file you 
> mean?
>

that's what you should see.


-- 
Chuq Von Rospach, Architech
chuqui@plaidworks.com -- http://www.plaidworks.com/chuqui/blog/

The first rule of holes: If you are in one, stop digging.


From mjm@michaelmeltzer.com  Thu Oct 17 20:22:14 2002
From: mjm@michaelmeltzer.com (Michael Meltzer)
Date: Thu, 17 Oct 2002 15:22:14 -0400
Subject: [Mailman-Developers] my encoding problem: fixed!
References: <00A8B67D-E201-11D6-99E3-0003934516A8@plaidworks.com>
Message-ID: <006801c27612$7ff61dc0$34f820c0@ix1x1000>

I am seeing only email-2.4.3.tar.gz but email 2.3 and email 2.4.3 directory's are both setup, misc/Makefile seem to be using
email-2.4.3, I do cvs update and ./configure.

Could you have missed a ./configure to regenerate misc/makefile?

MJM

----- Original Message ----- 
From: "Chuq Von Rospach" <chuqui@plaidworks.com>
To: <mailman-developers@python.org>
Sent: Thursday, October 17, 2002 2:48 PM
Subject: [Mailman-Developers] my encoding problem: fixed!


> 
> Barry, genius that he is, tracked down my encoding problem.
> 
> My mail libs were downrev, at 2.3 instead of 2.4.3
> 
> For some reason they weren't being upgraded by CVS, when they should 
> have been. We still dont' know why, but we need to see if it was "just 
> me", or if others are having this problem, too.
> 
> If you've been doing CVS updates, can you check src/mailman/misc and 
> see what release of the email directory you have? If it's NOT 2.4.3, 
> you have a similar problem and we need to know if anyone else is being 
> affected, so we can find out if we have to figure out a fix.
> 
> chuq
> 
> -- 
> Chuq Von Rospach, Architech
> chuqui@plaidworks.com -- http://www.plaidworks.com/chuqui/blog/
> 
> IMHO: Jargon. Acronym for In My Humble Opinion. Used to flag as an 
> opinion
> something that is clearly from context an opinion to everyone except the
> mentally dense. Opinions flagged by IMHO are actually rarely humble. 
> IMHO.
>      (source: third unabridged dictionary of chuqui-isms).
> 
> 
> _______________________________________________
> Mailman-Developers mailing list
> Mailman-Developers@python.org
> http://mail.python.org/mailman-21/listinfo/mailman-developers

From stoney@sb.org  Thu Oct 17 23:23:02 2002
From: stoney@sb.org (Stonewall Ballard)
Date: Thu, 17 Oct 2002 18:23:02 -0400
Subject: [Mailman-Developers] Rebuilding archives in 2.1b+
Message-ID: <B9D4B086.49365%stoney@sb.org>

I'm using yesterday's (10/16) CVS build.

I've managed to screw up my archives. Probably a permissions problem (I had
mailman_user set wrong for a while). Arch is segfaulting trying to rebuild
the date archives for a list.

Can I just delete all the html files and start over? Is that sufficient?
What do I need to keep? What should the owner/group of these files be?

TIA for any help.

 - Stoney

-- 
Stonewall Ballard 
stoney@sb.org           http://www.sb.org/stoney/



From bob@nleaudio.com  Fri Oct 18 00:50:56 2002
From: bob@nleaudio.com (Bob Puff@NLE)
Date: Thu, 17 Oct 2002 19:50:56 -0400
Subject: [Mailman-Developers] New TMDA-style service: spamarrest.com
Message-ID: <3DAF4CE0.4080408@nleaudio.com>

Hey Gang,

Just saw a user today who is using a service at http://spamarrest.com/ which looks just like the 
TMDA program, but on a commercial scale.  Naturally, they are claiming that they came up with the idea.

I am simply passing on the info, as it's gonna bring out a bunch more admin messages / mail loops 
for mailing lists.

Bob



From edlau@ucf.ics.uci.edu  Fri Oct 18 01:41:14 2002
From: edlau@ucf.ics.uci.edu (Edmund Lau)
Date: Thu, 17 Oct 2002 17:41:14 -0700 (PDT)
Subject: [Mailman-Developers] VERP code
Message-ID: <Pine.GSO.4.44.0210171731400.10831-100000@scotch.ucf.ics.uci.edu>

Hi All,

Mailman isn't processing any of my VERP messages from all my lists.  I've
been throwing in a lot of debug messages and found the area of code where
it isn't working for me.  This is from BounceRunner.py:155

            if bmailbox <> mo.group('bounces'):
                syslog('debug', 'mo.group= %s & bmailbox= %s',mo.
group('bounces'),bmailbox)
                continue                      # not a bounce to our list
            # All is good

The output I'm getting right there is
Oct 17 17:28:02 2002 (7175) mo.group= test & bmailbox= test-bounces

So as we can see, bmailbox & mo.group will never match and won't go to the
"All is good" section.  I replaced the "+" in the regex to "-" to get it
working with qmail.  The regex works otherwise I wouldn't be down in this
portion of code anyway.

Thoughts?  I'm using current CVS.

-Ed


From sb.list@sb.org  Fri Oct 18 03:20:29 2002
From: sb.list@sb.org (Stonewall Ballard)
Date: Thu, 17 Oct 2002 22:20:29 -0400
Subject: [Mailman-Developers] VERP code
In-Reply-To: <Pine.GSO.4.44.0210171731400.10831-100000@scotch.ucf.ics.uci.edu>
Message-ID: <B9D4E82D.49419%sb.list@sb.org>

On 10/17/02 8:41 PM, "Edmund Lau" <edlau@ucf.ics.uci.edu> wrote:

> Hi All,
> 
> Mailman isn't processing any of my VERP messages from all my lists.  I've
> been throwing in a lot of debug messages and found the area of code where
> it isn't working for me.  This is from BounceRunner.py:155
> 
>           if bmailbox <> mo.group('bounces'):
>               syslog('debug', 'mo.group= %s & bmailbox= %s',mo.
> group('bounces'),bmailbox)
>               continue                      # not a bounce to our list
>           # All is good
> 
> The output I'm getting right there is
> Oct 17 17:28:02 2002 (7175) mo.group= test & bmailbox= test-bounces
> 
> So as we can see, bmailbox & mo.group will never match and won't go to the
> "All is good" section.  I replaced the "+" in the regex to "-" to get it
> working with qmail.  The regex works otherwise I wouldn't be down in this
> portion of code anyway.
> 
> Thoughts?  I'm using current CVS.

I'm having the same problem. All of the bounces are "unrecognized". After
staring at this code for a while, I don't think that we have the same
problem, just the same symptoms.

Are you saying that your VERP_FORMAT is '%(bounces)s-%(mailbox)s=%(host)s' ?

If so, I don't think that would work very well since the mailbox name could
have a '-' in it, making part matching ambiguous.

If I misunderstood, can you clarify? What are your VERP_FORMAT and your
VERP_REGEXP?


 - Stoney


From chuqui@plaidworks.com  Fri Oct 18 06:44:17 2002
From: chuqui@plaidworks.com (Chuq Von Rospach)
Date: Thu, 17 Oct 2002 22:44:17 -0700
Subject: [Mailman-Developers] New TMDA-style service: spamarrest.com
In-Reply-To: <3DAF4CE0.4080408@nleaudio.com>
Message-ID: <A48F0E68-E25C-11D6-99E3-0003934516A8@plaidworks.com>


On Thursday, October 17, 2002, at 04:50  PM, Bob Puff@NLE wrote:

> Just saw a user today who is using a service at http://spamarrest.com/ 
> which looks just like the TMDA program, but on a commercial scale.  
> Naturally, they are claiming that they came up with the idea.
>

It's one of about four I've seen recently, by far the most popular of 
them, but they're sprouting like weeds.


-- 
Chuq Von Rospach, Architech
chuqui@plaidworks.com -- http://www.plaidworks.com/chuqui/blog/

IMHO: Jargon. Acronym for In My Humble Opinion. Used to flag as an 
opinion
something that is clearly from context an opinion to everyone except the
mentally dense. Opinions flagged by IMHO are actually rarely humble. 
IMHO.
     (source: third unabridged dictionary of chuqui-isms).


From chuqui@plaidworks.com  Fri Oct 18 06:58:11 2002
From: chuqui@plaidworks.com (Chuq Von Rospach)
Date: Thu, 17 Oct 2002 22:58:11 -0700
Subject: [Mailman-Developers] strange... digest volumes.
Message-ID: <9519E5F2-E25E-11D6-99E3-0003934516A8@plaidworks.com>



I just noticed something weird. All my digest volume/numbers are 
ratcheting up. When I installed 2.1b, I set all my volumes to 4 or 5, 
to put them above any previous versions of my list servers.

tonight, some of the digests I got were volume 20, #1, volume 16 #4, 
etc.

I don't know what's going on yet, but I THOUGHT I set the things to 
increment annually. I'll see waht I can find out.

anyone else seeing weird digest increment issues? Or have you looked 
for them?



-- 
Chuq Von Rospach, Architech
chuqui@plaidworks.com -- http://www.plaidworks.com/chuqui/blog/

Very funny, Scotty. Now beam my clothes down here, will you?


From tkikuchi@is.kochi-u.ac.jp  Fri Oct 18 07:53:40 2002
From: tkikuchi@is.kochi-u.ac.jp (Tokio Kikuchi)
Date: Fri, 18 Oct 2002 15:53:40 +0900
Subject: [Mailman-Developers] 
 output/input charset (was Re: encoding the subject line..)
References: <15790.59430.206821.692862@gargle.gargle.HOWL>
	<8413E943-E1F1-11D6-99E3-0003934516A8@plaidworks.com>
	<15790.62357.57583.142130@gargle.gargle.HOWL>
Message-ID: <3DAFAFF4.3070001@is.kochi-u.ac.jp>

Hi!

Sorry for confusing the discussion on the subject line matter.
While I was examing the CookHeaders.py myself, I noticed a possible
error in the charset determination.

Here is the diff:

diff -u CookHeaders.py.orig CookHeaders.py
--- CookHeaders.py.orig Fri Oct 18 15:34:47 2002
+++ CookHeaders.py      Fri Oct 18 15:36:06 2002
@@ -220,7 +220,7 @@
      # If prefix is a byte string and there are funky characters in it that
      # don't match the charset, we might as well replace them now.
      if not _isunicode(prefix):
-        prefix = unicode(prefix, charset.get_output_charset(), 'replace')
+        prefix = unicode(prefix, charset.input_charset, 'replace')
      # We purposefully leave no space b/w prefix and subject!
      h = Header(prefix, charset, header_name='Subject')
      for s, c in headerbits:

Explanation:
There is one language where output_charset and input_charset differ;
Japanese. In the web interface and internal processing, 'euc-jp' is used
while in the mail, 'iso-2022-jp'. The prefix encoding is 'euc-jp' because
it was input from web interface, and not 'iso-2022-jp', thus coding to
unicode with output_charset will result in 'funnier' character. The
charset.input_charset may be replaced with `charset` but I'm not sure.
There was a similar error in email/Header.py, also. Please check.

-- 
Tokio Kikuchi, tkikuchi@ is.kochi-u.ac.jp
http://weather.is.kochi-u.ac.jp/


From edlau@ucf.ics.uci.edu  Fri Oct 18 08:18:08 2002
From: edlau@ucf.ics.uci.edu (Edmund Lau)
Date: Fri, 18 Oct 2002 00:18:08 -0700 (PDT)
Subject: [Mailman-Developers] VERP code
In-Reply-To: <B9D4E82D.49419%sb.list@sb.org>
Message-ID: <Pine.GSO.4.44.0210180000050.12202-100000@scotch.ucf.ics.uci.edu>

On Thu, 17 Oct 2002, Stonewall Ballard wrote:

> I'm having the same problem. All of the bounces are "unrecognized". After
> staring at this code for a while, I don't think that we have the same
> problem, just the same symptoms.

We might actually be having the same problem.  I get the "unrecognized"
errors in my bounce logs too.  For instance:

Oct 16 21:57:27 2002 (32490) forwarding unrecognized, message-id:
<DNffhMoQ10000 984b@mc5-s4.law1.hotmail.com>

The only saving grace on why some addresses get recognized is because
addresses get checked after the initial verp_bounce call, if there aren't
any returned addrs on line 71.  The unrecognized ones come when some users
subscribe from accounts which forward messages.

>
> Are you saying that your VERP_FORMAT is '%(bounces)s-%(mailbox)s=%(host)s' ?
>
> If so, I don't think that would work very well since the mailbox name could
> have a '-' in it, making part matching ambiguous.
>
> If I misunderstood, can you clarify? What are your VERP_FORMAT and your
> VERP_REGEXP?
>

Yes, my VERP_FORMAT is '%(bounces)s-%(mailbox)s=%(host)s'.  However, my
understanding is that mailman only uses this variable to generate the
appropriate VERP string.  It's actually the VERP_REGEXP which matches when
a bounce is received.  Mine is:
r'^(?P<bounces>[^-]+?)-(?P<mailbox>[^=]+)=(?P<host>[^@]+)@.*$'

But like I said earlier, the REGEXP seems fine since my error is line 155
and the VERP_REGEXP has already been successfully matched on 151.  (If it
had failed, it would have hit 152-153 and never get around to 155.)  To be
sure, I made a quick driver and ran it against some known good and known
bad VERP strings.

So while it's true that extra hyphens may mess the matching up (which I do
have lists containing it), it's rather unlikely that's the problem, esp
since I have a list named 'test' and even that failed.

Any other ideas?
-Ed


From tkikuchi@is.kochi-u.ac.jp  Fri Oct 18 14:14:05 2002
From: tkikuchi@is.kochi-u.ac.jp (Tokio Kikuchi)
Date: Fri, 18 Oct 2002 22:14:05 +0900
Subject: [Mailman-Developers] strange... digest volumes.
References: <9519E5F2-E25E-11D6-99E3-0003934516A8@plaidworks.com>
Message-ID: <3DB0091D.6050009@is.kochi-u.ac.jp>

Chuq Von Rospach wrote:
> 
> 
> I just noticed something weird. All my digest volume/numbers are 
> ratcheting up. When I installed 2.1b, I set all my volumes to 4 or 5, to 
> put them above any previous versions of my list servers.
> 
> tonight, some of the digests I got were volume 20, #1, volume 16 #4, etc.
> 
> I don't know what's going on yet, but I THOUGHT I set the things to 
> increment annually. I'll see waht I can find out.
> 
> anyone else seeing weird digest increment issues? Or have you looked for 
> them?

Yes, I'm seeing this. This patch may solve, I hope.

diff -u ~/src/mailman/Mailman/Handlers/ToDigest.py ToDigest.py
--- /home/mailman/src/mailman/Mailman/Handlers/ToDigest.py      Tue Oct  8 06:35
:04 2002
+++ ToDigest.py Fri Oct 18 22:06:46 2002
@@ -113,8 +113,8 @@
              weeknum_now = int(time.strftime('%W', now))
              if weeknum_now > weeknum_last or timetup[0] > now[0]:
                  bump = 1
-        elif timetup[7] <> now[7]:
-            # assume daily
+        elif freq == 4 and timetup[7] <> now[7]:
+            # assume daily <-- Don't assume that way!
              bump = 1
          if bump:
              mlist.volume += 1

Better to test the variable freq only in the if .. elif chain.

-- 
Tokio Kikuchi, tkikuchi@ is.kochi-u.ac.jp
http://weather.is.kochi-u.ac.jp/


From sb.list@sb.org  Fri Oct 18 13:51:01 2002
From: sb.list@sb.org (Stonewall Ballard)
Date: Fri, 18 Oct 2002 08:51:01 -0400
Subject: [Mailman-Developers] VERP code
In-Reply-To: <Pine.GSO.4.44.0210180000050.12202-100000@scotch.ucf.ics.uci.edu>
Message-ID: <B9D57BF5.494FE%sb.list@sb.org>

On 10/18/02 3:18 AM, "Edmund Lau" <edlau@ucf.ics.uci.edu> wrote:

> On Thu, 17 Oct 2002, Stonewall Ballard wrote:
> 
>> I'm having the same problem. All of the bounces are "unrecognized". After
>> staring at this code for a while, I don't think that we have the same
>> problem, just the same symptoms.
> 
> We might actually be having the same problem.  I get the "unrecognized"
> errors in my bounce logs too.  For instance:
> 
> Oct 16 21:57:27 2002 (32490) forwarding unrecognized, message-id:
> <DNffhMoQ10000 984b@mc5-s4.law1.hotmail.com>

My bounce problems were due to the list being locked at the time the bounces
are processed. There's something whacked in my setup and I'm trying to
figure out what it is. I did something approximating a clean install, and
the bad locks disappeared, as did the bounce processing problems.

> Yes, my VERP_FORMAT is '%(bounces)s-%(mailbox)s=%(host)s'.  However, my
> understanding is that mailman only uses this variable to generate the
> appropriate VERP string.  It's actually the VERP_REGEXP which matches when
> a bounce is received.  Mine is:
> r'^(?P<bounces>[^-]+?)-(?P<mailbox>[^=]+)=(?P<host>[^@]+)@.*$'

Yes, but you need to unambiguously split the list name from the subscriber
address, and that VERP_REGEXP will always match 'bounces' up to the first
'-', giving you the results that you see.

If you used a regexp like this:

'^(?P<bounces>.+)-(?P<mailbox>[^=]+)=(?P<host>[^@]+)@.*$'

then 'bounces' would match up to the last '-', which would work unless the
subscriber address has a '-' in it. You could at least try this for testing.

You need to use a marker character that's not going to appear in a list
name, but is a legal component of an email address. '+' is a reasonable
choice. I don't know why it's not working for you. Maybe you could try
underscore.

> But like I said earlier, the REGEXP seems fine since my error is line 155
> and the VERP_REGEXP has already been successfully matched on 151.  (If it
> had failed, it would have hit 152-153 and never get around to 155.)  To be
> sure, I made a quick driver and ran it against some known good and known
> bad VERP strings.

Yes, it does match, but at the wrong '-'.
> 
> So while it's true that extra hyphens may mess the matching up (which I do
> have lists containing it), it's rather unlikely that's the problem, esp
> since I have a list named 'test' and even that failed.

Take a look at the value of mo.group('mailbox') after the match and see if
it has something like 'bounces-person' in it. That should prove that you're
matching the string incorrectly.

 - Stoney


From sb.list@sb.org  Fri Oct 18 14:51:42 2002
From: sb.list@sb.org (Stonewall Ballard)
Date: Fri, 18 Oct 2002 09:51:42 -0400
Subject: [Mailman-Developers] configure fails on OS X
Message-ID: <B9D58A2E.49524%sb.list@sb.org>

At least on Mac OS X Server 10.1.5, configure fails checking for the
existence of the 'mailman' user. The error says:

  sh: setgroups: Operation not permitted

This apparently caused my problems with permissions, since qrunner was set
to run under my username.

I have to manually patch config.status to set MAILMAN_USER to 'mailman'.

This was from a fresh CVS build.

 - Stoney

-- 
Stonewall Ballard 
stoney@sb.org           http://www.sb.org/stoney/



From barry@python.org  Fri Oct 18 16:22:24 2002
From: barry@python.org (Barry A. Warsaw)
Date: Fri, 18 Oct 2002 11:22:24 -0400
Subject: [Mailman-Developers] 
 Re: output/input charset (was Re: encoding the subject line..)
References: <15790.59430.206821.692862@gargle.gargle.HOWL>
	<8413E943-E1F1-11D6-99E3-0003934516A8@plaidworks.com>
	<15790.62357.57583.142130@gargle.gargle.HOWL>
	<3DAFAFF4.3070001@is.kochi-u.ac.jp>
Message-ID: <15792.10032.476559.56953@gargle.gargle.HOWL>


>>>>> "TK" == Tokio Kikuchi <tkikuchi@is.kochi-u.ac.jp> writes:

    TK> Sorry for confusing the discussion on the subject line matter.
    TK> While I was examing the CookHeaders.py myself, I noticed a
    TK> possible error in the charset determination.

I noticed this too and fixed it in CookHeaders.py 2.27.  Do a cvs
update. :)

    TK> Explanation: There is one language where output_charset and
    TK> input_charset differ; Japanese. In the web interface and
    TK> internal processing, 'euc-jp' is used while in the mail,
    TK> 'iso-2022-jp'. The prefix encoding is 'euc-jp' because it was
    TK> input from web interface, and not 'iso-2022-jp', thus coding
    TK> to unicode with output_charset will result in 'funnier'
    TK> character. The charset.input_charset may be replaced with
    TK> `charset` but I'm not sure.  There was a similar error in
    TK> email/Header.py, also. Please check.

Yes, Ben did a good job of explaining this to me and I think we got it
right in Header.py for the email 2.4.3 release.  I noticed the same
issue when I was reviewing CookHeaders.py.

Thanks,
-Barry

From sb.list@sb.org  Fri Oct 18 16:24:44 2002
From: sb.list@sb.org (Stonewall Ballard)
Date: Fri, 18 Oct 2002 11:24:44 -0400
Subject: [Mailman-Developers] make install creates empty bin/qrunner
Message-ID: <B9D59FFC.49564%sb.list@sb.org>

I just fetched a fresh CVS tree, configured, and installed it. I end up with
a 0-length bin/qrunner file. I have no idea why this is happening. I ended
up copying the one from the CVS tree over and editing the #! at the top.
Seems to work.

Mac OS X Server 10.1.5.

 - Stoney

-- 
Stonewall Ballard 
stoney@sb.org           http://www.sb.org/stoney/



From barry@python.org  Fri Oct 18 16:43:50 2002
From: barry@python.org (Barry A. Warsaw)
Date: Fri, 18 Oct 2002 11:43:50 -0400
Subject: [Mailman-Developers] strange... digest volumes.
References: <9519E5F2-E25E-11D6-99E3-0003934516A8@plaidworks.com>
	<3DB0091D.6050009@is.kochi-u.ac.jp>
Message-ID: <15792.11318.274160.669684@gargle.gargle.HOWL>


>>>>> "TK" == Tokio Kikuchi <tkikuchi@is.kochi-u.ac.jp> writes:

    TK> Yes, I'm seeing this. This patch may solve, I hope.

Dumb mistake on my part.  Thanks for the fix!
-Barry

From barry@python.org  Fri Oct 18 16:53:18 2002
From: barry@python.org (Barry A. Warsaw)
Date: Fri, 18 Oct 2002 11:53:18 -0400
Subject: [Mailman-Developers] qrunner
References: <200210170248.20057.philb@philb.us>
Message-ID: <15792.11886.511165.293574@gargle.gargle.HOWL>


>>>>> "PB" == Phil Barnett <philb@philb.us> writes:

    PB> What's the downside of qrunner not running each minute?

    PB> Would every 10 minutes be less cpu intensive and still
    PB> generally effective?

    PB> What am I trading off?

Running it less often makes your system less responsive.  It could be
as much as 10 minutes between the time your mail server gets the
message and Mailman processes and resends it.

But, on a quiet system, qrunner shouldn't take up much cpu.  With
MM2.1 it's even less because you don't have the costs of starting up
Python each time.

-Barry

From barry@python.org  Fri Oct 18 16:57:05 2002
From: barry@python.org (Barry A. Warsaw)
Date: Fri, 18 Oct 2002 11:57:05 -0400
Subject: [Mailman-Developers] make install creates empty bin/qrunner
References: <B9D59FFC.49564%sb.list@sb.org>
Message-ID: <15792.12113.372062.618737@gargle.gargle.HOWL>


>>>>> "SB" == Stonewall Ballard <sb.list@sb.org> writes:

    SB> I just fetched a fresh CVS tree, configured, and installed
    SB> it. I end up with a 0-length bin/qrunner file. I have no idea
    SB> why this is happening. I ended up copying the one from the CVS
    SB> tree over and editing the #! at the top.  Seems to work.

    SB> Mac OS X Server 10.1.5.

Weird.  Did you see any errors during configure or during make
install?  Does build/bin/qrunner look okay?  Maybe something about the
way configure creates that file has problems on OSX?  (Perhaps 10.1
specific -- I don't think Chuq's had a problem on Jaguar.)

-Barry

From sb.list@sb.org  Fri Oct 18 17:01:32 2002
From: sb.list@sb.org (Stonewall Ballard)
Date: Fri, 18 Oct 2002 12:01:32 -0400
Subject: [Mailman-Developers] ArchRunner is segfaulting
Message-ID: <B9D5A89C.4959D%sb.list@sb.org>

I've found that my lock problem is caused by the ArchRunner qrunner process
segfaulting and leaving a lock hanging around.

This was tested on a fresh CVS build, and with no pre-existing lists.

The same thing happens when I run bin/arch on a list. It prints out "Date",
and then segfaults.

Mac OS X Server 10.1.5.

Any idea what might be causing this?

 - Stoney

-- 
Stonewall Ballard 
stoney@sb.org           http://www.sb.org/stoney/



From barry@python.org  Fri Oct 18 17:22:30 2002
From: barry@python.org (Barry A. Warsaw)
Date: Fri, 18 Oct 2002 12:22:30 -0400
Subject: [Mailman-Developers] ArchRunner is segfaulting
References: <B9D5A89C.4959D%sb.list@sb.org>
Message-ID: <15792.13638.501731.406750@gargle.gargle.HOWL>


>>>>> "SB" == Stonewall Ballard <sb.list@sb.org> writes:

    SB> I've found that my lock problem is caused by the ArchRunner
    SB> qrunner process segfaulting and leaving a lock hanging around.

    SB> This was tested on a fresh CVS build, and with no pre-existing
    SB> lists.

    SB> The same thing happens when I run bin/arch on a list. It
    SB> prints out "Date", and then segfaults.

    SB> Mac OS X Server 10.1.5.

    SB> Any idea what might be causing this?

What version of Python are you using, and can you get a stack trace
from gdb?

-Barry

From sb.list@sb.org  Fri Oct 18 17:25:54 2002
From: sb.list@sb.org (Stonewall Ballard)
Date: Fri, 18 Oct 2002 12:25:54 -0400
Subject: [Mailman-Developers] make install creates empty bin/qrunner
In-Reply-To: <15792.12113.372062.618737@gargle.gargle.HOWL>
Message-ID: <B9D5AE52.495B0%sb.list@sb.org>

On 10/18/02 11:57 AM, "Barry A. Warsaw" <barry@python.org> wrote:

> 
>>>>>> "SB" == Stonewall Ballard <sb.list@sb.org> writes:
> 
>   SB> I just fetched a fresh CVS tree, configured, and installed
>   SB> it. I end up with a 0-length bin/qrunner file. I have no idea
>   SB> why this is happening. I ended up copying the one from the CVS
>   SB> tree over and editing the #! at the top.  Seems to work.
> 
>   SB> Mac OS X Server 10.1.5.
> 
> Weird.  Did you see any errors during configure or during make
> install?  Does build/bin/qrunner look okay?  Maybe something about the
> way configure creates that file has problems on OSX?  (Perhaps 10.1
> specific -- I don't think Chuq's had a problem on Jaguar.)

build/bin/qrunner was 0 length too. I just emptied my build directory and
rebuilt. This time build/bin/qrunner is ok.

I don't know why this happened.

Thanks.

 - Stoney 


From barry@python.org  Fri Oct 18 18:21:59 2002
From: barry@python.org (Barry A. Warsaw)
Date: Fri, 18 Oct 2002 13:21:59 -0400
Subject: [Mailman-Developers] make install creates empty bin/qrunner
References: <15792.12113.372062.618737@gargle.gargle.HOWL>
	<B9D5AE52.495B0%sb.list@sb.org>
Message-ID: <15792.17207.486138.369136@gargle.gargle.HOWL>


>>>>> "SB" == Stonewall Ballard <sb.list@sb.org> writes:

    SB> build/bin/qrunner was 0 length too. I just emptied my build
    SB> directory and rebuilt. This time build/bin/qrunner is ok.

    SB> I don't know why this happened.

Full moon.
-Barry

From edlau@ucf.ics.uci.edu  Fri Oct 18 19:40:40 2002
From: edlau@ucf.ics.uci.edu (Edmund Lau)
Date: Fri, 18 Oct 2002 11:40:40 -0700 (PDT)
Subject: [Mailman-Developers] VERP code
In-Reply-To: <B9D57BF5.494FE%sb.list@sb.org>
Message-ID: <Pine.GSO.4.44.0210181118500.17506-100000@scotch.ucf.ics.uci.edu>

You're absolutely right.  The explanation makes a lot of sense now.

So now I'm seemingly stuck between a rock and a hard place.  I've been
running stock qmail for a number of years now.  qmail likes using "-" as a
delimiter.  It's possible to recompile qmail to use "+" instead, but that
would break all my users' personal Email address lists.  (It's the same
thing as Postfix's .forward-<ext>)

qmail has a general catch-all address.  It's defined by a .qmail-default
file (or .qmail-ext-default).  For instance if I had a .qmail-blah-default
in my home directory, any Email with prefix to edlau-blah- gets processed
by it.  But note the trailing hyphen.  I can't have a .qmail-blah+default
as qmail won't recognize it, unless there's some heavy, magical trickery.

Back to mailman, I can use a "+" as a delimiter, but there needs to be a
"-" right before it.  Is there a regex that would get that?  So the new
VERP_FORMAT would be '%(bounces)s-+%(mailbox)s=%(host)s'.  I would assume
this is better than just matching the last "-" anyway as, like you said, a
user can have that character in their address.

Ideas?
-Ed

On Fri, 18 Oct 2002, Stonewall Ballard wrote:

> > Yes, my VERP_FORMAT is '%(bounces)s-%(mailbox)s=%(host)s'.  However, my
> > understanding is that mailman only uses this variable to generate the
> > appropriate VERP string.  It's actually the VERP_REGEXP which matches when
> > a bounce is received.  Mine is:
> > r'^(?P<bounces>[^-]+?)-(?P<mailbox>[^=]+)=(?P<host>[^@]+)@.*$'
>
> Yes, but you need to unambiguously split the list name from the subscriber
> address, and that VERP_REGEXP will always match 'bounces' up to the first
> '-', giving you the results that you see.
>
> If you used a regexp like this:
>
> '^(?P<bounces>.+)-(?P<mailbox>[^=]+)=(?P<host>[^@]+)@.*$'
>
> then 'bounces' would match up to the last '-', which would work unless the
> subscriber address has a '-' in it. You could at least try this for testing.
>
> You need to use a marker character that's not going to appear in a list
> name, but is a legal component of an email address. '+' is a reasonable
> choice. I don't know why it's not working for you. Maybe you could try
> underscore.
>
> > But like I said earlier, the REGEXP seems fine since my error is line 155
> > and the VERP_REGEXP has already been successfully matched on 151.  (If it
> > had failed, it would have hit 152-153 and never get around to 155.)  To be
> > sure, I made a quick driver and ran it against some known good and known
> > bad VERP strings.
>
> Yes, it does match, but at the wrong '-'.
> >
> > So while it's true that extra hyphens may mess the matching up (which I do
> > have lists containing it), it's rather unlikely that's the problem, esp
> > since I have a list named 'test' and even that failed.
>
> Take a look at the value of mo.group('mailbox') after the match and see if
> it has something like 'bounces-person' in it. That should prove that you're
> matching the string incorrectly.
>
>  - Stoney


From sb.list@sb.org  Fri Oct 18 20:26:03 2002
From: sb.list@sb.org (Stonewall Ballard)
Date: Fri, 18 Oct 2002 15:26:03 -0400
Subject: [Mailman-Developers] VERP code
In-Reply-To: <Pine.GSO.4.44.0210181118500.17506-100000@scotch.ucf.ics.uci.edu>
Message-ID: <B9D5D88B.4963E%sb.list@sb.org>

On 10/18/02 2:40 PM, "Edmund Lau" <edlau@ucf.ics.uci.edu> wrote:

> Back to mailman, I can use a "+" as a delimiter, but there needs to be a
> "-" right before it.  Is there a regex that would get that?  So the new
> VERP_FORMAT would be '%(bounces)s-+%(mailbox)s=%(host)s'.  I would assume
> this is better than just matching the last "-" anyway as, like you said, a
> user can have that character in their address.

That should work. Try this regexp:

'^(?P<bounces>.*?)-\+(?P<mailbox>[^=]+)=(?P<host>[^@]+)@.*$

 - Stoney


From gward@python.net  Fri Oct 18 21:49:52 2002
From: gward@python.net (Greg Ward)
Date: Fri, 18 Oct 2002 16:49:52 -0400
Subject: [Mailman-Developers] Doc patch: upgrading individual lists
Message-ID: <20021018204952.GA3354@cthulhu.gerg.ca>

Back in May, Barry posted this message on how to move/upgrade individual
lists to MM 2.1:

  http://mail.python.org/pipermail-21/mailman-developers/2002-May/011785.html

This seems like a useful incantation to me, so I've added it, with some
variations of my own, to the UPGRADING document.  Here's the patch:

------------------------------------------------------------------------
--- UPGRADING	1 Aug 2002 19:24:47 -0000	2.9
+++ UPGRADING	18 Oct 2002 20:49:07 -0000
@@ -96,6 +96,102 @@
     manually. 
 
 
+UPGRADING INDIVIDUAL LISTS
+
+    If you're nervous about upgrading all of your lists to 2.1 in one
+    go, you can move them and upgrade them one at a time.  Start by
+    doing a clean Mailman 2.1 installation in an empty directory --
+    call it $MM21.  (I'll assume your Mailman 2.0 installation is in
+    $MM20.)
+
+    Doing this means you'll have co-habiting Mailman 2.0 and 2.1
+    installations for a while, until you have moved all of your lists to
+    Mailman 2.1.  Depending on your MTA and web server, this could be
+    transparent and painless, or it could be an ongoing headache.  If
+    you use Apache with mod_rewrite, then it's fairly straightforward to
+    set things up so that both Mailman 2.0 and 2.1 inhabit the /mailman
+    and /pipermail URL-space of your server; this makes the transition
+    almost transparent to list admins and subscribers.  See below for
+    details.
+
+    Now, for each list that you want to move, you'll have to
+
+      * Shut down your MTA.
+
+        If you have a lot of outgoing list traffic, you might need to
+        leave your MTA up but only let it accept connections from
+        127.0.0.1, so Mailman 2.0 can flush its queue.  How to do this
+        is MTA-dependent; for Exim, you can set
+        "local_interfaces = 127.0.0.1" and "kill -HUP" the Exim daemon.
+
+      * Shut down your web server.  For a more professional look,
+        or if you want to allow people to keep accessing the rest
+        of your web site, you could make your web server respond to
+        all /mailman/ URLs with a "temporarily unavailable" message.
+
+        How to do this is web server-dependent; with Apache and
+        mod_rewrite, this does the trick:
+          RewriteRule ^/mailman/.* \
+                      /var/www/unavailable.html [L]
+
+        (Of course, you'll have to supply your own
+        /var/www/unavailable.html.)
+
+      * Force Mailman 2.0 to process its queue:
+          python -S $MM20/cron/qrunner
+
+        (This is only necessary if there are any files in $MM20/qfiles;
+        if you need to do this, make sure you left your MTA listening to
+        127.0.0.1.)
+
+      * Move the list:
+          cd $MM20
+          mv -i lists/foo-list $MM21/lists
+          mv -i archives/private/foo-list $MM21/archives/private
+          mv -i archives/private/foo-list.mbox $MM21/archives/private
+          rm archives/public/foo-list
+          rm archives/public/foo-list.mbox
+          cd $MM21
+          bin/withlist -l -r fix_url mylist
+
+        (The fix_url step will not be necessary if your Mailman 2.0
+        and 2.1 installations will be sharing the same URL-space.)
+
+      * Edit your web server config so the list's URLs continue to
+        work.  There are two possible approaches here; the simpler way
+        is to setup a new slice of URL-space that will be used by your
+        Mailman 2.1 installation, eg. /mailman-21:
+        With Apache and mod_rewrite:
+          RewriteRule /mailman/(.*)/(foo-list.*) \
+                      /mailman-21/$1/$2 [R=temp]
+
+        (The [R=temp] assumes that "/mailman-21/" is a temporary URL,
+        and you'll move all your lists to "/mailman/" when the
+        transition to Mailman 2.1 is complete.)
+
+        If you don't want to expose ugly temporary URLs like
+        "/mailman-21" to the world, it's only slightly more work to make
+        Mailman 2.0 and 2.1 share the same slices of URL-space.  Here's
+        how to do it with Apache and mod_rewrite:
+          RewriteRule ^/mailman/(.*)/(foo-list.*) \
+                      $MM21/cgi-bin/$1/$2 \
+                      [T=application/x-httpd-cgi]
+
+        Not only is this more aesthetically pleasing, it's faster -- no
+        redirects.
+
+        In either case, you'll want to rewrite the list's archive URLs
+        to Mailman 2.1's archive:
+          RewriteRule ^/pipermail/(foo-list.*) \
+                      $MM21/archives/public/$1
+
+      * Restart your web server (or disable the "temporarily
+        unavailable" stuff).
+
+      * Restart your MTA (or make it listen to more than just
+        127.0.0.1).
+
+
 UPGRADING FROM 2.0 to 2.0.1
 
     Nothing much more than running "make install" (after upgrading)
------------------------------------------------------------------------

Sound useful?  Big enough to go in a separate document?

        Greg
-- 
Greg Ward <gward@python.net>                         http://www.gerg.ca/
No man is an island, but some of us are long peninsulas.

From barry@python.org  Fri Oct 18 22:59:52 2002
From: barry@python.org (Barry A. Warsaw)
Date: Fri, 18 Oct 2002 17:59:52 -0400
Subject: [Mailman-Developers] ArchRunner is segfaulting
References: <B9D5A89C.4959D%sb.list@sb.org>
Message-ID: <15792.33880.504723.727269@gargle.gargle.HOWL>


>>>>> "SB" == Stonewall Ballard <sb.list@sb.org> writes:

    SB> I've found that my lock problem is caused by the ArchRunner
    SB> qrunner process segfaulting and leaving a lock hanging around.

We figured out what this was.  MacOSX users, listen up!

There is a known problem in Python related to deeply recursive regular
expressions on MacOSX, both 10.1 and 10.2.  The default stack size is
too small, which can cause Python to core dump.  Turns out, Stoney's
problem is related -- apparently pipermail tickles this bug.  We see
this as a crash in Python's test suite when runnng test_re.py and
test_sre.py.

The recommended work around is to boost your stack size to 2048 using
a tcsh or bash shell command, but that seems too inconvenient for a
daemon like Mailman.  Here is the Python code work around for problem:

http://sourceforge.net/tracker/download.php?group_id=5470&atid=305470&file_id=32837&aid=580869

I am going to add something like this to HyperArch.py, which should
run when either ArchRunner does its thing, or you run bin/arch.
Because it's wrapped in a "if sys.platform == 'darwin'" it should only
affect MacOSX users.  Could those of you running Mailman on OSX give
it a try and let me know if it fixes your problems?

Checkins coming momentarily.
-Barry

From barry@python.org  Sat Oct 19 00:53:35 2002
From: barry@python.org (Barry A. Warsaw)
Date: Fri, 18 Oct 2002 19:53:35 -0400
Subject: [Mailman-Developers] Bounce with verp on not recognized
References: <B9D31C5E.48F0C%stoney@sb.org>
Message-ID: <15792.40703.957154.241176@gargle.gargle.HOWL>


>>>>> "SB" == Stonewall Ballard <stoney@sb.org> writes:

    SB> Should I have expected that welcome messages would contain
    SB> verp'd reply-to's or is this likely to be a misconfiguration
    SB> somewhere on my part?

Until now, no.  But I think it makes sense to verp welcome messages
(and unsubscribe acks, and password reminders) if
VERP_PERSONALIZED_DELIVERIES is turned on.  I'll make this so.

    SB> BTW, this is a _wonderful_ feature.

VERP really moves the goal line for bounce detection, albeit at the
cost of some extra processing and bandwidth.

-Barry

From barry@python.org  Sat Oct 19 00:57:53 2002
From: barry@python.org (Barry A. Warsaw)
Date: Fri, 18 Oct 2002 19:57:53 -0400
Subject: [Mailman-Developers] VERP'd reply-to not being used by bounce
	detector
References: <B9D337FC.48F97%stoney@sb.org>
Message-ID: <15792.40961.98299.855335@gargle.gargle.HOWL>


>>>>> "SB" == Stonewall Ballard <stoney@sb.org> writes:

    SB> I'm using Mailman 2.1b+, from a CVS build earlier today.

    SB> I sent a batch of messages out, and got a dozen bounces. Each
    SB> one was forwarded to me as "uncaught", but the reply-to had
    SB> the verp'd address in it.

I recently changed the code in BounceRunner.py so that if the bounce
format (verp or otherwise) is recognized but yeilds no addresses which
are members of the list, the bouncing message is forwarded as
"uncaught".

Could this be what you're seeing?

-Barry

From barry@python.org  Sat Oct 19 01:00:49 2002
From: barry@python.org (Barry A. Warsaw)
Date: Fri, 18 Oct 2002 20:00:49 -0400
Subject: [Mailman-Developers] VERP code
References: <B9D57BF5.494FE%sb.list@sb.org>
	<Pine.GSO.4.44.0210181118500.17506-100000@scotch.ucf.ics.uci.edu>
Message-ID: <15792.41137.327805.809630@gargle.gargle.HOWL>


>>>>> "EL" == Edmund Lau <edlau@ucf.ics.uci.edu> writes:

    EL> So now I'm seemingly stuck between a rock and a hard place.
    EL> I've been running stock qmail for a number of years now.
    EL> qmail likes using "-" as a delimiter.  It's possible to
    EL> recompile qmail to use "+" instead, but that would break all
    EL> my users' personal Email address lists.  (It's the same thing
    EL> as Postfix's .forward-<ext>)

    EL> qmail has a general catch-all address.  It's defined by a
    EL> .qmail-default file (or .qmail-ext-default).  For instance if
    EL> I had a .qmail-blah-default in my home directory, any Email
    EL> with prefix to edlau-blah- gets processed by it.  But note the
    EL> trailing hyphen.  I can't have a .qmail-blah+default as qmail
    EL> won't recognize it, unless there's some heavy, magical
    EL> trickery.

    EL> Back to mailman, I can use a "+" as a delimiter, but there
    EL> needs to be a "-" right before it.  Is there a regex that
    EL> would get that?  So the new VERP_FORMAT would be
    EL> '%(bounces)s-+%(mailbox)s=%(host)s'.  I would assume this is
    EL> better than just matching the last "-" anyway as, like you
    EL> said, a user can have that character in their address.

I'm Cc'ing Norbert Bollow who wrote the README.QMAIL file.  Perhaps he
has some ideas about how to make this work.

Hi Norbert!
-Barry

From barry@python.org  Sat Oct 19 01:10:05 2002
From: barry@python.org (Barry A. Warsaw)
Date: Fri, 18 Oct 2002 20:10:05 -0400
Subject: [Mailman-Developers] Doc patch: upgrading individual lists
References: <20021018204952.GA3354@cthulhu.gerg.ca>
Message-ID: <15792.41693.202223.690831@gargle.gargle.HOWL>


>>>>> "GW" == Greg Ward <gward@python.net> writes:

    GW> Back in May, Barry posted this message on how to move/upgrade
    GW> individual lists to MM 2.1:

    GW> This seems like a useful incantation to me, so I've added it,
    GW> with some variations of my own, to the UPGRADING document.
    GW> Here's the patch:

Very cool Greg, thanks.  I really want to try your ideas about shared
url space when we upgrade python.org.

-Barry

From edlau@ucf.ics.uci.edu  Sat Oct 19 01:11:30 2002
From: edlau@ucf.ics.uci.edu (Edmund Lau)
Date: Fri, 18 Oct 2002 17:11:30 -0700 (PDT)
Subject: [Mailman-Developers] VERP code
In-Reply-To: <15792.41137.327805.809630@gargle.gargle.HOWL>
Message-ID: <Pine.GSO.4.44.0210181706020.18626-100000@scotch.ucf.ics.uci.edu>

On Fri, 18 Oct 2002, Barry A. Warsaw wrote:

> I'm Cc'ing Norbert Bollow who wrote the README.QMAIL file.  Perhaps he
> has some ideas about how to make this work.
>
> Hi Norbert!
> -Barry
>

I'm using Stonewall's suggestion to modify the regex.  It seems the
easiest(best?) solution so far.  I'll let it run on my lists for the
weekend and see what happens.  Initial tests suggest VERP now runs
properly.

Is Norbert actively maintaining the README.QMAIL file?  As soon as my VERP
stuff is more tested, I'd like to push some documentation changes for
review and eventual inclusion.  Nothing major, but better than nothing.

-Ed


From noreply@sourceforge.net  Sat Oct 19 02:24:04 2002
From: noreply@sourceforge.net (noreply@sourceforge.net)
Date: Fri, 18 Oct 2002 18:24:04 -0700
Subject: [Mailman-Developers] [ mailman-Patches-625482 ] i18n List-Id
Message-ID: <E182iLU-0006EV-00@usw-sf-web1.sourceforge.net>

Patches item #625482, was opened at 2002-10-19 01:24
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=300103&aid=625482&group_id=103

Category: internationalization
Group: Mailman 2.1
Status: Open
Resolution: None
Priority: 5
Submitted By: Tokio Kikuchi (tkikuchi)
Assigned to: Nobody/Anonymous (nobody)
Summary: i18n List-Id

Initial Comment:
List description is a short phrase which describes list
character and set by the list administrator. This is
added in every list-related mail in the 'List-Id'
header. With this patch, you can safely make this
phrase i18n. (Otherwise, break RFC)
This is a quick hack copying mostly from the code for
'Subject' header and may need optimization.


----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=300103&aid=625482&group_id=103

From barry@python.org  Sat Oct 19 02:41:26 2002
From: barry@python.org (Barry A. Warsaw)
Date: Fri, 18 Oct 2002 21:41:26 -0400
Subject: [Mailman-Developers] VERP code
References: <15792.41137.327805.809630@gargle.gargle.HOWL>
	<Pine.GSO.4.44.0210181706020.18626-100000@scotch.ucf.ics.uci.edu>
Message-ID: <15792.47174.81779.219625@gargle.gargle.HOWL>


>>>>> "EL" == Edmund Lau <edlau@ucf.ics.uci.edu> writes:

    >> I'm Cc'ing Norbert Bollow who wrote the README.QMAIL file.
    >> Perhaps he has some ideas about how to make this work.
    >> Hi Norbert!  -Barry

    EL> I'm using Stonewall's suggestion to modify the regex.  It
    EL> seems the easiest(best?) solution so far.  I'll let it run on
    EL> my lists for the weekend and see what happens.  Initial tests
    EL> suggest VERP now runs properly.

Cool.

    EL> Is Norbert actively maintaining the README.QMAIL file?

Not sure.  I haven't heard from Norbert in a while.
    
    EL> As soon as my VERP stuff is more tested, I'd like to push some
    EL> documentation changes for review and eventual inclusion.
    EL> Nothing major, but better than nothing.

+1

Thanks,
-Barry

From sb.list@sb.org  Sat Oct 19 03:46:46 2002
From: sb.list@sb.org (Stonewall Ballard)
Date: Fri, 18 Oct 2002 22:46:46 -0400
Subject: [Mailman-Developers] VERP'd reply-to not being used by bounce
	detector
In-Reply-To: <15792.40961.98299.855335@gargle.gargle.HOWL>
Message-ID: <B9D63FD6.4976C%sb.list@sb.org>

On 10/18/02 7:57 PM, "Barry A. Warsaw" <barry@python.org> wrote:

> 
>>>>>> "SB" == Stonewall Ballard <stoney@sb.org> writes:
> 
>   SB> I'm using Mailman 2.1b+, from a CVS build earlier today.
> 
>   SB> I sent a batch of messages out, and got a dozen bounces. Each
>   SB> one was forwarded to me as "uncaught", but the reply-to had
>   SB> the verp'd address in it.
> 
> I recently changed the code in BounceRunner.py so that if the bounce
> format (verp or otherwise) is recognized but yeilds no addresses which
> are members of the list, the bouncing message is forwarded as
> "uncaught".
> 
> Could this be what you're seeing?

I believe that this was caused by list locks being left by a crashed
archiver, therefore blocking the bounce detector from determining if it was
a list member.

 - Stoney


From noreply@sourceforge.net  Sat Oct 19 06:30:04 2002
From: noreply@sourceforge.net (noreply@sourceforge.net)
Date: Fri, 18 Oct 2002 22:30:04 -0700
Subject: [Mailman-Developers] 
 [ mailman-Bugs-625516 ] Archiver ends in NotLockedError
Message-ID: <E182mBY-0008Hw-00@usw-sf-web3.sourceforge.net>

Bugs item #625516, was opened at 2002-10-19 05:30
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=100103&aid=625516&group_id=103

Category: Pipermail
Group: 2.1 beta
Status: Open
Resolution: None
Priority: 5
Submitted By: Tokio Kikuchi (tkikuchi)
Assigned to: Nobody/Anonymous (nobody)
Summary: Archiver ends in NotLockedError

Initial Comment:
With most recent CVS version, archiver ends with
following error message in logs/error and the message
is shunted. No actual error is observed in the archived
messages.
Fixing unlock=0 to unlock=1 in Archiver/HyperArch.py as
was in rev 2.19 solved the problem. May be we should
tell the HyperArch that the mlist is locked already.
Here is the error log.

Oct 19 13:24:22 2002 (77321) Uncaught runner exception:
<LockFile 137238860: /ho
me/mailman4/locks/mm-test.lock [unlocked: 18000sec]
pid=77321>: None
Oct 19 13:24:22 2002 (77321) Traceback (most recent
call last):
  File "/home/mailman4/Mailman/Queue/Runner.py", line
105, in _oneloop
    self._onefile(msg, msgdata)
  File "/home/mailman4/Mailman/Queue/Runner.py", line
154, in _onefile
    keepqueued = self._dispose(mlist, msg, msgdata)
  File "/home/mailman4/Mailman/Queue/ArchRunner.py",
line 74, in _dispose
    mlist.Save()
  File "/home/mailman4/Mailman/MailList.py", line 490,
in Save
    self.__lock.refresh()
  File "/home/mailman4/Mailman/LockFile.py", line 222,
in refresh
    raise NotLockedError, '%s: %s' % (repr(self),
self.__read())
NotLockedError: <LockFile 137238860:
/home/mailman4/locks/mm-test.lock [unlocked
: 18000sec] pid=77321>: None



----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=100103&aid=625516&group_id=103

From spacey-mailman@lenin.nu  Sat Oct 19 16:15:22 2002
From: spacey-mailman@lenin.nu (Peter C. Norton)
Date: Sat, 19 Oct 2002 08:15:22 -0700
Subject: [Mailman-Developers] VERP code
In-Reply-To: <Pine.GSO.4.44.0210181118500.17506-100000@scotch.ucf.ics.uci.edu>
References: <B9D57BF5.494FE%sb.list@sb.org>
	<Pine.GSO.4.44.0210181118500.17506-100000@scotch.ucf.ics.uci.edu>
Message-ID: <20021019151522.GG1050@lenin.nu>

On Fri, Oct 18, 2002 at 11:40:40AM -0700, Edmund Lau wrote:
> qmail has a general catch-all address.  It's defined by a .qmail-default
> file (or .qmail-ext-default).  For instance if I had a .qmail-blah-default
> in my home directory, any Email with prefix to edlau-blah- gets processed
> by it.  But note the trailing hyphen.  I can't have a .qmail-blah+default
> as qmail won't recognize it, unless there's some heavy, magical trickery.
> 
> Back to mailman, I can use a "+" as a delimiter, but there needs to be a
> "-" right before it.  Is there a regex that would get that?  So the new
> VERP_FORMAT would be '%(bounces)s-+%(mailbox)s=%(host)s'.  I would assume
> this is better than just matching the last "-" anyway as, like you said, a
> user can have that character in their address.
> 
> Ideas?
> -Ed

man qmail-users

It documents /var/qmail/users/assign.  You can define the seperator
for the user in this file.  It isn't system-wide, it just takes effect
for deliveries targeted at the user specified.

-Peter

-- 
The 5 year plan:
In five years we'll make up another plan.
Or just re-use this one.

From noreply@sourceforge.net  Sat Oct 19 18:19:28 2002
From: noreply@sourceforge.net (noreply@sourceforge.net)
Date: Sat, 19 Oct 2002 10:19:28 -0700
Subject: [Mailman-Developers] 
 [ mailman-Patches-401270 ] Moderated newsgroup support
Message-ID: <E182xG4-0007lh-00@usw-sf-web5.sourceforge.net>

Patches item #401270, was opened at 2000-08-23 16:55
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=300103&aid=401270&group_id=103

Category: list administration
Group: None
>Status: Closed
Resolution: None
Priority: 5
Submitted By: Barry A. Warsaw (bwarsaw)
Assigned to: Barry A. Warsaw (bwarsaw)
Summary: Moderated newsgroup support

Initial Comment:
 

----------------------------------------------------------------------

>Comment By: Barry A. Warsaw (bwarsaw)
Date: 2002-10-19 13:19

Message:
Logged In: YES 
user_id=12800

I'm finally closing this because I think I have a much
simpler approach for Mailman 2.1

----------------------------------------------------------------------

Comment By: Barry A. Warsaw (bwarsaw)
Date: 2000-10-14 00:19

Message:
Re-open this so it's easier to find.

----------------------------------------------------------------------

Comment By: Barry A. Warsaw (bwarsaw)
Date: 2000-09-19 22:28

Message:
Provides hackish support for moderated newsgroups through the admindb interface.  Postponed until after 2.0 final.

----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=300103&aid=401270&group_id=103

From noreply@sourceforge.net  Sat Oct 19 22:04:08 2002
From: noreply@sourceforge.net (noreply@sourceforge.net)
Date: Sat, 19 Oct 2002 14:04:08 -0700
Subject: [Mailman-Developers] 
 [ mailman-Bugs-625516 ] Archiver ends in NotLockedError
Message-ID: <E1830lU-0008W3-00@usw-sf-web3.sourceforge.net>

Bugs item #625516, was opened at 2002-10-19 01:30
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=100103&aid=625516&group_id=103

Category: Pipermail
Group: 2.1 beta
>Status: Closed
>Resolution: Fixed
Priority: 5
Submitted By: Tokio Kikuchi (tkikuchi)
>Assigned to: Barry A. Warsaw (bwarsaw)
Summary: Archiver ends in NotLockedError

Initial Comment:
With most recent CVS version, archiver ends with
following error message in logs/error and the message
is shunted. No actual error is observed in the archived
messages.
Fixing unlock=0 to unlock=1 in Archiver/HyperArch.py as
was in rev 2.19 solved the problem. May be we should
tell the HyperArch that the mlist is locked already.
Here is the error log.

Oct 19 13:24:22 2002 (77321) Uncaught runner exception:
<LockFile 137238860: /ho
me/mailman4/locks/mm-test.lock [unlocked: 18000sec]
pid=77321>: None
Oct 19 13:24:22 2002 (77321) Traceback (most recent
call last):
  File "/home/mailman4/Mailman/Queue/Runner.py", line
105, in _oneloop
    self._onefile(msg, msgdata)
  File "/home/mailman4/Mailman/Queue/Runner.py", line
154, in _onefile
    keepqueued = self._dispose(mlist, msg, msgdata)
  File "/home/mailman4/Mailman/Queue/ArchRunner.py",
line 74, in _dispose
    mlist.Save()
  File "/home/mailman4/Mailman/MailList.py", line 490,
in Save
    self.__lock.refresh()
  File "/home/mailman4/Mailman/LockFile.py", line 222,
in refresh
    raise NotLockedError, '%s: %s' % (repr(self),
self.__read())
NotLockedError: <LockFile 137238860:
/home/mailman4/locks/mm-test.lock [unlocked
: 18000sec] pid=77321>: None



----------------------------------------------------------------------

>Comment By: Barry A. Warsaw (bwarsaw)
Date: 2002-10-19 17:04

Message:
Logged In: YES 
user_id=12800

I think this is occurring because the archiver pickles the
MailList and the MailList pickles the lock object.  When the
MailList -- and thus the lock -- is unpickled, and then
destroyed, the __del__ runs and unlocks the list out from
underneath us.  This happens because you now have two
MailList objects in the same process pointing to the same
lock.  That ordinarily can't happen, except in this
unpickling situation.

I think.

----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=100103&aid=625516&group_id=103

From noreply@sourceforge.net  Sun Oct 20 16:48:00 2002
From: noreply@sourceforge.net (noreply@sourceforge.net)
Date: Sun, 20 Oct 2002 08:48:00 -0700
Subject: [Mailman-Developers] 
 [ mailman-Bugs-621689 ] Subscribe requests dropped from admin db
Message-ID: <E183IJ6-0004we-00@usw-sf-web2.sourceforge.net>

Bugs item #621689, was opened at 2002-10-10 23:10
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=100103&aid=621689&group_id=103

Category: (un)subscribing
Group: 2.1 beta
Status: Open
Resolution: None
Priority: 5
Submitted By: Mark Dadgar (mdadgar)
Assigned to: Nobody/Anonymous (nobody)
Summary: Subscribe requests dropped from admin db

Initial Comment:
Occasionally, subscribe requests (via the web page) to a list that 
requires moderator approval for subscriptions never make it to the 
admin db for that list.  They are listed in the subscribe log file, but 
never show up on the admin db page.

Log file entries look like this:

Oct 10 18:43:43 2002 (31168) eventmasters: pending   
XXXXXX.swfla.rr.com

But the admin db page for the list shows no outstanding issues.

I have verified that there was an actual subscription request 
behind this log entry.

----------------------------------------------------------------------

>Comment By: Barry A. Warsaw (bwarsaw)
Date: 2002-10-20 11:48

Message:
Logged In: YES 
user_id=12800

What version of Mailman are you using.  It's import to know
the exact version, i.e. 2.1beta3, or if you're running from
cvs.  A lot of these problems have been fixed in cvs, which
will soon be released as 2.1 beta 4.

----------------------------------------------------------------------

Comment By: Mark Dadgar (mdadgar)
Date: 2002-10-11 00:29

Message:
Logged In: YES 
user_id=598228

I just verified, by cat'ing pending.pck, that the appropriate changes 
seem to be in there.  But the web interface never picks them up.

----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=100103&aid=621689&group_id=103

From noreply@sourceforge.net  Sun Oct 20 18:14:41 2002
From: noreply@sourceforge.net (noreply@sourceforge.net)
Date: Sun, 20 Oct 2002 10:14:41 -0700
Subject: [Mailman-Developers] 
 [ mailman-Bugs-621689 ] Subscribe requests dropped from admin db
Message-ID: <E183Jez-0001ux-00@usw-sf-web1.sourceforge.net>

Bugs item #621689, was opened at 2002-10-10 20:10
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=100103&aid=621689&group_id=103

Category: (un)subscribing
Group: 2.1 beta
Status: Open
Resolution: None
Priority: 5
Submitted By: Mark Dadgar (mdadgar)
Assigned to: Nobody/Anonymous (nobody)
Summary: Subscribe requests dropped from admin db

Initial Comment:
Occasionally, subscribe requests (via the web page) to a list that 
requires moderator approval for subscriptions never make it to the 
admin db for that list.  They are listed in the subscribe log file, but 
never show up on the admin db page.

Log file entries look like this:

Oct 10 18:43:43 2002 (31168) eventmasters: pending   
XXXXXX.swfla.rr.com

But the admin db page for the list shows no outstanding issues.

I have verified that there was an actual subscription request 
behind this log entry.

----------------------------------------------------------------------

>Comment By: Mark Dadgar (mdadgar)
Date: 2002-10-20 10:14

Message:
Logged In: YES 
user_id=598228

Sorry - I'm running 2.1b3.

----------------------------------------------------------------------

Comment By: Barry A. Warsaw (bwarsaw)
Date: 2002-10-20 08:48

Message:
Logged In: YES 
user_id=12800

What version of Mailman are you using.  It's import to know
the exact version, i.e. 2.1beta3, or if you're running from
cvs.  A lot of these problems have been fixed in cvs, which
will soon be released as 2.1 beta 4.

----------------------------------------------------------------------

Comment By: Mark Dadgar (mdadgar)
Date: 2002-10-10 21:29

Message:
Logged In: YES 
user_id=598228

I just verified, by cat'ing pending.pck, that the appropriate changes 
seem to be in there.  But the web interface never picks them up.

----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=100103&aid=621689&group_id=103

From fil@rezo.net  Sun Oct 20 20:05:11 2002
From: fil@rezo.net (Fil)
Date: Sun, 20 Oct 2002 21:05:11 +0200
Subject: [Mailman-Developers] email2.4.3
Message-ID: <20021020190510.GD23093@rezo.net>

Hi,

I'm just trying to update my MM cvs installation, and that means going up from email2.1
to email2.4.3 : here's the result :

miel:~/mailman/mailman/misc/email-2.4.3# python testall.py
........................................................EEEE..............................................................................................................
======================================================================
ERROR: test_add_header (email.test.test_email.TestMIMEAudio)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "email/test/test_email.py", line 767, in setUp
    fp = open(findfile('audiotest.au'), 'rb')
IOError: [Errno 2] No such file or directory: 'audiotest.au'
======================================================================
ERROR: test_custom_encoder (email.test.test_email.TestMIMEAudio)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "email/test/test_email.py", line 767, in setUp
    fp = open(findfile('audiotest.au'), 'rb')
IOError: [Errno 2] No such file or directory: 'audiotest.au'
======================================================================
ERROR: test_encoding (email.test.test_email.TestMIMEAudio)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "email/test/test_email.py", line 767, in setUp
    fp = open(findfile('audiotest.au'), 'rb')
IOError: [Errno 2] No such file or directory: 'audiotest.au'
======================================================================
ERROR: test_guess_minor_type (email.test.test_email.TestMIMEAudio)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "email/test/test_email.py", line 767, in setUp
    fp = open(findfile('audiotest.au'), 'rb')
IOError: [Errno 2] No such file or directory: 'audiotest.au'
----------------------------------------------------------------------
Ran 170 tests in 1.187s

FAILED (errors=4)


-- Fil


From fil@rezo.net  Sun Oct 20 20:07:18 2002
From: fil@rezo.net (Fil)
Date: Sun, 20 Oct 2002 21:07:18 +0200
Subject: [Mailman-Developers] email2.4.3
In-Reply-To: <20021020190510.GD23093@rezo.net>
References: <20021020190510.GD23093@rezo.net>
Message-ID: <20021020190718.GE23093@rezo.net>

> I'm just trying to update my MM cvs installation, and that means going up from email2.1

Then I get this when I 'make install' mailman :

/usr/local/mailman/pythonlib/korean/python/hangul.py:271: SyntaxWarning: local name 'kwargs' in 'format' shadows use of 'kwargs' as global in nested scope 'lambda'
  def format(fmtstr, *args, **kwargs):



-- Fil


From fil@rezo.net  Sun Oct 20 20:40:27 2002
From: fil@rezo.net (Fil)
Date: Sun, 20 Oct 2002 21:40:27 +0200
Subject: [Mailman-Developers] email2.4.3
In-Reply-To: <20021020190718.GE23093@rezo.net>
References: <20021020190510.GD23093@rezo.net>
	<20021020190718.GE23093@rezo.net>
Message-ID: <20021020194027.GA7174@rezo.net>

> Then I get this when I 'make install' mailman :
> /usr/local/mailman/pythonlib/korean/python/hangul.py:271: SyntaxWarning: local name 'kwargs' in 'format' shadows use of 'kwargs' as global in nested scope 'lambda'
>   def format(fmtstr, *args, **kwargs):

Seems like this was coming from not having rerun 'configure'

But I'm still getting this traceback quite often:

Oct 20 21:38:46 2002 (6966) Uncaught runner exception: ASCII decoding error:
ordinal not in range(128)
Oct 20 21:38:46 2002 (6966) Traceback (most recent call last):
  File "/usr/local/mailman/Mailman/Queue/Runner.py", line 105, in _oneloop
    self._onefile(msg, msgdata)
  File "/usr/local/mailman/Mailman/Queue/Runner.py", line 154, in _onefile
    keepqueued = self._dispose(mlist, msg, msgdata)
  File "/usr/local/mailman/Mailman/Queue/CommandRunner.py", line 207, in
_dispose
    res.send_response()
  File "/usr/local/mailman/Mailman/Queue/CommandRunner.py", line 159, in
send_response
    _('The results of your email commands'))
  File "/usr/local/mailman/Mailman/Message.py", line 203, in __init__
    self['Subject'] = Header(subject, charset, header_name='Subject')
  File "/usr/local/mailman/pythonlib/email/Header.py", line 164, in __init__
    self.append(s, charset)
  File "/usr/local/mailman/pythonlib/email/Header.py", line 224, in append
    unicode(s, charset.get_output_charset())
UnicodeError: ASCII decoding error: ordinal not in range(128)

-- Fil


From fil@rezo.net  Sun Oct 20 20:49:40 2002
From: fil@rezo.net (Fil)
Date: Sun, 20 Oct 2002 21:49:40 +0200
Subject: [Mailman-Developers] another traceback
Message-ID: <20021020194940.GC7174@rezo.net>


Oct 20 21:41:10 2002 (6965) Uncaught runner exception: 'Mailman.Queue.re'
module has no attribute 'search'
Oct 20 21:41:10 2002 (6965) Traceback (most recent call last):
  File "/usr/local/mailman/Mailman/Queue/Runner.py", line 105, in _oneloop
    self._onefile(msg, msgdata)
  File "/usr/local/mailman/Mailman/Queue/Runner.py", line 154, in _onefile
    keepqueued = self._dispose(mlist, msg, msgdata)
  File "/usr/local/mailman/Mailman/Queue/BounceRunner.py", line 71, in
_dispose
    addrs = verp_bounce(mlist, msg)
  File "/usr/local/mailman/Mailman/Queue/BounceRunner.py", line 151, in
verp_bounce
    mo = re.search(mm_cfg.VERP_REGEXP, to)
AttributeError: 'Mailman.Queue.re' module has no attribute 'search'


(current CVS)  This one is weird, I must admit -- what's the difference
betwenn re.search and re.match by the way?


-- Fil


From fil@rezo.net  Sun Oct 20 20:52:49 2002
From: fil@rezo.net (Fil)
Date: Sun, 20 Oct 2002 21:52:49 +0200
Subject: [Mailman-Developers] another traceback
In-Reply-To: <20021020194940.GC7174@rezo.net>
References: <20021020194940.GC7174@rezo.net>
Message-ID: <20021020195249.GD7174@rezo.net>

> AttributeError: 'Mailman.Queue.re' module has no attribute 'search'

There was a re.pyc file in my Mailman/Queue/ directory, can you believe it?

-- Fil


From fil@rezo.net  Sun Oct 20 22:04:43 2002
From: fil@rezo.net (Fil)
Date: Sun, 20 Oct 2002 23:04:43 +0200
Subject: [Mailman-Developers] mime filtering
Message-ID: <20021020210443.GA14456@rezo.net>


Dear all,

what mime-filtering is doing - can do - could do is not clear to me:

* how do I get such a message when things happen, or, say, when is it
  silent, when does it say what it does ?
    "[ Picked text/plain from multipart/alternative ]"

* Is it possible to plug in more mime converters : for instance, I'm on a
 Mailman list (one which I don't run) where msword files are deleted. Is it
 possible not to delete them, but to filter them through a doc2text filter?

* has anyone a "good combination" of mime filters that could be recommended
  to my listmasters? -- whatever "good" means to you ;-)

that's all for now!

-- Fil


From Daniel.Buchmann@bibsys.no  Mon Oct 21 00:04:06 2002
From: Daniel.Buchmann@bibsys.no (Daniel Buchmann)
Date: 21 Oct 2002 01:04:06 +0200
Subject: [Mailman-Developers] A quick patch and a .po entry question
Message-ID: <1035155049.6229.18.camel@fornax.bibsys.no>


---------------------- multipart/signed attachment

--=-PGV4PHC+OTeib7IxnrNj
Hi, attaching a quick patch for some invalid html in ContentFilter.py.
(an "<a href>" tag was not properly terminated and there was no linked
text...)

Also, something seems to be weird about the last sentence in
#: Mailman/Queue/BounceRunner.py:173, or maybe my english interpretation
is not good ;)

"You, as the list administrators have requested to receive all
unrecognized bounce messages."

Should it be:

"You, as the list administrator, have requested to receive all
unrecognized bounce messages."

or something like:

"You have received this notice, as the list administrators have
requested to receive all unrecognized bounce messages."

??


-Daniel, norwegian translator for Mailman


--=-PGV4PHC+OTeib7IxnrNj
A non-text attachment was scrubbed...
Name: contentfilter.patch
Type: text/x-diff
Size: 730 bytes
Desc: not available
Url : http://mail.python.org/pipermail-21/mailman-developers/attachments/20021021/530fa1f4/contentfilter.bin

--=-PGV4PHC+OTeib7IxnrNj--

---------------------- multipart/signed attachment
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 232 bytes
Desc: This is a digitally signed message part
Url : http://mail.python.org/pipermail-21/mailman-developers/attachments/20021021/530fa1f4/attachment.bin

---------------------- multipart/signed attachment--



From noreply@sourceforge.net  Mon Oct 21 00:33:55 2002
From: noreply@sourceforge.net (noreply@sourceforge.net)
Date: Sun, 20 Oct 2002 16:33:55 -0700
Subject: [Mailman-Developers] 
 [ mailman-Bugs-626125 ] listinfo page doesn't show newsgrp name
Message-ID: <E183PZz-0000bB-00@usw-sf-web3.sourceforge.net>

Bugs item #626125, was opened at 2002-10-21 09:33
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=100103&aid=626125&group_id=103

Category: Web/CGI
Group: None
Status: Open
Resolution: None
Priority: 5
Submitted By: Ben Elliston (elliston)
Assigned to: Nobody/Anonymous (nobody)
Summary: listinfo page doesn't show newsgrp name

Initial Comment:
If a mailing list is gatewayed to a newsgroup, the
listinfo page does not help a user discover what the
newsgroup is, or where to read it!

Can this be added?


----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=100103&aid=626125&group_id=103

From jwblist@olympus.net  Mon Oct 21 02:23:31 2002
From: jwblist@olympus.net (John W Baxter)
Date: Sun, 20 Oct 2002 18:23:31 -0700
Subject: [Mailman-Developers] another traceback
In-Reply-To: <20021020194940.GC7174@rezo.net>
References: <20021020194940.GC7174@rezo.net>
Message-ID: <p05200a00b9d90623f684@[192.168.1.50]>

At 21:49 +0200 10/20/2002, Fil wrote:
>(current CVS)  This one is weird, I must admit -- what's the difference
>betwenn re.search and re.match by the way?

re.search searches the target string for a match to the regular expression.
It's what one coming from Perl (and others) is used to.

re.match succeeds only if the RE matches at the start of the target string
(or at the (optional) starting position within the string).

The difference, I think, is inherited from Python's old and deprecated
string module, with string.search and string.match behaving as above.
These are now more often seen as the search() and match() methods on
string-like objects (such as strings).

  --John
-- 
John Baxter   jwblist@olympus.net      Port Ludlow, WA, USA

From gward@python.net  Mon Oct 21 16:22:30 2002
From: gward@python.net (Greg Ward)
Date: Mon, 21 Oct 2002 11:22:30 -0400
Subject: [Mailman-Developers] A quick patch and a .po entry question
In-Reply-To: <1035155049.6229.18.camel@fornax.bibsys.no>
References: <1035155049.6229.18.camel@fornax.bibsys.no>
Message-ID: <20021021152230.GA13454@cthulhu.gerg.ca>

On 21 October 2002, Daniel Buchmann said:
> Also, something seems to be weird about the last sentence in
> #: Mailman/Queue/BounceRunner.py:173, or maybe my english interpretation
> is not good ;)

I think I would tweak that entire message to:

  The attached message was received as a bounce, but either the bounce format
  was not recognized, or no member addresses could be extracted from it.
  (This list has been configured to send all unrecognized bounce messages
  to the list administrator(s).)

Native English speaker and a rather picky one at that --

        Greg
-- 
Greg Ward <gward@python.net>                         http://www.gerg.ca/
An elephant: a mouse built to government specifications.

From barry@python.org  Mon Oct 21 17:16:58 2002
From: barry@python.org (Barry A. Warsaw)
Date: Mon, 21 Oct 2002 12:16:58 -0400
Subject: [Mailman-Developers] A quick patch and a .po entry question
References: <1035155049.6229.18.camel@fornax.bibsys.no>
	<20021021152230.GA13454@cthulhu.gerg.ca>
Message-ID: <15796.10362.948684.23079@gargle.gargle.HOWL>


>>>>> "GW" == Greg Ward <gward@python.net> writes:

    GW> Native English speaker and a rather picky one at that --

Thanks Greg.

naive-english-speaker-who-likes-to-pick-things-ly y'rs,
-Barry

From barry@python.org  Mon Oct 21 19:35:36 2002
From: barry@python.org (Barry A. Warsaw)
Date: Mon, 21 Oct 2002 14:35:36 -0400
Subject: [Mailman-Developers] another traceback
References: <20021020194940.GC7174@rezo.net>
	<20021020195249.GD7174@rezo.net>
Message-ID: <15796.18680.257079.23161@gargle.gargle.HOWL>


>>>>> "fil" ==   <fil@rezo.net> writes:

    >> AttributeError: 'Mailman.Queue.re' module has no attribute
    >> 'search'

    fil> There was a re.pyc file in my Mailman/Queue/ directory, can
    fil> you believe it?

That's a weird one, alright!  Any idea how that got there?

-Barry

From fil@rezo.net  Mon Oct 21 19:38:38 2002
From: fil@rezo.net (Fil)
Date: Mon, 21 Oct 2002 20:38:38 +0200
Subject: [Mailman-Developers] another traceback
In-Reply-To: <15796.18680.257079.23161@gargle.gargle.HOWL>
References: <20021020194940.GC7174@rezo.net> <20021020195249.GD7174@rezo.net>
	<15796.18680.257079.23161@gargle.gargle.HOWL>
Message-ID: <20021021183837.GB26130@rezo.net>

>     fil> There was a re.pyc file in my Mailman/Queue/ directory, can
>     fil> you believe it?
> 
> That's a weird one, alright!  Any idea how that got there?

Nope, and I was stupid enough to delete it (I just wanted it to work!)
instead of moving it to the lab for further analysis. Sorry! (next time
maybe).

-- Fil


From barry@python.org  Mon Oct 21 19:41:50 2002
From: barry@python.org (Barry A. Warsaw)
Date: Mon, 21 Oct 2002 14:41:50 -0400
Subject: [Mailman-Developers] another traceback
References: <20021020194940.GC7174@rezo.net>
	<20021020195249.GD7174@rezo.net>
	<15796.18680.257079.23161@gargle.gargle.HOWL>
	<20021021183837.GB26130@rezo.net>
Message-ID: <15796.19054.598265.675136@gargle.gargle.HOWL>


>>>>> "fil" ==   <fil@rezo.net> writes:

    fil> Nope, and I was stupid enough to delete it (I just wanted it
    fil> to work!)  instead of moving it to the lab for further
    fil> analysis. Sorry! (next time maybe).

n.p.

From barry@python.org  Mon Oct 21 19:51:26 2002
From: barry@python.org (Barry A. Warsaw)
Date: Mon, 21 Oct 2002 14:51:26 -0400
Subject: [Mailman-Developers] mime filtering
References: <20021020210443.GA14456@rezo.net>
Message-ID: <15796.19630.825613.393195@gargle.gargle.HOWL>


>>>>> "fil" ==   <fil@rezo.net> writes:


    fil> * how do I get such a message when things happen, or, say,
    fil> when is it silent, when does it say what it does ?  "[ Picked
    fil> text/plain from multipart/alternative ]"

MimeDel isn't very chatty about what it's doing, other than by
sticking a header in telling you it /did/ change the message.

    fil> * Is it possible to plug in more mime converters : for
    fil> instance, I'm on a Mailman list (one which I don't run) where
    fil> msword files are deleted. Is it possible not to delete them,
    fil> but to filter them through a doc2text filter?

Anything's possible <wink>, but MimeDel isn't easily modular this
way.  It wouldn't be a bad idea to add something like this for a
future version.

    fil> * has anyone a "good combination" of mime filters that could
    fil> be recommended to my listmasters? -- whatever "good" means to
    fil> you ;-)

Chuq's posted some recommendations recently.

-Barry

From jon@csh.rit.edu  Mon Oct 21 19:56:25 2002
From: jon@csh.rit.edu (Jon Parise)
Date: Mon, 21 Oct 2002 14:56:25 -0400
Subject: [Mailman-Developers] usage() going to stderr
Message-ID: <20021021185625.GB22948@csh.rit.edu>

Most of the scripts in the bin/ directory send their usage() output to
sys.stderr.  Unfortunately, this makes it difficult to pipe the usage
output to a program such as more(1) (without redirecting stderr to
stdout).

Thoughts?

-- 
Jon Parise (jon@csh.rit.edu)  ::  http://www.csh.rit.edu/~jon/

From barry@python.org  Mon Oct 21 19:56:26 2002
From: barry@python.org (Barry A. Warsaw)
Date: Mon, 21 Oct 2002 14:56:26 -0400
Subject: [Mailman-Developers] email2.4.3
References: <20021020190510.GD23093@rezo.net>
Message-ID: <15796.19930.789759.982356@gargle.gargle.HOWL>


>>>>> "fil" ==   <fil@rezo.net> writes:

    fil> I'm just trying to update my MM cvs installation, and that
    fil> means going up from email2.1 to email2.4.3 : here's the
    fil> result :

Is your Python installed from RPM or built from source?  I'm guessing
the former, otherwise the audiotest.au file would have been picked up
from your Python installation.  Because it lives in a different part
of the tree, it doesn't get packaged up with the email package.

For now, you can safely ignore these errors.

-Barry

From fil@rezo.net  Mon Oct 21 20:02:40 2002
From: fil@rezo.net (Fil)
Date: Mon, 21 Oct 2002 21:02:40 +0200
Subject: [Mailman-Developers] email2.4.3
In-Reply-To: <15796.19930.789759.982356@gargle.gargle.HOWL>
References: <20021020190510.GD23093@rezo.net>
	<15796.19930.789759.982356@gargle.gargle.HOWL>
Message-ID: <20021021190240.GA699@rezo.net>

>     fil> I'm just trying to update my MM cvs installation, and that
>     fil> means going up from email2.1 to email2.4.3 : here's the
>     fil> result :
> 
> Is your Python installed from RPM or built from source? 

debian package

> I'm guessing the former, otherwise the audiotest.au file would have been
> picked up from your Python installation.  Because it lives in a different
> part of the tree, it doesn't get packaged up with the email package.
> For now, you can safely ignore these errors.

Good!

-- Fil


From barry@python.org  Mon Oct 21 20:07:32 2002
From: barry@python.org (Barry A. Warsaw)
Date: Mon, 21 Oct 2002 15:07:32 -0400
Subject: [Mailman-Developers] email2.4.3
References: <20021020190510.GD23093@rezo.net>
	<15796.19930.789759.982356@gargle.gargle.HOWL>
	<20021021190240.GA699@rezo.net>
Message-ID: <15796.20596.883128.350260@gargle.gargle.HOWL>


>>>>> "fil" ==   <fil@rezo.net> writes:

    >> Is your Python installed from RPM or built from source?

    fil> debian package

    >> I'm guessing the former, otherwise the audiotest.au file would
    >> have been picked up from your Python installation.  Because it
    >> lives in a different part of the tree, it doesn't get packaged
    >> up with the email package.  For now, you can safely ignore
    >> these errors.

    fil> Good!

For kicks, you might try installing a -devel package, if they're
separate, to see if that gets you the audiotest.au file.

-Barry

From barry@python.org  Mon Oct 21 20:10:54 2002
From: barry@python.org (Barry A. Warsaw)
Date: Mon, 21 Oct 2002 15:10:54 -0400
Subject: [Mailman-Developers] email2.4.3
References: <20021020190510.GD23093@rezo.net>
	<20021020190718.GE23093@rezo.net>
Message-ID: <15796.20798.350012.67867@gargle.gargle.HOWL>


>>>>> "fil" ==   <fil@rezo.net> writes:

    >> I'm just trying to update my MM cvs installation, and that
    >> means going up from email2.1

    fil> Then I get this when I 'make install' mailman :

    fil> /usr/local/mailman/pythonlib/korean/python/hangul.py:271:
    fil> SyntaxWarning: local name 'kwargs' in 'format' shadows use of
    fil> 'kwargs' as global in nested scope 'lambda' def
    fil> format(fmtstr, *args, **kwargs):

That looks like a (minor) incompatibility with the Korean codecs
package and the version of Python you're using.  It's not a horrible
problem (since it's just a warning).  Would you mind contacting the
Korean codec development team about this?

-Barry

From fil@rezo.net  Mon Oct 21 20:14:15 2002
From: fil@rezo.net (Fil)
Date: Mon, 21 Oct 2002 21:14:15 +0200
Subject: [Mailman-Developers] email2.4.3
In-Reply-To: <15796.20798.350012.67867@gargle.gargle.HOWL>
References: <20021020190510.GD23093@rezo.net>
	<20021020190718.GE23093@rezo.net>
	<15796.20798.350012.67867@gargle.gargle.HOWL>
Message-ID: <20021021191415.GA2065@rezo.net>

>     fil> /usr/local/mailman/pythonlib/korean/python/hangul.py:271:
>     fil> SyntaxWarning: local name 'kwargs' in 'format' shadows use of
>     fil> 'kwargs' as global in nested scope 'lambda' def
>     fil> format(fmtstr, *args, **kwargs):
> 
> That looks like a (minor) incompatibility with the Korean codecs
> package and the version of Python you're using.  It's not a horrible
> problem (since it's just a warning).  Would you mind contacting the
> Korean codec development team about this?

This disappeared after I re-ran 'configure'; I don't think I have more to
say than this traceback (which can be forwarded to the Korean codec
development team... except that I don't know who/where to send it to).

-- Fil


From barry@python.org  Mon Oct 21 20:20:38 2002
From: barry@python.org (Barry A. Warsaw)
Date: Mon, 21 Oct 2002 15:20:38 -0400
Subject: [Mailman-Developers] email2.4.3
References: <20021020190510.GD23093@rezo.net>
	<20021020190718.GE23093@rezo.net>
	<15796.20798.350012.67867@gargle.gargle.HOWL>
	<20021021191415.GA2065@rezo.net>
Message-ID: <15796.21382.758323.772213@gargle.gargle.HOWL>


>>>>> "fil" ==   <fil@rezo.net> writes:

    fil> This disappeared after I re-ran 'configure'; I don't think I
    fil> have more to say than this traceback (which can be forwarded
    fil> to the Korean codec development team... except that I don't
    fil> know who/where to send it to).

Ok, I won't sweat it then.
-Barry

From barry@python.org  Mon Oct 21 20:32:08 2002
From: barry@python.org (Barry A. Warsaw)
Date: Mon, 21 Oct 2002 15:32:08 -0400
Subject: [Mailman-Developers] email2.4.3
References: <20021020190510.GD23093@rezo.net>
	<20021020190718.GE23093@rezo.net>
	<20021020194027.GA7174@rezo.net>
Message-ID: <15796.22072.609641.308438@gargle.gargle.HOWL>


>>>>> "fil" ==   <fil@rezo.net> writes:

    >> Then I get this when I 'make install' mailman :
    >> /usr/local/mailman/pythonlib/korean/python/hangul.py:271:
    >> SyntaxWarning: local name 'kwargs' in 'format' shadows use of
    >> 'kwargs' as global in nested scope 'lambda' def format(fmtstr,
    >> *args, **kwargs):

    fil> Seems like this was coming from not having rerun 'configure'

Ah.

    fil> But I'm still getting this traceback quite often:

    fil> Oct 20 21:38:46 2002 (6966) Uncaught runner exception: ASCII
    fil> decoding error: ordinal not in range(128) Oct 20 21:38:46
    fil> 2002 (6966) Traceback (most recent call last):
    fil>   File "/usr/local/mailman/Mailman/Queue/Runner.py", line
    fil> 105, in _oneloop self._onefile(msg, msgdata) File
    fil> "/usr/local/mailman/Mailman/Queue/Runner.py", line 154, in
    fil> _onefile keepqueued = self._dispose(mlist, msg, msgdata) File
    fil> "/usr/local/mailman/Mailman/Queue/CommandRunner.py", line
    fil> 207, in
    fil> _dispose
    fil>     res.send_response() File
    fil> "/usr/local/mailman/Mailman/Queue/CommandRunner.py", line
    fil> 159, in
    fil> send_response
    fil>     _('The results of your email commands')) File
    fil> "/usr/local/mailman/Mailman/Message.py", line 203, in
    fil> __init__ self['Subject'] = Header(subject, charset,
    fil> header_name='Subject') File
    fil> "/usr/local/mailman/pythonlib/email/Header.py", line 164, in
    fil> __init__ self.append(s, charset) File
    fil> "/usr/local/mailman/pythonlib/email/Header.py", line 224, in
    fil> append unicode(s, charset.get_output_charset())
    fil> UnicodeError: ASCII decoding error: ordinal not in range(128)

Can you explain what you're doing to provide these errors?  E.g. what
commands are you sending, and what language is your list set to?

-Barry

From fil@rezo.net  Mon Oct 21 20:44:40 2002
From: fil@rezo.net (Fil)
Date: Mon, 21 Oct 2002 21:44:40 +0200
Subject: [Mailman-Developers] email2.4.3
In-Reply-To: <15796.22072.609641.308438@gargle.gargle.HOWL>
References: <20021020190510.GD23093@rezo.net>
	<20021020190718.GE23093@rezo.net> <20021020194027.GA7174@rezo.net>
	<15796.22072.609641.308438@gargle.gargle.HOWL>
Message-ID: <20021021194440.GC2065@rezo.net>

>     fil> append unicode(s, charset.get_output_charset())
>     fil> UnicodeError: ASCII decoding error: ordinal not in range(128)
> 
> Can you explain what you're doing to provide these errors?  E.g. what
> commands are you sending, and what language is your list set to?

I'm just writing a blank mail to test-unsubscribe. The server's default
language is French, and the test list is also in French.

I then receive the confirmation request from Mailman, answer it, and when
the server gets the confirmation it barks again (below, sorry for
formatting). And this time I don't get a receipt or anything, and I'm not
unsubscribed. Confirming via the web interface works though.

Now let me try with the list set up in English (USA) and my user prefs in
English too: ah ah! this time I receive not only the confirmation request
but also this message:
    """
    Subject: The results of your email commands
    [-- Attachement #1 --]
    [-- Type : text/plain, Codage : 7bit, Taille : 0,2K --]
    The results of your email command are provided below. Attached is your
    original message.
    - Results:
        A removal confirmation message has been sent.
    - Unprocessed:
        -- Fil
    - Done.
    """



>>>>>>>>>>>>>>>>>>>>>>>>>>

Oct 21 21:37:50 2002 (13586) Uncaught runner exception: ASCII decoding
error: ordinal not in range(128)
Oct 21 21:37:50 2002 (13586) Traceback (most recent call last):
  File "/usr/local/mailman/Mailman/Queue/Runner.py", line 105, in _oneloop
    self._onefile(msg, msgdata)
  File "/usr/local/mailman/Mailman/Queue/Runner.py", line 154, in _onefile
    keepqueued = self._dispose(mlist, msg, msgdata)
  File "/usr/local/mailman/Mailman/Queue/CommandRunner.py", line 198, in
_dispose   
    res.process()
  File "/usr/local/mailman/Mailman/Queue/CommandRunner.py", line 93, in
process
    stop = self.do_command(cmd, args)
  File "/usr/local/mailman/Mailman/Queue/CommandRunner.py", line 117, in
do_command 
    return self.do_command(cmd, args)
  File "/usr/local/mailman/Mailman/Queue/CommandRunner.py", line 119, in
do_command 
    return handler.process(self, args)
  File "/usr/local/mailman/Mailman/Commands/cmd_confirm.py", line 44, in
process
    results = mlist.ProcessConfirmation(cookie, res.msg)
  File "/usr/local/mailman/Mailman/MailList.py", line 1108, in
ProcessConfirmation  
    self.ApprovedDeleteMember(addr, whence='web confirmation')
  File "/usr/local/mailman/Mailman/MailList.py", line 944, in
ApprovedDeleteMember  
    self.SendUnsubscribeAck(name)
  File "/usr/local/mailman/Mailman/Deliverer.py", line 78, in
SendUnsubscribeAck
    Utils.wrap(self.goodbye_msg), self.getMemberLanguage(addr))
  File "/usr/local/mailman/Mailman/Message.py", line 203, in __init__
    self['Subject'] = Header(subject, charset, header_name='Subject')
  File "/usr/local/mailman/pythonlib/email/Header.py", line 164, in __init__
    self.append(s, charset)
  File "/usr/local/mailman/pythonlib/email/Header.py", line 224, in append
    unicode(s, charset.get_output_charset())
UnicodeError: ASCII decoding error: ordinal not in range(128)



-- Fil


From chuqui@plaidworks.com  Mon Oct 21 21:03:23 2002
From: chuqui@plaidworks.com (Chuq Von Rospach)
Date: Mon, 21 Oct 2002 13:03:23 -0700
Subject: [Mailman-Developers] usage() going to stderr
In-Reply-To: <20021021185625.GB22948@csh.rit.edu>
Message-ID: <2732E68C-E530-11D6-87E7-0003934516A8@plaidworks.com>

> Most of the scripts in the bin/ directory send their usage() output to
> sys.stderr.  Unfortunately, this makes it difficult to pipe the usage
> output to a program such as more(1) (without redirecting stderr to
> stdout).
>
> Thoughts?

that's standard programming. Feature, not bug.


-- 
Chuq Von Rospach, Architech
chuqui@plaidworks.com -- http://www.plaidworks.com/chuqui/blog/



From jon@csh.rit.edu  Mon Oct 21 21:13:55 2002
From: jon@csh.rit.edu (Jon Parise)
Date: Mon, 21 Oct 2002 16:13:55 -0400
Subject: [Mailman-Developers] usage() going to stderr
In-Reply-To: <2732E68C-E530-11D6-87E7-0003934516A8@plaidworks.com>
References: <20021021185625.GB22948@csh.rit.edu>
	<2732E68C-E530-11D6-87E7-0003934516A8@plaidworks.com>
Message-ID: <20021021201355.GD22948@csh.rit.edu>

On Mon, Oct 21, 2002 at 01:03:23PM -0700, Chuq Von Rospach wrote:

> >Most of the scripts in the bin/ directory send their usage() output to
> >sys.stderr.  Unfortunately, this makes it difficult to pipe the usage
> >output to a program such as more(1) (without redirecting stderr to
> >stdout).
> 
> that's standard programming. Feature, not bug.
 
'vim --help' spits its usage output to stdout, as does 'python -h'.

I can understand errors being printed to syserr (duh), but usage
information is more consumer grade, I think.

-- 
Jon Parise (jon@csh.rit.edu)  ::  http://www.csh.rit.edu/~jon/

From barry@python.org  Mon Oct 21 21:20:15 2002
From: barry@python.org (Barry A. Warsaw)
Date: Mon, 21 Oct 2002 16:20:15 -0400
Subject: [Mailman-Developers] usage() going to stderr
References: <20021021185625.GB22948@csh.rit.edu>
	<2732E68C-E530-11D6-87E7-0003934516A8@plaidworks.com>
	<20021021201355.GD22948@csh.rit.edu>
Message-ID: <15796.24959.411000.135837@gargle.gargle.HOWL>


>>>>> "JP" == Jon Parise <jon@csh.rit.edu> writes:

    JP> 'vim --help' spits its usage output to stdout, as does 'python
    JP> -h'.

    JP> I can understand errors being printed to syserr (duh), but
    JP> usage information is more consumer grade, I think.

I could almost be convinced that usage(0) should go to stdout, but I'm
not sure it's worth changing all the scripts.  At least we're
consistent. :)

-Barry

From barry@python.org  Mon Oct 21 21:36:31 2002
From: barry@python.org (Barry A. Warsaw)
Date: Mon, 21 Oct 2002 16:36:31 -0400
Subject: [Mailman-Developers] email2.4.3
References: <20021020190510.GD23093@rezo.net>
	<20021020190718.GE23093@rezo.net>
	<20021020194027.GA7174@rezo.net>
	<15796.22072.609641.308438@gargle.gargle.HOWL>
	<20021021194440.GC2065@rezo.net>
Message-ID: <15796.25935.172890.765152@gargle.gargle.HOWL>


>>>>> "fil" ==   <fil@rezo.net> writes:

    fil> I'm just writing a blank mail to test-unsubscribe. The
    fil> server's default language is French, and the test list is
    fil> also in French.

    fil> I then receive the confirmation request from Mailman, answer
    fil> it, and when the server gets the confirmation it barks again
    fil> (below, sorry for formatting). And this time I don't get a
    fil> receipt or anything, and I'm not unsubscribed. Confirming via
    fil> the web interface works though.

Thanks, I can reproduce this.  Now for the fun part!
-Barry

From barry@python.org  Mon Oct 21 21:53:46 2002
From: barry@python.org (Barry A. Warsaw)
Date: Mon, 21 Oct 2002 16:53:46 -0400
Subject: [Mailman-Developers] email2.4.3
References: <20021020190510.GD23093@rezo.net>
	<20021020190718.GE23093@rezo.net>
	<20021020194027.GA7174@rezo.net>
	<15796.22072.609641.308438@gargle.gargle.HOWL>
	<20021021194440.GC2065@rezo.net>
Message-ID: <15796.26970.516713.646192@gargle.gargle.HOWL>


>>>>> "fil" ==   <fil@rezo.net> writes:

    fil> I'm just writing a blank mail to test-unsubscribe. The
    fil> server's default language is French, and the test list is
    fil> also in French.

I found the bug, and will check in a fix shortly.

-Barry

From donn@u.washington.edu  Mon Oct 21 21:58:16 2002
From: donn@u.washington.edu (Donn Cave)
Date: Mon, 21 Oct 2002 13:58:16 -0700
Subject: [Mailman-Developers] usage() going to stderr
Message-ID: <200210212058.g9LKwGmT017289@mailhost1.u.washington.edu>

Quoth barry@python.org (Barry A. Warsaw):
| >>>>> "JP" == Jon Parise <jon@csh.rit.edu> writes:
|
|     JP> 'vim --help' spits its usage output to stdout, as does 'python
|     JP> -h'.
|
|     JP> I can understand errors being printed to syserr (duh), but
|     JP> usage information is more consumer grade, I think.
|
| I could almost be convinced that usage(0) should go to stdout, but I'm
| not sure it's worth changing all the scripts.  At least we're
| consistent. :)

Consistent with general practice, too, the way I see it.  The clearest
cases are programs that generate useful output data, like list_members.
Those should never mix that data with diagnostic output of any kind, so
the user can safely redirect into some other program with the expectation
of getting only that output data.

"rmlist 2>&1 | more" isn't all that convenient, but it's far easier
than separating outputs that have been mixed inappropriately.  If it's
useful, a shell function like "amore () { "$@" 2>&1 | more; }" would be
easier on the fingers.

Incidentally, I note an inconsistency - "mailmanctl" is coded to write
usage to sys.stderr, but stderr is gimmicked by LogStdErr and actually
goes to unit 1, not unit 2.

	Donn Cave, donn@u.washington.edu

From dgc@uchicago.edu  Mon Oct 21 21:59:04 2002
From: dgc@uchicago.edu (David Champion)
Date: Mon, 21 Oct 2002 15:59:04 -0500
Subject: [Mailman-Developers] usage() going to stderr
In-Reply-To: <2732E68C-E530-11D6-87E7-0003934516A8@plaidworks.com>
References: <20021021185625.GB22948@csh.rit.edu>
	<2732E68C-E530-11D6-87E7-0003934516A8@plaidworks.com>
Message-ID: <20021021205904.GA25431@dust.uchicago.edu>

* On 2002.10.21, in <2732E68C-E530-11D6-87E7-0003934516A8@plaidworks.com>,
*	"Chuq Von Rospach" <chuqui@plaidworks.com> wrote:
> >Most of the scripts in the bin/ directory send their usage() output to
> >sys.stderr.  Unfortunately, this makes it difficult to pipe the usage
> >output to a program such as more(1) (without redirecting stderr to
> >stdout).
> >
> >Thoughts?
> 
> that's standard programming. Feature, not bug.

My feeling on the topic, having seen many programs that swing each way,
is that usage() should go to stderr when it's caused by a usage error,
but stdout when it's requested via "-h", "-?", "--help", or whatever.
At least this keeps with the general rule of thumb that errors go to
stderr, and output requested by a user goes to stdout.

Not that I adhere to this reliably, myself. :/

-- 
 -D.                    We establised a fine coffee. What everybody can say
 Sun Project, APC/UCCO  TASTY! It's fresh, so-mild, with some special coffee's
 University of Chicago  bitter and sourtaste. "LET'S HAVE SUCH A COFFEE! NOW!"
 dgc@uchicago.edu       Please love CAFE MIAMI. Many thanks.

From noreply@sourceforge.net  Mon Oct 21 22:12:37 2002
From: noreply@sourceforge.net (noreply@sourceforge.net)
Date: Mon, 21 Oct 2002 14:12:37 -0700
Subject: [Mailman-Developers] 
 [ mailman-Patches-626550 ] Add "Date" field to admindb summary
Message-ID: <E183jqn-0002Cb-00@usw-sf-web2.sourceforge.net>

Patches item #626550, was opened at 2002-10-21 17:12
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=300103&aid=626550&group_id=103

Category: Web UI
Group: Mailman 2.1
Status: Open
Resolution: None
Priority: 5
Submitted By: Jon Parise (jparise)
Assigned to: Nobody/Anonymous (nobody)
Summary: Add "Date" field to admindb summary

Initial Comment:
The attached patch adds the message's "Date" value to
the summary fields (which already include Subject,
Size, and Reason).

----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=300103&aid=626550&group_id=103

From barry@python.org  Mon Oct 21 22:14:20 2002
From: barry@python.org (Barry A. Warsaw)
Date: Mon, 21 Oct 2002 17:14:20 -0400
Subject: [Mailman-Developers] usage() going to stderr
References: <200210212058.g9LKwGmT017289@mailhost1.u.washington.edu>
Message-ID: <15796.28204.628059.847559@gargle.gargle.HOWL>


>>>>> "DC" == Donn Cave <donn@u.washington.edu> writes:

    DC> Incidentally, I note an inconsistency - "mailmanctl" is coded
    DC> to write usage to sys.stderr, but stderr is gimmicked by
    DC> LogStdErr and actually goes to unit 1, not unit 2.

qrunner is the same.  Ug, did someone open up a can of worms around
here?  I think I'm going to change LogStdErr() so that it will
optionally log to the real stderr as well as the log file.  That's
probably the best fix and can be done in a mostly transparent way, I
think.

>>>>> "DC" == David Champion <dgc@uchicago.edu> writes:

    DC> My feeling on the topic, having seen many programs that swing
    DC> each way, is that usage() should go to stderr when it's caused
    DC> by a usage error, but stdout when it's requested via "-h",
    DC> "-?", "--help", or whatever.  At least this keeps with the
    DC> general rule of thumb that errors go to stderr, and output
    DC> requested by a user goes to stdout.

I'd agree in general, but it would mean we'd have to patch every
usage() function in cron/* and bin/*.  Not something I want to spend
time on.

-Barry

From bob@nleaudio.com  Mon Oct 21 22:45:20 2002
From: bob@nleaudio.com (Bob Puff@NLE)
Date: Mon, 21 Oct 2002 17:45:20 -0400
Subject: [Mailman-Developers] usage() going to stderr
References: <20021021185625.GB22948@csh.rit.edu>
	<2732E68C-E530-11D6-87E7-0003934516A8@plaidworks.com>
	<20021021205904.GA25431@dust.uchicago.edu>
Message-ID: <3DB47570.23595C7D@nleaudio.com>

Yeah, it's a real pain when you do 'program -?', and the output scrolls way more than one screen's worth, and you can't more it.

Bob


David Champion wrote:
> 
> * On 2002.10.21, in <2732E68C-E530-11D6-87E7-0003934516A8@plaidworks.com>,
> *       "Chuq Von Rospach" <chuqui@plaidworks.com> wrote:
> > >Most of the scripts in the bin/ directory send their usage() output to
> > >sys.stderr.  Unfortunately, this makes it difficult to pipe the usage
> > >output to a program such as more(1) (without redirecting stderr to
> > >stdout).
> > >
> > >Thoughts?
> >
> > that's standard programming. Feature, not bug.
> 
> My feeling on the topic, having seen many programs that swing each way,
> is that usage() should go to stderr when it's caused by a usage error,
> but stdout when it's requested via "-h", "-?", "--help", or whatever.
> At least this keeps with the general rule of thumb that errors go to
> stderr, and output requested by a user goes to stdout.
> 
> Not that I adhere to this reliably, myself. :/
> 
> --
>  -D.                    We establised a fine coffee. What everybody can say
>  Sun Project, APC/UCCO  TASTY! It's fresh, so-mild, with some special coffee's
>  University of Chicago  bitter and sourtaste. "LET'S HAVE SUCH A COFFEE! NOW!"
>  dgc@uchicago.edu       Please love CAFE MIAMI. Many thanks.
> 
> _______________________________________________
> Mailman-Developers mailing list
> Mailman-Developers@python.org
> http://mail.python.org/mailman-21/listinfo/mailman-developers

From jon@csh.rit.edu  Mon Oct 21 22:57:44 2002
From: jon@csh.rit.edu (Jon Parise)
Date: Mon, 21 Oct 2002 17:57:44 -0400
Subject: [Mailman-Developers] usage() going to stderr
In-Reply-To: <15796.28204.628059.847559@gargle.gargle.HOWL>
References: <200210212058.g9LKwGmT017289@mailhost1.u.washington.edu>
	<15796.28204.628059.847559@gargle.gargle.HOWL>
Message-ID: <20021021215744.GB26639@csh.rit.edu>


---------------------- multipart/mixed attachment
On Mon, Oct 21, 2002 at 05:14:20PM -0400, Barry A. Warsaw wrote:

> >>>>> "DC" == David Champion <dgc@uchicago.edu> writes:
> 
>     DC> My feeling on the topic, having seen many programs that swing
>     DC> each way, is that usage() should go to stderr when it's caused
>     DC> by a usage error, but stdout when it's requested via "-h",
>     DC> "-?", "--help", or whatever.  At least this keeps with the
>     DC> general rule of thumb that errors go to stderr, and output
>     DC> requested by a user goes to stdout.
> 
> I'd agree in general, but it would mean we'd have to patch every
> usage() function in cron/* and bin/*.  Not something I want to spend
> time on.
 
I don't know what possessed me to do this, but see the attached patch.

-- 
Jon Parise (jon@csh.rit.edu)  ::  http://www.csh.rit.edu/~jon/

---------------------- multipart/mixed attachment
Index: bin/add_members
===================================================================
RCS file: /cvsroot/mailman/mailman/bin/add_members,v
retrieving revision 2.12
diff -u -r2.12 add_members
--- bin/add_members	6 Sep 2002 02:31:43 -0000	2.12
+++ bin/add_members	21 Oct 2002 21:56:28 -0000
@@ -90,9 +90,13 @@
 
 
 def usage(status, msg=''):
-    print >> sys.stderr, _(__doc__)
+    if status:
+        fd = sys.stderr
+    else:
+        fd = sys.stdout
+    print >> fd, _(__doc__)
     if msg:
-        print >> sys.stderr, msg
+        print >> fd, msg
     sys.exit(status)
 
 
Index: bin/arch
===================================================================
RCS file: /cvsroot/mailman/mailman/bin/arch,v
retrieving revision 2.8
diff -u -r2.8 arch
--- bin/arch	9 Oct 2002 04:22:54 -0000	2.8
+++ bin/arch	21 Oct 2002 21:56:28 -0000
@@ -69,9 +69,13 @@
 
 
 def usage(code, msg=''):
-    print >> sys.stderr, _(__doc__)
+    if code:
+        fd = sys.stderr
+    else:
+        fd = sys.stdout
+    print >> fd, _(__doc__)
     if msg:
-        print >> sys.stderr, msg
+        print >> fd, msg
     sys.exit(code)
 
 
Index: bin/change_pw
===================================================================
RCS file: /cvsroot/mailman/mailman/bin/change_pw,v
retrieving revision 2.3
diff -u -r2.3 change_pw
--- bin/change_pw	11 Feb 2002 23:38:36 -0000	2.3
+++ bin/change_pw	21 Oct 2002 21:56:28 -0000
@@ -83,9 +83,13 @@
 
 
 def usage(code, msg=''):
-    print >> sys.stderr, _(__doc__)
+    if code:
+        fd = sys.stderr
+    else:
+        fd = sys.stdout
+    print >> fd, _(__doc__)
     if msg:
-        print >> sys.stderr, msg
+        print >> fd, msg
     sys.exit(code)
 
 
Index: bin/check_db
===================================================================
RCS file: /cvsroot/mailman/mailman/bin/check_db,v
retrieving revision 2.4
diff -u -r2.4 check_db
--- bin/check_db	5 Mar 2002 16:24:22 -0000	2.4
+++ bin/check_db	21 Oct 2002 21:56:28 -0000
@@ -66,9 +66,13 @@
 
 
 def usage(code, msg=''):
-    print >> sys.stderr, _(__doc__)
+    if code:
+        fd = sys.stderr
+    else:
+        fd = sys.stdout
+    print >> fd, _(__doc__)
     if msg:
-        print >> sys.stderr, msg
+        print >> fd, msg
     sys.exit(code)
 
 
Index: bin/check_perms
===================================================================
RCS file: /cvsroot/mailman/mailman/bin/check_perms,v
retrieving revision 2.12
diff -u -r2.12 check_perms
--- bin/check_perms	20 Oct 2002 01:58:40 -0000	2.12
+++ bin/check_perms	21 Oct 2002 21:56:28 -0000
@@ -295,9 +295,13 @@
 
 
 def usage(code, msg=''):
-    print >> sys.stderr, _(__doc__)
+    if code:
+        fd = sys.stderr
+    else:
+        fd = sys.stdout
+    print >> fd, _(__doc__)
     if msg:
-        print >> sys.stderr, msg
+        print >> fd, msg
     sys.exit(code)
 
 
Index: bin/cleanarch
===================================================================
RCS file: /cvsroot/mailman/mailman/bin/cleanarch,v
retrieving revision 2.3
diff -u -r2.3 cleanarch
--- bin/cleanarch	26 Aug 2002 16:37:42 -0000	2.3
+++ bin/cleanarch	21 Oct 2002 21:56:28 -0000
@@ -64,9 +64,13 @@
 
 
 def usage(code, msg=''):
-    print >> sys.stderr, _(__doc__)
+    if code:
+        fd = sys.stderr
+    else:
+        fd = sys.stdout
+    print >> fd, _(__doc__)
     if msg:
-        print >> sys.stderr, msg
+        print >> fd, msg
     sys.exit(code)
 
 
Index: bin/clone_member
===================================================================
RCS file: /cvsroot/mailman/mailman/bin/clone_member,v
retrieving revision 2.7
diff -u -r2.7 clone_member
--- bin/clone_member	16 Mar 2002 06:57:37 -0000	2.7
+++ bin/clone_member	21 Oct 2002 21:56:28 -0000
@@ -77,9 +77,13 @@
 
 
 def usage(code, msg=''):
-    print >> sys.stderr, _(__doc__)
+    if code:
+        fd = sys.stderr
+    else:
+        fd = sys.stdout
+    print >> fd, _(__doc__)
     if msg:
-        print >> sys.stderr, msg
+        print >> fd, msg
     sys.exit(code)
 
 
Index: bin/config_list
===================================================================
RCS file: /cvsroot/mailman/mailman/bin/config_list,v
retrieving revision 2.10
diff -u -r2.10 config_list
--- bin/config_list	17 Oct 2002 17:14:56 -0000	2.10
+++ bin/config_list	21 Oct 2002 21:56:29 -0000
@@ -79,9 +79,13 @@
 
 
 def usage(code, msg=''):
-    print >> sys.stderr, _(__doc__)
+    if code:
+        fd = sys.stderr
+    else:
+        fd = sys.stdout
+    print >> fd, _(__doc__)
     if msg:
-        print >> sys.stderr, msg
+        print >> fd, msg
     sys.exit(code)
 
 
Index: bin/dumpdb
===================================================================
RCS file: /cvsroot/mailman/mailman/bin/dumpdb,v
retrieving revision 2.11
diff -u -r2.11 dumpdb
--- bin/dumpdb	23 Jul 2002 06:14:57 -0000	2.11
+++ bin/dumpdb	21 Oct 2002 21:56:29 -0000
@@ -64,9 +64,13 @@
 
 
 def usage(code, msg=''):
-    print >> sys.stderr, _(__doc__) % globals()
+    if code:
+        fd = sys.stderr
+    else:
+        fd = sys.stdout
+    print >> fd, _(__doc__) % globals()
     if msg:
-        print >> sys.stderr, msg
+        print >> fd, msg
     sys.exit(code)
 
 
Index: bin/find_member
===================================================================
RCS file: /cvsroot/mailman/mailman/bin/find_member,v
retrieving revision 2.4
diff -u -r2.4 find_member
--- bin/find_member	16 Mar 2002 06:57:37 -0000	2.4
+++ bin/find_member	21 Oct 2002 21:56:29 -0000
@@ -72,9 +72,13 @@
 
 
 def usage(code, msg=''):
-    print >> sys.stderr, _(__doc__)
+    if code:
+        fd = sys.stderr
+    else:
+        fd = sys.stdout
+    print >> fd, _(__doc__)
     if msg:
-        print >> sys.stderr, msg
+        print >> fd, msg
     sys.exit(code)
 
 
Index: bin/genaliases
===================================================================
RCS file: /cvsroot/mailman/mailman/bin/genaliases,v
retrieving revision 2.10
diff -u -r2.10 genaliases
--- bin/genaliases	16 Mar 2002 06:57:37 -0000	2.10
+++ bin/genaliases	21 Oct 2002 21:56:29 -0000
@@ -42,9 +42,13 @@
 
 
 def usage(code, msg=''):
-    print >> sys.stderr, _(__doc__)
+    if code:
+        fd = sys.stderr
+    else:
+        fd = sys.stdout
+    print >> fd, _(__doc__)
     if msg:
-        print >> sys.stderr, msg
+        print >> fd, msg
     sys.exit(code)
 
 
Index: bin/inject
===================================================================
RCS file: /cvsroot/mailman/mailman/bin/inject,v
retrieving revision 2.0
diff -u -r2.0 inject
--- bin/inject	26 Jan 2002 23:32:13 -0000	2.0
+++ bin/inject	21 Oct 2002 21:56:29 -0000
@@ -52,9 +52,13 @@
 
 
 def usage(code, msg=''):
-    print >> sys.stderr, _(__doc__)
+    if code:
+        fd = sys.stderr
+    else:
+        fd = sys.stdout
+    print >> fd, _(__doc__)
     if msg:
-        print >> sys.stderr, msg
+        print >> fd, msg
     sys.exit(code)
 
 
Index: bin/list_admins
===================================================================
RCS file: /cvsroot/mailman/mailman/bin/list_admins,v
retrieving revision 2.3
diff -u -r2.3 list_admins
--- bin/list_admins	25 May 2002 07:07:33 -0000	2.3
+++ bin/list_admins	21 Oct 2002 21:56:29 -0000
@@ -53,9 +53,13 @@
 
 
 def usage(code, msg=''):
-    print >> sys.stderr, _(__doc__)
+    if code:
+        fd = sys.stderr
+    else:
+        fd = sys.stdout
+    print >> fd, _(__doc__)
     if msg:
-        print >> sys.stderr, msg
+        print >> fd, msg
     sys.exit(code)
 
 
Index: bin/list_lists
===================================================================
RCS file: /cvsroot/mailman/mailman/bin/list_lists,v
retrieving revision 2.3
diff -u -r2.3 list_lists
--- bin/list_lists	16 Mar 2002 06:57:37 -0000	2.3
+++ bin/list_lists	21 Oct 2002 21:56:29 -0000
@@ -51,9 +51,13 @@
 program = sys.argv[0]
 
 def usage(code, msg=''):
-    print >> sys.stderr, _(__doc__)
+    if code:
+        fd = sys.stderr
+    else:
+        fd = sys.stdout
+    print >> fd, _(__doc__)
     if msg:
-        print >> sys.stderr, msg
+        print >> fd, msg
     sys.exit(code)
 
 
Index: bin/list_members
===================================================================
RCS file: /cvsroot/mailman/mailman/bin/list_members,v
retrieving revision 2.10
diff -u -r2.10 list_members
--- bin/list_members	22 Aug 2002 03:48:25 -0000	2.10
+++ bin/list_members	21 Oct 2002 21:56:29 -0000
@@ -81,9 +81,13 @@
 
 
 def usage(code, msg=''):
-    print >> sys.stderr, _(__doc__)
+    if code:
+        fd = sys.stderr
+    else:
+        fd = sys.stdout
+    print >> fd, _(__doc__)
     if msg:
-        print >> sys.stderr, msg
+        print >> fd, msg
     sys.exit(code)
 
 
Index: bin/mailmanctl
===================================================================
RCS file: /cvsroot/mailman/mailman/bin/mailmanctl,v
retrieving revision 2.22
diff -u -r2.22 mailmanctl
--- bin/mailmanctl	21 Oct 2002 21:34:25 -0000	2.22
+++ bin/mailmanctl	21 Oct 2002 21:56:29 -0000
@@ -122,9 +122,13 @@
 
 
 def usage(code, msg=''):
-    print >> sys.stderr, _(__doc__)
+    if code:
+        fd = sys.stderr
+    else:
+        fd = sys.stdout
+    print >> fd, _(__doc__)
     if msg:
-        print >> sys.stderr, msg
+        print >> fd, msg
     sys.exit(code)
 
 
Index: bin/mmsitepass
===================================================================
RCS file: /cvsroot/mailman/mailman/bin/mmsitepass,v
retrieving revision 2.5
diff -u -r2.5 mmsitepass
--- bin/mmsitepass	16 Mar 2002 06:57:37 -0000	2.5
+++ bin/mmsitepass	21 Oct 2002 21:56:29 -0000
@@ -50,9 +50,13 @@
 
 
 def usage(code, msg=''):
-    print >> sys.stderr, _(__doc__)
+    if code:
+        fd = sys.stderr
+    else:
+        fd = sys.stdout
+    print >> fd, _(__doc__)
     if msg:
-        print >> sys.stderr, msg
+        print >> fd, msg
     sys.exit(code)
 
 
Index: bin/newlist
===================================================================
RCS file: /cvsroot/mailman/mailman/bin/newlist,v
retrieving revision 2.14
diff -u -r2.14 newlist
--- bin/newlist	6 Oct 2002 01:35:57 -0000	2.14
+++ bin/newlist	21 Oct 2002 21:56:29 -0000
@@ -85,9 +85,13 @@
 
 
 def usage(code, msg=''):
-    print >> sys.stderr, _(__doc__)
+    if code:
+        fd = sys.stderr
+    else:
+        fd = sys.stdout
+    print >> fd, _(__doc__)
     if msg:
-        print >> sys.stderr, msg
+        print >> fd, msg
     sys.exit(code)
 
 
Index: bin/pygettext.py
===================================================================
RCS file: /cvsroot/mailman/mailman/bin/pygettext.py,v
retrieving revision 2.5
diff -u -r2.5 pygettext.py
--- bin/pygettext.py	7 Sep 2001 23:18:47 -0000	2.5
+++ bin/pygettext.py	21 Oct 2002 21:56:30 -0000
@@ -183,9 +183,13 @@
 
 
 def usage(code, msg=''):
-    print >> sys.stderr, _(__doc__) % globals()
+    if code:
+        fd = sys.stderr
+    else:
+        fd = sys.stdout
+    print >> fd, _(__doc__) % globals()
     if msg:
-        print >> sys.stderr, msg
+        print >> fd, msg
     sys.exit(code)
 
 
Index: bin/qrunner
===================================================================
RCS file: /cvsroot/mailman/mailman/bin/qrunner,v
retrieving revision 2.8
diff -u -r2.8 qrunner
--- bin/qrunner	21 Oct 2002 21:34:41 -0000	2.8
+++ bin/qrunner	21 Oct 2002 21:56:30 -0000
@@ -89,9 +89,13 @@
 
 
 def usage(code, msg=''):
-    print >> sys.stderr, _(__doc__)
+    if code:
+        fd = sys.stderr
+    else:
+        fd = sys.stdout
+    print >> fd, _(__doc__)
     if msg:
-        print >> sys.stderr, msg
+        print >> fd, msg
     sys.exit(code)
 
 
Index: bin/remove_members
===================================================================
RCS file: /cvsroot/mailman/mailman/bin/remove_members,v
retrieving revision 2.5
diff -u -r2.5 remove_members
--- bin/remove_members	6 Mar 2002 15:18:42 -0000	2.5
+++ bin/remove_members	21 Oct 2002 21:56:30 -0000
@@ -61,9 +61,13 @@
 
 
 def usage(code, msg=''):
-    print >> sys.stderr, _(__doc__)
+    if code:
+        fd = sys.stderr
+    else:
+        fd = sys.stdout
+    print >> fd, _(__doc__)
     if msg:
-        print >> sys.stderr, msg
+        print >> fd, msg
     sys.exit(code)
 
 
Index: bin/rmlist
===================================================================
RCS file: /cvsroot/mailman/mailman/bin/rmlist,v
retrieving revision 2.5
diff -u -r2.5 rmlist
--- bin/rmlist	2 Jan 2002 02:02:11 -0000	2.5
+++ bin/rmlist	21 Oct 2002 21:56:30 -0000
@@ -49,9 +49,13 @@
 
 
 def usage(code, msg=''):
-    print >> sys.stderr, _(__doc__)
+    if code:
+        fd = sys.stderr
+    else:
+        fd = sys.stdout
+    print >> fd, _(__doc__)
     if msg:
-        print >> sys.stderr, msg
+        print >> fd, msg
     sys.exit(code)
 
 
Index: bin/sync_members
===================================================================
RCS file: /cvsroot/mailman/mailman/bin/sync_members,v
retrieving revision 2.10
diff -u -r2.10 sync_members
--- bin/sync_members	8 Oct 2002 19:57:38 -0000	2.10
+++ bin/sync_members	21 Oct 2002 21:56:30 -0000
@@ -93,9 +93,13 @@
 PROGRAM = sys.argv[0]
 
 def usage(code, msg=''):
-    print >> sys.stderr, _(__doc__)
+    if code:
+        fd = sys.stderr
+    else:
+        fd = sys.stdout
+    print >> fd, _(__doc__)
     if msg:
-        print >> sys.stderr, msg
+        print >> fd, msg
     sys.exit(code)
 
 
Index: bin/transcheck
===================================================================
RCS file: /cvsroot/mailman/mailman/bin/transcheck,v
retrieving revision 2.1
diff -u -r2.1 transcheck
--- bin/transcheck	15 Aug 2002 16:42:19 -0000	2.1
+++ bin/transcheck	21 Oct 2002 21:56:30 -0000
@@ -40,9 +40,13 @@
 
 
 def usage(code, msg):
-    print >> sys.stderr, _(__doc__)
+    if code:
+        fd = sys.stderr
+    else:
+        fd = sys.stdout
+    print >> fd, _(__doc__)
     if msg:
-        print >> sys.stderr, msg
+        print >> fd, msg
     sys.exit(code)
 
 
Index: bin/unshunt
===================================================================
RCS file: /cvsroot/mailman/mailman/bin/unshunt,v
retrieving revision 2.3
diff -u -r2.3 unshunt
--- bin/unshunt	21 Oct 2002 16:36:36 -0000	2.3
+++ bin/unshunt	21 Oct 2002 21:56:31 -0000
@@ -40,9 +40,13 @@
 
 
 def usage(code, msg=''):
-    print >> sys.stderr, _(__doc__)
+    if code:
+        fd = sys.stderr
+    else:
+        fd = sys.stdout
+    print >> fd, _(__doc__)
     if msg:
-        print >> sys.stderr, msg
+        print >> fd, msg
     sys.exit(code)
 
 
Index: bin/update
===================================================================
RCS file: /cvsroot/mailman/mailman/bin/update,v
retrieving revision 2.20
diff -u -r2.20 update
--- bin/update	20 Oct 2002 15:39:46 -0000	2.20
+++ bin/update	21 Oct 2002 21:56:31 -0000
@@ -523,9 +523,13 @@
 
 
 def usage(code, msg=''):
-    print >> sys.stderr, _(__doc__) % globals()
+    if code:
+        fd = sys.stderr
+    else:
+        fd = sys.stdout
+    print >> fd, _(__doc__) % globals()
     if msg:
-        print msg
+        print >> sys.stderr, msg
     sys.exit(code)
 
 
Index: bin/withlist
===================================================================
RCS file: /cvsroot/mailman/mailman/bin/withlist,v
retrieving revision 2.9
diff -u -r2.9 withlist
--- bin/withlist	14 Mar 2002 23:42:20 -0000	2.9
+++ bin/withlist	21 Oct 2002 21:56:31 -0000
@@ -137,9 +137,13 @@
 
 
 def usage(code, msg=''):
-    print >> sys.stderr, _(__doc__)
+    if code:
+        fd = sys.stderr
+    else:
+        fd = sys.stdout
+    print >> fd, _(__doc__)
     if msg:
-        print >> sys.stderr, msg
+        print >> fd, msg
     sys.exit(code)
 
 
Index: cron/bumpdigests
===================================================================
RCS file: /cvsroot/mailman/mailman/cron/bumpdigests,v
retrieving revision 2.6
diff -u -r2.6 bumpdigests
--- cron/bumpdigests	16 Mar 2002 06:57:37 -0000	2.6
+++ cron/bumpdigests	21 Oct 2002 21:56:31 -0000
@@ -48,9 +48,13 @@
 
 
 def usage(code, msg=''):
-    print >> sys.stderr, _(__doc__)
+    if code:
+        fd = sys.stderr
+    else:
+        fd = sys.stdout
+    print >> fd, _(__doc__)
     if msg:
-        print >> sys.stderr, msg
+        print >> fd, msg
     sys.exit(code)
 
 
Index: cron/disabled
===================================================================
RCS file: /cvsroot/mailman/mailman/cron/disabled,v
retrieving revision 2.7
diff -u -r2.7 disabled
--- cron/disabled	9 Aug 2002 21:32:20 -0000	2.7
+++ cron/disabled	21 Oct 2002 21:56:31 -0000
@@ -86,9 +86,13 @@
 
 
 def usage(code, msg=''):
-    print >> sys.stderr, _(__doc__)
+    if code:
+        fd = sys.stderr
+    else:
+        fd = sys.stdout
+    print >> fd, _(__doc__)
     if msg:
-        print >> sys.stderr, msg
+        print >> fd, msg
     sys.exit(code)
 
 
Index: cron/gate_news
===================================================================
RCS file: /cvsroot/mailman/mailman/cron/gate_news,v
retrieving revision 2.14
diff -u -r2.14 gate_news
--- cron/gate_news	21 Oct 2002 21:35:01 -0000	2.14
+++ cron/gate_news	21 Oct 2002 21:56:31 -0000
@@ -69,10 +69,14 @@
 
 
 def usage(status, msg=''):
-    print >> sys.stderr, _(__doc__)
+    if code:
+        fd = sys.stderr
+    else:
+        fd = sys.stdout
+    print >> fd, _(__doc__)
     if msg:
-        print msg
-    sys.exit(status)
+        print >> fd, msg
+    sys.exit(code)
 
 
 
Index: cron/mailpasswds
===================================================================
RCS file: /cvsroot/mailman/mailman/cron/mailpasswds,v
retrieving revision 2.11
diff -u -r2.11 mailpasswds
--- cron/mailpasswds	22 May 2002 02:39:42 -0000	2.11
+++ cron/mailpasswds	21 Oct 2002 21:56:31 -0000
@@ -67,9 +67,13 @@
 
 
 def usage(code, msg=''):
-    print >> sys.stderr, _(__doc__)
+    if code:
+        fd = sys.stderr
+    else:
+        fd = sys.stdout
+    print >> fd, _(__doc__)
     if msg:
-        print >> sys.stderr, msg
+        print >> fd, msg
     sys.exit(code)
 
 
Index: cron/nightly_gzip
===================================================================
RCS file: /cvsroot/mailman/mailman/cron/nightly_gzip,v
retrieving revision 2.5
diff -u -r2.5 nightly_gzip
--- cron/nightly_gzip	16 Mar 2002 06:57:37 -0000	2.5
+++ cron/nightly_gzip	21 Oct 2002 21:56:32 -0000
@@ -63,9 +63,13 @@
 VERBOSE = 0
 
 def usage(code, msg=''):
-    print __doc__ % globals()
+    if code:
+        fd = sys.stderr
+    else:
+        fd = sys.stdout
+    print >> fd, _(__doc__) % globals()
     if msg:
-        print msg
+        print >> fd, msg
     sys.exit(code)
 
 
Index: cron/senddigests
===================================================================
RCS file: /cvsroot/mailman/mailman/cron/senddigests,v
retrieving revision 2.7
diff -u -r2.7 senddigests
--- cron/senddigests	28 Aug 2002 16:06:15 -0000	2.7
+++ cron/senddigests	21 Oct 2002 21:56:32 -0000
@@ -48,9 +48,13 @@
 
 
 def usage(code, msg=''):
-    print >> sys.stderr, _(__doc__)
+    if code:
+        fd = sys.stderr
+    else:
+        fd = sys.stdout
+    print >> fd, _(__doc__)
     if msg:
-        print >> sys.stderr, msg
+        print >> fd, msg
     sys.exit(code)
 
 

---------------------- multipart/mixed attachment--

From barry@python.org  Mon Oct 21 23:05:13 2002
From: barry@python.org (Barry A. Warsaw)
Date: Mon, 21 Oct 2002 18:05:13 -0400
Subject: [Mailman-Developers] Rebuilding archives in 2.1b+
References: <B9D4B086.49365%stoney@sb.org>
Message-ID: <15796.31257.505831.634741@gargle.gargle.HOWL>


>>>>> "SB" == Stonewall Ballard <stoney@sb.org> writes:

    SB> I'm using yesterday's (10/16) CVS build.

    SB> I've managed to screw up my archives. Probably a permissions
    SB> problem (I had mailman_user set wrong for a while). Arch is
    SB> segfaulting trying to rebuild the date archives for a list.

    SB> Can I just delete all the html files and start over? Is that
    SB> sufficient?  What do I need to keep? What should the
    SB> owner/group of these files be?

As you might imagine, I tend to regenerate archives /a lot/, and
here's my recipe:

- cd /usr/local/mailman
- \rm -rf archives/private/mylist/*

(this last one is important to get just right)

- bin/arch mylist

As long as you don't mess with
archives/private/mylist.mbox/mylist.mbox everything will be cool.

BTW, I like your site Stoney.  I think we need to put together The
Mailman Jam Band. :)

-Barry

From barry@python.org  Mon Oct 21 23:32:42 2002
From: barry@python.org (Barry A. Warsaw)
Date: Mon, 21 Oct 2002 18:32:42 -0400
Subject: [Mailman-Developers] usage() going to stderr
References: <200210212058.g9LKwGmT017289@mailhost1.u.washington.edu>
	<15796.28204.628059.847559@gargle.gargle.HOWL>
	<20021021215744.GB26639@csh.rit.edu>
Message-ID: <15796.32906.395444.974876@gargle.gargle.HOWL>


>>>>> "JP" == Jon Parise <jon@csh.rit.edu> writes:

    JP> I don't know what possessed me to do this, but see the
    JP> attached patch.

Wow, cool!   Thanks,
-Barry

From jon@csh.rit.edu  Mon Oct 21 23:59:08 2002
From: jon@csh.rit.edu (Jon Parise)
Date: Mon, 21 Oct 2002 18:59:08 -0400
Subject: [Mailman-Developers] SpamAssassin handler for Mailman 2.1
Message-ID: <20021021225908.GC27873@csh.rit.edu>

I've taken James Henstridge's SpamAssassin handler[1] and updated it
for Mailman 2.1.  It's been running on the lists.horde.org mailing
lists since last night with no ill effects.  I'm going to keep it
running for a few more days before I'm confident in its performance,
but I'll be submitting it (or adding it to the open patch request)
soon.

[1] http://sourceforge.net/tracker/index.php?func=detail&aid=534577&group_id=103&atid=300103

-- 
Jon Parise (jon@csh.rit.edu)  ::  http://www.csh.rit.edu/~jon/

From noreply@sourceforge.net  Tue Oct 22 01:35:50 2002
From: noreply@sourceforge.net (noreply@sourceforge.net)
Date: Mon, 21 Oct 2002 17:35:50 -0700
Subject: [Mailman-Developers] 
 [ mailman-Bugs-214156 ] MIME quoted printable subject lines (PR#190)
Message-ID: <E183n1S-0006dF-00@usw-sf-web4.sourceforge.net>

Bugs item #214156, was opened at 2000-09-11 13:22
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=100103&aid=214156&group_id=103

Category: Pipermail
Group: None
Status: Closed
Resolution: Fixed
Priority: 6
Submitted By: Nobody/Anonymous (nobody)
Assigned to: Jeremy Hylton (jhylton)
Summary: MIME quoted printable subject lines (PR#190)

Initial Comment:
Jitterbug-Id: 190
Submitted-By: ziegler@ttt-atm.ttt.bme.hu
Date: Thu, 27 Jan 2000 10:11:10 -0500 (EST)
Version: v.1.0rc2
OS: Debian 2.1


The MIME quoted-printable decoding does not work for mail header lines!
It should be!

If I use accented national characters in mails (i.e., MIME quoted printable),
e.g.
in the subject line, or in the full name of sender/addressee, 
then the archive index is almost unreadable, since I see only
garbage instead of meaningful subject and sender lines, like this:

[Hsn-empl] =?iso-8859-1?Q?=DCdv?= =?iso-8859-1?Q?G=E1bor?= Ziegler 

It is also true in the plain text downloadable version, see the archive text
version
of the message above:

>From ziegler@ttt-atm.ttt.bme.hu Thu, 27 Jan 2000 14:36:19 +0100
Date: Thu, 27 Jan 2000 14:36:19 +0100
From: =?iso-8859-1?Q?G=E1bor?= Ziegler ziegler@ttt-atm.ttt.bme.hu
Subject: [Hsn-empl] =?iso-8859-1?Q?=DCdv?=

Üdvözlet mindenkinek!

Ez a lista kiserleti jelleggel jött létre, a HSNLabos oktatok
tajékoztatására.
A levelforgalom kicsi lesz, csak is a labor mukodesevel kapcsolatos
infokat tervezem ide.

Ez a lista zártkörû, tehát:
* a lista tagjainak listajat csak a tagok lathatjak
* a feliratkozasokat a adminisztrator kell jovahagyja.
* listara levelet kuldeni csak tagok, valamint masok (egyelore) a 
*.ttt.bme.hu,
  *ericsson.se domainekbol tudnak.

Lehetoseg van:
* a levelek atmeneti letiltasara, 
* napi egyszeri osszefoglalot kerni csak (digest), ami persze csak akkor
jon, ha volt forgalom,
* az archivumban bongeszni
* egyedi beállitasokat megvaltoztatni
* leiratkozni, feliratkozni, 
s mindezt egyszerüen a Weben keresztül! 

Remelem elnyeri tetszeseteket!

Udv:
                Gabor





====================================================================
Audit trail:
None

----------------------------------------------------------------------

Comment By: Mike Brown (mike_j_brown)
Date: 2002-10-21 17:35

Message:
Logged In: YES 
user_id=371366

Is there any chance this can be made to work in the indices 
as well, so that authors and subjects there are decoded in 
the same way?

Another suggestion -- If you can get the Unicode code point 
of the non-ASCII characters (difficult in Python 1.5.2, not too 
hard in 1.6, very easy in 2.0+), emit them in HTML as 
numeric character references so you don't have to worry 
about conflicts with the rest of the message.


----------------------------------------------------------------------

Comment By: Jeremy Hylton (jhylton)
Date: 2000-09-21 21:12

Message:
Fixed in rev. 1.30 of Mailman/Archiver/HyperArch.py.  If the charset of the header matches the charset of the message body, it will be decoded.


----------------------------------------------------------------------

Comment By: Barry A. Warsaw (bwarsaw)
Date: 2000-09-14 12:37

Message:
Unpostponing, reopening, and assigning to Jeremy

----------------------------------------------------------------------

Comment By: Barry A. Warsaw (bwarsaw)
Date: 2000-09-13 05:49

Message:
Yes, Pipermail should be improved in this regard.  Current recommendation is to use an external archiver.

----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=100103&aid=214156&group_id=103

From fil@rezo.net  Tue Oct 22 08:42:15 2002
From: fil@rezo.net (Fil)
Date: Tue, 22 Oct 2002 09:42:15 +0200
Subject: [Mailman-Developers] email2.4.3
In-Reply-To: <20021021194440.GC2065@rezo.net>
References: <20021020190510.GD23093@rezo.net>
	<20021020190718.GE23093@rezo.net> <20021020194027.GA7174@rezo.net>
	<15796.22072.609641.308438@gargle.gargle.HOWL>
	<20021021194440.GC2065@rezo.net>
Message-ID: <20021022074215.GD24706@rezo.net>

@ Fil <fil@rezo.net> :
> >     fil> append unicode(s, charset.get_output_charset())
> >     fil> UnicodeError: ASCII decoding error: ordinal not in range(128)

Here's another instance of the same bug, I suppose:


    Subject: Cron <mailman@miel> /usr/local/bin/python -S /usr/local/mailman/cron/checkdbs

Traceback (most recent call last):
  File "/usr/local/mailman/cron/checkdbs", line 116, in ?
    main()
  File "/usr/local/mailman/cron/checkdbs", line 75, in main
    text += '\n' + pending_requests(mlist)
UnicodeError: ASCII decoding error: ordinal not in range(128)



-- Fil


From ashank@rogers.com  Fri Oct 18 02:04:38 2002
From: ashank@rogers.com (Andre Shank)
Date: Thu, 17 Oct 2002 21:04:38 -0400
Subject: [Mailman-Developers] CF Scripts for Custom User Interface
Message-ID: <002901c27642$553a5300$4601a8c0@ashank.phub.net.cable.rogers.com>

This is a multi-part message in MIME format.

---------------------- multipart/alternative attachment
Hello,

We have very recently moved to Mailman - all seems to be working fine to =
date.  We're using mailman as the MLM for several of our corporate =
contact lists that are maintained in a database on a different server.  =
The database interface for the contact lists database is in Cold Fusion. =
 Basically, I'd like our Contact List app to serve as the UI for the =
Mailman lists.  I've written scripts which ensure that modifications to =
the Contact List data are also processed and updated by Mailman.  The =
scripts which emulate the list-user subscribe and unsubscribe functions =
work fine.  I would also like to be able to emulate the Membership =
Management functions of the Admin UI but am not having much luck.  I =
suspect this might have something to do with authentication...

Here's a scaled down verison of the script used to pass subscribe =
requests.

<cfhttp method=3D"post" =
url=3D"http://myserver.com/cgi-bin/mailman/subscribe/mylist">
  <cfhttpparam type=3D"FormField" name=3D"email"    =
value=3Dnewsubscriber@org.com>
  <cfhttpparam type=3D"FormField" name=3D"pw"    value =3D "test">
  <cfhttpparam type=3D"FormField" name=3D"pw-conf"   value =3D "test">
  <cfhttpparam type=3D"FormField" name=3D"digest"    value =3D "0">
</cfhttp>

The above is fine when used in cases where new subscribers would use my =
custom UI to subscribe to a list which requires subscription approvals.

This second script would allow subscribe requests to be done the same =
way as the "Mass Subscribe" of the Admin UI option does (e.g. bypass any =
approval step). It just won't work.

<cfhttp method=3D"post" =
url=3D"http://myserver.com/cgi-bin/mailman/admin/mylist/members">
 <cfhttpparam type=3D"FormField" name=3D"send_welcome_msg_to_this_batch" =
value=3D"0">
 <cfhttpparam type=3D"FormField" name=3D"subscribees" =
value=3Dnewsubscriber@org.com>
</cfhttp>

You can see from my little problem that I'm not a hard-core programmer =
but would greatly appreciate any assistance from those out there who =
are.


Regards,

Andr=E9 Shank
ashank@rogers.com


---------------------- multipart/alternative attachment--


From colinp@waikato.ac.nz  Fri Oct 18 05:23:35 2002
From: colinp@waikato.ac.nz (Colin Palmer)
Date: Fri, 18 Oct 2002 17:23:35 +1300 (NZDT)
Subject: [Mailman-Developers] VERP code
In-Reply-To: <B9D4E82D.49419%sb.list@sb.org>
Message-ID: <Pine.LNX.4.21.0210181714090.13577-100000@firefox.cc.waikato.ac.nz>

On Thu, 17 Oct 2002, Stonewall Ballard wrote:

> Are you saying that your VERP_FORMAT is '%(bounces)s-%(mailbox)s=%(host)s' ?
> 
> If so, I don't think that would work very well since the mailbox name could
> have a '-' in it, making part matching ambiguous.
> 
> If I misunderstood, can you clarify? What are your VERP_FORMAT and your
> VERP_REGEXP?

I'm using Mailman 2.1b3 here and I've been tinkering with the VERP support
too, trying to tie it in with qmail's, so I also went through the changing
+ to - before realising that since I was using a hacked version of
qmail-to-mailman.py I didn't need to... oh well.  Anyway, here's what I
have in mm_cfg.py

VERP_FORMAT = '%(bounces)s-%(mailbox)s=%(host)s'
VERP_REGEXP = r'^(?P<bounces>[^-]+?)-(?P<mailbox>[^=]+)=(?P<host>[^@]+)@.*$'

and I haven't seen any messages about 'unrecognised bounces', though it
hasn't been tested much at this point.

For 'normal' list deliveries though, I've set VERP_INTERVAL to 0 and use a
hacked-about version of Sendmail.py that forces the envelope sender to be
'listname-bounces-@mailhost.dom-@[]' and injects via qmail-inject

-- 
** Colin Palmer, Systems and Development Group, University of Waikato, NZ **


From Jost.Krieger+mailman@rub.de  Tue Oct 22 12:26:30 2002
From: Jost.Krieger+mailman@rub.de (Jost Krieger)
Date: Tue, 22 Oct 2002 13:26:30 +0200
Subject: [Mailman-Developers] 2.1b3 passing most(?) posts to the request
	address
Message-ID: <20021022112630.GV4725@ruhr-uni-bochum.de>

I haven't found this in the archives, but that may be my fault.

Since I upgraded to 2.1b3, my "vette" logfile ist filled with the following
type of messages:

Oct 22 11:20:31 2002 (2195) Precedence: bulk message ignored by: security-request@lists.ruhr-uni-bochum.de
Oct 22 11:32:30 2002 (2195) Precedence: bulk message ignored by: security-request@lists.ruhr-uni-bochum.de
Oct 22 12:29:52 2002 (2195) Precedence: bulk message ignored by: security-request@lists.ruhr-uni-bochum.de

These coincide with messages to this "security" list, and it has *not*
activated the administrivia setting.  (It is just redistribution of security
notices.) The messages to the list *are* "Precedence: bulk".

What may be going on?

Thanks for any help
Jost
-- 
| Jost.Krieger+sig@ruhr-uni-bochum.de  Please help stamp out spam! |
| Postmaster, JAPH, resident answer machine          am RZ der RUB |
| Pluralitas non est ponenda sine necessitate                      |
|                                 William of Ockham (1285-1347/49) |

From skye@ffwd.cx  Thu Oct 17 23:27:22 2002
From: skye@ffwd.cx (Skye Poier)
Date: Thu, 17 Oct 2002 15:27:22 -0700
Subject: [Mailman-Developers] (forw) Re: [Mailman-Users] bounce detection
	broken?
Message-ID: <20021017222722.GG54297@ffwd.cx>

Hello, I sent this to -users yesterday, but didn't get much response
so I'm copying it to -developers.

Thanks,
Skye

----- Forwarded message from Skye Poier <skye@ffwd.cx> -----

From: Skye Poier <skye@ffwd.cx>
To: mailman-users@python.org
Subject: Re: [Mailman-Users] bounce detection broken?
Date: Wed, 16 Oct 2002 12:58:47 -0700

Well, since nobody helped, I went ahead and started debugging the stuff
in Bouncers/*.py

First I took a look at why Hotmail bounces weren't being caught in
DSN.py.  Here's an example with my debugging:

Oct 16 12:46:01 2002 (11207) DSN parsing header {'action': 'failed',
'final-recipient': 'rfc822;djmattie@hotmail.com', 'status': '5.0.0',
'original-recipient': ''}
Oct 16 12:46:01 2002 (11207) DSN got val
Oct 16 12:46:01 2002 (11207) DSN parsing header {'action': 'failed',
'final-recipient': 'rfc822;primordialnature@hotmail.com', 'status':
'5.0.0', 'original-recipient': ''}
Oct 16 12:46:01 2002 (11207) DSN got val
Oct 16 12:46:01 2002 (11207) DSN parsing header {'action': 'failed',
'final-recipient': 'rfc822;j_j_vernon@hotmail.com', 'status': '5.0.0',
'original-recipient': ''}
Oct 16 12:46:01 2002 (11207) DSN got val

So for some reason the final-recipient was there but it wasn't picking
it up.  So on a hunch I changed this line:

       val = headers.get('original-recipient',
                         headers.get('final-recipient'))

to simply this:

        val = headers.get('final-recipient')

and now val is filled, and DSN bounce detection works in this case.

Why was the header not being found????????
Skye


Word on the street is Skye Poier said:
> Hello,
> 
> I just upgraded from mailman 2.0beta4 to 2.0.13 (python 2.2.1) and
> everything is going pretty smoothly, except certain kinds of bounces
> aren't being picked up by the bounce detector.
> 
> I've verified that the Bouncers/*.py process() functions are being
> called, and it picks up some bounces, but not others (one is from
> MSEXCH)
> 
> I can collect a bunch of un-detected bounces and send them to someone
> for analysis.  Please help, my list owners are being flooded with mail.
> It's like being a majordomo owner all over again :)
> 
> Thanks,
> Skye
> 

------------------------------------------------------
Mailman-Users mailing list
Mailman-Users@python.org
http://mail.python.org/mailman/listinfo/mailman-users
Mailman FAQ: http://www.python.org/cgi-bin/faqw-mm.py
Searchable Archives: http://www.mail-archive.com/mailman-users%40python.org/

----- End forwarded message -----

From skye@ffwd.cx  Thu Oct 17 23:29:16 2002
From: skye@ffwd.cx (Skye Poier)
Date: Thu, 17 Oct 2002 15:29:16 -0700
Subject: [Mailman-Developers] (forw) Re: [Mailman-Users] bounce detection
	broken?
Message-ID: <20021017222916.GH54297@ffwd.cx>

Here's another -users post that may be interesting.
I'm not sure if MSEXCH bounces are supposed to be picked up -
if not, feel free to integrate my patch.

Skye

----- Forwarded message from Skye Poier <skye@ffwd.cx> -----

From: Skye Poier <skye@ffwd.cx>
To: mailman-users@python.org
Subject: Re: [Mailman-Users] bounce detection broken?
Date: Wed, 16 Oct 2002 17:00:47 -0700

Another followup.
I added the following to SimpleMatch.py

    # MSEXCH
    (_c('did not reach the following recipient'),
     _c('MSEXCH:'),
     _c('(?P<addr>[^ ]*) on ')),

to catch bounces like this:

Your message

  To:      test@ffwd.cx
  Subject: [Test] test8
  Sent:    Tue, 15 Oct 2002 19:51:22 -0500

did not reach the following recipient(s):

martind@icpas.org on Tue, 15 Oct 2002 19:57:11 -0500
    The recipient name is not recognized
        The MTS-ID of the original message is: c=us;a= ;p=illinois cpa
soc;l=ICPASNT102101600574W9M1ATW
    MSEXCH:IMS:Illinois CPA Society:ICPAS:ICPASNT1 0 (000C05A6) Unknown
Recipient

However the DSN match problem is unexplained, and I'm not sure why my
OLD version of mailman 2.0beta4 did NOT spam list-admin with these.

Should I be directing this to mailman-developers?  I feel like I'm
talking to myself...

Skye


------------------------------------------------------
Mailman-Users mailing list
Mailman-Users@python.org
http://mail.python.org/mailman/listinfo/mailman-users
Mailman FAQ: http://www.python.org/cgi-bin/faqw-mm.py
Searchable Archives: http://www.mail-archive.com/mailman-users%40python.org/

----- End forwarded message -----

From barry@python.org  Tue Oct 22 14:51:02 2002
From: barry@python.org (Barry A. Warsaw)
Date: Tue, 22 Oct 2002 09:51:02 -0400
Subject: [Mailman-Developers] (forw) Re: [Mailman-Users] bounce detection
	broken?
References: <20021017222916.GH54297@ffwd.cx>
Message-ID: <15797.22470.339245.86733@gargle.gargle.HOWL>


>>>>> "SP" == Skye Poier <skye@ffwd.cx> writes:

    SP> Here's another -users post that may be interesting.  I'm not
    SP> sure if MSEXCH bounces are supposed to be picked up - if not,
    SP> feel free to integrate my patch.

BTW, I integrated the MSEXCH patches into cvs.

Thanks,
-Barry

From barry@python.org  Tue Oct 22 14:52:20 2002
From: barry@python.org (Barry A. Warsaw)
Date: Tue, 22 Oct 2002 09:52:20 -0400
Subject: [Mailman-Developers] (forw) Re: [Mailman-Users] bounce detection
	broken?
References: <20021017222722.GG54297@ffwd.cx>
Message-ID: <15797.22548.598922.296769@gargle.gargle.HOWL>


>>>>> "SP" == Skye Poier <skye@ffwd.cx> writes:

    SP> Well, since nobody helped, I went ahead and started debugging
    SP> the stuff in Bouncers/*.py

    SP> First I took a look at why Hotmail bounces weren't being
    SP> caught in DSN.py.  Here's an example with my debugging:

As is the case with all bounce formats, I need a preserved sample of
the bounce.  Please forward me all examples that are missed, but you
need to include all the headers too.

-Barry

From barry@python.org  Tue Oct 22 14:53:07 2002
From: barry@python.org (Barry A. Warsaw)
Date: Tue, 22 Oct 2002 09:53:07 -0400
Subject: [Mailman-Developers] 2.1b3 passing most(?) posts to the request
	address
References: <20021022112630.GV4725@ruhr-uni-bochum.de>
Message-ID: <15797.22595.196551.540195@gargle.gargle.HOWL>


>>>>> "JK" == Jost Krieger <Jost.Krieger+mailman@rub.de> writes:

    JK> I haven't found this in the archives, but that may be my
    JK> fault.

    JK> Since I upgraded to 2.1b3, my "vette" logfile ist filled with
    JK> the following type of messages:

    JK> These coincide with messages to this "security" list, and it
    JK> has *not* activated the administrivia setting.  (It is just
    JK> redistribution of security notices.) The messages to the list
    JK> *are* "Precedence: bulk".

    JK> What may be going on?

Chuq helped me track down a bug related to this, which is now fixed in
cvs.

-Barry

From chk@pobox.com  Tue Oct 22 14:59:39 2002
From: chk@pobox.com (Harald Koch)
Date: Tue, 22 Oct 2002 09:59:39 -0400
Subject: [Mailman-Developers] 
 Re: 2.1b3 passing most(?) posts to the request address 
In-Reply-To: Your message of "Tue, 22 Oct 2002 13:26:30 +0200".
	 <20021022112630.GV4725@ruhr-uni-bochum.de> 
References: <20021022112630.GV4725@ruhr-uni-bochum.de>
Message-ID: <23442.1035295179@elisabeth.cfrq.net>

Of all the gin joints in all the towns in all the world, Jost Krieger
had to walk into mine and say:
> 
> Since I upgraded to 2.1b3, my "vette" logfile ist filled with the following
> type of messages:
> 
> Oct 22 11:20:31 2002 (2195) Precedence: bulk message ignored by: security-request@lists.ruhr-uni-bochum.de
> Oct 22 11:32:30 2002 (2195) Precedence: bulk message ignored by: security-request@lists.ruhr-uni-bochum.de
> Oct 22 12:29:52 2002 (2195) Precedence: bulk message ignored by: security-request@lists.ruhr-uni-bochum.de

I've been seeing this too; they always appear 30 seconds after other
messages in the logs, and do *not* correspond to anything in my sendmail
logs. Whatever emails are triggering this are internal to mailman...

<http://mail.python.org/pipermail-21/mailman-developers/2002-September/013292.html>
was my post on the subject :-)

-- 
Harald Koch     <chk@pobox.com>

"It takes a child to raze a village."
		-Michael T. Fry

From noreply@sourceforge.net  Tue Oct 22 15:03:43 2002
From: noreply@sourceforge.net (noreply@sourceforge.net)
Date: Tue, 22 Oct 2002 07:03:43 -0700
Subject: [Mailman-Developers] 
 [ mailman-Bugs-214156 ] MIME quoted printable subject lines (PR#190)
Message-ID: <E183zdH-0000J7-00@usw-sf-web5.sourceforge.net>

Bugs item #214156, was opened at 2000-09-11 16:22
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=100103&aid=214156&group_id=103

Category: Pipermail
Group: None
Status: Closed
Resolution: Fixed
Priority: 6
Submitted By: Nobody/Anonymous (nobody)
Assigned to: Jeremy Hylton (jhylton)
Summary: MIME quoted printable subject lines (PR#190)

Initial Comment:
Jitterbug-Id: 190
Submitted-By: ziegler@ttt-atm.ttt.bme.hu
Date: Thu, 27 Jan 2000 10:11:10 -0500 (EST)
Version: v.1.0rc2
OS: Debian 2.1


The MIME quoted-printable decoding does not work for mail header lines!
It should be!

If I use accented national characters in mails (i.e., MIME quoted printable),
e.g.
in the subject line, or in the full name of sender/addressee, 
then the archive index is almost unreadable, since I see only
garbage instead of meaningful subject and sender lines, like this:

[Hsn-empl] =?iso-8859-1?Q?=DCdv?= =?iso-8859-1?Q?G=E1bor?= Ziegler 

It is also true in the plain text downloadable version, see the archive text
version
of the message above:

>From ziegler@ttt-atm.ttt.bme.hu Thu, 27 Jan 2000 14:36:19 +0100
Date: Thu, 27 Jan 2000 14:36:19 +0100
From: =?iso-8859-1?Q?G=E1bor?= Ziegler ziegler@ttt-atm.ttt.bme.hu
Subject: [Hsn-empl] =?iso-8859-1?Q?=DCdv?=

Üdvözlet mindenkinek!

Ez a lista kiserleti jelleggel jött létre, a HSNLabos oktatok
tajékoztatására.
A levelforgalom kicsi lesz, csak is a labor mukodesevel kapcsolatos
infokat tervezem ide.

Ez a lista zártkörû, tehát:
* a lista tagjainak listajat csak a tagok lathatjak
* a feliratkozasokat a adminisztrator kell jovahagyja.
* listara levelet kuldeni csak tagok, valamint masok (egyelore) a 
*.ttt.bme.hu,
  *ericsson.se domainekbol tudnak.

Lehetoseg van:
* a levelek atmeneti letiltasara, 
* napi egyszeri osszefoglalot kerni csak (digest), ami persze csak akkor
jon, ha volt forgalom,
* az archivumban bongeszni
* egyedi beállitasokat megvaltoztatni
* leiratkozni, feliratkozni, 
s mindezt egyszerüen a Weben keresztül! 

Remelem elnyeri tetszeseteket!

Udv:
                Gabor





====================================================================
Audit trail:
None

----------------------------------------------------------------------

>Comment By: Barry A. Warsaw (bwarsaw)
Date: 2002-10-22 10:03

Message:
Logged In: YES 
user_id=12800

That's what you'd have to do.  I don't have time to work on
this myself, but I would accept some clean patches against
MM2.1 to fix this.

----------------------------------------------------------------------

Comment By: Mike Brown (mike_j_brown)
Date: 2002-10-21 20:35

Message:
Logged In: YES 
user_id=371366

Is there any chance this can be made to work in the indices 
as well, so that authors and subjects there are decoded in 
the same way?

Another suggestion -- If you can get the Unicode code point 
of the non-ASCII characters (difficult in Python 1.5.2, not too 
hard in 1.6, very easy in 2.0+), emit them in HTML as 
numeric character references so you don't have to worry 
about conflicts with the rest of the message.


----------------------------------------------------------------------

Comment By: Jeremy Hylton (jhylton)
Date: 2000-09-22 00:12

Message:
Fixed in rev. 1.30 of Mailman/Archiver/HyperArch.py.  If the charset of the header matches the charset of the message body, it will be decoded.


----------------------------------------------------------------------

Comment By: Barry A. Warsaw (bwarsaw)
Date: 2000-09-14 15:37

Message:
Unpostponing, reopening, and assigning to Jeremy

----------------------------------------------------------------------

Comment By: Barry A. Warsaw (bwarsaw)
Date: 2000-09-13 08:49

Message:
Yes, Pipermail should be improved in this regard.  Current recommendation is to use an external archiver.

----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=100103&aid=214156&group_id=103

From chk@pobox.com  Tue Oct 22 15:07:07 2002
From: chk@pobox.com (Harald Koch)
Date: Tue, 22 Oct 2002 10:07:07 -0400
Subject: [Mailman-Developers] 
 Re: 2.1b3 passing most(?) posts to the request address 
In-Reply-To: barry's message of "Tue, 22 Oct 2002 09:53:07 -0400".
	 <15797.22595.196551.540195@gargle.gargle.HOWL> 
References: <20021022112630.GV4725@ruhr-uni-bochum.de>
            <15797.22595.196551.540195@gargle.gargle.HOWL>
Message-ID: <23555.1035295627@elisabeth.cfrq.net>

Of all the gin joints in all the towns in all the world, Barry A. Warsaw
had to walk into mine and say:
> 
> Chuq helped me track down a bug related to this, which is now fixed in
> cvs.

Excellent!

(Two messages that pass in the night... :-)

-- 
Harald Koch     <chk@pobox.com>

From noreply@sourceforge.net  Tue Oct 22 15:11:14 2002
From: noreply@sourceforge.net (noreply@sourceforge.net)
Date: Tue, 22 Oct 2002 07:11:14 -0700
Subject: [Mailman-Developers] 
 [ mailman-Bugs-621689 ] Subscribe requests dropped from admin db
Message-ID: <E183zkY-0000Z1-00@usw-sf-web5.sourceforge.net>

Bugs item #621689, was opened at 2002-10-10 23:10
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=100103&aid=621689&group_id=103

Category: (un)subscribing
Group: 2.1 beta
Status: Open
Resolution: None
Priority: 5
Submitted By: Mark Dadgar (mdadgar)
Assigned to: Nobody/Anonymous (nobody)
Summary: Subscribe requests dropped from admin db

Initial Comment:
Occasionally, subscribe requests (via the web page) to a list that 
requires moderator approval for subscriptions never make it to the 
admin db for that list.  They are listed in the subscribe log file, but 
never show up on the admin db page.

Log file entries look like this:

Oct 10 18:43:43 2002 (31168) eventmasters: pending   
XXXXXX.swfla.rr.com

But the admin db page for the list shows no outstanding issues.

I have verified that there was an actual subscription request 
behind this log entry.

----------------------------------------------------------------------

>Comment By: Barry A. Warsaw (bwarsaw)
Date: 2002-10-22 10:11

Message:
Logged In: YES 
user_id=12800

I'm going to defer this one until 2.1b4 goes out.  I've
fixed some related bugs in cvs, so hopefully this will be
fixed in the new version.

----------------------------------------------------------------------

Comment By: Mark Dadgar (mdadgar)
Date: 2002-10-20 13:14

Message:
Logged In: YES 
user_id=598228

Sorry - I'm running 2.1b3.

----------------------------------------------------------------------

Comment By: Barry A. Warsaw (bwarsaw)
Date: 2002-10-20 11:48

Message:
Logged In: YES 
user_id=12800

What version of Mailman are you using.  It's import to know
the exact version, i.e. 2.1beta3, or if you're running from
cvs.  A lot of these problems have been fixed in cvs, which
will soon be released as 2.1 beta 4.

----------------------------------------------------------------------

Comment By: Mark Dadgar (mdadgar)
Date: 2002-10-11 00:29

Message:
Logged In: YES 
user_id=598228

I just verified, by cat'ing pending.pck, that the appropriate changes 
seem to be in there.  But the web interface never picks them up.

----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=100103&aid=621689&group_id=103

From noreply@sourceforge.net  Tue Oct 22 15:31:31 2002
From: noreply@sourceforge.net (noreply@sourceforge.net)
Date: Tue, 22 Oct 2002 07:31:31 -0700
Subject: [Mailman-Developers] 
 [ mailman-Bugs-609692 ] Non-member hold reason is wrong
Message-ID: <E18404B-0001Fr-00@usw-sf-web5.sourceforge.net>

Bugs item #609692, was opened at 2002-09-15 18:07
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=100103&aid=609692&group_id=103

Category: mail delivery
Group: 2.1 beta
>Status: Closed
>Resolution: Fixed
Priority: 5
Submitted By: Les Niles (lniles)
Assigned to: Nobody/Anonymous (nobody)
Summary: Non-member hold reason is wrong

Initial Comment:
Maybe I'm missing something, but it would seem 
that Handlers.Moderate.process() should call 
hold.hold_for_approval() with a fourth arg of 
Hold.NonMemberPost instead of Hold.ModeratedPost 
at line 85.  The latter causes confusion for the typical 
case of a member (person) posting from other than 
her subscribed email address, since she gets back a 
message saying that the post was to a moderated 
list.  And the default reject message for the 
moderator, something about the posting being 
deemed inappropriate, is not appropriate. :)


----------------------------------------------------------------------

>Comment By: Barry A. Warsaw (bwarsaw)
Date: 2002-10-22 10:31

Message:
Logged In: YES 
user_id=12800

Actually, I think tkikuchi's patch is correct and both
places should be changed.  BTW, it's always better to upload
patches instead of paste them into the comment area! 

Thanks.

----------------------------------------------------------------------

Comment By: Tokio Kikuchi (tkikuchi)
Date: 2002-10-16 02:30

Message:
Logged In: YES 
user_id=67709

I am sending a patch to raise attention: sorry for the
inconvenience with the folded lines but only two lines. (TK)

--- /home/mailman/src/mailman/Mailman/Handlers/Moderate.py 
    Wed Jul 10 04:29
:53 2002
+++ Moderate.py Wed Oct 16 15:25:26 2002
@@ -82,7 +82,7 @@
     if matches_p(sender, mlist.accept_these_nonmembers):
         return
     if matches_p(sender, mlist.hold_these_nonmembers):
-        Hold.hold_for_approval(mlist, msg, msgdata,
Hold.ModeratedPost)
+        Hold.hold_for_approval(mlist, msg, msgdata,
Hold.NonMemberPost)
         # No return
     if matches_p(sender, mlist.reject_these_nonmembers):
         do_reject(mlist)
@@ -98,7 +98,7 @@
         # Accept
         return
     elif mlist.generic_nonmember_action == 1:
-        Hold.hold_for_approval(mlist, msg, msgdata,
Hold.ModeratedPost)
+        Hold.hold_for_approval(mlist, msg, msgdata,
Hold.NonMemberPost)
     elif mlist.generic_nonmember_action == 2:
         do_reject(mlist)
     elif mlist.generic_nonmember_action == 3:


----------------------------------------------------------------------

Comment By: Les Niles (lniles)
Date: 2002-09-18 12:56

Message:
Logged In: YES 
user_id=92931

Oops -- Hold.NonMemberPost should replace 
Hold.ModeratedPost at line 101, not line 85.  (This is 
in Sept 3 2002 CVS.)


----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=100103&aid=609692&group_id=103

From noreply@sourceforge.net  Tue Oct 22 16:18:25 2002
From: noreply@sourceforge.net (noreply@sourceforge.net)
Date: Tue, 22 Oct 2002 08:18:25 -0700
Subject: [Mailman-Developers] 
 [ mailman-Bugs-620033 ] gate_news should catch lost connection
Message-ID: <E1840nZ-0001jU-00@usw-sf-web1.sourceforge.net>

Bugs item #620033, was opened at 2002-10-08 02:07
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=100103&aid=620033&group_id=103

Category: nntp/news
Group: 2.1 beta
Status: Closed
Resolution: Accepted
Priority: 5
Submitted By: Ron Jarrell (jarrell)
Assigned to: Nobody/Anonymous (nobody)
Summary: gate_news should catch lost connection

Initial Comment:
Sep 20 10:51:35 2002 gate_news(9301): Traceback 
(most recent call last):
Sep 20 10:51:35 2002 gate_news(9301):   
File "/home/mailman/cron/gate_news", line 270, 
in ?
Sep 20 10:51:35 2002 gate_news(9301):      main()
Sep 20 10:51:35 2002 gate_news(9301):   
File "/home/mailman/cron/gate_news", line 250, in 
main
Sep 20 10:51:35 2002 gate_news(9301):      
process_lists(lock)
Sep 20 10:51:35 2002 gate_news(9301):   
File "/home/mailman/cron/gate_news", line 194, in 
process_lists
Sep 20 10:51:35 2002 gate_news(9301):      conn, 
first, last = open_newsgroup(mlist)
Sep 20 10:51:35 2002 gate_news(9301):   
File "/home/mailman/cron/gate_news", line 89, in 
open_newsgroup
Sep 20 10:51:35 2002 gate_news(9301):      
password=mm_cfg.NNTP_PASSWORD)
Sep 20 10:51:35 2002 gate_news(9301):   
File "/usr/local/lib/python2.2/nntplib.py", line 
117, in __init__
Sep 20 10:51:35 2002 gate_news(9301):      
self.welcome = self.getresp()
Sep 20 10:51:35 2002 gate_news(9301):   
File "/usr/local/lib/python2.2/nntplib.py", line 
203, in getresp
Sep 20 10:51:35 2002 gate_news(9301):      resp = 
self.getline()
Sep 20 10:51:35 2002 gate_news(9301):   
File "/usr/local/lib/python2.2/nntplib.py", line 
192, in getline
Sep 20 10:51:35 2002 gate_news(9301):      line = 
self.file.readline()
Sep 20 10:51:35 2002 gate_news(9301): IOError :  
[Errno 131] Connection reset by peer 


----------------------------------------------------------------------

Comment By: Barry A. Warsaw (bwarsaw)
Date: 2002-10-15 19:00

Message:
Logged In: YES 
user_id=12800

I'm applying this patch

----------------------------------------------------------------------

Comment By: Barry A. Warsaw (bwarsaw)
Date: 2002-10-15 18:19

Message:
Logged In: YES 
user_id=12800

I didn't realize that nntplib.NNTP() can raise an IOError --
I thought it would have been a socket.error.  Oh well,
here's an untested patch.  I think I'll check this in just
because I don't see how it could break anything.

----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=100103&aid=620033&group_id=103

From noreply@sourceforge.net  Tue Oct 22 16:19:29 2002
From: noreply@sourceforge.net (noreply@sourceforge.net)
Date: Tue, 22 Oct 2002 08:19:29 -0700
Subject: [Mailman-Developers] 
 [ mailman-Bugs-620033 ] gate_news should catch lost connection
Message-ID: <E1840ob-0001kz-00@usw-sf-web1.sourceforge.net>

Bugs item #620033, was opened at 2002-10-08 02:07
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=100103&aid=620033&group_id=103

Category: nntp/news
Group: 2.1 beta
Status: Closed
Resolution: Accepted
Priority: 5
Submitted By: Ron Jarrell (jarrell)
Assigned to: Nobody/Anonymous (nobody)
Summary: gate_news should catch lost connection

Initial Comment:
Sep 20 10:51:35 2002 gate_news(9301): Traceback 
(most recent call last):
Sep 20 10:51:35 2002 gate_news(9301):   
File "/home/mailman/cron/gate_news", line 270, 
in ?
Sep 20 10:51:35 2002 gate_news(9301):      main()
Sep 20 10:51:35 2002 gate_news(9301):   
File "/home/mailman/cron/gate_news", line 250, in 
main
Sep 20 10:51:35 2002 gate_news(9301):      
process_lists(lock)
Sep 20 10:51:35 2002 gate_news(9301):   
File "/home/mailman/cron/gate_news", line 194, in 
process_lists
Sep 20 10:51:35 2002 gate_news(9301):      conn, 
first, last = open_newsgroup(mlist)
Sep 20 10:51:35 2002 gate_news(9301):   
File "/home/mailman/cron/gate_news", line 89, in 
open_newsgroup
Sep 20 10:51:35 2002 gate_news(9301):      
password=mm_cfg.NNTP_PASSWORD)
Sep 20 10:51:35 2002 gate_news(9301):   
File "/usr/local/lib/python2.2/nntplib.py", line 
117, in __init__
Sep 20 10:51:35 2002 gate_news(9301):      
self.welcome = self.getresp()
Sep 20 10:51:35 2002 gate_news(9301):   
File "/usr/local/lib/python2.2/nntplib.py", line 
203, in getresp
Sep 20 10:51:35 2002 gate_news(9301):      resp = 
self.getline()
Sep 20 10:51:35 2002 gate_news(9301):   
File "/usr/local/lib/python2.2/nntplib.py", line 
192, in getline
Sep 20 10:51:35 2002 gate_news(9301):      line = 
self.file.readline()
Sep 20 10:51:35 2002 gate_news(9301): IOError :  
[Errno 131] Connection reset by peer 


----------------------------------------------------------------------

>Comment By: Ron Jarrell (jarrell)
Date: 2002-10-22 15:19

Message:
Logged In: YES 
user_id=34619

Didn't work...  This comes up, btw, not during initial 
connection, but if the connection goes away during a 
gate_news run, like the daemon on the other end dies, 
or the news server reboots, etc.  You get an IOerror on 
the mailman side.

----------------------------------------------------------------------

Comment By: Barry A. Warsaw (bwarsaw)
Date: 2002-10-15 19:00

Message:
Logged In: YES 
user_id=12800

I'm applying this patch

----------------------------------------------------------------------

Comment By: Barry A. Warsaw (bwarsaw)
Date: 2002-10-15 18:19

Message:
Logged In: YES 
user_id=12800

I didn't realize that nntplib.NNTP() can raise an IOError --
I thought it would have been a socket.error.  Oh well,
here's an untested patch.  I think I'll check this in just
because I don't see how it could break anything.

----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=100103&aid=620033&group_id=103

From noreply@sourceforge.net  Tue Oct 22 17:03:10 2002
From: noreply@sourceforge.net (noreply@sourceforge.net)
Date: Tue, 22 Oct 2002 09:03:10 -0700
Subject: [Mailman-Developers] 
 [ mailman-Bugs-620033 ] gate_news should catch lost connection
Message-ID: <E1841Us-0003CR-00@usw-sf-web1.sourceforge.net>

Bugs item #620033, was opened at 2002-10-07 22:07
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=100103&aid=620033&group_id=103

Category: nntp/news
Group: 2.1 beta
>Status: Open
Resolution: Accepted
Priority: 5
Submitted By: Ron Jarrell (jarrell)
Assigned to: Nobody/Anonymous (nobody)
Summary: gate_news should catch lost connection

Initial Comment:
Sep 20 10:51:35 2002 gate_news(9301): Traceback 
(most recent call last):
Sep 20 10:51:35 2002 gate_news(9301):   
File "/home/mailman/cron/gate_news", line 270, 
in ?
Sep 20 10:51:35 2002 gate_news(9301):      main()
Sep 20 10:51:35 2002 gate_news(9301):   
File "/home/mailman/cron/gate_news", line 250, in 
main
Sep 20 10:51:35 2002 gate_news(9301):      
process_lists(lock)
Sep 20 10:51:35 2002 gate_news(9301):   
File "/home/mailman/cron/gate_news", line 194, in 
process_lists
Sep 20 10:51:35 2002 gate_news(9301):      conn, 
first, last = open_newsgroup(mlist)
Sep 20 10:51:35 2002 gate_news(9301):   
File "/home/mailman/cron/gate_news", line 89, in 
open_newsgroup
Sep 20 10:51:35 2002 gate_news(9301):      
password=mm_cfg.NNTP_PASSWORD)
Sep 20 10:51:35 2002 gate_news(9301):   
File "/usr/local/lib/python2.2/nntplib.py", line 
117, in __init__
Sep 20 10:51:35 2002 gate_news(9301):      
self.welcome = self.getresp()
Sep 20 10:51:35 2002 gate_news(9301):   
File "/usr/local/lib/python2.2/nntplib.py", line 
203, in getresp
Sep 20 10:51:35 2002 gate_news(9301):      resp = 
self.getline()
Sep 20 10:51:35 2002 gate_news(9301):   
File "/usr/local/lib/python2.2/nntplib.py", line 
192, in getline
Sep 20 10:51:35 2002 gate_news(9301):      line = 
self.file.readline()
Sep 20 10:51:35 2002 gate_news(9301): IOError :  
[Errno 131] Connection reset by peer 


----------------------------------------------------------------------

Comment By: Ron Jarrell (jarrell)
Date: 2002-10-22 11:19

Message:
Logged In: YES 
user_id=34619

Didn't work...  This comes up, btw, not during initial 
connection, but if the connection goes away during a 
gate_news run, like the daemon on the other end dies, 
or the news server reboots, etc.  You get an IOerror on 
the mailman side.

----------------------------------------------------------------------

Comment By: Barry A. Warsaw (bwarsaw)
Date: 2002-10-15 15:00

Message:
Logged In: YES 
user_id=12800

I'm applying this patch

----------------------------------------------------------------------

Comment By: Barry A. Warsaw (bwarsaw)
Date: 2002-10-15 14:19

Message:
Logged In: YES 
user_id=12800

I didn't realize that nntplib.NNTP() can raise an IOError --
I thought it would have been a socket.error.  Oh well,
here's an untested patch.  I think I'll check this in just
because I don't see how it could break anything.

----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=100103&aid=620033&group_id=103

From claw@kanga.nu  Tue Oct 22 18:35:28 2002
From: claw@kanga.nu (J C Lawrence)
Date: Tue, 22 Oct 2002 10:35:28 -0700
Subject: [Mailman-Developers] usage() going to stderr 
In-Reply-To: Message from "Bob Puff@NLE" <bob@nleaudio.com> 
   of "Mon, 21 Oct 2002 17:45:20 EDT." <3DB47570.23595C7D@nleaudio.com> 
References: <20021021185625.GB22948@csh.rit.edu>
	<2732E68C-E530-11D6-87E7-0003934516A8@plaidworks.com>
	<20021021205904.GA25431@dust.uchicago.edu>  <3DB47570.23595C7D@nleaudio.com> 
Message-ID: <32113.1035308128@kanga.nu>

On Mon, 21 Oct 2002 17:45:20 -0400 
bob  <Puff@NLE" <bob@nleaudio.com>> wrote:

> Yeah, it's a real pain when you do 'program -?', and the output
> scrolls way more than one screen's worth, and you can't more it.  

man screen

-- 
J C Lawrence                
---------(*)                Satan, oscillate my metallic sonatas. 
claw@kanga.nu               He lived as a devil, eh?		  
http://www.kanga.nu/~claw/  Evil is a name of a foeman, as I live.


From Dirk.DeConinck@inno.com  Wed Oct 23 09:50:46 2002
From: Dirk.DeConinck@inno.com (Dirk De Coninck)
Date: Wed, 23 Oct 2002 10:50:46 +0200
Subject: [Mailman-Developers] Password security vulnerability
Message-ID: <000301c27a71$473e3ef0$040ba8c0@hq.inno.com>

Dear developers,

It seems there is a bug in the listadmin password checking. I have
version 2.0.11 running at my site and to my experience mailman only
looks to the first 8 characters of a password. This was a real problem
for me as all the listadmin passwords began with the name of the list...
As a result one could login be typing the name of the list (for those
lists with a name longer then 8 characters). I have searched the
archives if this is a known problem or if there is a fix available but
without success. I would appreciate any feedback regarding this issue.

Kind regards,
Dirk.

P.S. Thank you all for your contributions and the development of
mailman. Keep up the good work !!!

________________________________________________________

Dirk De Coninck                                 Inno.com
                                       Heiststeenweg 131
                                  2580 Beerzel - Belgium
Mobile +32 486 50 12 85              Tel +32 15 25 82 00
                                     Fax +32 15 25 82 01
mailto:Dirk.DeConinck@inno.com       http://www.inno.com
________________________________________________________
 


From barry@python.org  Wed Oct 23 13:18:46 2002
From: barry@python.org (Barry A. Warsaw)
Date: Wed, 23 Oct 2002 08:18:46 -0400
Subject: [Mailman-Developers] Password security vulnerability
References: <000301c27a71$473e3ef0$040ba8c0@hq.inno.com>
Message-ID: <15798.37798.811650.180428@gargle.gargle.HOWL>


>>>>> "DDC" == Dirk De Coninck <Dirk.DeConinck@inno.com> writes:

    DDC> It seems there is a bug in the listadmin password checking. I
    DDC> have version 2.0.11 running at my site and to my experience
    DDC> mailman only looks to the first 8 characters of a
    DDC> password. This was a real problem for me as all the listadmin
    DDC> passwords began with the name of the list...  As a result one
    DDC> could login be typing the name of the list (for those lists
    DDC> with a name longer then 8 characters). I have searched the
    DDC> archives if this is a known problem or if there is a fix
    DDC> available but without success. I would appreciate any
    DDC> feedback regarding this issue.

Set USE_CRYPT=0 to use md5 hashed passwords over crypt()'d passwords.

Note that MM2.1 uses sha hashes for all passwords.

    DDC> P.S. Thank you all for your contributions and the development
    DDC> of mailman. Keep up the good work !!!

You're welcome!
-Barry

From fil@rezo.net  Wed Oct 23 19:06:06 2002
From: fil@rezo.net (Fil)
Date: Wed, 23 Oct 2002 20:06:06 +0200
Subject: [Mailman-Developers] email2.4.3
In-Reply-To: <20021022074215.GD24706@rezo.net>
References: <20021020190510.GD23093@rezo.net>
	<20021020190718.GE23093@rezo.net> <20021020194027.GA7174@rezo.net>
	<15796.22072.609641.308438@gargle.gargle.HOWL>
	<20021021194440.GC2065@rezo.net> <20021022074215.GD24706@rezo.net>
Message-ID: <20021023180606.GB4035@rezo.net>

> UnicodeError: ASCII decoding error: ordinal not in range(128)

In fact this bug prevents email confirmation of
subscriptions/unsubscriptions, so I'm getting quite a mess here, with people
thinking they are subscribed - or unsubscribed, and not understanding why
they don't get/still get emails.

-- Fil


From mss@mawhrin.net  Wed Oct 23 20:29:04 2002
From: mss@mawhrin.net (Mikhail Sobolev)
Date: Wed, 23 Oct 2002 20:29:04 +0100
Subject: [Mailman-Developers] Re: [Mailman-checkins] mailman TODO,2.11,2.12
In-Reply-To: <E184LkG-0006a6-00@usw-pr-cvs1.sourceforge.net>
References: <E184LkG-0006a6-00@usw-pr-cvs1.sourceforge.net>
Message-ID: <20021023192904.GA28789@mawhrin.net>


---------------------- multipart/signed attachment
On Wed, Oct 23, 2002 at 06:40:24AM -0700, bwarsaw@users.sourceforge.net wrote:
> +    - Have an option to sort the list of members by real name or email
> +      address, /if/ you could figure out how to sort a list with real
> +      names in multiple languages.
Can't just the unicode representations be compared?  I am pretty sure
that'd be enough...

--
Misha

---------------------- multipart/signed attachment
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: not available
Url : http://mail.python.org/pipermail-21/mailman-developers/attachments/20021023/f9464e06/attachment.bin

---------------------- multipart/signed attachment--

From barry@python.org  Wed Oct 23 20:32:12 2002
From: barry@python.org (Barry A. Warsaw)
Date: 23 Oct 2002 15:32:12 -0400
Subject: [Mailman-Developers] Re: email2.4.3
References: <20021020190510.GD23093@rezo.net>
	<20021020190718.GE23093@rezo.net> <20021020194027.GA7174@rezo.net>
	<15796.22072.609641.308438@gargle.gargle.HOWL>
	<20021021194440.GC2065@rezo.net> <20021022074215.GD24706@rezo.net>
	<20021023180606.GB4035@rezo.net>
Message-ID: <m3bs5l2bpf.fsf@zope.com>


>>>>> "fil" ==   <fil@rezo.net> writes:

    >> UnicodeError: ASCII decoding error: ordinal not in range(128)

    fil> In fact this bug prevents email confirmation of
    fil> subscriptions/unsubscriptions, so I'm getting quite a mess
    fil> here, with people thinking they are subscribed - or
    fil> unsubscribed, and not understanding why they don't get/still
    fil> get emails.

Just to be sure, are you up to date w/ cvs and did you do a "configure
; make install ; mailmanctl restart"?

-Barry

    



From fil@rezo.net  Wed Oct 23 21:50:59 2002
From: fil@rezo.net (Fil)
Date: Wed, 23 Oct 2002 22:50:59 +0200
Subject: [Mailman-Developers] Re: email2.4.3
In-Reply-To: <m3bs5l2bpf.fsf@zope.com>
References: <20021020190510.GD23093@rezo.net>
	<20021020190718.GE23093@rezo.net> <20021020194027.GA7174@rezo.net>
	<15796.22072.609641.308438@gargle.gargle.HOWL>
	<20021021194440.GC2065@rezo.net> <20021022074215.GD24706@rezo.net>
	<20021023180606.GB4035@rezo.net> <m3bs5l2bpf.fsf@zope.com>
Message-ID: <20021023205059.GB3931@rezo.net>

>     >> UnicodeError: ASCII decoding error: ordinal not in range(128)
> 
>     fil> In fact this bug prevents email confirmation of
>     fil> subscriptions/unsubscriptions, so I'm getting quite a mess
>     fil> here, with people thinking they are subscribed - or
>     fil> unsubscribed, and not understanding why they don't get/still
>     fil> get emails.
> 
> Just to be sure, are you up to date w/ cvs and did you do a "configure
> ; make install ; mailmanctl restart"?

Oh no! Not again! In order to be sure, I just did a 'cvs update -A' and a
lot of diffs have arrived!! How comes I had sticky tags - I don't think I've
never done anything but cvs update (?) ; well, the good news is that the
bug's squashed.

-- Fil


From fil@rezo.net  Wed Oct 23 21:52:44 2002
From: fil@rezo.net (Fil)
Date: Wed, 23 Oct 2002 22:52:44 +0200
Subject: [Mailman-Developers] wrong log
Message-ID: <20021023205244.GH9387@rezo.net>

the log reads >>>>
Oct 23 22:49:40 2002 (9350) spip: deleted fil+09@rezo.net; web confirmation

But I confirmed by email ;)

-- Fil


From fil@rezo.net  Wed Oct 23 21:57:52 2002
From: fil@rezo.net (Fil)
Date: Wed, 23 Oct 2002 22:57:52 +0200
Subject: [Mailman-Developers] listname not defined
Message-ID: <20021023205752.GA12292@rezo.net>


I can't believe it's a Mailman bug: must be coming from my stupid
installation. However, just in case...

mailman@miel:~$ bin/newlist zoom-ecolo
Enter the email of the person running the list: n.herve@laposte.net
Initial zoom-ecolo password: xxxxxxx

To finish creating your mailing list, you must edit your /etc/aliases (or
equivalent) file by adding the following lines, and possibly running the
=0Eewaliases' program:

## %(listname)s mailing list
Traceback (most recent call last):
  File "bin/newlist", line 219, in ?
    main()
  File "bin/newlist", line 186, in main
    sys.modules[modname].create(mlist)
  File "/usr/local/mailman/Mailman/MTA/Manual.py", line 73, in create
    for k, v in makealiases(listname):
NameError: global name 'listname' is not defined

-- Fil


From fil@rezo.net  Wed Oct 23 23:01:53 2002
From: fil@rezo.net (Fil)
Date: Thu, 24 Oct 2002 00:01:53 +0200
Subject: [Mailman-Developers] wouldnitbenice to have URLs in admin mails?
Message-ID: <20021023220153.GA23069@rezo.net>

I suggest we add an URL just after these kind of sentences:

> The attached message was received as a bounce, but either the bounce
> format was not recognized, or no member addresses could be extracted
> from it.  This mailing list has been configured to send all
> unrecognized bounce messages to the list administrator(s).
+ http://HOSTNAME/mailman/admin/listname/...


-- Fil


From chuqui@plaidworks.com  Thu Oct 24 06:36:29 2002
From: chuqui@plaidworks.com (Chuq Von Rospach)
Date: Wed, 23 Oct 2002 22:36:29 -0700
Subject: [Mailman-Developers] The Life Cycle of Mailing Lists.
Message-ID: <8BD1A59E-E712-11D6-B94D-0003934516A8@plaidworks.com>

a friend passed this to me today. I thought some of you might find it 
interesting.....

<http://www.plaidworks.com/chuqui/blog/000053.html#000053>


-- 
Chuq Von Rospach, Architech
chuqui@plaidworks.com -- http://www.plaidworks.com/chuqui/blog/

Stress is when you wake up screaming and you realize you haven't fallen 
asleep yet.


From chuqui@plaidworks.com  Thu Oct 24 18:11:42 2002
From: chuqui@plaidworks.com (Chuq Von Rospach)
Date: Thu, 24 Oct 2002 10:11:42 -0700
Subject: [Mailman-Developers] some mime filter notes...
Message-ID: <AA963AA6-E773-11D6-B383-0003934516A8@plaidworks.com>


I know someone asked about this the other day and Barry pointed at me, 
so...

I just made a change to how I do my mime filtering in 2.1b3+++, and I 
thought I'd pass it along.

Since the text digests don't deal with html well right now, I've turned 
all that off, so I'm using the mime filtering to get to a usable 
text/plain part without destroying the message.

So on the filtering page of my lists, I'm set this way:

filter_content: yes.

filter_mime_types: empty

pass_mime_types:

multipart/alternative
multipart/mixed
text/plain

convert_html_to_plaintext: no

the change I made this morning was add the multipart/mixed to 
pass_mime_types. turns out there are a couple of e-mail clients that 
send out mail not as M/A, but as M/M, because they're attaching 
non-text "things" to a text message (in this case, it was a .vcf file). 
You need to allow in the multipart or the message is rejected, and then 
all of the non text/plain pieces get stripped out of it, leading to the 
text message (wrapped with various mime pieces).

This will probably be my default for 2.1 on my lists unless I run into 
a new glitch. If I do, I'll pass it along, but I think the missing 
M/mixed was the thing causing the last of my "messages are 
disappearing" issues.


-- 
Chuq Von Rospach, Architech
chuqui@plaidworks.com -- http://www.plaidworks.com/chuqui/blog/

IMHO: Jargon. Acronym for In My Humble Opinion. Used to flag as an 
opinion
something that is clearly from context an opinion to everyone except the
mentally dense. Opinions flagged by IMHO are actually rarely humble. 
IMHO.
     (source: third unabridged dictionary of chuqui-isms).


From fil@rezo.net  Fri Oct 25 09:29:16 2002
From: fil@rezo.net (Fil)
Date: Fri, 25 Oct 2002 10:29:16 +0200
Subject: [Mailman-Developers] checkdbs
Message-ID: <20021025082916.GB32050@rezo.net>


CVS from two days ago... maybe I'm still bugged down by my bad practice of
cvs, but?  ('French' is my server's default language)

----- Forwarded message from Cron Daemon <root@miel.brainstorm.fr> -----

Date: Fri, 25 Oct 2002 08:00:05 +0200 (CEST)
From: root@miel.brainstorm.fr (Cron Daemon)
To: mailman@miel.brainstorm.fr
Subject: Cron <mailman@miel> /usr/local/bin/python -S /usr/local/mailman/cron/checkdbs

Traceback (most recent call last):
  File "/usr/local/mailman/cron/checkdbs", line 116, in ?
    main()
  File "/usr/local/mailman/cron/checkdbs", line 75, in main
    text += '\n' + pending_requests(mlist)
UnicodeError: ASCII decoding error: ordinal not in range(128)

----- End forwarded message -----

-- Fil


From barry@python.org  Fri Oct 25 20:04:20 2002
From: barry@python.org (Barry A. Warsaw)
Date: Fri, 25 Oct 2002 15:04:20 -0400
Subject: [Mailman-Developers] checkdbs
References: <20021025082916.GB32050@rezo.net>
Message-ID: <15801.38324.701830.815870@gargle.gargle.HOWL>


>>>>> "fil" ==   <fil@rezo.net> writes:

    fil> CVS from two days ago... maybe I'm still bugged down by my
    fil> bad practice of cvs, but?  ('French' is my server's default
    fil> language)

Ok, I'm about to check in a fix for this.  I think there may be a
better (more general) patch than what I'm going to commit, but I'm
afraid of repercussions, so I'm going with a local-to-checkdbs patch.

The problem happens when you have encoded names in the From field.
These get converted to Unicode in the requests.pck file, which would
return a Unicode from pending_requests().

But if maketext returns a byte string with non-ASCII characters in
them, you get this exception when trying to concatenate the two
strings together.  Once I fixed that, I noticed that if you then
change the list's language back to English, you'll get a different
exception because the notification's payload text contains Unicode
characters that don't match the list's charset.

So I'm changing checkdbs so that we make sure the payload is always a
byte string in charset of the list's language (with bad characters
replaced with '?').

The better fix might be for maketext() to always return a Unicode and
for us to use Unicode everywhere, but that's a Big Change.  Let's see
if we can worm around this for now.

-Barry

From barry@python.org  Fri Oct 25 21:07:54 2002
From: barry@python.org (Barry A. Warsaw)
Date: Fri, 25 Oct 2002 16:07:54 -0400
Subject: [Mailman-Developers] some mime filter notes...
References: <AA963AA6-E773-11D6-B383-0003934516A8@plaidworks.com>
Message-ID: <15801.42138.344811.957862@gargle.gargle.HOWL>


>>>>> "CVR" == Chuq Von Rospach <chuqui@plaidworks.com> writes:

    CVR> So on the filtering page of my lists, I'm set this way:

    CVR> filter_content: yes.

    CVR> filter_mime_types: empty

    CVR> pass_mime_types:

    | multipart/alternative
    | multipart/mixed
    | text/plain

    CVR> convert_html_to_plaintext: no

I'm wondering if there are good values for defaults here.  I'm
thinking specifically adding m/a, m/m, t/p to DEFAULT_PASS_MIME_TYPES
in Defaults.py.in.

-Barry

From barry@python.org  Fri Oct 25 21:15:06 2002
From: barry@python.org (Barry A. Warsaw)
Date: Fri, 25 Oct 2002 16:15:06 -0400
Subject: [Mailman-Developers] 
 Re: [Mailman-Users] speeding up archiver in mailman 2.1b3?
References: <176920000.1035416106@modulus.oit.ucsb.edu>
Message-ID: <15801.42570.962199.334290@gargle.gargle.HOWL>


>>>>> "ADC" == Andrew D Clark <andrew.clark@ucsb.edu> writes:

    ADC> Since I'm hopelessly backlogged in my archive queue (1644
    ADC> files), does anyone have any suggestions for speeding up
    ADC> archiving?  The qrunner process is certainly eating up CPU
    ADC> and memory, but is only archiving about 1 msg per minute.
    ADC> All the other queues move at a decent pace.

Here's a thought, if you're interested in hacking some code.

In Mailman/Archiver/Archiver.py, ArchiveMail() we create a new
HyperArchive instance each time we want to add a new message to the
archive.  That in turn creates a new HyperDatabase instance, which in
turn un-marshals all the state of the archiver.

I wonder if it wouldn't make more sense if the archiver stored the
HyperArchive instance on self and re-used it.  That might save a lot
of i/o, although I don't know if it would help much with overall
performance, and I don't know if Pipermail would still operate
correctly.  It's worth a shot.

Another idea would to change the scheme the Pipermail archiver used
from a one-file-per-message scheme to a Unix mailbox scheme.  The
basic idea would be for ToArchive.py to append the message to a Unix
mbox, and then have ArchRunner.py slurp a multi-message mbox into the
archive instead of doing one message at a time.

I don't have time to play with these ideas, so I'm cc'ing
mailman-developers, in case anyone wants to do some hacking and
profiling.

-Barry

From barry@python.org  Fri Oct 25 21:29:49 2002
From: barry@python.org (Barry A. Warsaw)
Date: Fri, 25 Oct 2002 16:29:49 -0400
Subject: [Mailman-Developers] wouldnitbenice to have URLs in admin mails?
References: <20021023220153.GA23069@rezo.net>
Message-ID: <15801.43453.699642.465603@gargle.gargle.HOWL>


>>>>> "fil" ==   <fil@rezo.net> writes:

    fil> I suggest we add an URL just after these kind of sentences:

    >> The attached message was received as a bounce, but either the
    >> bounce format was not recognized, or no member addresses could
    >> be extracted from it.  This mailing list has been configured to
    >> send all unrecognized bounce messages to the list
    >> administrator(s).
    fil> + http://HOSTNAME/mailman/admin/listname/...

Not a bad idea.  Thanks,
-Barry

From barry@python.org  Fri Oct 25 21:40:24 2002
From: barry@python.org (Barry A. Warsaw)
Date: Fri, 25 Oct 2002 16:40:24 -0400
Subject: [Mailman-Developers] listname not defined
References: <20021023205752.GA12292@rezo.net>
Message-ID: <15801.44088.980829.195703@gargle.gargle.HOWL>


>>>>> "fil" ==   <fil@rezo.net> writes:

    fil> I can't believe it's a Mailman bug: must be coming from my
    fil> stupid installation. However, just in case...

I think I accidently deleted a few lines in my last checkin.  Sorry
about that, fixed now in cvs.

-Barry

From barry@python.org  Fri Oct 25 22:09:47 2002
From: barry@python.org (Barry A. Warsaw)
Date: Fri, 25 Oct 2002 17:09:47 -0400
Subject: [Mailman-Developers] wrong log
References: <20021023205244.GH9387@rezo.net>
Message-ID: <15801.45851.747364.63893@gargle.gargle.HOWL>


>>>>> "fil" ==   <fil@rezo.net> writes:

    fil> the log reads >>>> Oct 23 22:49:40 2002 (9350) spip: deleted
    fil> fil+09@rezo.net; web confirmation

    fil> But I confirmed by email ;)

Fixed.
-Barry

From barry@python.org  Fri Oct 25 22:12:31 2002
From: barry@python.org (Barry A. Warsaw)
Date: Fri, 25 Oct 2002 17:12:31 -0400
Subject: [Mailman-Developers] Re: email2.4.3
References: <20021020190510.GD23093@rezo.net>
	<20021020190718.GE23093@rezo.net>
	<20021020194027.GA7174@rezo.net>
	<15796.22072.609641.308438@gargle.gargle.HOWL>
	<20021021194440.GC2065@rezo.net>
	<20021022074215.GD24706@rezo.net>
	<20021023180606.GB4035@rezo.net>
	<m3bs5l2bpf.fsf@zope.com>
	<20021023205059.GB3931@rezo.net>
Message-ID: <15801.46015.378103.298177@gargle.gargle.HOWL>


>>>>> "fil" ==   <fil@rezo.net> writes:

    fil> Oh no! Not again! In order to be sure, I just did a 'cvs
    fil> update -A' and a lot of diffs have arrived!! How comes I had
    fil> sticky tags - I don't think I've never done anything but cvs
    fil> update (?) ; well, the good news is that the bug's squashed.

Do you have a CVS/Tag file?  Do you have anything in a .cvsrc file
either in the current directory or in your home directory?

-Barry

From barry@python.org  Fri Oct 25 22:13:22 2002
From: barry@python.org (Barry A. Warsaw)
Date: Fri, 25 Oct 2002 17:13:22 -0400
Subject: [Mailman-Developers] Re: [Mailman-checkins] mailman
	TODO,2.11,2.12
References: <E184LkG-0006a6-00@usw-pr-cvs1.sourceforge.net>
	<20021023192904.GA28789@mawhrin.net>
Message-ID: <15801.46066.587235.786119@gargle.gargle.HOWL>


>>>>> "MS" == Mikhail Sobolev <mss@mawhrin.net> writes:

    MS> Can't just the unicode representations be compared?  I am
    MS> pretty sure that'd be enough...

I think you're right.  I'm beginning to think that MM3 will have to
use Unicode everywhere.

-Barry

From fil@rezo.net  Fri Oct 25 23:18:17 2002
From: fil@rezo.net (Fil)
Date: Sat, 26 Oct 2002 00:18:17 +0200
Subject: [Mailman-Developers] Re: email2.4.3
In-Reply-To: <15801.46015.378103.298177@gargle.gargle.HOWL>
References: <20021020190510.GD23093@rezo.net>
	<20021020190718.GE23093@rezo.net> <20021020194027.GA7174@rezo.net>
	<15796.22072.609641.308438@gargle.gargle.HOWL>
	<20021021194440.GC2065@rezo.net> <20021022074215.GD24706@rezo.net>
	<20021023180606.GB4035@rezo.net> <m3bs5l2bpf.fsf@zope.com>
	<20021023205059.GB3931@rezo.net>
	<15801.46015.378103.298177@gargle.gargle.HOWL>
Message-ID: <20021025221817.GA12927@rezo.net>

>     fil> Oh no! Not again! In order to be sure, I just did a 'cvs
>     fil> update -A' and a lot of diffs have arrived!! How comes I had
>     fil> sticky tags - I don't think I've never done anything but cvs
>     fil> update (?) ; well, the good news is that the bug's squashed.
> 
> Do you have a CVS/Tag file?  Do you have anything in a .cvsrc file
> either in the current directory or in your home directory?

No CVS/Tag, no .cvsrc; I really don't know! Must be a ghost in the shell.
I'm updating (with 'cvs update -A' again), but I still get

mailman@miel:~$ /usr/local/bin/python -S /usr/local/mailman/cron/checkdbs
Traceback (most recent call last):
  File "/usr/local/mailman/cron/checkdbs", line 116, in ?
    main()
  File "/usr/local/mailman/cron/checkdbs", line 75, in main
    text += '\n' + pending_requests(mlist)
UnicodeError: ASCII decoding error: ordinal not in range(128)


-- Fil


From barry@python.org  Fri Oct 25 23:24:46 2002
From: barry@python.org (Barry A. Warsaw)
Date: Fri, 25 Oct 2002 18:24:46 -0400
Subject: [Mailman-Developers] Re: email2.4.3
References: <20021020190510.GD23093@rezo.net>
	<20021020190718.GE23093@rezo.net>
	<20021020194027.GA7174@rezo.net>
	<15796.22072.609641.308438@gargle.gargle.HOWL>
	<20021021194440.GC2065@rezo.net>
	<20021022074215.GD24706@rezo.net>
	<20021023180606.GB4035@rezo.net>
	<m3bs5l2bpf.fsf@zope.com>
	<20021023205059.GB3931@rezo.net>
	<15801.46015.378103.298177@gargle.gargle.HOWL>
	<20021025221817.GA12927@rezo.net>
Message-ID: <15801.50350.664098.21259@gargle.gargle.HOWL>


>>>>> "fil" ==   <fil@rezo.net> writes:

    fil> No CVS/Tag, no .cvsrc; I really don't know! Must be a ghost
    fil> in the shell.

Any aliases?
    
    fil> I'm updating (with 'cvs update -A' again), but I still get

    fil> mailman@miel:~$ /usr/local/bin/python -S
    fil> /usr/local/mailman/cron/checkdbs Traceback (most recent call
    fil> last):
    |   File "/usr/local/mailman/cron/checkdbs", line 116, in ?
    |     main()
    |   File "/usr/local/mailman/cron/checkdbs", line 75, in main
    |     text += '\n' + pending_requests(mlist)
    fil> UnicodeError: ASCII decoding error: ordinal not in range(128)

./config.status ; make install

-Barry

From fil@rezo.net  Fri Oct 25 23:30:09 2002
From: fil@rezo.net (Fil)
Date: Sat, 26 Oct 2002 00:30:09 +0200
Subject: [Mailman-Developers] Re: email2.4.3
In-Reply-To: <20021025221817.GA12927@rezo.net>
References: <20021020190718.GE23093@rezo.net> <20021020194027.GA7174@rezo.net>
	<15796.22072.609641.308438@gargle.gargle.HOWL>
	<20021021194440.GC2065@rezo.net> <20021022074215.GD24706@rezo.net>
	<20021023180606.GB4035@rezo.net> <m3bs5l2bpf.fsf@zope.com>
	<20021023205059.GB3931@rezo.net>
	<15801.46015.378103.298177@gargle.gargle.HOWL>
	<20021025221817.GA12927@rezo.net>
Message-ID: <20021025223009.GB12927@rezo.net>

> UnicodeError: ASCII decoding error: ordinal not in range(128)

Sorry! The cron/checkdbs file in my cvs directory contained the latest
checkin you made, but not the file in mailman's directory. The mistake comes
from the way I did 'make install'. I forgot to do ./config.status
beforehand. Case closed.


-- Fil


From fil@rezo.net  Fri Oct 25 23:35:01 2002
From: fil@rezo.net (Fil)
Date: Sat, 26 Oct 2002 00:35:01 +0200
Subject: [Mailman-Developers] Re: email2.4.3
In-Reply-To: <15801.50350.664098.21259@gargle.gargle.HOWL>
References: <20021020194027.GA7174@rezo.net>
	<15796.22072.609641.308438@gargle.gargle.HOWL>
	<20021021194440.GC2065@rezo.net> <20021022074215.GD24706@rezo.net>
	<20021023180606.GB4035@rezo.net> <m3bs5l2bpf.fsf@zope.com>
	<20021023205059.GB3931@rezo.net>
	<15801.46015.378103.298177@gargle.gargle.HOWL>
	<20021025221817.GA12927@rezo.net>
	<15801.50350.664098.21259@gargle.gargle.HOWL>
Message-ID: <20021025223501.GC12927@rezo.net>

>     fil> No CVS/Tag, no .cvsrc; I really don't know! Must be a ghost
>     fil> in the shell.
> 
> Any aliases?

How would I verify this? 

-- Fil


From barry@python.org  Fri Oct 25 23:34:58 2002
From: barry@python.org (Barry A. Warsaw)
Date: Fri, 25 Oct 2002 18:34:58 -0400
Subject: [Mailman-Developers] I'm going to cut MM2.1b4 in a few hours
Message-ID: <15801.50962.600970.92713@gargle.gargle.HOWL>


Folks,

I'm going to cut MM2.1b4 later tonight.  I figure it's going to be in
about 4 or 5 hours from now (say around midnight -0400).  I'm going to
update python.org to the latest snapshot and watch it for a few hours
while we have a little dinner party :).  If everything looks okay,
I'll roll out the beta.  There have been tons of patches since 2.1b3
so even if things aren't perfect yet (ha ha) the new beta is still
overdue.

I'm letting i18n'ers know because I'm updating the catalogs right
now.  Which means you have about 4 hours to make last minute updates
for 2.1b4 or send them to me to check in.

It's always possible I'll get side tracked and you'll get a reprieve
of course. :)

-Barry

From barry@python.org  Fri Oct 25 23:45:45 2002
From: barry@python.org (Barry A. Warsaw)
Date: Fri, 25 Oct 2002 18:45:45 -0400
Subject: [Mailman-Developers] Re: email2.4.3
References: <20021020194027.GA7174@rezo.net>
	<15796.22072.609641.308438@gargle.gargle.HOWL>
	<20021021194440.GC2065@rezo.net>
	<20021022074215.GD24706@rezo.net>
	<20021023180606.GB4035@rezo.net>
	<m3bs5l2bpf.fsf@zope.com>
	<20021023205059.GB3931@rezo.net>
	<15801.46015.378103.298177@gargle.gargle.HOWL>
	<20021025221817.GA12927@rezo.net>
	<15801.50350.664098.21259@gargle.gargle.HOWL>
	<20021025223501.GC12927@rezo.net>
Message-ID: <15801.51609.77979.200307@gargle.gargle.HOWL>


>>>>> "fil" ==   <fil@rezo.net> writes:

    >> Any aliases?

    fil> How would I verify this?

Actually, I meant any shell aliases.  E.g. here's what I tend to use:

% alias sup
alias sup='cvs -q up -P -d'

-Barry

From noreply@sourceforge.net  Fri Oct 25 23:50:39 2002
From: noreply@sourceforge.net (noreply@sourceforge.net)
Date: Fri, 25 Oct 2002 15:50:39 -0700
Subject: [Mailman-Developers] 
 [ mailman-Patches-626550 ] Add "Date" field to admindb summary
Message-ID: <E185DHr-0001AR-00@usw-sf-web4.sourceforge.net>

Patches item #626550, was opened at 2002-10-21 17:12
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=300103&aid=626550&group_id=103

Category: Web UI
Group: Mailman 2.1
Status: Open
Resolution: None
Priority: 5
Submitted By: Jon Parise (jparise)
Assigned to: Nobody/Anonymous (nobody)
>Summary: Add "Date" field to admindb summary

Initial Comment:
The attached patch adds the message's "Date" value to
the summary fields (which already include Subject,
Size, and Reason).

----------------------------------------------------------------------

>Comment By: Jon Parise (jparise)
Date: 2002-10-25 18:50

Message:
Logged In: YES 
user_id=485579

Updated the patch to apply to the latest Mailman CVS code.

----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=300103&aid=626550&group_id=103

From fil@rezo.net  Sat Oct 26 00:04:28 2002
From: fil@rezo.net (Fil)
Date: Sat, 26 Oct 2002 01:04:28 +0200
Subject: [Mailman-Developers] Re: email2.4.3
In-Reply-To: <15801.51609.77979.200307@gargle.gargle.HOWL>
References: <20021021194440.GC2065@rezo.net> <20021022074215.GD24706@rezo.net>
	<20021023180606.GB4035@rezo.net> <m3bs5l2bpf.fsf@zope.com>
	<20021023205059.GB3931@rezo.net>
	<15801.46015.378103.298177@gargle.gargle.HOWL>
	<20021025221817.GA12927@rezo.net>
	<15801.50350.664098.21259@gargle.gargle.HOWL>
	<20021025223501.GC12927@rezo.net>
	<15801.51609.77979.200307@gargle.gargle.HOWL>
Message-ID: <20021025230428.GB22106@rezo.net>

> Actually, I meant any shell aliases.  E.g. here's what I tend to use:
> 
> % alias sup
> alias sup='cvs -q up -P -d'

All right. No such thing here. It's just that I had forgotten
./config.status

-- Fil


From chuqui@plaidworks.com  Sat Oct 26 02:53:53 2002
From: chuqui@plaidworks.com (Chuq Von Rospach)
Date: Fri, 25 Oct 2002 18:53:53 -0700
Subject: [Mailman-Developers] some mime filter notes...
In-Reply-To: <15801.42138.344811.957862@gargle.gargle.HOWL>
Message-ID: <C7AB5BBA-E885-11D6-B383-0003934516A8@plaidworks.com>


On Friday, October 25, 2002, at 01:07  PM, Barry A. Warsaw wrote:

> I'm wondering if there are good values for defaults here.  I'm
> thinking specifically adding m/a, m/m, t/p to DEFAULT_PASS_MIME_TYPES
> in Defaults.py.in.
>

I think so. It seems to pass along what is "the best compromise" for 
turning things into a text/plain message going out the other end, which 
is (I think) what most list users really want.

I've seen one funky thing, which is that if you take a M/M and strip it 
to a text/plain by removing the vcards and whatever else the mail 
server was attached, I've been told that the message when sent to an (I 
think) exchange server will turn the message footer into an attachment. 
That seems to be an exchange/outlook "feature". Whether we want to 
worry about that or not, I dunno.

But IMHO, that's the only glitch I've seen recently. My lists now seem 
rock solid, and I'm quite happy with them. I've had one guy have 
messages eaten because he was sending a text/html without a comparable 
text/plain, so his messages were showing up blank. When I pointed it 
out to him, he admitted he was simply hacking a web page into his 
email, so he realizes that's his problem, not mine.

The personalization is confusing a few people and messing up some 
people's filtering, but that's all working ot get people used to 
things, and I've been telling people to use List-ID for filtering for a 
year anyway... (grin).

I don't see any issues on my end. It's really stabilized nicely.


-- 
Chuq Von Rospach, Architech
chuqui@plaidworks.com -- http://www.plaidworks.com/chuqui/blog/

Stress is when you wake up screaming and you realize you haven't fallen 
asleep yet.


From chuqui@plaidworks.com  Sat Oct 26 04:06:12 2002
From: chuqui@plaidworks.com (Chuq Von Rospach)
Date: Fri, 25 Oct 2002 20:06:12 -0700
Subject: [Mailman-Developers] 2.1b3 issue with text/html email.
Message-ID: <E2050686-E88F-11D6-B383-0003934516A8@plaidworks.com>


I've got one user who's email has been disappearing. turns out he's 
sending email in text/html. Not M/A with a text and a plain part, but 
strictly text/html.

The problem is, in this case, his messages are being thrown away by 
mailman, but it logs absolutely nothing, sends back no message to the 
author, and doesn't tell the admin. it completely blackholes. The only 
reason I know these messages got here is because I archive all incoming 
mail, so I can tell that mailman saw this, and then ate it without a 
trace...


-- 
Chuq Von Rospach, Architech
chuqui@plaidworks.com -- http://www.plaidworks.com/chuqui/blog/

The Cliff's Notes Cliff's Notes on Hamlet:
     And they all died happily ever after


From tkikuchi@is.kochi-u.ac.jp  Sat Oct 26 13:03:06 2002
From: tkikuchi@is.kochi-u.ac.jp (Tokio Kikuchi)
Date: Sat, 26 Oct 2002 21:03:06 +0900
Subject: [Mailman-Developers] 
 Re: [Mailman-i18n] I'm going to cut MM2.1b4 in a few hours
References: <15801.50962.600970.92713@gargle.gargle.HOWL>
Message-ID: <3DBA847A.8010905@is.kochi-u.ac.jp>

Is it too late ? It was already weekend here in Japan.
Please download the latest japanese translation from

http://mm.tkikuchi.net/mailman-2.1.ja/mailman.i18n.ja.tar.gz

Barry A. Warsaw wrote:
> Folks,
> 
> I'm going to cut MM2.1b4 later tonight.  I figure it's going to be in
> about 4 or 5 hours from now (say around midnight -0400).  I'm going to
> update python.org to the latest snapshot and watch it for a few hours
> while we have a little dinner party :).  If everything looks okay,
> I'll roll out the beta.  There have been tons of patches since 2.1b3
> so even if things aren't perfect yet (ha ha) the new beta is still
> overdue.
> 
> I'm letting i18n'ers know because I'm updating the catalogs right
> now.  Which means you have about 4 hours to make last minute updates
> for 2.1b4 or send them to me to check in.
> 
> It's always possible I'll get side tracked and you'll get a reprieve
> of course. :)
> 
> -Barry
> 
> _______________________________________________
> Mailman-i18n mailing list
> Mailman-i18n@python.org
> http://mail.python.org/mailman/listinfo/mailman-i18n
> 
> 


-- 
Tokio Kikuchi, tkikuchi@ is.kochi-u.ac.jp
http://weather.is.kochi-u.ac.jp/


From barry@python.org  Sat Oct 26 18:17:05 2002
From: barry@python.org (Barry A. Warsaw)
Date: Sat, 26 Oct 2002 13:17:05 -0400
Subject: [Mailman-Developers] RELEASED Mailman 2.1 beta 4
Message-ID: <15802.52753.139705.927733@gargle.gargle.HOWL>


I've released Mailman 2.1 beta 4; see below for a list of changes
since version 2.1 beta 2.  There have been tons of bug fixes, added
support for Swedish, and some minor new features.

Remember that discussions about version 2.1 are still preferred on
mailman-developers.  It's also not a bad idea to follow the cvs
checkins if you're using 2.1b4.  I plan on moving the remaining
Mailman discussion lists (mailman-users, mailman-announce,
mailman-i18n, mailman-docs, and mailman-checkins) over to the new
version asap.

I'm also hoping that this will be the last beta release.  I've gotten
tons of great feedback since 2.1b3, and there are still a bunch of
mostly minor bugs to fix, but it's great to see more and more people
testing and deploying the new version.  I think we're getting really
close.

Enjoy,
-Barry

-------------------- snip snip --------------------
2.1 beta 4 (26-Oct-2002)

    The usual assortment of bug fixes and language updates, some u/i
    tweaks, as well as the following:

    - Configuring / building / installing
        o Tightened up some configure checks; it will now bark loudly
          if you don't have the Python distutils package available
          (some Linux distros only include distutils in their "devel"
          packages).

        o Mailman's username/group security assertions are now done by
          symbolic name instead of numeric id.  This provides a level
          of indirection that makes it much easier to move or package
          Mailman.  --with-mail-gid and --with-cgi-gid are retained,
          but they control the group names used instead.

    - Command line scripts
        o A new script, bin/transcheck that language teams can use to
          check their .po files.

        o bin/list_members grew a --fullnames/-f option to print the
          full names along with the addresses.

        o cron/senddigests grew --help/-h and --listname/-l options.

        o bin/fix_url.py grew some command line options to support moving
          a list to a specific virtual domain.

    - Pipermail / archiving
        o Reworked the directory layout for archive attachments to be
          less susceptible to inode overload.  Attachments are now
          placed in

          archives/private/<listname>/attachments/<YYYYMMDD>/<msgidhash>

        o Internationalization support in the archiver has been improved.

    - Internationalization
        o New languages: Swedish.

    - Mail handling
        o Content filtering now has a pass_mime_type variable, which
          is a whitelist of MIME types to allow in postings.  See the
          details of the variable in the Content Filtering category
          for more information.

        o If a member has enabled their DontReceiveDuplicates option,
          we'll also strip their addresses from the Cc headers in the
          copy of the message sent to the list.  This helps keep the
          Cc lines from growing astronomically.

        o Bounce messages are now forwarded to the list administrators
          both if they are unrecognized, and if no list member's
          address could be extracted.

        o Content filtering now has a filter_action variable which
          controls what happens when a message matches the content
          filter rules.  The default is still to discard the message.

        o When searching for an Approve/Approved header, the first
          non-whitespace line of the body of the message is also
          checked, if the body has a MIME type of text/plain.

        o If a list is personalized, and the list's posting address is
          not included in a Reply-To header, the posting address is
          copied into a Cc header, otherwise there was no (easy) way a
          recipient could reply back to the list.

        o Added a MS Exchange bounce recognizer.

        o New configuration variable news_moderation which allows the
          mail->news gateway to properly post to moderated newsgroups.

        o Messages sent to a list's owners now comes from the site
          list to prevent mail loops when list owners or moderators
          having bouncing addresses.

    - Miscellaneous
        o mailanctl prevents runaway restarts by imposing a maximum
          restart value (defaulting to 10) for restarting the
          qrunners.  If you hit this limit, do "mailmanctl stop"
          followed by "mailmanctl start".

        o The Membership Management page's search feature now includes
          searching on members real names.

        o The start of a manual for list administrators is given in
          Python HOWTO format (LaTeX).  It's in doc/mailman-admin.tex
          but it still needs lots of fleshing out.

        o More protections against creating a list with an invalid name.

From andrew.clark@ucsb.edu  Thu Oct 24 18:54:12 2002
From: andrew.clark@ucsb.edu (Andrew D. Clark)
Date: Thu, 24 Oct 2002 10:54:12 -0700
Subject: [Mailman-Developers] amazingly slow ArchRunner performance
Message-ID: <208930000.1035482052@modulus.oit.ucsb.edu>

Hello all,

I'm using mailman 2.1b3, on FreeBSD 4.6 (PIII 700mhz with 1GB RAM and fast 
disks) with Exim 4 and a local caching named server.  Qrunner performance 
seems vastly improved over mailman 2.0.x, but archiving is amazingly slow 
(ArchRunner pushes about 1 msg. every 1 to 2 minutes).  I've posted on the 
mailman-users list, to no avail.  I've got about twenty lists, but the 
problem one is a particularly bursty high volume list with only three 
members.  There's currently 2072 files in qfiles/archive (so about 1036 
messages), the oldest of which is a little over 24 hours old.  Restarting 
the queue runners didn't help, and there aren't any stale locks sitting 
around.  The mbox file for this list is 540MB in size.  I've grubbed 
through the code a bit (but I'm a Perl hacker, not a Python guru).  Help?
I'm not sure if there's a memory leak in the ArchRunner, or what exactly 
it's trying to slurp up, but it sure does eat RAM:

PID USERNAME PRI NICE  SIZE    RES STATE    TIME   WCPU    CPU COMMAND
42813 mailman   60   0    99M    98M RUN    921:25 96.09% 96.09% python

--
Andrew Clark
Campus Network Programmer
Office of Information Technology
University of California, Santa Barbara
andrew.clark@ucsb.edu (805) 893-5311

From andrew.clark@ucsb.edu  Fri Oct 25 21:28:09 2002
From: andrew.clark@ucsb.edu (Andrew D. Clark)
Date: Fri, 25 Oct 2002 13:28:09 -0700
Subject: [Mailman-Developers] 
 Re: [Mailman-Users] speeding up archiver in mailman 2.1b3?
In-Reply-To: <15801.42570.962199.334290@gargle.gargle.HOWL>
References: <176920000.1035416106@modulus.oit.ucsb.edu>
 <15801.42570.962199.334290@gargle.gargle.HOWL>
Message-ID: <6260000.1035577689@modulus.oit.ucsb.edu>

I've actually got over 3000 files in the archive queue now :)
But, setting ARCHIVE_TO_MBOX = 1 (skipping pipermail archiving), seems to 
speed things up vastly (at least, running qrunner -r Arch -o actually 
processes one message and takes less than 1 second (as opposed to over 5 
minutes)).  Running it right now without -o (endless loop) and it's 
chugging along at a wonderful pace.  Just finished in about 10 seconds.  So 
there ya go.  It's definately pipermail archiving that's the very expensive 
operation.
I can always run bin/arch to build the pipermail archive after the fact.  I 
wanted to look into using mhonarc anyhow..
I'll look into fixing mailman too ;)
Thanks very much for your help.

--
Andrew Clark
Campus Network Programmer
Office of Information Technology
University of California, Santa Barbara
andrew.clark@ucsb.edu (805) 893-5311

--On Friday, October 25, 2002 16:15:06 -0400 "Barry A. Warsaw" 
<barry@python.org> wrote:

>
>>>>>> "ADC" == Andrew D Clark <andrew.clark@ucsb.edu> writes:
>
>     ADC> Since I'm hopelessly backlogged in my archive queue (1644
>     ADC> files), does anyone have any suggestions for speeding up
>     ADC> archiving?  The qrunner process is certainly eating up CPU
>     ADC> and memory, but is only archiving about 1 msg per minute.
>     ADC> All the other queues move at a decent pace.
>
> Here's a thought, if you're interested in hacking some code.
>
> In Mailman/Archiver/Archiver.py, ArchiveMail() we create a new
> HyperArchive instance each time we want to add a new message to the
> archive.  That in turn creates a new HyperDatabase instance, which in
> turn un-marshals all the state of the archiver.
>
> I wonder if it wouldn't make more sense if the archiver stored the
> HyperArchive instance on self and re-used it.  That might save a lot
> of i/o, although I don't know if it would help much with overall
> performance, and I don't know if Pipermail would still operate
> correctly.  It's worth a shot.
>
> Another idea would to change the scheme the Pipermail archiver used
> from a one-file-per-message scheme to a Unix mailbox scheme.  The
> basic idea would be for ToArchive.py to append the message to a Unix
> mbox, and then have ArchRunner.py slurp a multi-message mbox into the
> archive instead of doing one message at a time.
>
> I don't have time to play with these ideas, so I'm cc'ing
> mailman-developers, in case anyone wants to do some hacking and
> profiling.
>
> -Barry




From danny@terweij.nl  Sat Oct 26 22:27:45 2002
From: danny@terweij.nl (Danny Terweij)
Date: Sat, 26 Oct 2002 23:27:45 +0200
Subject: [Mailman-Developers] Upgrade beta 1 to beta 4
Message-ID: <0b0d01c27d36$86d2c0a0$1e00a8c0@onsnet.org>

Hi,

I am upgrading now from b1 to b4. The compiling and installing without any
error messages!

But check_perms gives this :
directory permissions must be 02775: /usr/local/mailman/templates/pt_BR
directory permissions must be 02775: /usr/local/mailman/templates/sv
directory permissions must be 02775: /usr/local/mailman/pythonlib/lib
directory permissions must be 02775:
/usr/local/mailman/pythonlib/lib/python2.1
directory permissions must be 02775:
/usr/local/mailman/pythonlib/lib/python2.1/site-packages
directory permissions must be 02775: /usr/local/mailman/messages/pt_BR
directory permissions must be 02775: /usr/local/mailman/messages/sv

Maybe this is a little bug... if you can call it a bug.. maybe a typo orso.

I am running now the check_perms -f as root to fix.
(This check_perms process is running for 40 minutes here :( )

If any error found after the perms fix i let it know :)

Danny Terweij
Mailman - Dutch Translator (Still translating!!)




From danny@terweij.nl  Sat Oct 26 23:21:57 2002
From: danny@terweij.nl (Danny Terweij)
Date: Sun, 27 Oct 2002 00:21:57 +0200
Subject: [Mailman-Developers] transcheck
Message-ID: <0b8a01c27d3e$191f20f0$1e00a8c0@onsnet.org>


transcheck from beta4 gives a error

./transcheck
  File "./transcheck", line 47
    print >> fd, _(__doc__)
           ^
SyntaxError: invalid syntax


From gary.frederick@jsoft.com  Thu Oct 24 11:52:09 2002
From: gary.frederick@jsoft.com (Gary Frederick)
Date: Thu, 24 Oct 2002 05:52:09 -0500
Subject: [Mailman-Developers] The Life Cycle of Mailing Lists.
References: <8BD1A59E-E712-11D6-B94D-0003934516A8@plaidworks.com>
Message-ID: <3DB7D0D9.1080305@jsoft.com>

My wife includes it in the unsubscribe messages.

Gary

Chuq Von Rospach wrote:
> a friend passed this to me today. I thought some of you might find it 
> interesting.....
> 
> <http://www.plaidworks.com/chuqui/blog/000053.html#000053>
> 
> 


From danny@terweij.nl  Sun Oct 27 02:15:47 2002
From: danny@terweij.nl (Danny Terweij)
Date: Sun, 27 Oct 2002 03:15:47 +0100
Subject: [Mailman-Developers] Upgrade beta 1 to beta 4
References: <0b0d01c27d36$86d2c0a0$1e00a8c0@onsnet.org>
Message-ID: <0c1601c27d5e$c3cf8470$1e00a8c0@onsnet.org>

From: "Danny Terweij" <danny@terweij.nl>
> I am upgrading now from b1 to b4. The compiling and installing without any
> error messages!


> If any error found after the perms fix i let it know :)

All went fine until i ran bin/arch

Seems that the install did not make all the *.py files to *.pyc in
~/mailman/Mailman/Archiver/
/arch has some errors, it can not find /Archiver/HyperArch.pyc
How to convert *.py to *.pyc by hand?

Danny Terweij
Dutch - Translator


From fil@rezo.net  Sun Oct 27 10:34:26 2002
From: fil@rezo.net (Fil)
Date: Sun, 27 Oct 2002 11:34:26 +0100
Subject: [Mailman-Developers] checkdbs
Message-ID: <20021027103426.GB21333@rezo.net>

Now I'm getting these ;(

----- Forwarded message from Cron Daemon <root@miel.brainstorm.fr> -----

Date: Sun, 27 Oct 2002 08:00:02 +0100 (CET)
From: root@miel.brainstorm.fr (Cron Daemon)
To: mailman@miel.brainstorm.fr
Subject: Cron <mailman@miel> /usr/local/bin/python -S /usr/local/mailman/cron/checkdbs

Traceback (most recent call last):
  File "/usr/local/mailman/cron/checkdbs", line 26, in ?
    from email.Charset import Charset
ImportError: No module named email.Charset

----- End forwarded message -----

-- Fil


From pioppo@ferrara.linux.it  Sun Oct 27 13:20:04 2002
From: pioppo@ferrara.linux.it (Simone Piunno)
Date: Sun, 27 Oct 2002 14:20:04 +0100
Subject: [Mailman-Developers] transcheck
In-Reply-To: <0b8a01c27d3e$191f20f0$1e00a8c0@onsnet.org>
References: <0b8a01c27d3e$191f20f0$1e00a8c0@onsnet.org>
Message-ID: <20021027132004.GA13935@ferrara.linux.it>

On Sun, Oct 27, 2002 at 12:21:57AM +0200, Danny Terweij wrote:

> transcheck from beta4 gives a error
> 
> ./transcheck
>   File "./transcheck", line 47
>     print >> fd, _(__doc__)
>            ^
> SyntaxError: invalid syntax

me too. 
in my case the problem was that the 1st line hasn't been substituted
with the right python location (I've configured and compiled
--with-python=/usr/bin/python2).  
Probably you are in the same situation.

If you manually change the 1st line, another error immediatly appears,
introduced by the stderr->stdout conversion for usage().

This is the fix for both:

Index: bin/transcheck
===================================================================
RCS file: /cvsroot/mailman/mailman/bin/transcheck,v
retrieving revision 2.2
diff -u -r2.2 transcheck
--- bin/transcheck      21 Oct 2002 22:36:56 -0000      2.2
+++ bin/transcheck      27 Oct 2002 13:19:34 -0000
@@ -1,6 +1,6 @@
-#! /usr/bin/env python
+#! @PYTHON@
 #
-# transcheck - (c) 2002 by Simone Piunno <pioppo@ferrara.linux.it>
+# Copyright (C) 1998,1999,2000,2001,2002 by the Free Software
Foundation, Inc.
 #
 # This program is free software; you can redistribute it and/or modify
 # it
 # under the terms of the version 2.0 of the GNU General Public License
 # as
@@ -39,7 +39,7 @@
 
 
 

-def usage(code, msg):
+def usage(code, msg=''):
     if code:
         fd = sys.stderr
     else:


-- 
Adde parvum parvo magnus acervus erit.
Simone Piunno, FerraraLUG - http://members.ferrara.linux.it/pioppo

From danny@terweij.nl  Sun Oct 27 13:38:45 2002
From: danny@terweij.nl (Danny Terweij)
Date: Sun, 27 Oct 2002 14:38:45 +0100
Subject: [Mailman-Developers] transcheck
References: <0b8a01c27d3e$191f20f0$1e00a8c0@onsnet.org>
	<20021027132004.GA13935@ferrara.linux.it>
Message-ID: <0ca201c27dbe$2c6f11d0$1e00a8c0@onsnet.org>

From: "Simone Piunno" <pioppo@ferrara.linux.it>

> > transcheck from beta4 gives a error

> in my case the problem was that the 1st line hasn't been substituted
> with the right python location (I've configured and compiled
> --with-python=/usr/bin/python2).  
> Probably you are in the same situation.

Yes indeed.
 
> If you manually change the 1st line, another error immediatly appears,
> introduced by the stderr->stdout conversion for usage().
> 
> This is the fix for both:

Thanx, i'll add the fix right now :)

Danny.



From danny@terweij.nl  Sun Oct 27 13:44:07 2002
From: danny@terweij.nl (Danny Terweij)
Date: Sun, 27 Oct 2002 14:44:07 +0100
Subject: [Mailman-Developers] transcheck
References: 
	<0b8a01c27d3e$191f20f0$1e00a8c0@onsnet.org><20021027132004.GA13935@ferrara.linux.it>
	<0ca201c27dbe$2c6f11d0$1e00a8c0@onsnet.org>
Message-ID: <0caa01c27dbe$ec658690$1e00a8c0@onsnet.org>

From: "Danny Terweij" <danny@terweij.nl>
> Thanx, i'll add the fix right now :)

Now i got this error :

  File "./transcheck", line 42
    def usage(code, msg="):
                                       ^
SyntaxError: invalid token



From pioppo@ferrara.linux.it  Sun Oct 27 13:47:32 2002
From: pioppo@ferrara.linux.it (Simone Piunno)
Date: Sun, 27 Oct 2002 14:47:32 +0100
Subject: [Mailman-Developers] transcheck
In-Reply-To: <0caa01c27dbe$ec658690$1e00a8c0@onsnet.org>
References: <0ca201c27dbe$2c6f11d0$1e00a8c0@onsnet.org>
	<0caa01c27dbe$ec658690$1e00a8c0@onsnet.org>
Message-ID: <20021027134732.GA14238@ferrara.linux.it>

On Sun, Oct 27, 2002 at 02:44:07PM +0100, Danny Terweij wrote:

> Now i got this error :
> 
>   File "./transcheck", line 42
>     def usage(code, msg="):
-------------------------^^^

they are 2 single quotes, e.g. '', and not one double quote, e.g. "

-- 
Adde parvum parvo magnus acervus erit.
Simone Piunno, FerraraLUG - http://members.ferrara.linux.it/pioppo

From danny@terweij.nl  Sun Oct 27 14:31:03 2002
From: danny@terweij.nl (Danny Terweij)
Date: Sun, 27 Oct 2002 15:31:03 +0100
Subject: [Mailman-Developers] transcheck
References: <0ca201c27dbe$2c6f11d0$1e00a8c0@onsnet.org>
	<0caa01c27dbe$ec658690$1e00a8c0@onsnet.org>
	<20021027134732.GA14238@ferrara.linux.it>
Message-ID: <0d0501c27dc5$7a826370$1e00a8c0@onsnet.org>

From: "Simone Piunno" <pioppo@ferrara.linux.it>


> they are 2 single quotes, e.g. '', and not one double quote, e.g. "

:) '' and " looks almost the same on windows platform :(

But after running :

---
[root@lnx mailman]# ./bin/transcheck -q nl
578 warnings in 5 files
---

Witch warnings and how to fix ?

Danny.





From pioppo@ferrara.linux.it  Sun Oct 27 14:34:20 2002
From: pioppo@ferrara.linux.it (Simone Piunno)
Date: Sun, 27 Oct 2002 15:34:20 +0100
Subject: [Mailman-Developers] transcheck
In-Reply-To: <0d0501c27dc5$7a826370$1e00a8c0@onsnet.org>
References: <0ca201c27dbe$2c6f11d0$1e00a8c0@onsnet.org>
	<0caa01c27dbe$ec658690$1e00a8c0@onsnet.org>
	<20021027134732.GA14238@ferrara.linux.it>
	<0d0501c27dc5$7a826370$1e00a8c0@onsnet.org>
Message-ID: <20021027143420.GB14608@ferrara.linux.it>

On Sun, Oct 27, 2002 at 03:31:03PM +0100, Danny Terweij wrote:

> ---
> [root@lnx mailman]# ./bin/transcheck -q nl
> 578 warnings in 5 files
> ---
> 
> Witch warnings and how to fix ?

try re-running without -q

-- 
Adde parvum parvo magnus acervus erit.
Simone Piunno, FerraraLUG - http://members.ferrara.linux.it/pioppo

From danny@terweij.nl  Sun Oct 27 21:20:06 2002
From: danny@terweij.nl (Danny Terweij)
Date: Sun, 27 Oct 2002 22:20:06 +0100
Subject: [Mailman-Developers] Bounce notifications
Message-ID: <0e7301c27dfe$9fb79910$1e00a8c0@onsnet.org>

Hi,

On the admin.bounce webpage :

Should Mailman send you, the list owner, any bounce messages that failed to
be detected by the bounce processor? Yes is recommended.
(Details voor bounce_unrecognized_goes_to_list_owner)

If set to "No", it still sends unrecognized bounce notifications to the list
owner.

Since the upgrade from b1 to b4 i have so many not recognized bounce
notifications that my outgoing mailserver sends much more mail than ever. I
need to turn this option off!!

Any one a quick fix/patch?

Danny Terweij.



From noreply@sourceforge.net  Mon Oct 28 00:26:28 2002
From: noreply@sourceforge.net (noreply@sourceforge.net)
Date: Sun, 27 Oct 2002 16:26:28 -0800
Subject: [Mailman-Developers] 
 [ mailman-Bugs-571636 ] Traceback in error log - bounce handling
Message-ID: <E185xjg-0003Rs-00@usw-sf-web3.sourceforge.net>

Bugs item #571636, was opened at 2002-06-20 15:19
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=100103&aid=571636&group_id=103

Category: bounce detection
Group: 2.1 beta
>Status: Closed
Resolution: None
Priority: 4
Submitted By: Daniel Buchmann (avalon)
Assigned to: Nobody/Anonymous (nobody)
Summary: Traceback in error log - bounce handling

Initial Comment:
I got this traceback in my error log:

Jun 20 14:58:56 2002 (1992) Traceback (most recent call
last):
  File "/home/mailman/Mailman/Queue/Runner.py", line
105, in __oneloop
    self.__onefile(msg, msgdata)
  File "/home/mailman/Mailman/Queue/Runner.py", line
154, in __onefile
    keepqueued = self._dispose(mlist, msg, msgdata)
  File "/home/mailman/Mailman/Queue/BounceRunner.py",
line 71, in _dispose
    addrs = BouncerAPI.ScanMessages(mlist, msg)
  File "/home/mailman/Mailman/Bouncers/BouncerAPI.py",
line 60, in ScanMessages
    addrs = sys.modules[modname].process(msg)
  File "/home/mailman/Mailman/Bouncers/Microsoft.py",
line 35, in process
    body = StringIO(subpart.get_payload())
TypeError: expected read buffer, list found

The following had probably happened:
1. A spam mail was sent to one of our lists.
2. The list tried to send a "waiting for moderator
approval" mail to the sender of the spam mail.
3. The "waiting for moderator approval" mail bounced
(because the sender address was, of course, invalid). 

The bounce in (3) gave me the traceback and was moved
to qfiles/shunt/. PCK file is attached to this bug
report. Let me know if you need the DB file too... :)

----------------------------------------------------------------------

>Comment By: Daniel Buchmann (avalon)
Date: 2002-10-28 01:26

Message:
Logged In: YES 
user_id=184577

Okay, never mind, I've discarded the message. 
It was probably spam anyway..
:)

----------------------------------------------------------------------

Comment By: Barry A. Warsaw (bwarsaw)
Date: 2002-08-23 05:31

Message:
Logged In: YES 
user_id=12800

If this message has been in the queue since before you
upgraded, it's still got a bogus representation, which won't
be automatically fixed.  You'd need to do some manual
hacking on the object representation and then re-save it. 
If the message is important to you and you can't fix it,
upload it or email it and I'll attempt to do so.  If the
message isn't important to you, just discard it.  Messages
sent after the upgrade should be fine.

----------------------------------------------------------------------

Comment By: Daniel Buchmann (avalon)
Date: 2002-08-19 16:20

Message:
Logged In: YES 
user_id=184577

Well, at least something must have happened, because now the 
traceback looks different, this is the output of dumpdb on it:

Traceback (most recent call last):
  File "/home/mailman/bin/dumpdb", line 130, in ?
    msg = main()
  File "/home/mailman/bin/dumpdb", line 124, in main
    pp.pprint(m)
  File "/usr/lib/python2.2/pprint.py", line 110, in pprint
    self.__stream.write(self.pformat(object) + "\n")
  File "/usr/lib/python2.2/pprint.py", line 114, in pformat
    self.__format(object, sio, 0, 0, {}, 0)
  File "/usr/lib/python2.2/pprint.py", line 136, in __format
    rep = self.__repr(object, context, level - 1)
  File "/usr/lib/python2.2/pprint.py", line 200, in __repr
    self.__depth, level)
  File "/usr/lib/python2.2/pprint.py", line 287, in _safe_repr
    rep = `object`
  File "/home/mailman/Mailman/Message.py", line 41, in __repr__
    return self.__str__()
  File "/home/mailman/pythonlib/email/Message.py", line 89, in __str__
    return self.as_string(unixfrom=1)
  File "/home/mailman/pythonlib/email/Message.py", line 99, in as_string
    g.flatten(self, unixfrom=unixfrom)
  File "/home/mailman/pythonlib/email/Generator.py", line 81, in flatten
    self._write(msg)
  File "/home/mailman/pythonlib/email/Generator.py", line 109, in _write
    self._dispatch(msg)
  File "/home/mailman/pythonlib/email/Generator.py", line 136, in 
_dispatch
    meth(msg)
  File "/home/mailman/pythonlib/email/Generator.py", line 219, in 
_handle_multipart
    g.flatten(part, unixfrom=0)
  File "/home/mailman/pythonlib/email/Generator.py", line 81, in flatten
    self._write(msg)
  File "/home/mailman/pythonlib/email/Generator.py", line 109, in _write
    self._dispatch(msg)
  File "/home/mailman/pythonlib/email/Generator.py", line 136, in 
_dispatch
    meth(msg)
  File "/home/mailman/pythonlib/email/Generator.py", line 278, in 
_handle_message
    g.flatten(msg.get_payload(0), unixfrom=0)
  File "/home/mailman/pythonlib/email/Message.py", line 167, in 
get_payload
    raise TypeError, i
TypeError: 0

(Using latest CVS checked out today 19-aug-2002)

----------------------------------------------------------------------

Comment By: Barry A. Warsaw (bwarsaw)
Date: 2002-08-09 21:26

Message:
Logged In: YES 
user_id=12800

I think we resolved this one, didn't we?

----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=100103&aid=571636&group_id=103

From noreply@sourceforge.net  Mon Oct 28 00:43:31 2002
From: noreply@sourceforge.net (noreply@sourceforge.net)
Date: Sun, 27 Oct 2002 16:43:31 -0800
Subject: [Mailman-Developers] [ mailman-Bugs-629616 ] Invalid admin
	notification
Message-ID: <E185y0B-0000sz-00@usw-sf-web5.sourceforge.net>

Bugs item #629616, was opened at 2002-10-28 01:43
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=100103&aid=629616&group_id=103

Category: (un)subscribing
Group: 2.1 beta
Status: Open
Resolution: None
Priority: 5
Submitted By: Daniel Buchmann (avalon)
Assigned to: Nobody/Anonymous (nobody)
Summary: Invalid admin notification

Initial Comment:
Using latest CVS (2.1b4).
When someone subscribes through the web interface (listinfo page), the 
list admin will receive an invalid email if subscription notification has 
been turned on. The email contains the content of 
adminsubscribeack.txt, but no string interpolation is done. This is not 
the case if  subscribing using the mail command interface or subscribed 
by the admin using the admin page.

Also, if the subscribing user chooses a different language than the list 
language, the email notification sent to the admin will have a Subject: 
according to the list language, but body according to what the 
subscribing user chose. Which is probably not what we want... ;)

----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=100103&aid=629616&group_id=103

From keichwa@gmx.net  Sun Oct 27 05:52:27 2002
From: keichwa@gmx.net (Karl Eichwalder)
Date: Sun, 27 Oct 2002 06:52:27 +0100
Subject: [Mailman-Developers] 
 Translations via TP Robot? (Re: [Mailman-Announce] RELEASED Mailman
 2.1 beta 4)
In-Reply-To: <15802.52753.139705.927733@gargle.gargle.HOWL>
	(barry@python.org's message of "Sat, 26 Oct 2002 13:17:05 -0400")
References: <15802.52753.139705.927733@gargle.gargle.HOWL>
Message-ID: <sh3cqsh1ic.fsf@tux.gnu.franken.de>

barry@python.org (Barry A. Warsaw) writes:

This reminds me:

>     The usual assortment of bug fixes and language updates, some u/i
>     tweaks, as well as the following:

[...]

>     - Internationalization
>         o New languages: Swedish.

If you want to make use of the so call Translation Project Robot (TP
Robot) we can help you to find translators and to manage translation
updates.  Basically it enough to tell us from where to fetch the
.tar.bz2 (.tar.gz) archive file containing the .pot file.  All the
glory details you can read here:

    http://www.iro.umontreal.ca/contrib/po/HTML/maintainers.html

-- 
ke@suse.de (work) / keichwa@gmx.net (home):              |
http://www.suse.de/~ke/                                  |      ,__o
Free Translation Project:                                |    _-\_<,
http://www.iro.umontreal.ca/contrib/po/HTML/             |   (*)/'(*)

From narnett@mccmedia.com  Mon Oct 28 00:47:45 2002
From: narnett@mccmedia.com (Nick Arnett)
Date: Sun, 27 Oct 2002 16:47:45 -0800
Subject: [Mailman-Developers] Startup problems
Message-ID: <PHEDLIMPMIAGAEPEFBGCAEAKDJAA.narnett@mccmedia.com>

I asked this on mailman-users and got one good hint, but I guess this is a
more appropriate place.

After upgrading to 2.1b2 (on Red Hat 7.2), during which I changed
installation to /usr/local (from /var/local, I think), mailman is not
starting when I boot, and when I use Linuxconf to do anything, on exit, it
thinks that Mailman isn't running (it is) and tries to start it, but kills
it instead, somehow.

Looking at the Linuxconf logs, I find something like this:

Service Mailman is not running
/etc/rc5.d/S98mailman start

-- and --

No such file or directory:
'master-qrunner.www.mccmedia.com.3516'

I really don't know enough about this part of Linux to guess where to look
for the problem, though I suspect that perhaps it has something to do with
leftovers from the earlier version (whose version number I do not recall,
but it was a 1.x, I think).

Any suggestions?  The startup script appears to be in the right place
(/etc/init.d/mailman).

Mailman starts manually and runs just fine -- I just have to remember to
start it after booting or letting Linuxconf do its thing.  And
unfortunately, I forget, since those are rare events.

Thanks in advance,

Nick


--
Nick Arnett
Phone/fax: (408) 904-7198
narnett@mccmedia.com



From narnett@mccmedia.com  Mon Oct 28 01:03:38 2002
From: narnett@mccmedia.com (Nick Arnett)
Date: Sun, 27 Oct 2002 17:03:38 -0800
Subject: [Mailman-Developers] 2.1b3 issue with text/html email.
In-Reply-To: <E2050686-E88F-11D6-B383-0003934516A8@plaidworks.com>
Message-ID: <PHEDLIMPMIAGAEPEFBGCCEALDJAA.narnett@mccmedia.com>

> -----Original Message-----
> From: mailman-developers-bounces@python.org
> [mailto:mailman-developers-bounces@python.org]On Behalf Of Chuq Von
> Rospach
...

> I've got one user who's email has been disappearing. turns out he's
> sending email in text/html. Not M/A with a text and a plain part, but
> strictly text/html.
>
> The problem is, in this case, his messages are being thrown away by
> mailman, but it logs absolutely nothing, sends back no message to the
> author, and doesn't tell the admin. it completely blackholes. The only
> reason I know these messages got here is because I archive all incoming
> mail, so I can tell that mailman saw this, and then ate it without a
> trace...

Aha!  I have the same problem and I suspect that was the reason.  My user is
on AOL and I've sent him instructions on how to change his mail
configuration appropriately.  The complete lack of notification is a
problem...

Nick


From barry@python.org  Mon Oct 28 03:20:46 2002
From: barry@python.org (Barry A. Warsaw)
Date: Sun, 27 Oct 2002 22:20:46 -0500
Subject: [Mailman-Developers] transcheck
References: <0b8a01c27d3e$191f20f0$1e00a8c0@onsnet.org>
	<20021027132004.GA13935@ferrara.linux.it>
Message-ID: <15804.44302.425448.732047@gargle.gargle.HOWL>


Fixed, thanks.

From barry@python.org  Mon Oct 28 03:44:26 2002
From: barry@python.org (Barry A. Warsaw)
Date: Sun, 27 Oct 2002 22:44:26 -0500
Subject: [Mailman-Developers] checkdbs
References: <20021027103426.GB21333@rezo.net>
Message-ID: <15804.45722.60296.928538@gargle.gargle.HOWL>


Fixed in cvs.

From noreply@sourceforge.net  Mon Oct 28 03:47:40 2002
From: noreply@sourceforge.net (noreply@sourceforge.net)
Date: Sun, 27 Oct 2002 19:47:40 -0800
Subject: [Mailman-Developers] [ mailman-Bugs-615040 ] problem creating newlist
Message-ID: <E1860sO-0007pY-00@usw-sf-web4.sourceforge.net>

Bugs item #615040, was opened at 2002-09-26 12:22
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=100103&aid=615040&group_id=103

Category: command line scripts
>Group: 2.1 beta
>Status: Closed
>Resolution: Out of Date
Priority: 5
Submitted By: Tim White (reverend_timmyd)
Assigned to: Nobody/Anonymous (nobody)
Summary: problem creating newlist

Initial Comment:

I get the following when I try to run "newlist":

Traceback (innermost last):
  File "/home/admin/src/mailman-2.0.13/bin/newlist", line 
52, in ?
    import paths
ImportError: No module named paths

any ideas?


----------------------------------------------------------------------

>Comment By: Barry A. Warsaw (bwarsaw)
Date: 2002-10-27 22:47

Message:
Logged In: YES 
user_id=12800

I've since learned that another reason this error can happen
is if the Python you've got doesn't have the distutils
package available.  This happens on Linux when Python is
installed via package manager and the Python "devel" package
isn't installed.  Mailman 2.1b4 has a check for this in its
configure script.


----------------------------------------------------------------------

Comment By: Tim White (reverend_timmyd)
Date: 2002-09-27 15:57

Message:
Logged In: YES 
user_id=618626

thanks for the help. However, the guy that installed it for me 
had to hack it up so much to make it work with plesk (my 
virtual host management software, that we've had to write 
some custom scripts to use instead of the ones that came w/ 
mailman...

----------------------------------------------------------------------

Comment By: Tomas Lindroos (skitta)
Date: 2002-09-26 12:43

Message:
Logged In: YES 
user_id=613582

Don't run "newlist" from the source directory, run it from the installation  
directory. This is true for some other scripts as well.  
  
Check the installation directory, and you will find a file "paths.py", which is  
correctly imported from "newlist" (and other) scripts. The same library can 
be found in other directories as well (i.e. $install_dir/cron/). 
  
/skitta  

----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=100103&aid=615040&group_id=103

From gward@mems-exchange.org  Mon Oct 28 16:12:44 2002
From: gward@mems-exchange.org (Greg Ward)
Date: Mon, 28 Oct 2002 11:12:44 -0500
Subject: [Mailman-Developers] Slightly lame logging: web subscription request
Message-ID: <20021028161244.GA30884@cthulhu.gerg.ca>

It appears that MM 2.1b4's logging of web subscription requests is a bit
lame -- it doesn't actually log the email address involved, just the
real name.  Is this deliberate?

        Greg
-- 
Greg Ward - software developer                gward@mems-exchange.org
MEMS Exchange                            http://www.mems-exchange.org

From barry@python.org  Mon Oct 28 16:16:16 2002
From: barry@python.org (Barry A. Warsaw)
Date: Mon, 28 Oct 2002 11:16:16 -0500
Subject: [Mailman-Developers] Slightly lame logging: web subscription
	request
References: <20021028161244.GA30884@cthulhu.gerg.ca>
Message-ID: <15805.25296.355815.747438@gargle.gargle.HOWL>


>>>>> "GW" == Greg Ward <gward@mems-exchange.org> writes:

    GW> It appears that MM 2.1b4's logging of web subscription
    GW> requests is a bit lame -- it doesn't actually log the email
    GW> address involved, just the real name.  Is this deliberate?

No.  I'll have to double check this when I get a chance.
-Barry

From gward@mems-exchange.org  Mon Oct 28 16:22:21 2002
From: gward@mems-exchange.org (Greg Ward)
Date: Mon, 28 Oct 2002 11:22:21 -0500
Subject: [Mailman-Developers] Slightly lame logging: web subscription
	request
In-Reply-To: <15805.25296.355815.747438@gargle.gargle.HOWL>
References: <20021028161244.GA30884@cthulhu.gerg.ca>
	<15805.25296.355815.747438@gargle.gargle.HOWL>
Message-ID: <20021028162221.GB30906@cthulhu.gerg.ca>

On 28 October 2002, Barry A. Warsaw said:
> No.  I'll have to double check this when I get a chance.

OK.  If it helps, here are some log excerpts.  First, Apache:

24.202.9.108 - - [28/Oct/2002:11:09:27 -0500] "GET /mailman/listinfo/mems-announce HTTP/1.1" 200 6451 "http://mail.mems-exchange.org/mailman/listinfo/" "Opera/6.1 (Linux 2.4.19-gw i686; U)  [en]" "SSL -" 0sec
24.202.9.108 - - [28/Oct/2002:11:12:04 -0500] "POST /mailman/subscribe/mems-announce HTTP/1.1" 200 1202 "http://mail.mems-exchange.org/mailman/listinfo/mems-announce" "Opera/6.1 (Linux 2.4.19-gw i686; U)  [en]" "SSL -" 1sec

I entered the subscribe form with email="xxx@gerg.ca" (censored since I
don't like revealing addresses in my personal domain) and fullname="Greg
Ward".  Here's what Mailman logged:

==> subscribe <==
Oct 28 11:12:04 2002 (26884) mems-announce: pending Greg Ward  24.202.9.108

==> smtp <==
Oct 28 11:12:05 2002 (24845) <mailman.0.1035821524.26884.mems-announce@memsnet.org> smtp for 1 recips, completed in 0.157 seconds

and here's what Exim logged for the outgoing confirmation mail:

2002-10-28 11:12:05 186CUm-0006zf-00 <= mems-announce-bounces@memsnet.org H=localhost (kronos.mems-exchange.org) [127.0.0.1] P=esmtp S=1884 id=mailman.0.1035821524.26884.mems-announce@memsnet.org
2002-10-28 11:12:06 186CUm-0006zf-00 => xxx@gerg.ca R=special_lookuphost T=special_remote_smtp H=cthulhu.gerg.ca [24.202.9.108]
2002-10-28 11:12:06 186CUm-0006zf-00 Completed

        Greg

From gward@mems-exchange.org  Mon Oct 28 16:45:24 2002
From: gward@mems-exchange.org (Greg Ward)
Date: Mon, 28 Oct 2002 11:45:24 -0500
Subject: [Mailman-Developers] qfiles/shunt full of posts
Message-ID: <20021028164524.GA31049@cthulhu.gerg.ca>

I just noticed that qfiles/shunt in my MM 2.1 installation contains a
large subset of all messages posted since I upgraded to 2.1b3+ a few
weeks ago.  (It's entirely possible that every single list post we've
received lives on in qfiles/shunt -- I haven't investigated thoroughly
enough to know if that's the case.)  We're now running 2.1b4, and the
problem continues -- most list posts received (hmmm, apparently not all
of them) sit in qfiles/shunt and live there indefinitely.  It looks as
though everything in qfiles/shunt was in fact successfully processed by
Mailman and posted to the relevant list.  I don't see any bounces,
confirmation notices/responses, or other bureaucracy.

Anyone want to sell me a clue here?  Seen anything similar?
Suggestions for what I should investigate?  Things to look for in the
logs?

        Greg
-- 
Greg Ward - software developer                gward@mems-exchange.org
MEMS Exchange                            http://www.mems-exchange.org

From gward@mems-exchange.org  Mon Oct 28 16:54:08 2002
From: gward@mems-exchange.org (Greg Ward)
Date: Mon, 28 Oct 2002 11:54:08 -0500
Subject: [Mailman-Developers] Contrib: bin/show_qfiles
Message-ID: <20021028165408.GA31086@cthulhu.gerg.ca>


---------------------- multipart/mixed attachment
The attached script is handy for reading Mailman 2.1 queue files, eg.

  bin/show_qfiles qfiles/shunt/*.pck

Help yourselves.

        Greg
-- 
Greg Ward - software developer                gward@mems-exchange.org
MEMS Exchange                            http://www.mems-exchange.org

---------------------- multipart/mixed attachment
#! /www/python/bin/python

"""Show the contents of one or more Mailman queue files.

Usage: show_qfiles qfile ...

Example: show_qfiles qfiles/shunt/*.pck
"""

import sys
from cPickle import load
import paths

from Mailman.i18n import _


def usage(code, msg=''):
    if code:
        fd = sys.stderr
    else:
        fd = sys.stdout
    print >> fd, _(__doc__)
    if msg:
        print >> fd, msg
    sys.exit(code)


def main():
    if len(sys.argv) < 2:
	usage(1, "Not enough arguments")

    for filename in sys.argv[1:]:
	print "** %s **" % filename
        file = open(filename)
	if filename.endswith(".pck"):
	    msg = load(file)
	    sys.stdout.write(str(msg))
	else:
	    sys.stdout.write(file.read())


if __name__ == '__main__':
    main()

---------------------- multipart/mixed attachment--

From blair@orcaware.com  Tue Oct 29 00:52:05 2002
From: blair@orcaware.com (Blair Zajac)
Date: Mon, 28 Oct 2002 16:52:05 -0800
Subject: [Mailman-Developers] Re: [Mailman-Announce] Welcome to Mailman 2.1
References: <15805.55591.47010.608367@gargle.gargle.HOWL>
Message-ID: <3DBDDBB5.934F349A@orcaware.com>

"Barry A. Warsaw" wrote:
> 
> Hi folks,
> 
> Just a quick note to let you know that I moved these four lists to
> Mailman 2.1b4+.  Things seem to be working okay, but of course, now
> the fun begins.  I'll be watching the logs, but please do let me know
> if you notice anything strange.
> 
> -Barry
> 
> P.S. to mailman-users-owner... Hope you like the new moderation
> interface!
> 
> _______________________________________________
> Mailman-announce mailing list
> Mailman-announce@python.org
> http://mail.python.org/mailman/listinfo/mailman-announce

Selecting the URL

    http://mail.python.org/mailman/listinfo/mailman-announce

gets a page that says at the top:

    "No such list mailman-announce"

Best,
Blair

-- 
Blair Zajac <blair@orcaware.com>
Web and OS performance plots - http://www.orcaware.com/orca/

From barry@python.org  Tue Oct 29 01:11:58 2002
From: barry@python.org (Barry A. Warsaw)
Date: Mon, 28 Oct 2002 20:11:58 -0500
Subject: [Mailman-Developers] Re: [Mailman-Announce] Welcome to Mailman
	2.1
References: <15805.55591.47010.608367@gargle.gargle.HOWL>
	<3DBDDBB5.934F349A@orcaware.com>
Message-ID: <15805.57438.11694.478206@gargle.gargle.HOWL>


>>>>> "BZ" == Blair Zajac <blair@orcaware.com> writes:

    BZ> Selecting the URL

    BZ>     http://mail.python.org/mailman/listinfo/mailman-announce

    BZ> gets a page that says at the top:

    BZ>     "No such list mailman-announce"

Thanks, should be fixed now.  I messed up a url in the Apache
RewriteRule.

-Barry

From chuqui@plaidworks.com  Tue Oct 29 01:17:16 2002
From: chuqui@plaidworks.com (Chuq Von Rospach)
Date: Mon, 28 Oct 2002 17:17:16 -0800
Subject: [Mailman-Developers] future enhancement thought...
Message-ID: <2972A1CC-EADC-11D6-A331-0003934516A8@plaidworks.com>


Just had a weird thought.

One of the things I run into at times is when someone mails me and says 
"am I still on the list? Is something wrong?"

And what I generally do is do a find_members and send them the result. 
Much of the time, it's a list that's gone quiet, sometimes it's a 
problem on their end.

but I was thinking today, as I did this, that it'd be nice if 
find_members had a verbose flag that would optionally print out a 
summary of the user settings, so I could tell at a glance (and forward 
to the user) things like digest mode or nomail settings, since a person 
could be nomail and you can't really tell without using the web 
interface....


-- 
Chuq Von Rospach, Architech
chuqui@plaidworks.com -- http://www.plaidworks.com/chuqui/blog/


From chuqui@plaidworks.com  Tue Oct 29 07:23:53 2002
From: chuqui@plaidworks.com (Chuq Von Rospach)
Date: Mon, 28 Oct 2002 23:23:53 -0800
Subject: [Mailman-Developers] Way into the future -- XML and Mailman?
Message-ID: <60CF0B18-EB0F-11D6-A331-0003934516A8@plaidworks.com>


Have folks seen Jon Udell's piece on XML?

<http://www.infoworld.com/articles/pl/xml/02/10/28/ 
021028plxmlclient.xml>

Some fascinating stuff in there. and I think in large part he's right.

But it's got me wondering, in a very blue-sky kind of way, how a move  
to XML as a content model for e-mail and general data transfers is  
going to impact mailing list systems like mailman. how would a  
transition be handled?

I don't have a clue. It's something to be looked at for maybe Mailman  
5.0 or whatever, some forever into the future (where, in tech time,  
forever is > 18 months).

But maybe it's soemthing folks ought to gnaw on a bit, just to start  
thinking the issue over...

What if the future of email isn't HTML or style text (per se), but a  
mime-encoded XML piece? How does that affect list servers? do they  
mutate into specialized versions of a jabber server?


-- 
Chuq Von Rospach, Architech
chuqui@plaidworks.com -- http://www.plaidworks.com/chuqui/blog/

Someday, we'll look back on this, laugh
nervously and change the subject.


From noreply@sourceforge.net  Tue Oct 29 09:03:54 2002
From: noreply@sourceforge.net (noreply@sourceforge.net)
Date: Tue, 29 Oct 2002 01:03:54 -0800
Subject: [Mailman-Developers] 
 [ mailman-Patches-630339 ] create list with multiple owners
Message-ID: <E186SHy-0001je-00@usw-sf-web1.sourceforge.net>

Patches item #630339, was opened at 2002-10-29 18:03
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=300103&aid=630339&group_id=103

Category: list administration
Group: Mailman 2.1
Status: Open
Resolution: None
Priority: 5
Submitted By: Hatuka*nezumi (hatukanezumi)
Assigned to: Nobody/Anonymous (nobody)
Summary: create list with multiple owners

Initial Comment:
Create list with multiple list owners, by 'create' CGI
script and 'newlist' command line script.


----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=300103&aid=630339&group_id=103

From gary.frederick@jsoft.com  Tue Oct 29 12:12:37 2002
From: gary.frederick@jsoft.com (Gary Frederick)
Date: Tue, 29 Oct 2002 06:12:37 -0600
Subject: [Mailman-Developers] Way into the future -- XML and Mailman?
References: <60CF0B18-EB0F-11D6-A331-0003934516A8@plaidworks.com>
Message-ID: <3DBE7B35.1050004@jsoft.com>

I poked around with XMLizing email a while ago. I did a bit of testing with using J2EE's JavaMail to read email and turning headers and the various mime parts into xml. I also did a bit with making MHonArc XHTML.

My feel was that if email were XML, it would simplify parsing and that software that treated email as another source of information would start to become more common. I saw mailing list software as tapping into a stream of email and then doing what it does best. I did not see a big change on how stuff works. We would still get a message, apply whatever filters and send the message out. It would probably be a bit easier to 'see' what was happening with XML. I saw XMLizing email as a benefit for archiving. We could use XSLT to translate the email to whatever archive format without much work.

Then I went on to other things :-)

Gary

Chuq Von Rospach wrote:
> 
> Have folks seen Jon Udell's piece on XML?
> 
> <http://www.infoworld.com/articles/pl/xml/02/10/28/ 021028plxmlclient.xml>
> 
> Some fascinating stuff in there. and I think in large part he's right.
> 
> But it's got me wondering, in a very blue-sky kind of way, how a move  
> to XML as a content model for e-mail and general data transfers is  
> going to impact mailing list systems like mailman. how would a  
> transition be handled?
> 
> I don't have a clue. It's something to be looked at for maybe Mailman  
> 5.0 or whatever, some forever into the future (where, in tech time,  
> forever is > 18 months).
> 
> But maybe it's soemthing folks ought to gnaw on a bit, just to start  
> thinking the issue over...
> 
> What if the future of email isn't HTML or style text (per se), but a  
> mime-encoded XML piece? How does that affect list servers? do they  
> mutate into specialized versions of a jabber server?
> 
> 


From chk@pobox.com  Tue Oct 29 14:49:32 2002
From: chk@pobox.com (Harald Koch)
Date: Tue, 29 Oct 2002 09:49:32 -0500
Subject: [Mailman-Developers] 2.1b4: cron/checkdbs tweak
Message-ID: <19817.1035902972@elisabeth.cfrq.net>

Contains the following:

	import sys
	import time
	from types import UnicodeType
	from email.Charset import Charset

	import paths

Alas, if you haven't installed the latest email package in your python
system directory, this doesn't work.

The line "from email.Charset import Charset" should come after "import
paths", n'est pas?

-- 
Harald Koch     <chk@pobox.com>

From noreply@sourceforge.net  Tue Oct 29 14:41:18 2002
From: noreply@sourceforge.net (noreply@sourceforge.net)
Date: Tue, 29 Oct 2002 06:41:18 -0800
Subject: [Mailman-Developers] [ mailman-Bugs-630497 ] Traceback in error file
Message-ID: <E186XYU-0008MD-00@usw-sf-web4.sourceforge.net>

Bugs item #630497, was opened at 2002-10-29 14:41
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=100103&aid=630497&group_id=103

Category: bounce detection
Group: 2.1 beta
Status: Open
Resolution: None
Priority: 5
Submitted By: Andy Carpenter (acarpenter)
Assigned to: Nobody/Anonymous (nobody)
Summary: Traceback in error file

Initial Comment:
Since I upgraded to 2.1b4, I have started to get the 
following tracebacks in the error log file.  It is a 
characteristics of our busiest mail list that many of the posts 
come from the far east with unicode characters in their 
subject line.


Oct 29 14:19:35 2002 (26533) Uncaught runner exception: 
ASCII decoding error: or
dinal not in range(128)
Oct 29 14:19:35 2002 (26533) Traceback (most recent call 
last):
  File "/usr/local/mailman/Mailman/Queue/Runner.py", line 
105, in _oneloop
    self._onefile(msg, msgdata)
  File "/usr/local/mailman/Mailman/Queue/Runner.py", line 
154, in _onefile
    keepqueued = self._dispose(mlist, msg, msgdata)
  
File "/usr/local/mailman/Mailman/Queue/IncomingRunner.py"
, line 129, in _dispo
se
    status = self._dopipeline(mlist, msg, msgdata, pipeline)
  
File "/usr/local/mailman/Mailman/Queue/IncomingRunner.py"
, line 167, in _dopip
eline
    mlist.BounceMessage(msg, msgdata, e)
  File "/usr/local/mailman/Mailman/Bouncer.py", line 273, in 
BounceMessage
    lang=self.preferred_language)
  File "/usr/local/mailman/Mailman/Message.py", line 203, in 
__init__
    self['Subject'] = Header(subject, charset, 
header_name='Subject')
  File "/usr/local/mailman/pythonlib/email/Header.py", line 
164, in __init__
    self.append(s, charset)
  File "/usr/local/mailman/pythonlib/email/Header.py", line 
230, in append
    ustr = unicode(s, incodec)
UnicodeError: ASCII decoding error: ordinal not in range
(128)

----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=100103&aid=630497&group_id=103

From gward@python.net  Tue Oct 29 15:10:47 2002
From: gward@python.net (Greg Ward)
Date: Tue, 29 Oct 2002 10:10:47 -0500
Subject: [Mailman-Developers] VERP interval: why not time-based?
Message-ID: <20021029151047.GA613@cthulhu.gerg.ca>

Seems to me as though the obvious way to decide when it's time to VERP
another outgoing message is, "Has it been ~1 day since the last VERP
message", rather than, "Has it been N messages since the last VERP
message?"  This is based on the fact that, by default, only one bounce
per day counts.  Ie. if a list's bounce-removal parameters have not been
tweaked, then there's no difference between detecting 1 bounce or 10
bounces in a given day, but detecting 5 bounces in 5 days is highly
useful information.

Surely I'm not the only one who has thought of this -- was this wrangled
over months and months ago, or what?

        Greg
-- 
Greg Ward <gward@python.net>                         http://www.gerg.ca/
"Question authority!"  "Oh yeah?  Says who?"

From barry@python.org  Tue Oct 29 16:49:21 2002
From: barry@python.org (Barry A. Warsaw)
Date: Tue, 29 Oct 2002 11:49:21 -0500
Subject: [Mailman-Developers] 2.1b4: cron/checkdbs tweak
References: <19817.1035902972@elisabeth.cfrq.net>
Message-ID: <15806.48145.621004.759628@gargle.gargle.HOWL>


>>>>> "HK" == Harald Koch <chk@pobox.com> writes:

    HK> The line "from email.Charset import Charset" should come after
    HK> "import paths", n'est pas?

Yes, exactly.  Already fixed in cvs.
-Barry

From JasonR.Mastaler  Tue Oct 29 19:07:51 2002
From: JasonR.Mastaler (JasonR.Mastaler)
Date: Tue, 29 Oct 2002 12:07:51 -0700
Subject: [Mailman-Developers] Re: New TMDA-style service: spamarrest.com
References: <3DAF4CE0.4080408@nleaudio.com>
Message-ID: <hhy98hqd14.fsf@hrothgar.la.mastaler.com>

"Bob Puff@NLE" <bob@nleaudio.com> writes:

> I am simply passing on the info, as it's gonna bring out a bunch
> more admin messages / mail loops for mailing lists.

Does spamarrest not supress auto-responses to mailing list messages?
Does it not have any auto-response rate limiting to prevent mail
loops?



From Dan.Mick@sun.com  Tue Oct 29 19:11:40 2002
From: Dan.Mick@sun.com (Dan Mick)
Date: Tue, 29 Oct 2002 11:11:40 -0800
Subject: [Mailman-Developers] Re: email2.4.3
References: <20021020190510.GD23093@rezo.net>
	<20021020190718.GE23093@rezo.net> <20021020194027.GA7174@rezo.net>
	<15796.22072.609641.308438@gargle.gargle.HOWL>
	<20021021194440.GC2065@rezo.net> <20021022074215.GD24706@rezo.net>
	<20021023180606.GB4035@rezo.net> <m3bs5l2bpf.fsf@zope.com>
	<20021023205059.GB3931@rezo.net>
Message-ID: <3DBEDD6C.6070206@sun.com>

Fil wrote:
 > How comes I had sticky tags - I don't think I've
> never done anything but cvs update (?) 

I believe that falls under the heading of "CVS sucks".

I *never* do any update other than "cvs -q update -dAP".



From danny@terweij.nl  Wed Oct 30 10:58:57 2002
From: danny@terweij.nl (Danny Terweij)
Date: Wed, 30 Oct 2002 11:58:57 +0100
Subject: [Mailman-Developers] B4 changed the To: line?
Message-ID: <134901c28003$58d1d900$1e00a8c0@onsnet.org>

Hi all,

I do not know if this is the same as discussed now on the users list about
"List Id".
But all my users from all lists are complaining since i upgraded to B4/B4+ .

The Short list description is placed at every outgoing mail at the TO:
field.

If  the short list desciption is: "Welcome at this fine list" (You can see
this at the /listinfo/ web page) Then all mail looks like this:

>From : Danny Terweij
To: Welcome at this fine list

How to stop this or how to get back what it was before B4 version?

Danny Terweij.




From noreply@sourceforge.net  Wed Oct 30 11:39:26 2002
From: noreply@sourceforge.net (noreply@sourceforge.net)
Date: Wed, 30 Oct 2002 03:39:26 -0800
Subject: [Mailman-Developers] 
 [ mailman-Patches-444879 ] Archive indexer control to improve index
Message-ID: <E186rC2-0006Yd-00@usw-sf-web4.sourceforge.net>

Patches item #444879, was opened at 2001-07-26 18:01
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=300103&aid=444879&group_id=103

Category: None
Group: None
Status: Open
Resolution: None
Priority: 7
Submitted By: Richard Barrett (ppsys)
Assigned to: Barry A. Warsaw (bwarsaw)
Summary: Archive indexer control to improve index

Initial Comment:
This patch is applicable to Mailman 2.0.6 release and 
supercedes ealier patches 401669 and 402422.

This patch should improve the quality of search 
results returned by search engines
such as htdig (http://www.htdig.org) where the seach 
engine's index builder responds 
to strings embedded in the html pages that are the 
subject of the indexing. The 
changes in this patch: 

1. allow strings for enabling and disabling indexing 
to be defined in mm_cfg.py.

2. embeds those strings in the pages generated as the 
html version of a list's
   archive.

By default nothing in the html changes. To get the 
desired effect, you must
define ARCHIVE_INDEXING_ENABLE and 
ARCHIVE_INDEXING_DISABLE in mm_cfg.py

You probably want to run this patch as follows:

cd <mailman 2.0.6 untarred and unzipped directory>
patch -p1 < <this patch file>

See also the associated patch for integrating the 
htdig search software with mailman's internal archiver 
ouput.



----------------------------------------------------------------------

>Comment By: Richard Barrett (ppsys)
Date: 2002-10-30 11:39

Message:
Logged In: YES 
user_id=75166

indexing-2.1b4-0.1.patch is a revised version of the patch 
that is compatible with MM 2.1b4

----------------------------------------------------------------------

Comment By: Richard Barrett (ppsys)
Date: 2002-08-14 16:46

Message:
Logged In: YES 
user_id=75166

indexing-2.1b3-0.1.patch is a revised version of the patch 
that is compatible with MM 2.1b3

----------------------------------------------------------------------

Comment By: Richard Barrett (ppsys)
Date: 2002-08-08 17:32

Message:
Logged In: YES 
user_id=75166

An additional file, README.NOINDEXtags is added to 
indexing-2.0.13-0.3.patch version that discusses the issue of 
what tags to use for controlling various search engine 
indexers.

----------------------------------------------------------------------

Comment By: Richard Barrett (ppsys)
Date: 2002-08-08 17:19

Message:
Logged In: YES 
user_id=75166

An error when building the indexing-2.1b2-0.1.patch meant 
that copies of the originals of two of the files modified by this 
version of the patch were added when the patch was run. 
indexing-2.1b2-0.1.patch removes this error. However, the 
original error is benign and can be corrected by deleting the 
extra files HyperArch.py.orig and Defaults.py.in.orig.

An additional file, README.NOINDEXtags is added that 
discusses the issue of what tags to use for controlling various 
search engine indexers.

----------------------------------------------------------------------

Comment By: Barry A. Warsaw (bwarsaw)
Date: 2002-08-08 14:19

Message:
Logged In: YES 
user_id=12800

Another question: is there no standard (de-facto or
otherwise) for generic markup that tells indexers not to
index a particular section?  IOW, for
ARCHIVE_INDEXING_ENABLE and ARCHIVE_INDEXING_DISABLE, is
there some generic value that would instruct most (all?)
indexers to ignore that section?  Or does it necessarily
have to be indexer specific?  

I'm thinking of the situation where you might have ht://Dig
installed locally, but your archives are still being
spidered by external indexers.  It would be good if
something more generic could be added to Defaults.py.in

----------------------------------------------------------------------

Comment By: Barry A. Warsaw (bwarsaw)
Date: 2002-08-08 14:14

Message:
Logged In: YES 
user_id=12800

Looking at the 2.1b2 patch, why does it try to create
HyperArch.py.orig and Defaults.py.in.orig?  Are those
included in the patch by mistake?

----------------------------------------------------------------------

Comment By: Richard Barrett (ppsys)
Date: 2002-08-05 10:08

Message:
Logged In: YES 
user_id=75166

indexing-2.0.13-0.2.patch just adds a GPL notice to the patch

----------------------------------------------------------------------

Comment By: Richard Barrett (ppsys)
Date: 2002-08-01 16:33

Message:
Logged In: YES 
user_id=75166

indexing-2.1b2-0.1.patch is a revised version of the patch 
that is compatible with MM 2.1b2

----------------------------------------------------------------------

Comment By: Richard Barrett (ppsys)
Date: 2002-07-30 11:23

Message:
Logged In: YES 
user_id=75166

indexing-2.0.13-0.1.patch is purely cosmetic to get no 
mumble application to MM 2.0.13

----------------------------------------------------------------------

Comment By: Richard Barrett (ppsys)
Date: 2002-07-25 14:11

Message:
Logged In: YES 
user_id=75166

indexing-2.0.11-0.1.patch should apply without problems to 
MM 2.0.12

----------------------------------------------------------------------

Comment By: Richard Barrett (ppsys)
Date: 2002-05-23 09:50

Message:
Logged In: YES 
user_id=75166

indexing-2.0.11-0.1.patch is a revised version of the patch 
that is compatible with MM 2.0.11

----------------------------------------------------------------------

Comment By: Richard Barrett (ppsys)
Date: 2002-04-19 10:53

Message:
Logged In: YES 
user_id=75166

indexing-2.0.9-0.1.patch should apply without problems to 
MM 2.0.10

----------------------------------------------------------------------

Comment By: Richard Barrett (ppsys)
Date: 2002-04-08 17:43

Message:
Logged In: YES 
user_id=75166

indexing-2.0.9-0.1.patch is a revised version of the patch 
that is compatible with MM 2.0.9

----------------------------------------------------------------------

Comment By: Richard Barrett (ppsys)
Date: 2002-03-06 16:14

Message:
Logged In: YES 
user_id=75166

indexing-2.1cvs-20020306.patch is a revised version of the  patch that is compatible with the code 
published in mailman CVS on sourceforge as 12:30 GMT 6 Mar 2002.

Corrects problem noted or 5 Mar 2002 by nobody

----------------------------------------------------------------------

Comment By: Nobody/Anonymous (nobody)
Date: 2002-03-05 21:41

Message:
Logged In: NO 

When applying this patch I get an error with Hunk 1 and 
Defaults.py is not updated. This happens with the a clean 
download of the latest cvs installation (5 Mar 2002). Any 
ideas what the problem is?

----------------------------------------------------------------------

Comment By: Richard Barrett (ppsys)
Date: 2001-12-17 16:53

Message:
Logged In: YES 
user_id=75166

indexing-2.1cvs-20011217.patch is a revised version of the 
patch that is compatible with the code published in mailman 
CVS on sourceforge as 11:50 GMT 17 Dec 2001

----------------------------------------------------------------------

Comment By: Richard Barrett (ppsys)
Date: 2001-12-13 16:48

Message:
Logged In: YES 
user_id=75166

indexing-2.1a3-0.1.patch is a revised version of the patch that is compatible with the code published in 
mailman-2.1a3.tgz on sourceforge. 

----------------------------------------------------------------------

Comment By: Richard Barrett (ppsys)
Date: 2001-11-28 11:07

Message:
Logged In: YES 
user_id=75166

This patch should also apply without problems to MM 2.0.8

----------------------------------------------------------------------

Comment By: Richard Barrett (ppsys)
Date: 2001-11-27 12:03

Message:
Logged In: YES 
user_id=75166

This patch should also apply without problems to MM 2.0.7

----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=300103&aid=444879&group_id=103

From noreply@sourceforge.net  Wed Oct 30 11:40:45 2002
From: noreply@sourceforge.net (noreply@sourceforge.net)
Date: Wed, 30 Oct 2002 03:40:45 -0800
Subject: [Mailman-Developers] 
 [ mailman-Patches-444879 ] Archive indexer control to improve index
Message-ID: <E186rDJ-0006Zc-00@usw-sf-web4.sourceforge.net>

Patches item #444879, was opened at 2001-07-26 18:01
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=300103&aid=444879&group_id=103

Category: None
Group: None
Status: Open
Resolution: None
Priority: 7
Submitted By: Richard Barrett (ppsys)
Assigned to: Barry A. Warsaw (bwarsaw)
Summary: Archive indexer control to improve index

Initial Comment:
This patch is applicable to Mailman 2.0.6 release and 
supercedes ealier patches 401669 and 402422.

This patch should improve the quality of search 
results returned by search engines
such as htdig (http://www.htdig.org) where the seach 
engine's index builder responds 
to strings embedded in the html pages that are the 
subject of the indexing. The 
changes in this patch: 

1. allow strings for enabling and disabling indexing 
to be defined in mm_cfg.py.

2. embeds those strings in the pages generated as the 
html version of a list's
   archive.

By default nothing in the html changes. To get the 
desired effect, you must
define ARCHIVE_INDEXING_ENABLE and 
ARCHIVE_INDEXING_DISABLE in mm_cfg.py

You probably want to run this patch as follows:

cd <mailman 2.0.6 untarred and unzipped directory>
patch -p1 < <this patch file>

See also the associated patch for integrating the 
htdig search software with mailman's internal archiver 
ouput.



----------------------------------------------------------------------

>Comment By: Richard Barrett (ppsys)
Date: 2002-10-30 11:40

Message:
Logged In: YES 
user_id=75166

indexing-2.1b4-0.1.patch is a revised version of the patch 
that is compatible with MM 2.1b4

----------------------------------------------------------------------

Comment By: Richard Barrett (ppsys)
Date: 2002-10-30 11:39

Message:
Logged In: YES 
user_id=75166

indexing-2.1b4-0.1.patch is a revised version of the patch 
that is compatible with MM 2.1b4

----------------------------------------------------------------------

Comment By: Richard Barrett (ppsys)
Date: 2002-08-14 16:46

Message:
Logged In: YES 
user_id=75166

indexing-2.1b3-0.1.patch is a revised version of the patch 
that is compatible with MM 2.1b3

----------------------------------------------------------------------

Comment By: Richard Barrett (ppsys)
Date: 2002-08-08 17:32

Message:
Logged In: YES 
user_id=75166

An additional file, README.NOINDEXtags is added to 
indexing-2.0.13-0.3.patch version that discusses the issue of 
what tags to use for controlling various search engine 
indexers.

----------------------------------------------------------------------

Comment By: Richard Barrett (ppsys)
Date: 2002-08-08 17:19

Message:
Logged In: YES 
user_id=75166

An error when building the indexing-2.1b2-0.1.patch meant 
that copies of the originals of two of the files modified by this 
version of the patch were added when the patch was run. 
indexing-2.1b2-0.1.patch removes this error. However, the 
original error is benign and can be corrected by deleting the 
extra files HyperArch.py.orig and Defaults.py.in.orig.

An additional file, README.NOINDEXtags is added that 
discusses the issue of what tags to use for controlling various 
search engine indexers.

----------------------------------------------------------------------

Comment By: Barry A. Warsaw (bwarsaw)
Date: 2002-08-08 14:19

Message:
Logged In: YES 
user_id=12800

Another question: is there no standard (de-facto or
otherwise) for generic markup that tells indexers not to
index a particular section?  IOW, for
ARCHIVE_INDEXING_ENABLE and ARCHIVE_INDEXING_DISABLE, is
there some generic value that would instruct most (all?)
indexers to ignore that section?  Or does it necessarily
have to be indexer specific?  

I'm thinking of the situation where you might have ht://Dig
installed locally, but your archives are still being
spidered by external indexers.  It would be good if
something more generic could be added to Defaults.py.in

----------------------------------------------------------------------

Comment By: Barry A. Warsaw (bwarsaw)
Date: 2002-08-08 14:14

Message:
Logged In: YES 
user_id=12800

Looking at the 2.1b2 patch, why does it try to create
HyperArch.py.orig and Defaults.py.in.orig?  Are those
included in the patch by mistake?

----------------------------------------------------------------------

Comment By: Richard Barrett (ppsys)
Date: 2002-08-05 10:08

Message:
Logged In: YES 
user_id=75166

indexing-2.0.13-0.2.patch just adds a GPL notice to the patch

----------------------------------------------------------------------

Comment By: Richard Barrett (ppsys)
Date: 2002-08-01 16:33

Message:
Logged In: YES 
user_id=75166

indexing-2.1b2-0.1.patch is a revised version of the patch 
that is compatible with MM 2.1b2

----------------------------------------------------------------------

Comment By: Richard Barrett (ppsys)
Date: 2002-07-30 11:23

Message:
Logged In: YES 
user_id=75166

indexing-2.0.13-0.1.patch is purely cosmetic to get no 
mumble application to MM 2.0.13

----------------------------------------------------------------------

Comment By: Richard Barrett (ppsys)
Date: 2002-07-25 14:11

Message:
Logged In: YES 
user_id=75166

indexing-2.0.11-0.1.patch should apply without problems to 
MM 2.0.12

----------------------------------------------------------------------

Comment By: Richard Barrett (ppsys)
Date: 2002-05-23 09:50

Message:
Logged In: YES 
user_id=75166

indexing-2.0.11-0.1.patch is a revised version of the patch 
that is compatible with MM 2.0.11

----------------------------------------------------------------------

Comment By: Richard Barrett (ppsys)
Date: 2002-04-19 10:53

Message:
Logged In: YES 
user_id=75166

indexing-2.0.9-0.1.patch should apply without problems to 
MM 2.0.10

----------------------------------------------------------------------

Comment By: Richard Barrett (ppsys)
Date: 2002-04-08 17:43

Message:
Logged In: YES 
user_id=75166

indexing-2.0.9-0.1.patch is a revised version of the patch 
that is compatible with MM 2.0.9

----------------------------------------------------------------------

Comment By: Richard Barrett (ppsys)
Date: 2002-03-06 16:14

Message:
Logged In: YES 
user_id=75166

indexing-2.1cvs-20020306.patch is a revised version of the  patch that is compatible with the code 
published in mailman CVS on sourceforge as 12:30 GMT 6 Mar 2002.

Corrects problem noted or 5 Mar 2002 by nobody

----------------------------------------------------------------------

Comment By: Nobody/Anonymous (nobody)
Date: 2002-03-05 21:41

Message:
Logged In: NO 

When applying this patch I get an error with Hunk 1 and 
Defaults.py is not updated. This happens with the a clean 
download of the latest cvs installation (5 Mar 2002). Any 
ideas what the problem is?

----------------------------------------------------------------------

Comment By: Richard Barrett (ppsys)
Date: 2001-12-17 16:53

Message:
Logged In: YES 
user_id=75166

indexing-2.1cvs-20011217.patch is a revised version of the 
patch that is compatible with the code published in mailman 
CVS on sourceforge as 11:50 GMT 17 Dec 2001

----------------------------------------------------------------------

Comment By: Richard Barrett (ppsys)
Date: 2001-12-13 16:48

Message:
Logged In: YES 
user_id=75166

indexing-2.1a3-0.1.patch is a revised version of the patch that is compatible with the code published in 
mailman-2.1a3.tgz on sourceforge. 

----------------------------------------------------------------------

Comment By: Richard Barrett (ppsys)
Date: 2001-11-28 11:07

Message:
Logged In: YES 
user_id=75166

This patch should also apply without problems to MM 2.0.8

----------------------------------------------------------------------

Comment By: Richard Barrett (ppsys)
Date: 2001-11-27 12:03

Message:
Logged In: YES 
user_id=75166

This patch should also apply without problems to MM 2.0.7

----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=300103&aid=444879&group_id=103

From noreply@sourceforge.net  Wed Oct 30 11:43:11 2002
From: noreply@sourceforge.net (noreply@sourceforge.net)
Date: Wed, 30 Oct 2002 03:43:11 -0800
Subject: [Mailman-Developers] 
 [ mailman-Patches-444884 ] Integration of Mailman & htdig for archi
Message-ID: <E186rFf-0006bD-00@usw-sf-web4.sourceforge.net>

Patches item #444884, was opened at 2001-07-26 18:27
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=300103&aid=444884&group_id=103

Category: None
Group: None
Status: Open
Resolution: None
Priority: 7
Submitted By: Richard Barrett (ppsys)
Assigned to: Barry A. Warsaw (bwarsaw)
Summary: Integration of Mailman & htdig for archi

Initial Comment:
This patch is applicable to Mailman 2.0.6 release that 
has had search enhancement patch 444879 patch 
installed - if your Defaults.py has the 
ARCHIVE_INDEXING_ENABLE and ARCHIVE_INDEXING_DISABLE 
in it then you've got that patch.

It replaces earlier patches 401670 and 402423 and is 
mainly to correct some problems arising from fixes 
introduced into Mailman by bug fix releases since the 
402423 patch. 

This patch integrates htdig with Mailman and provides:

1. per list search facility with a search form on the 
list's TOC page.

2. maintenance of privacy of private archives which 
requires the user to establish their credentials via 
the normal private archive access before any access 
via htdig is allowed.
   
3. a common base URL for both public and private 
archive access via htsearch results so that htdig 
indices are unaffected by changingan archive from 
private to public and vice versa. All access to 
archives via htdig is controlled by a new wrapped cgi-
bin script called htdig.py.
   
4. a new cron activated script and extra crontab entry 
which runs htdig regularly to maintain the per list 
search indices.
   
5. automatic creation, deletion and maintenance of 
htdig configuration files and such. Beyond installing 
htdig and telling Mailman where it is via mm_cfg you 
do not have to do any other setup. Well not quite you 
do have to set up a single per installation symlink to 
allow htdig to find the automatically generated per 
list htdig configuration files.

You probably want to run this patch as follows:

cd <mailman 2.0.6 untarred and unzipped directory>
patch -p1 < <this patch file>  

----------------------------------------------------------------------

>Comment By: Richard Barrett (ppsys)
Date: 2002-10-30 11:43

Message:
Logged In: YES 
user_id=75166

htdig-2.1b4-0.1.patch is a revised version of the patch that is 
compatible with MM 2.1b4

----------------------------------------------------------------------

Comment By: Richard Barrett (ppsys)
Date: 2002-10-14 11:50

Message:
Logged In: YES 
user_id=75166

htdig-2.1b3-0.3.patch removes use of the file() function, used 
instead of the open() function, in three cron scripts added by 
the patch. Use of the file() function created an unnecessary 
dependency on Python 2.2

----------------------------------------------------------------------

Comment By: Colin Mackinlay (cmackinlay)
Date: 2002-10-12 16:51

Message:
Logged In: YES 
user_id=624179

Got a workaround!

The line referred to in the traceback:
 file(rundig_run_file, 'w').close()
is used to create a 'rundig_last_run' file of lenght 0 bytes
Creating this manually (or copying it) means the line isn't 
called and everything seems to work.

Either file() is not a valid function call or my python is broken - 
I'm not literate enough in python to know the answer though! 

----------------------------------------------------------------------

Comment By: Colin Mackinlay (cmackinlay)
Date: 2002-10-06 14:18

Message:
Logged In: YES 
user_id=624179

Just rebuilt MM as 2.1b3 with htdig.
Upgraded lists which had htdig before work fine
New lists give the obvious error:
  Unable to read word database file
  Did you run htmerge?
Running the cronjob doesn't fix as it used to, message is:
  Output from command /usr/bin/python -
S /usr/local/mailman/cron/nightly_htdig ..

Traceback (most recent call last):
  File "/usr/local/mailman/cron/nightly_htdig", line 153, in ?
    main()
  File "/usr/local/mailman/cron/nightly_htdig", line 118, in main
    file(rundig_run_file, 'w').close()
NameError: global name 'file' is not defined

The archive/htdig folder only contains the xx.conf file, but no 
db.xx files

If I copy in db.xx files from another list then the problem goes 
away (except I've now got an invalid set of references!)

Is this my elementary error or is it more sinister?!


----------------------------------------------------------------------

Comment By: Richard Barrett (ppsys)
Date: 2002-08-15 11:02

Message:
Logged In: YES 
user_id=75166

htdig-2.1b3-0.2.patch corrects a dumb syntax error in htdig-
2.1b3-0.1.patch which will typically show up as logged errors 
in the operation of the ArchRunner qrunner at line 721 of 
HyperArch.py

----------------------------------------------------------------------

Comment By: Richard Barrett (ppsys)
Date: 2002-08-14 16:51

Message:
Logged In: YES 
user_id=75166

htdig-2.1b3-0.1.patch is a revised version of the patch that is 
compatible with MM 2.1b3

----------------------------------------------------------------------

Comment By: Barry A. Warsaw (bwarsaw)
Date: 2002-08-08 16:33

Message:
Logged In: YES 
user_id=12800

I've sent Richard some comments off-line about this patch.

Meta comments: the 2.0.x patches can't be officially
supported, but I'm going to create an unofficial patches
page off the wiki for where the 2.0 patches can be migrated.

I think this patch set is too big for MM2.1, but if it's
cleaned up as per my private message, let's re-evaluate it
for MM2.2 (or 3.0).

----------------------------------------------------------------------

Comment By: Richard Barrett (ppsys)
Date: 2002-08-05 10:11

Message:
Logged In: YES 
user_id=75166

htdig-2.0.13-0.2.patch just adds a GPL notice to the patch


----------------------------------------------------------------------

Comment By: Richard Barrett (ppsys)
Date: 2002-08-01 16:35

Message:
Logged In: YES 
user_id=75166

htdig-2.1b2-0.1.patch is a revised version of the patch 
that is compatible with MM 2.1b2

----------------------------------------------------------------------

Comment By: Richard Barrett (ppsys)
Date: 2002-07-30 11:25

Message:
Logged In: YES 
user_id=75166

htdig-2.0.13-0.1.patch is purely cosmetic to get no mumble 
application to MM 2.0.13

----------------------------------------------------------------------

Comment By: Richard Barrett (ppsys)
Date: 2002-07-25 15:07

Message:
Logged In: YES 
user_id=75166

Do not use htdig-2.0.12-0.1.patch there is an error in it.
Use htdig-2.0.12-0.2.patch instead

----------------------------------------------------------------------

Comment By: Richard Barrett (ppsys)
Date: 2002-07-25 14:10

Message:
Logged In: YES 
user_id=75166

htdig-2.0.12-0.1.patch is a revised version of the patch that 
applies without complaint to MM 2.0.12.

It also add a facility for adding site wide htdig configuration 
attributes to all list specific htdig configuration files. 



----------------------------------------------------------------------

Comment By: Richard Barrett (ppsys)
Date: 2002-05-23 09:59

Message:
Logged In: YES 
user_id=75166

htdig-2.0.11-0.1.patch is a revised version of the patch that 
is compatible with MM 2.0.11

This version removes an incompatibility with Python 2.2 
which caused warning messages to be generated when any 
of the family cron/nightly_htdig scripts were run.

Some guidance on file access permissions for some htdig 
database files needed by rundig have been added to 
installation notes.

----------------------------------------------------------------------

Comment By: Richard Barrett (ppsys)
Date: 2002-04-19 10:59

Message:
Logged In: YES 
user_id=75166

htdig-2.0.10-0.1.patch is a revised version of the patch 
that is compatible with MM 2.0.10

----------------------------------------------------------------------

Comment By: Richard Barrett (ppsys)
Date: 2002-04-08 17:46

Message:
Logged In: YES 
user_id=75166

htdig-2.0.9-0.1.patch is a revised version of the patch 
that is compatible with MM 2.0.9

----------------------------------------------------------------------

Comment By: Richard Barrett (ppsys)
Date: 2002-03-06 16:22

Message:
Logged In: YES 
user_id=75166

htdig-2.1cvs-20020306.patch is a revised version of the patch that is compatible with the code published in 
mailman CVS on sourceforge as 12:30 GMT 6 Mar 2002

Known deficiency is that the non-English versions of files under $build/templates still contain text in English 
and need translations I cannot do. Also the necessary pygettext activity and subsequent translations in 
files under $build/messages remain to be done.

----------------------------------------------------------------------

Comment By: Richard Barrett (ppsys)
Date: 2001-12-17 16:56

Message:
Logged In: YES 
user_id=75166

htdig-2.1cvs-20011217.patch is a revised version of the 
patch that is compatible with the code published in mailman 
CVS on sourceforge as 11:50 GMT 17 Dec 2001

The only known deficiency is that the non-English versions 
of files under $build/templates still contain text in 
English and need translations I cannot do. Also the 
necessary pygettext activity and subsequent translations in 
files under $build/messages remain to be done.

----------------------------------------------------------------------

Comment By: Richard Barrett (ppsys)
Date: 2001-12-13 16:58

Message:
Logged In: YES 
user_id=75166

htdig-2.1a3-0.1.patch is a revised version of the patch that is compatible with the code published in 
mailman-2.1a3.tgz on sourceforge.

The only known deficiency is that the non-English versions of files under $build/templates still contain text 
in English and need translations I cannot do. Also the necessary pygettext activity and subsequent 
translations in files under $build/messages remain to be done. 

----------------------------------------------------------------------

Comment By: Richard Barrett (ppsys)
Date: 2001-11-28 17:33

Message:
Logged In: YES 
user_id=75166

The htdig-2.0.8-0.1.patch version of the patch resolves a problem that can arise with htdig indexing if the 
web_page_url for a list uses other than the http addressing (some folks want to use https). While specified 
as for MM 2.0.8 the revised patch should work OK with 2.0.7, 2.0.6 and probably back as far as 2.0.3. If 
you do not have the requirement for using other than http addressing in you lists web_page_urls it probably 
isn't worth the trouble of upgrading to this patch level.

----------------------------------------------------------------------

Comment By: Richard Barrett (ppsys)
Date: 2001-11-28 11:08

Message:
Logged In: YES 
user_id=75166

This patch should also apply without problems to MM 2.0.8

----------------------------------------------------------------------

Comment By: Richard Barrett (ppsys)
Date: 2001-11-27 12:00

Message:
Logged In: YES 
user_id=75166

This patch should also apply without problems to Mm 2.0.7

----------------------------------------------------------------------

Comment By: Richard Barrett (ppsys)
Date: 2001-11-09 11:54

Message:
Logged In: YES 
user_id=75166

The htdig-2.0.6-03.patch version of the patch makes some 
previously hard-coded things configurable and enhances the 
capability to run the htdig searches and indexing on a 
different machine to the one delivering Mailman and 
Mailman's web UI.

----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=300103&aid=444884&group_id=103

From danny@terweij.nl  Wed Oct 30 14:32:07 2002
From: danny@terweij.nl (Danny Terweij)
Date: Wed, 30 Oct 2002 15:32:07 +0100
Subject: [Mailman-Developers] strange things
Message-ID: <001a01c28021$201ea2f0$1e00a8c0@onsnet.org>

Still the bounce messages are not solved in latest cvs version.

Oct 30 13:40:59 2002 (18497) forwarding unrecognized, message-id:
<0a72d5740121ea2NET014S@hetnet.nl>
Oct 30 13:43:34 2002 (18497) foreverfriends: miss.aqua@uku.co.uk already
scored a bounce for today
Oct 30 14:31:37 2002 (18497) foreverfriends: miss.aqua@uku.co.uk already
scored a bounce for today
Oct 30 14:33:56 2002 (18497) bounce message with non-members of
psp-club-plaatjes-enzo: tm485569@mint.xo.nl
Oct 30 14:33:57 2002 (18497) forwarding unrecognized, message-id:
<20021030133219.E8C5754CA@mint.nl.gxn.net>
Oct 30 14:36:08 2002 (18497) bounce message with non-members of
psp-club-plaatjes-enzo: tm485569@mint.xo.nl
Oct 30 14:36:09 2002 (18497) forwarding unrecognized, message-id:
<20021030133439.2FFD1531B@mint.nl.gxn.net>
Oct 30 14:36:11 2002 (18497) bounce message with non-members of
psp-club-plaatjes-enzo: tm485569@mint.xo.nl
Oct 30 14:36:11 2002 (18497) forwarding unrecognized, message-id:
<20021030133446.6F3A553FD@mint.nl.gxn.net>
Oct 30 14:37:42 2002 (18497) bounce message with non-members of psp-club:
jahens@hetnet.nl

And i have al lot of :

Oct 30 15:26:34 2002 (1604) psp-club: couldn't get list lock

And many of these :

Oct 30 12:38:27 2002 (18499) Traceback (most recent call last):
  File "/usr/local/mailman/Mailman/Queue/Runner.py", line 105, in _oneloop
    self._onefile(msg, msgdata)
  File "/usr/local/mailman/Mailman/Queue/Runner.py", line 154, in _onefile
    keepqueued = self._dispose(mlist, msg, msgdata)
  File "/usr/local/mailman/Mailman/Queue/IncomingRunner.py", line 129, in
_dispose
    status = self._dopipeline(mlist, msg, msgdata, pipeline)
  File "/usr/local/mailman/Mailman/Queue/IncomingRunner.py", line 152, in
_dopipeline
    sys.modules[modname].process(mlist, msg, msgdata)
  File "/usr/local/mailman/Mailman/Handlers/CookHeaders.py", line 57, in
process
    prefix_subject(mlist, msg, msgdata)
  File "/usr/local/mailman/Mailman/Handlers/CookHeaders.py", line 235, in
prefix_subject
    s = unicode(s, codec, 'replace')
LookupError: unknown encoding

Oct 30 14:44:56 2002 (18496) Uncaught runner exception: <LockFile 137086212:
/usr/local/mailman/locks/foreverfri
ends.lock [unlocked: 18000sec] pid=18496>: None
Oct 30 14:45:00 2002 (18496) Traceback (most recent call last):
  File "/usr/local/mailman/Mailman/Queue/Runner.py", line 105, in _oneloop
    self._onefile(msg, msgdata)
  File "/usr/local/mailman/Mailman/Queue/Runner.py", line 154, in _onefile
    keepqueued = self._dispose(mlist, msg, msgdata)
  File "/usr/local/mailman/Mailman/Queue/ArchRunner.py", line 74, in
_dispose
    mlist.Save()
  File "/usr/local/mailman/Mailman/MailList.py", line 508, in Save
    self.__lock.refresh()
  File "/usr/local/mailman/Mailman/LockFile.py", line 223, in refresh
    raise NotLockedError, '%s: %s' % (repr(self), self.__read())
NotLockedError: <LockFile 137086212:
/usr/local/mailman/locks/foreverfriends.lock [unlocked: 18000sec] pid=18496
>: None

Take a look at it please, my users are getting angry :) ( i do not mind they
know it is a beta version hehe)

Danny Terweij




From danny@terweij.nl  Wed Oct 30 16:05:01 2002
From: danny@terweij.nl (Danny Terweij)
Date: Wed, 30 Oct 2002 17:05:01 +0100
Subject: [Mailman-Developers] No archive updates b4+
Message-ID: <00b101c2802e$1a5a1f40$1e00a8c0@onsnet.org>

Hi,

b4+ does not update the archives with new messages (pipermail).

My /qfiles/archives directory is filled with a lot of messages.

Looks like that the Archrunner is not working.

Danny Terweij



From Daniel.Buchmann@bibsys.no  Wed Oct 30 16:21:34 2002
From: Daniel.Buchmann@bibsys.no (Daniel Buchmann)
Date: 30 Oct 2002 17:21:34 +0100
Subject: [Mailman-Developers] No archive updates b4+
Message-ID: <1035994895.2312.62.camel@fornax.bibsys.no>


---------------------- multipart/signed attachment
It can't be a general b4+ problem, because my 2.1b4+ installation
doesn't have this problem, neither does python.org...
(look at the mailman-developers archive.. your latest messages are in
it... ;)

Danny wrote:

> Hi,
>
> b4+ does not update the archives with new messages (pipermail).
>
> My /qfiles/archives directory is filled with a lot of messages.
>
> Looks like that the Archrunner is not working.
>
> Danny Terweij



-Daniel, norwegian translator

---------------------- multipart/signed attachment
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 232 bytes
Desc: This is a digitally signed message part
Url : http://mail.python.org/pipermail/mailman-developers/attachments/20021030/a6554d4c/attachment.bin

---------------------- multipart/signed attachment--



From danny@terweij.nl  Wed Oct 30 16:36:48 2002
From: danny@terweij.nl (Danny Terweij)
Date: Wed, 30 Oct 2002 17:36:48 +0100
Subject: [Mailman-Developers] No archive updates b4+
References: <1035994895.2312.62.camel@fornax.bibsys.no>
Message-ID: <00d701c28032$8b2e9080$1e00a8c0@onsnet.org>


>From: "Daniel Buchmann" <Daniel.Buchmann@bibsys.no>

>It can't be a general b4+ problem, because my 2.1b4+ installation
>doesn't have this problem, neither does python.org...
>(look at the mailman-developers archive.. your latest messages are in
>it... ;)
>Danny wrote:
> Hi,
> b4+ does not update the archives with new messages (pipermail).

Well, it is exacly the date/time when i did the upgrade from the b4 to b4+.
Just the messages sits there and nothing happens. Even no loggings.
With a restart of mailmanctl is still hanging arround in the qfiles/archives

I must say, beta 2 did work better than this latest cvs.
I have many problems now.
One problem i temp. solved is by setting all mailing lists to not check any
bounces but just discard it. This is not what i want but my list owners are
much happier now without the "Bounce mailbomb".

Danny.




From barry@python.org  Wed Oct 30 16:41:07 2002
From: barry@python.org (Barry A. Warsaw)
Date: Wed, 30 Oct 2002 11:41:07 -0500
Subject: [Mailman-Developers] No archive updates b4+
References: <00b101c2802e$1a5a1f40$1e00a8c0@onsnet.org>
Message-ID: <15808.2979.712884.645422@gargle.gargle.HOWL>


>>>>> "DT" == Danny Terweij <danny@terweij.nl> writes:

    DT> b4+ does not update the archives with new messages
    DT> (pipermail).

    DT> My /qfiles/archives directory is filled with a lot of
    DT> messages.

    DT> Looks like that the Archrunner is not working.

I thought there was code there to not pickle and unpickle the lock
object when the archiver unpickles the MailList.  Maybe that wasn't
the real bug. :/

You could always blow away your archives and regen from scratch.  I
believe that will fix the problem.

-Barry

From martin.whinnery@sbc.ac.uk  Tue Oct 29 13:04:11 2002
From: martin.whinnery@sbc.ac.uk (Martin Whinnery)
Date: Tue, 29 Oct 2002 13:04:11 -0000
Subject: [Mailman-Developers] Help programming MemberAdaptor...
Message-ID: <002d01c27f4b$b2b02c00$c805010a@MARTLAPIBM>

Hello guru folks.

Er, I've been trying to make Mailman populate lists from our LDAP
directory.

I've never used python before, and so I'm probably doing stuff ass fust.
Here's what I've been doing:

I've copied OldStyleMemberships.py to LDAPMemberships.py

In MailList.pl I've changed the import line to:

from Mailman.LDAPMemberships import OldStyleMemberships

And I'm hacking getRegularMemberKeys to add stuff to self.__mlist before
getRegularMemberKeys returns.

A few questions:

Is this how I ought to be going? If not, how ought I to approach this?

What I'd really like to do is have lists EITHER populate in the normal
file way OR populated from LDAP. I was thinking I might use an address
like

'ldap@cn=groupname,ou=groups,dc=example,dc=com' as the initial and only
subscriber address, and use this to flag LDAP status.

But I'm sure that's really crap.


What ought I to do? Will somebody cleverer than me do the work so I
don't have to?

HELP!!!

Martin Whinnery
Assistant Network Manager
South Birmingham College


From miller_brett@bah.com  Tue Oct 29 17:50:28 2002
From: miller_brett@bah.com (Miller Brett)
Date: Tue, 29 Oct 2002 12:50:28 -0500
Subject: [Mailman-Developers] Problem with checkdbs script in Beta 4
Message-ID: <02c201c27f73$ae5a5f40$3f80509c@nitrox>

Hello,

I just upgraded to Mailman beta 4 from beta 3 and I am receiving this
error when a mailman cron job runs. Has anyone seen this error and does
anyone know how to fix it. This checkdbs script seemed to work fine in
beta 3 but not beta 4. Could someone tell me what this function does?
Thank you for your time.


Traceback (most recent call last):
  File "/usr/local/mailman/cron/checkdbs", line 26, in ?
    from email.Charset import Charset
ImportError: No module named email.Charset


Brett Miller 
Booz | Allen | Hamilton




From rge1@egwpartners.com  Wed Oct 30 16:55:42 2002
From: rge1@egwpartners.com (Rick Ennis)
Date: Wed, 30 Oct 2002 11:55:42 -0500
Subject: [Mailman-Developers] Can't Approve/Reject/Discard from Admin Web
	Interface
Message-ID: <019301c28035$2e80f320$0202a8c0@anakin>

Hi...  I have a quick question.

I just started playing with 2.1 yesterday (snagged 2.1b4+ from cvs) and have
been searching the mailing lists all morning.  What I can't seem to figure
out is...  should the approve/reject/discard interface on the admin screen
be working (in this beta)?   I click one of the radio buttons followed by
"submit all data" but it just reloads the same page.  Logging out and back
in shows no change in the db state.  None of my files in the mailman/logs
directory have been updated, nor my /var/log/messages file.  Is this not
implemented yet or did I just do something brain-dead during the
installation?

I should mention that I can successfully send mail to lists, get
notifications for moderator approval, and passed the bin/check_perms script.
So I'm a little stuck.  Any suggestions?

Thanks.
--
Rick Ennis




From martin.whinnery@sbc.ac.uk  Wed Oct 30 16:58:42 2002
From: martin.whinnery@sbc.ac.uk (Martin Whinnery)
Date: Wed, 30 Oct 2002 16:58:42 -0000
Subject: [Mailman-Developers] More extend.py questions
Message-ID: <005c01c28035$99df5e90$c805010a@MARTLAPIBM>

Folks,

Please indulge a python virgin. Barry has advised someone to 'overload
the Load & Save function' in extend.py . Would someone help me with a
little sample? It doesn't need to do anything useful, just replace the
original function.

Martin Whinnery
Assistant Network Manager
South Birmingham College


From danny@terweij.nl  Wed Oct 30 17:58:35 2002
From: danny@terweij.nl (Danny Terweij)
Date: Wed, 30 Oct 2002 18:58:35 +0100
Subject: [Mailman-Developers] No archive updates b4+
References: <00b101c2802e$1a5a1f40$1e00a8c0@onsnet.org>
	<15808.2979.712884.645422@gargle.gargle.HOWL>
Message-ID: <014c01c2803d$f7c728a0$1e00a8c0@onsnet.org>

From: "Barry A. Warsaw" <barry@python.org>

>     DT> My /qfiles/archives directory is filled with a lot of
>     DT> messages.
> 
>     DT> Looks like that the Archrunner is not working.
 
> I thought there was code there to not pickle and unpickle the lock
> object when the archiver unpickles the MailList.  Maybe that wasn't
> the real bug. :/
 
> You could always blow away your archives and regen from scratch.  I
> believe that will fix the problem.

What you mean? run /bin/arch listname?
or delete the mbox and the whole archive?

Danny.




From barry@python.org  Wed Oct 30 17:59:45 2002
From: barry@python.org (Barry A. Warsaw)
Date: Wed, 30 Oct 2002 12:59:45 -0500
Subject: [Mailman-Developers] No archive updates b4+
References: <00b101c2802e$1a5a1f40$1e00a8c0@onsnet.org>
	<15808.2979.712884.645422@gargle.gargle.HOWL>
	<014c01c2803d$f7c728a0$1e00a8c0@onsnet.org>
Message-ID: <15808.7697.146295.964412@gargle.gargle.HOWL>


>>>>> "DT" == Danny Terweij <danny@terweij.nl> writes:

    DT> What you mean? run /bin/arch listname?
    DT> or delete the mbox and the whole archive?

Before you do anything, can you tar up
$prefix/archives/private/listname/ and send it to me?

-Barry




From gward@python.net  Wed Oct 30 18:05:52 2002
From: gward@python.net (Greg Ward)
Date: Wed, 30 Oct 2002 13:05:52 -0500
Subject: [Mailman-Developers] Help programming MemberAdaptor...
In-Reply-To: <002d01c27f4b$b2b02c00$c805010a@MARTLAPIBM>
References: <002d01c27f4b$b2b02c00$c805010a@MARTLAPIBM>
Message-ID: <20021030180552.GA5220@cthulhu.gerg.ca>

On 29 October 2002, Martin Whinnery said:
> In MailList.pl I've changed the import line to:
> 
> from Mailman.LDAPMemberships import OldStyleMemberships

Hacking on the Mailman modules is a bad idea -- next time you upgrade,
your change will be lost.

> Is this how I ought to be going? If not, how ought I to approach this?
> 
> What I'd really like to do is have lists EITHER populate in the normal
> file way OR populated from LDAP. I was thinking I might use an address
> like

Err, define "populate".  Is this a one-time thing?  Or do you want the
list membership to always come from an LDAP lookup?

If the former, I would just write a script that does the LDAP query and
sets the list population from it.  Use bin/add_members as an example.

If the latter, why use Mailman at all?  Why not just use an MTA that can
lookup aliases using LDAP, and then send mail to that alias?  I'm pretty
sure both Exim and postfix can do that.

Note that I am *completely* ignorant of LDAP, not very clueful about
Mailman internals.  I'm just trying to help you ask the right
question(s).

        Greg
-- 
Greg Ward <gward@python.net>                         http://www.gerg.ca/
Monday is an awful way to spend one seventh of your life.

From gward@python.net  Wed Oct 30 18:08:37 2002
From: gward@python.net (Greg Ward)
Date: Wed, 30 Oct 2002 13:08:37 -0500
Subject: [Mailman-Developers] Problem with checkdbs script in Beta 4
In-Reply-To: <02c201c27f73$ae5a5f40$3f80509c@nitrox>
References: <02c201c27f73$ae5a5f40$3f80509c@nitrox>
Message-ID: <20021030180837.GB5220@cthulhu.gerg.ca>

On 29 October 2002, Miller Brett said:
> I just upgraded to Mailman beta 4 from beta 3 and I am receiving this
> error when a mailman cron job runs. Has anyone seen this error and does
> anyone know how to fix it. This checkdbs script seemed to work fine in
> beta 3 but not beta 4.

Silly bug in 2.1b4, fixed in CVS.  You could either upgrade to the CVS
version, or apply this patch (which I just pulled out of CVS):

--- checkdbs	25 Oct 2002 19:06:33 -0000	2.14
+++ checkdbs	28 Oct 2002 03:43:42 -0000	2.17
@@ -6,14 +6,14 @@
 # modify it under the terms of the GNU General Public License
 # as published by the Free Software Foundation; either version 2
 # of the License, or (at your option) any later version.
-# 
+#
 # This program is distributed in the hope that it will be useful,
 # but WITHOUT ANY WARRANTY; without even the implied warranty of
 # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
 # GNU General Public License for more details.
-# 
+#
 # You should have received a copy of the GNU General Public License
-# along with this program; if not, write to the Free Software 
+# along with this program; if not, write to the Free Software
 # Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
 
 """Invoked by cron, this checks for pending moderation requests and mails the
@@ -23,12 +23,12 @@
 import sys
 import time
 from types import UnicodeType
-from email.Charset import Charset
 
 import paths
-# mm_cfg must be imported before the other modules, due to the side-effect of
-# it hacking sys.paths to include site-packages.  Without this, running this
-# script from cron with python -S will fail.
+
+# Import this after paths so we get Mailman's copy of the email package
+from email.Charset import Charset
+
 from Mailman import mm_cfg
 from Mailman import Utils
 from Mailman import MailList
@@ -52,7 +52,7 @@
 def main():
     for name in Utils.list_names():
         # the list must be locked in order to open the requests database
-	mlist = MailList.MailList(name)
+        mlist = MailList.MailList(name)
         try:
             count = mlist.NumRequestsPending()
             # While we're at it, let's evict yesterday's autoresponse data
@@ -87,8 +87,8 @@
                 msg.send(mlist, **{'tomoderators': 1})
         finally:
             mlist.Unlock()
-            
-            
+
+
 
 
 def pending_requests(mlist):
@@ -111,8 +111,10 @@
         info = mlist.GetRecord(id)
         when, sender, subject, reason, text, msgdata = mlist.GetRecord(id)
         date = time.ctime(when)
-        pending.append(_(
-            '     From: %(sender)s on %(date)s\n    Cause: %(reason)s'))
+        pending.append(_("""\
+From: %(sender)s on %(date)s
+Subject: %(subject)s
+Cause: %(reason)s"""))
         pending.append('')
     # Make sure that the text we return from here can be encoded to a byte
     # string in the charset of the list's language.  This could fail if for


-- 
Greg Ward <gward@python.net>                         http://www.gerg.ca/
Shape shifting reptilians are about to achieve complete control over
this planet!

From gward@python.net  Wed Oct 30 18:15:41 2002
From: gward@python.net (Greg Ward)
Date: Wed, 30 Oct 2002 13:15:41 -0500
Subject: [Mailman-Developers] Can't Approve/Reject/Discard from Admin Web
	Interface
In-Reply-To: <019301c28035$2e80f320$0202a8c0@anakin>
References: <019301c28035$2e80f320$0202a8c0@anakin>
Message-ID: <20021030181541.GC5220@cthulhu.gerg.ca>

On 30 October 2002, Rick Ennis said:
> Hi...  I have a quick question.

Sorry, I don't have any quick answers -- just more questions.

> I just started playing with 2.1 yesterday (snagged 2.1b4+ from cvs) and have
> been searching the mailing lists all morning.  What I can't seem to figure
> out is...  should the approve/reject/discard interface on the admin screen
> be working (in this beta)?   I click one of the radio buttons followed by
> "submit all data" but it just reloads the same page.  Logging out and back
> in shows no change in the db state.

Which web browser, and on which OS?  Which web server, and on which OS?
What does your web server log say?  Anything weird like an HTTP proxy
server in the way?  Have you tried running a packet tracer (I like
ethereal myself) to the traffic between web browser and server to make
sure it looks sane?

> None of my files in the mailman/logs
> directory have been updated, nor my /var/log/messages file.

AFAIK Mailman does not use syslog, so /var/log/* should be irrelevant.

Do look in ~mailman/data though -- that's where it stores pending
messages.  When you approve/reject/discard a message, a file should
disappear from that directory (I think).

> Is this not
> implemented yet or did I just do something brain-dead during the
> installation?

        Greg
-- 
Greg Ward <gward@python.net>                         http://www.gerg.ca/
Question authority!

From gward@python.net  Wed Oct 30 18:16:56 2002
From: gward@python.net (Greg Ward)
Date: Wed, 30 Oct 2002 13:16:56 -0500
Subject: [Mailman-Developers] More extend.py questions
In-Reply-To: <005c01c28035$99df5e90$c805010a@MARTLAPIBM>
References: <005c01c28035$99df5e90$c805010a@MARTLAPIBM>
Message-ID: <20021030181656.GD5220@cthulhu.gerg.ca>

On 30 October 2002, Martin Whinnery said:
> Please indulge a python virgin. Barry has advised someone to 'overload
> the Load & Save function' in extend.py . Would someone help me with a
> little sample? It doesn't need to do anything useful, just replace the
> original function.

Huh?  There's no file called extend.py in the Mailman source tree.  Just
what are you trying to accomplish?

        Greg
-- 
Greg Ward <gward@python.net>                         http://www.gerg.ca/
Question authority!

From noreply@sourceforge.net  Wed Oct 30 19:02:21 2002
From: noreply@sourceforge.net (noreply@sourceforge.net)
Date: Wed, 30 Oct 2002 11:02:21 -0800
Subject: [Mailman-Developers] [ mailman-Bugs-631186 ] Postfix msg.type
Message-ID: <E186y6f-0008Ft-00@usw-sf-web4.sourceforge.net>

Bugs item #631186, was opened at 2002-10-30 14:02
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=100103&aid=631186&group_id=103

Category: bounce detection
Group: 2.0.x
Status: Open
Resolution: None
Priority: 5
Submitted By: Gregory Karpinsky, TIV.NET Inc. (tivnet)
Assigned to: Nobody/Anonymous (nobody)
Summary: Postfix msg.type

Initial Comment:
In Postfix.py, I replaced

'multipart/mixed'

with

'multipart/report'

-- and now the bouncing works for me.

I have:
postfix-20011210_1
mailman-2.0.13

gregory
at
tiv.net

----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=100103&aid=631186&group_id=103

From noreply@sourceforge.net  Wed Oct 30 19:28:59 2002
From: noreply@sourceforge.net (noreply@sourceforge.net)
Date: Wed, 30 Oct 2002 11:28:59 -0800
Subject: [Mailman-Developers] 
 [ mailman-Bugs-631200 ] qrunner in loop, keeps eating memory
Message-ID: <E186yWR-0003uU-00@usw-sf-web1.sourceforge.net>

Bugs item #631200, was opened at 2002-10-30 17:28
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=100103&aid=631200&group_id=103

Category: command line scripts
Group: 2.0.x
Status: Open
Resolution: None
Priority: 5
Submitted By: Andreas Hasenack (pandora)
Assigned to: Nobody/Anonymous (nobody)
Summary: qrunner in loop, keeps eating memory

Initial Comment:
mailman-2.0.13
python-2.2.1

Every now and then qrunner goes nuts and starts eating
up all available memory on the system. I'm attaching a
.msg and .db file which seems to reproduce the problem
at all times, and also a log trace which showed up in
the error log when I pressed CTRL-C. This last log
shows where qrunner was looping, and it's in
pickle.loads (log below).

Oct 30 16:06:47 2002 qrunner(25698): Traceback (most
recent call last):
Oct 30 16:06:47 2002 qrunner(25698):   File
"/usr/lib/mailman/Mailman/Archiver/Archiver.py", line
222, in ArchiveMail
Oct 30 16:06:47 2002 qrunner(25698):     h.close()
Oct 30 16:06:47 2002 qrunner(25698):   File
"/usr/lib/mailman/Mailman/Archiver/pipermail.py", line
304, in close
Oct 30 16:06:47 2002 qrunner(25698):    
self.update_dirty_archives()
Oct 30 16:06:47 2002 qrunner(25698):   File
"/usr/lib/mailman/Mailman/Archiver/HyperArch.py", line
935, in update_dirty
_archives
Oct 30 16:06:47 2002 qrunner(25698):    
self.__super_update_dirty_archives()
Oct 30 16:06:47 2002 qrunner(25698):   File
"/usr/lib/mailman/Mailman/Archiver/pipermail.py", line
513, in update_dirty
_archives
Oct 30 16:06:47 2002 qrunner(25698):    
self.update_archive(i)
Oct 30 16:06:47 2002 qrunner(25698):   File
"/usr/lib/mailman/Mailman/Archiver/HyperArch.py", line
938, in update_archi
ve
Oct 30 16:06:47 2002 qrunner(25698):    
self.__super_update_archive(archive)
Oct 30 16:06:47 2002 qrunner(25698):   File
"/usr/lib/mailman/Mailman/Archiver/pipermail.py", line
420, in update_archi
ve
Oct 30 16:06:47 2002 qrunner(25698):    
self._update_simple_index(hdr, archive, arcdir)
Oct 30 16:06:47 2002 qrunner(25698):   File
"/usr/lib/mailman/Mailman/Archiver/pipermail.py", line
437, in _update_simp
le_index
Oct 30 16:06:47 2002 qrunner(25698):     article =
self.database.getArticle(self.archive, msgid)
Oct 30 16:06:47 2002 qrunner(25698):   File
"/usr/lib/mailman/Mailman/Archiver/HyperDatabase.py",
line 281, in getArtic
le
Oct 30 16:06:47 2002 qrunner(25698):     article =
self.__cache[msgid] = pickle.loads(buf)
Oct 30 16:06:47 2002 qrunner(25698):   File "<string>",
line 0, in ?
Oct 30 16:06:47 2002 qrunner(25698): KeyboardInterrupt
Oct 30 16:06:47 2002 (25698) CORRUPT ARCHIVE FOR LIST:
commits

I don't know if the CORRUPT ARCHIVE message is due to
me pressing CTRL-C or if there is actually something
wrong with it.

Any help would be greatly appreciated. Thanks.

----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=100103&aid=631200&group_id=103

From jon@csh.rit.edu  Thu Oct 31 05:03:16 2002
From: jon@csh.rit.edu (Jon Parise)
Date: Thu, 31 Oct 2002 00:03:16 -0500
Subject: [Mailman-Developers] Recent cmd_confirm.py error
Message-ID: <20021031050315.GB2877@csh.rit.edu>

I've just noticed two instances of the following in my errors log:

Oct 30 10:32:09 2002 (27335) Uncaught runner exception: 'NoneType' object has no attribute 'lstrip'
Oct 30 10:32:09 2002 (27335) Traceback (most recent call last):
  File "/usr/local/mailman/Mailman/Queue/Runner.py", line 105, in _oneloop
    self._onefile(msg, msgdata)
  File "/usr/local/mailman/Mailman/Queue/Runner.py", line 154, in _onefile
    keepqueued = self._dispose(mlist, msg, msgdata)
  File "/usr/local/mailman/Mailman/Queue/CommandRunner.py", line 199, in _dispose
    res.process()
  File "/usr/local/mailman/Mailman/Queue/CommandRunner.py", line 93, in process
    stop = self.do_command(cmd, args)
  File "/usr/local/mailman/Mailman/Queue/CommandRunner.py", line 119, in do_command
    return handler.process(self, args)
  File "/usr/local/mailman/Mailman/Commands/cmd_confirm.py", line 72, in process
    if line.lstrip() == match:
AttributeError: 'NoneType' object has no attribute 'lstrip'

This is with an up-to-date installation of the Mailman 2.1b4+.  I'm
not yet familiar with this part of the Mailman code, so I don't have a
fix.

I suppose the simple workaround would involve checking 'line' against
None, but I'm not sure if the fact that None has trickled down into
res.commands is a problem unto itself (and I don't want to simply mask
the problem).

-- 
Jon Parise (jon@csh.rit.edu)  ::  http://www.csh.rit.edu/~jon/

From noreply@sourceforge.net  Thu Oct 31 05:53:44 2002
From: noreply@sourceforge.net (noreply@sourceforge.net)
Date: Wed, 30 Oct 2002 21:53:44 -0800
Subject: [Mailman-Developers] 
 [ mailman-Patches-601117 ] add sequencial number in subject prefix
Message-ID: <E1878H2-0003Nk-00@usw-sf-web2.sourceforge.net>

Patches item #601117, was opened at 2002-08-28 05:07
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=300103&aid=601117&group_id=103

Category: mail delivery
Group: Mailman 2.1
Status: Open
Resolution: None
Priority: 5
Submitted By: Tokio Kikuchi (tkikuchi)
Assigned to: Nobody/Anonymous (nobody)
Summary: add sequencial number in subject prefix

Initial Comment:
This patch for 'CookHeaders.py' add an ability to add a
sequencial number in the subject prefix. You can define
a subject prefix like:
   [listname %d]
Then, the subject line of delivered mail becomes:
   Subject: [listname 123] Hoge hoge
When someone replied this mail, mailman receives a
messages with:
   Subject: Re: [listname 123] Hoge hoge
Then, this patch removes [listname \d+] part and
deliver it with:
   Subject: [listname 124] Re: Hoge hoge
And next, another person replies with
   Subject: Re: [listname 124] Re: Hoge hoge
Then, (magically!) you get:
   Subject: [listname 125] Re: Hoge hoge
Not with Re: Re: Hoge hoge.

Looks like complicated but this patch has been working
well with Japanese-enhanced Mailman for more than a year.

Without %d, this patch works like current version, I
believe.



----------------------------------------------------------------------

>Comment By: Tokio Kikuchi (tkikuchi)
Date: 2002-10-31 05:53

Message:
Logged In: YES 
user_id=67709

Patch ID 625482 (i18n List-Id) and this was merged for 2.1b4+



----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=300103&aid=601117&group_id=103

From noreply@sourceforge.net  Thu Oct 31 05:59:29 2002
From: noreply@sourceforge.net (noreply@sourceforge.net)
Date: Wed, 30 Oct 2002 21:59:29 -0800
Subject: [Mailman-Developers] 
 [ mailman-Patches-601117 ] add sequencial number in subject prefix
Message-ID: <E1878Mb-0003TJ-00@usw-sf-web2.sourceforge.net>

Patches item #601117, was opened at 2002-08-28 05:07
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=300103&aid=601117&group_id=103

Category: mail delivery
Group: Mailman 2.1
Status: Open
Resolution: None
Priority: 5
Submitted By: Tokio Kikuchi (tkikuchi)
Assigned to: Nobody/Anonymous (nobody)
Summary: add sequencial number in subject prefix

Initial Comment:
This patch for 'CookHeaders.py' add an ability to add a
sequencial number in the subject prefix. You can define
a subject prefix like:
   [listname %d]
Then, the subject line of delivered mail becomes:
   Subject: [listname 123] Hoge hoge
When someone replied this mail, mailman receives a
messages with:
   Subject: Re: [listname 123] Hoge hoge
Then, this patch removes [listname \d+] part and
deliver it with:
   Subject: [listname 124] Re: Hoge hoge
And next, another person replies with
   Subject: Re: [listname 124] Re: Hoge hoge
Then, (magically!) you get:
   Subject: [listname 125] Re: Hoge hoge
Not with Re: Re: Hoge hoge.

Looks like complicated but this patch has been working
well with Japanese-enhanced Mailman for more than a year.

Without %d, this patch works like current version, I
believe.



----------------------------------------------------------------------

>Comment By: Tokio Kikuchi (tkikuchi)
Date: 2002-10-31 05:59

Message:
Logged In: YES 
user_id=67709

I have uploaded the patch with the same name as the old one.
Please download the upper one because it's newer.
Sorry for the inconvenience.


----------------------------------------------------------------------

Comment By: Tokio Kikuchi (tkikuchi)
Date: 2002-10-31 05:53

Message:
Logged In: YES 
user_id=67709

Patch ID 625482 (i18n List-Id) and this was merged for 2.1b4+



----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=300103&aid=601117&group_id=103

From chuqui@plaidworks.com  Thu Oct 31 06:06:00 2002
From: chuqui@plaidworks.com (Chuq Von Rospach)
Date: Wed, 30 Oct 2002 22:06:00 -0800
Subject: [Mailman-Developers] updated to CVS and...
Message-ID: <D4340328-EC96-11D6-9FAC-0003934516A8@plaidworks.com>


when I try to start, I get:

plaidworks.com 239# bin/mailmanctl start
Traceback (most recent call last):
   File "bin/mailmanctl", line 120, in ?
     LogStdErr('error', 'mailmanctl', manual_reprime=0, tee_to_stdout=1)
TypeError: LogStdErr() got an unexpected keyword argument 
'tee_to_stdout'


I'm off looking, but there seems to be a glitch.


-- 
Chuq Von Rospach, Architech
chuqui@plaidworks.com -- http://www.plaidworks.com/chuqui/blog/

Yes, I am an agent of Satan, but my duties
are largely ceremonial.


From chuqui@plaidworks.com  Thu Oct 31 06:15:22 2002
From: chuqui@plaidworks.com (Chuq Von Rospach)
Date: Wed, 30 Oct 2002 22:15:22 -0800
Subject: [Mailman-Developers] CVS glitch.
Message-ID: <2373BAE0-EC98-11D6-9FAC-0003934516A8@plaidworks.com>


Hmm. Barry, in current CVS, mailmanctl and qrunner both call LogStdErr 
with a parameter of tee_to_stdout. but in the library they include, 
that parameter is tee_to_real_stderr. Something's a bit out of sync in 
CVS

so does cron/gate-news

everything else seems okay, just those three files.


-- 
Chuq Von Rospach, Architech
chuqui@plaidworks.com -- http://www.plaidworks.com/chuqui/blog/

The Cliff's Notes Cliff's Notes on Hamlet:
     And they all died happily ever after


From barry@python.org  Thu Oct 31 06:50:29 2002
From: barry@python.org (Barry A. Warsaw)
Date: Thu, 31 Oct 2002 01:50:29 -0500
Subject: [Mailman-Developers] CVS glitch.
References: <2373BAE0-EC98-11D6-9FAC-0003934516A8@plaidworks.com>
Message-ID: <15808.53941.660814.92001@gargle.gargle.HOWL>


>>>>> "CVR" == Chuq Von Rospach <chuqui@plaidworks.com> writes:

    CVR> Hmm. Barry, in current CVS, mailmanctl and qrunner both call
    CVR> LogStdErr with a parameter of tee_to_stdout. but in the
    CVR> library they include, that parameter is
    CVR> tee_to_real_stderr. Something's a bit out of sync in CVS

    CVR> so does cron/gate-news

    CVR> everything else seems okay, just those three files.

But I think it's your cvs tree!

-------------------- snip snip --------------------
% find -type f | xargs grep tee_to
./Mailman/Logging/Utils.py:def LogStdErr(category, label, manual_reprime=1, tee_to_real_stderr=1):
./Mailman/Logging/Utils.py:    If tee_to_real_stderr is true, then the real standard error also gets
./Mailman/Logging/Utils.py:        if tee_to_real_stderr:
-------------------- snip snip --------------------
% find -type f | xargs grep LogStdErr
./Mailman/Logging/Utils.py:def LogStdErr(category, label, manual_reprime=1, tee_to_real_stderr=1):
./bin/mailmanctl:from Mailman.Logging.Utils import LogStdErr
./bin/mailmanctl:LogStdErr('error', 'mailmanctl', manual_reprime=0)
./bin/qrunner:from Mailman.Logging.Utils import LogStdErr
./bin/qrunner:LogStdErr('error', 'qrunner', manual_reprime=0)
./contrib/auto:from Mailman.Logging.Utils import LogStdErr
./contrib/auto:LogStdErr('auto', 'auto')
./cron/gate_news:from Mailman.Logging.Utils import LogStdErr
./cron/gate_news:LogStdErr('error', 'gate_news', manual_reprime=0)
./scripts/bounces:from Mailman.Logging.Utils import LogStdErr
./scripts/bounces:LogStdErr('error', 'bounces')
./scripts/confirm:from Mailman.Logging.Utils import LogStdErr
./scripts/confirm:LogStdErr('error', 'confirm')
./scripts/join:from Mailman.Logging.Utils import LogStdErr
./scripts/join:LogStdErr('error', 'join')
./scripts/leave:from Mailman.Logging.Utils import LogStdErr
./scripts/leave:LogStdErr('error', 'leave')
./scripts/owner:from Mailman.Logging.Utils import LogStdErr
./scripts/owner:LogStdErr('error', 'mailowner')
./scripts/post:from Mailman.Logging.Utils import LogStdErr
./scripts/post:LogStdErr("error", "post")
./scripts/request:from Mailman.Logging.Utils import LogStdErr
./scripts/request:LogStdErr('error', 'mailcmd')
[skipping the build subdir]
-------------------- snip snip --------------------

What happens when you do "cvs -q up -P -d -A" ?

-Barry

From chuqui@plaidworks.com  Thu Oct 31 06:55:26 2002
From: chuqui@plaidworks.com (Chuq Von Rospach)
Date: Wed, 30 Oct 2002 22:55:26 -0800
Subject: [Mailman-Developers] CVS glitch.
In-Reply-To: <15808.53941.660814.92001@gargle.gargle.HOWL>
Message-ID: <BC3B43CA-EC9D-11D6-9FAC-0003934516A8@plaidworks.com>

well, that's bloody weird. CVS doesn't think so.

plaidworks.com 162# cvs -z3 
-d:pserver:anonymous@cvs.mailman.sourceforge.net:/cvsroot/mailman 
update mailman
cvs server: Updating mailman
cvs server: Updating mailman/Archiver
cvs server: Updating mailman/Bouncers
cvs server: Updating mailman/Cgi
cvs server: Updating mailman/Commands
cvs server: Updating mailman/Gui
cvs server: Updating mailman/Handlers
cvs server: Updating mailman/Logging
cvs server: Updating mailman/MTA
cvs server: Updating mailman/Queue

here we go again, Barry. rmemeber when my cvs didn't udate the mail 
packages?

weirdo.


On Wednesday, October 30, 2002, at 10:50  PM, Barry A. Warsaw wrote:

>
>>>>>> "CVR" == Chuq Von Rospach <chuqui@plaidworks.com> writes:
>
>     CVR> Hmm. Barry, in current CVS, mailmanctl and qrunner both call
>     CVR> LogStdErr with a parameter of tee_to_stdout. but in the
>     CVR> library they include, that parameter is
>     CVR> tee_to_real_stderr. Something's a bit out of sync in CVS
>
>     CVR> so does cron/gate-news
>
>     CVR> everything else seems okay, just those three files.
>
> But I think it's your cvs tree!
>
> -------------------- snip snip --------------------
> % find -type f | xargs grep tee_to
> ./Mailman/Logging/Utils.py:def LogStdErr(category, label, 
> manual_reprime=1, tee_to_real_stderr=1):
> ./Mailman/Logging/Utils.py:    If tee_to_real_stderr is true, then the 
> real standard error also gets
> ./Mailman/Logging/Utils.py:        if tee_to_real_stderr:
> -------------------- snip snip --------------------
> % find -type f | xargs grep LogStdErr
> ./Mailman/Logging/Utils.py:def LogStdErr(category, label, 
> manual_reprime=1, tee_to_real_stderr=1):
> ./bin/mailmanctl:from Mailman.Logging.Utils import LogStdErr
> ./bin/mailmanctl:LogStdErr('error', 'mailmanctl', manual_reprime=0)
> ./bin/qrunner:from Mailman.Logging.Utils import LogStdErr
> ./bin/qrunner:LogStdErr('error', 'qrunner', manual_reprime=0)
> ./contrib/auto:from Mailman.Logging.Utils import LogStdErr
> ./contrib/auto:LogStdErr('auto', 'auto')
> ./cron/gate_news:from Mailman.Logging.Utils import LogStdErr
> ./cron/gate_news:LogStdErr('error', 'gate_news', manual_reprime=0)
> ./scripts/bounces:from Mailman.Logging.Utils import LogStdErr
> ./scripts/bounces:LogStdErr('error', 'bounces')
> ./scripts/confirm:from Mailman.Logging.Utils import LogStdErr
> ./scripts/confirm:LogStdErr('error', 'confirm')
> ./scripts/join:from Mailman.Logging.Utils import LogStdErr
> ./scripts/join:LogStdErr('error', 'join')
> ./scripts/leave:from Mailman.Logging.Utils import LogStdErr
> ./scripts/leave:LogStdErr('error', 'leave')
> ./scripts/owner:from Mailman.Logging.Utils import LogStdErr
> ./scripts/owner:LogStdErr('error', 'mailowner')
> ./scripts/post:from Mailman.Logging.Utils import LogStdErr
> ./scripts/post:LogStdErr("error", "post")
> ./scripts/request:from Mailman.Logging.Utils import LogStdErr
> ./scripts/request:LogStdErr('error', 'mailcmd')
> [skipping the build subdir]
> -------------------- snip snip --------------------
>
> What happens when you do "cvs -q up -P -d -A" ?
>
> -Barry
>
-- 
Chuq Von Rospach, Architech
chuqui@plaidworks.com -- http://www.plaidworks.com/chuqui/blog/

Someday, we'll look back on this, laugh
nervously and change the subject.


From claw@kanga.nu  Thu Oct 31 08:14:18 2002
From: claw@kanga.nu (J C Lawrence)
Date: Thu, 31 Oct 2002 00:14:18 -0800
Subject: [Mailman-Developers] CVS glitch. 
In-Reply-To: Message from Chuq Von Rospach <chuqui@plaidworks.com> 
	<BC3B43CA-EC9D-11D6-9FAC-0003934516A8@plaidworks.com> 
References: <BC3B43CA-EC9D-11D6-9FAC-0003934516A8@plaidworks.com> 
Message-ID: <13054.1036052058@kanga.nu>

On Wed, 30 Oct 2002 22:55:26 -0800 
Chuq Von Rospach <chuqui@plaidworks.com> wrote:

> well, that's bloody weird. CVS doesn't think so.  
...
> weirdo.

General CVS advice:

  When in doubt, concern with CVS or even just the wrong day of the week
  (ie anything between Sunday and Saturday inclusive), do a fresh
  checkout, not an update.  Its saved my bacon and shorts far too many
  times to ignore.  CVS has a tendency to get these sorts of rare and
  complex operations (eg update) wrong now and then.

<Elide rant on the gloriously broken state and quality of CVS>

-- 
J C Lawrence                
---------(*)                Satan, oscillate my metallic sonatas. 
claw@kanga.nu               He lived as a devil, eh?		  
http://www.kanga.nu/~claw/  Evil is a name of a foeman, as I live.


From noreply@sourceforge.net  Thu Oct 31 09:18:55 2002
From: noreply@sourceforge.net (noreply@sourceforge.net)
Date: Thu, 31 Oct 2002 01:18:55 -0800
Subject: [Mailman-Developers] 
 [ mailman-Bugs-631447 ] error during subscription process
Message-ID: <E187BTb-0000B9-00@usw-sf-web3.sourceforge.net>

Bugs item #631447, was opened at 2002-10-31 10:18
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=100103&aid=631447&group_id=103

Category: (un)subscribing
Group: 2.0.x
Status: Open
Resolution: None
Priority: 5
Submitted By: Carlo Meme' (cmeme)
Assigned to: Nobody/Anonymous (nobody)
Summary: error during subscription process

Initial Comment:
One user trying to subscribe got the following email 
message: 

Bug in Mailman version 2.0.13
We're sorry, we hit a bug!

The information in the log are:

admin(8347): [----- Mailman Version: 2.0.13 -----]
admin(8347): [----- Traceback ------]
admin(8347): Traceback (innermost last):
admin(8347):   File "/var/mailman/scripts/driver", line 96, 
in run_main
admin(8347):     main()
admin(8347):   
File "/var/mailman/Mailman/Cgi/subscribe.py", line 76, in 
main
admin(8347):     process_form(mlist, doc)
admin(8347):   
File "/var/mailman/Mailman/Cgi/subscribe.py", line 104, 
in process_form
admin(8347):     if form.has_key("UserOptions") or \
admin(8347):   
File "/var/mailman/Mailman/pythonlib/cgi.py", line 564, 
in has_key
admin(8347):     raise TypeError, "not indexable"
admin(8347): TypeError: not indexable

admin(8347): [----- Python Information -----]
admin(8347): sys.version    = 1.5.2 (#1, Jul  5 2001, 
03:02:19)  [GCC 2.96 20000731 (Red Hat Linux 7.1 2
admin(8347): sys.executable = /usr/bin/python
admin(8347): sys.prefix     = /usr
admin(8347): sys.exec_prefix= /usr
admin(8347): sys.path       = /usr
admin(8347): sys.platform   = linux-i386


----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=100103&aid=631447&group_id=103

From barry@python.org  Thu Oct 31 14:04:04 2002
From: barry@python.org (Barry A. Warsaw)
Date: Thu, 31 Oct 2002 09:04:04 -0500
Subject: [Mailman-Developers] CVS glitch.
References: <15808.53941.660814.92001@gargle.gargle.HOWL>
	<BC3B43CA-EC9D-11D6-9FAC-0003934516A8@plaidworks.com>
Message-ID: <15809.14420.439191.907147@gargle.gargle.HOWL>


>>>>> "CVR" == Chuq Von Rospach <chuqui@plaidworks.com> writes:

    CVR> well, that's bloody weird. CVS doesn't think so.

I just tried an anonymous cvs co on MacOS 10.2.1 and got the expected
files.  I thought maybe it could be a OSX thing, or maybe a pserver
thing (remember that I always work in an authenticated checkout).  But
nope, I got the expected files.

>>>>> "JCL" == J C Lawrence <claw@kanga.nu> writes:

    JCL> <Elide rant on the gloriously broken state and quality of
    JCL> CVS>

Sigh.  Greg Wilson said at a Python conference many years ago
(paraphrasing): cvs is the worst tool all of us use on a daily basis.

subversive-ly y'rs,
-Barry

From noreply@sourceforge.net  Thu Oct 31 14:33:24 2002
From: noreply@sourceforge.net (noreply@sourceforge.net)
Date: Thu, 31 Oct 2002 06:33:24 -0800
Subject: [Mailman-Developers] 
 [ mailman-Bugs-631569 ] conn.quit() -> SMTPServerDisconnected
Message-ID: <E187GNw-0008Ts-00@usw-sf-web3.sourceforge.net>

Bugs item #631569, was opened at 2002-10-31 15:33
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=100103&aid=631569&group_id=103

Category: mail delivery
Group: 2.1 beta
Status: Open
Resolution: None
Priority: 5
Submitted By: Denzel D'aviagio (uplink2sf)
Assigned to: Nobody/Anonymous (nobody)
Summary: conn.quit() -> SMTPServerDisconnected

Initial Comment:
well this is driving me insane. had it since  2.1b3 on
RedHat 7.2 using python 2.2 (stock rpm install).

there's a few lists active but mail gets delivered
rarely - seems when sending mail,
Mailman/Handlers/SMTPDirect.py fails to conn.quit()
thus raising a
SMTPServerDisconnected: please run connect() first

...weird!

please find attached looots of logfiles that hopefully
make sense to you. I have too little knowledge of
python (<=0) to investigate more.

---
root [tweak] # $(which python2) -v
# /usr/lib/python2.2/site.pyc matches
/usr/lib/python2.2/site.py
import site # precompiled from /usr/lib/python2.2/site.pyc
# /usr/lib/python2.2/os.pyc matches
/usr/lib/python2.2/os.py
import os # precompiled from /usr/lib/python2.2/os.pyc
import posix # builtin
# /usr/lib/python2.2/posixpath.pyc matches
/usr/lib/python2.2/posixpath.py
import posixpath # precompiled from
/usr/lib/python2.2/posixpath.pyc
# /usr/lib/python2.2/stat.pyc matches
/usr/lib/python2.2/stat.py
import stat # precompiled from /usr/lib/python2.2/stat.pyc
# /usr/lib/python2.2/UserDict.pyc matches
/usr/lib/python2.2/UserDict.py
import UserDict # precompiled from
/usr/lib/python2.2/UserDict.pyc
Python 2.2 (#1, Apr 12 2002, 15:29:57)
[GCC 2.96 20000731 (Red Hat Linux 7.2 2.96-109)] on linux2
Type "help", "copyright", "credits" or "license" for
more information.
dlopen("/usr/lib/python2.2/lib-dynload/readline.so", 2);
import readline # dynamically loaded from
/usr/lib/python2.2/lib-dynload/readline.so
>>>
---
the following logfile mix is obfuscated a bit (email
and IP addresses)

08017 >>> 220 companyMail ESMTP
08017 <<< EHLO company.at
08017 >>> 250-mx.company.at Hello fqdn.domain.net
[2.3.4.5], pleased to meet you
08017 >>> 250-ENHANCEDSTATUSCODES
08017 >>> 250-PIPELINING
08017 >>> 250-8BITMIME
08017 >>> 250-SIZE
08017 >>> 250-DSN
08017 >>> 250-ETRN
08017 >>> 250-AUTH DIGEST-MD5 CRAM-MD5
08017 >>> 250-STARTTLS
08017 >>> 250-DELIVERBY
08017 >>> 250 HELP
08017 <<< STARTTLS
08017 >>> 220 2.0.0 Ready to start TLS
08018 >>> 220 companyMail ESMTP
08018 <<< HELO company.at
08018 >>> 250 mx.company.at Hello fqdn.domain.net
[2.3.4.5], pleased to meet you
08018 <<< MAIL FROM:<SENDER@DOMAIN.NET>
08018 >>> 250 2.1.0 <SENDER@DOMAIN.NET>... Sender ok
08018 <<< RCPT TO:<apfelsucht@company.at>
08018 >>> 250 2.1.5 <apfelsucht@company.at>... Recipient ok
08018 <<< DATA
08018 >>> 354 Enter mail, end with "." on a line by itself
08018 <<< Date: Thu, 31 Oct 2002 15:05:15 +0100
08018 <<< Mime-Version: 1.0 (Apple Message framework v546)
08018 <<< Content-Type: text/plain; charset=US-ASCII;
format=flowed
08018 <<< Subject: test
08018 <<< From: SENDER REALNAME <SENDER@DOMAIN.NET>
08018 <<< To: apfelsucht@company.at
08018 <<< Content-Transfer-Encoding: 7bit
08018 <<< Message-Id:
<C795F2CE-ECD9-11D6-B0B2-000393920DA6@company.at>
08018 <<< X-Mailer: Apple Mail (2.546)
08018 <<<
08018 <<< ENDOFTEST
08018 <<<
08018 <<< .
08018 >>> 250 2.0.0 g9VDs6Nu008018 Message accepted for
delivery
08018 <<< QUIT
08018 >>> 221 2.0.0 mx.company.at closing connection
08019 === EXEC smrsh -c
/opt/mailman/current/mail/mailman post apfelsucht
08019 >>> From SENDER@DOMAIN.NET Thu Oct 31 14:54:06 2002
08019 >>> Received: from company.at (fqdn.domain.net
[2.3.4.5]) by mx.company.at (company.at/20021013) with
SMTP id g9VDs6Nu008018 for <apfelsucht@company.at>;
Thu, 31 Oct 2002 14:54:06 +0100
08019 >>> Date: Thu, 31 Oct 2002 15:05:15 +0100
08019 >>> Mime-Version: 1.0 (Apple Message framework v546)
08019 >>> Content-Type: text/plain; charset=US-ASCII;
format=flowed
08019 >>> Subject: test
08019 >>> From: SENDER REALNAME <SENDER@DOMAIN.NET>
08019 >>> To: apfelsucht@company.at
08019 >>> Content-Transfer-Encoding: 7bit
08019 >>> Message-Id:
<C795F2CE-ECD9-11D6-B0B2-000393920DA6@company.at>
08019 >>> X-Mailer: Apple Mail (2.546)
08019 >>>
08019 >>> ENDOFTEST
08019 >>>
Oct 31 14:54:06 tweak sm-mta[8018]: g9VDs6Nu008018:
from=<SENDER@DOMAIN.NET>, size=377, class=0, nrcpts=1,
msgid=<C795F2CE-ECD9-11D6-B0B2-000393920DA6@company.at>,
proto=SMTP, daemon=MTA, relay=fqdn.domain.net [2.3.4.5]
08019 <<< [EOF]
Oct 31 14:54:07 tweak sm-mta[8019]: g9VDs6Nu008018:
to="|/opt/mailman/current/mail/mailman post
apfelsucht", ctladdr=<apfelsucht@company.at> (47/0),
delay=00:00:01, xdelay=00:00:01, mailer=prog,
pri=30568, dsn=2.0.0, stat=Sent
connect: ('localhost', 25)
connect: ('localhost', 25)
reply: '220 companyMail ESMTP\r\n'
reply: retcode (220); Msg: companyMail ESMTP
connect: companyMail ESMTP
send: 'ehlo tweak.company.at\r\n'
reply: '250-mx.company.at Hello localhost [127.0.0.1],
pleased to meet you\r\n'
reply: '250-ENHANCEDSTATUSCODES\r\n'
reply: '250-PIPELINING\r\n'
reply: '250-8BITMIME\r\n'
reply: '250-SIZE\r\n'
reply: '250-DSN\r\n'
reply: '250-ETRN\r\n'
reply: '250-AUTH DIGEST-MD5 CRAM-MD5\r\n'
reply: '250-STARTTLS\r\n'
reply: '250-DELIVERBY\r\n'
reply: '250 HELP\r\n'
reply: retcode (250); Msg: mx.company.at Hello
localhost [127.0.0.1], pleased to meet you
ENHANCEDSTATUSCODES
PIPELINING
8BITMIME
SIZE
DSN
ETRN
AUTH DIGEST-MD5 CRAM-MD5
STARTTLS
DELIVERBY
HELP
send: 'mail FROM:<apfelsucht-bounces@company.at>
size=1168\r\n'
reply: '250 2.1.0 <apfelsucht-bounces@company.at>...
Sender ok\r\n'
reply: retcode (250); Msg: 2.1.0
<apfelsucht-bounces@company.at>... Sender ok
send: 'rcpt TO:<recipient@subdomain.dhs.org>\r\n'
reply: '250 2.1.5 <recipient@subdomain.dhs.org>...
Recipient ok\r\n'
reply: retcode (250); Msg: 2.1.5
<recipient@subdomain.dhs.org>... Recipient ok
send: 'data\r\n'
reply: '354 Enter mail, end with "." on a line by
itself\r\n'
reply: retcode (354); Msg: Enter mail, end with "." on
a line by itself
data: (354, 'Enter mail, end with "." on a line by itself')
send: "Received: from company.at (fqdn.domain.net
[2.3.4.5]) by mx.company.at\r\n\t(company.at/20021013)
with SMTP id g9VDs6Nu008018 for
<apfelsucht@company.at>;\r\n\tThu, 31 Oct 2002 14:54:06
+0100\r\nDate: Thu, 31 Oct 2002 15:05:15
+0100\r\nMime-Version: 1.0 (Apple Message framework
v546)\r\nContent-Type: text/plain; charset=US-ASCII;
format=flowed\r\nFrom: SENDER REALNAME
<SENDER@DOMAIN.NET>\r\nTo:
apfelsucht@company.at\r\nContent-Transfer-Encoding:
7bit\r\nMessage-Id:
<C795F2CE-ECD9-11D6-B0B2-000393920DA6@company.at>\r\nX-Mailer:
Apple Mail (2.546)\r\nSubject: [Apfelsucht]
test\r\nX-BeenThere:
apfelsucht@company.at\r\nPrecedence: list\r\nList-Id:
Appl'o'holics, MacLovers & Switchers 
<apfelsucht.company.at>\r\nList-Post:
<mailto:apfelsucht@company.at>\r\nList-Subscribe:
<https://webmail.company.at/mailman/listinfo/apfelsucht>,\r\n\t<mailto:apfelsucht-request@company.at?subject=subscribe>\r\nList-Unsubscribe:
<https://webmail.company.at/mailman/listinfo/apfelsucht>,\r\n\t<mailto:apfelsucht-request@company.at?subject=unsubscribe>\r\nList-Archive:
</pipermail>\r\nList-Help:
<mailto:apfelsucht-request@company.at?subject=help>\r\nSender:
apfelsucht-bounces@company.at\r\nErrors-To:
apfelsucht-bounces@company.at\r\n\r\nENDOFTEST\r\n\r\n.\r\n"
reply: '250 2.0.0 g9VDs9Nu008025 Message accepted for
delivery\r\n'
reply: retcode (250); Msg: 2.0.0 g9VDs9Nu008025 Message
accepted for delivery
data: (250, '2.0.0 g9VDs9Nu008025 Message accepted for
delivery')
send: 'mail FROM:<apfelsucht-bounces@company.at>
size=1168\r\n'
reply: '250 2.1.0 <apfelsucht-bounces@company.at>...
Sender ok\r\n'
reply: retcode (250); Msg: 2.1.0
<apfelsucht-bounces@company.at>... Sender ok
send: 'rcpt TO:<al7tzliugfwefg@daemons.at>\r\n'
send: 'quit\r\n'
send: 'quit\r\n'
Oct 31 14:54:09 2002 (7608) All recipients refused:
please run connect() first
08025 >>> 220 companyMail ESMTP
08025 <<< ehlo tweak.company.at
08025 >>> 250-mx.company.at Hello localhost
[127.0.0.1], pleased to meet you
08025 >>> 250-ENHANCEDSTATUSCODES
08025 >>> 250-PIPELINING
08025 >>> 250-8BITMIME
08025 >>> 250-SIZE
08025 >>> 250-DSN
08025 >>> 250-ETRN
08025 >>> 250-AUTH DIGEST-MD5 CRAM-MD5
08025 >>> 250-STARTTLS
08025 >>> 250-DELIVERBY
08025 >>> 250 HELP
08025 <<< mail FROM:<apfelsucht-bounces@company.at>
size=1168
08025 >>> 250 2.1.0 <apfelsucht-bounces@company.at>...
Sender ok
08025 <<< rcpt TO:<recipient@subdomain.dhs.org>
08025 >>> 250 2.1.5 <recipient@subdomain.dhs.org>...
Recipient ok
08025 <<< data
08025 >>> 354 Enter mail, end with "." on a line by itself
08025 <<< Received: from company.at (fqdn.domain.net
[2.3.4.5]) by mx.company.at
08025 <<<       (company.at/20021013) with SMTP id
g9VDs6Nu008018 for <apfelsucht@company.at>;
08025 <<<       Thu, 31 Oct 2002 14:54:06 +0100
08025 <<< Date: Thu, 31 Oct 2002 15:05:15 +0100
08025 <<< Mime-Version: 1.0 (Apple Message framework v546)
08025 <<< Content-Type: text/plain; charset=US-ASCII;
format=flowed
08025 <<< From: SENDER REALNAME <SENDER@DOMAIN.NET>
08025 <<< To: apfelsucht@company.at
08025 <<< Content-Transfer-Encoding: 7bit
08025 <<< Message-Id:
<C795F2CE-ECD9-11D6-B0B2-000393920DA6@company.at>
08025 <<< X-Mailer: Apple Mail (2.546)
08025 <<< Subject: [Apfelsucht] test
08025 <<< X-BeenThere: apfelsucht@company.at
08025 <<< Precedence: list
08025 <<< List-Id: Appl'o'holics, MacLovers & Switchers
 <apfelsucht.company.at>
08025 <<< List-Post: <mailto:apfelsucht@company.at>
08025 <<< List-Subscribe:
<https://webmail.company.at/mailman/listinfo/apfelsucht>,
08025 <<<      
<mailto:apfelsucht-request@company.at?subject=subscribe>
08025 <<< List-Unsubscribe:
<https://webmail.company.at/mailman/listinfo/apfelsucht>,
08025 <<<      
<mailto:apfelsucht-request@company.at?subject=unsubscribe>
08025 <<< List-Archive: </pipermail>
08025 <<< List-Help:
<mailto:apfelsucht-request@company.at?subject=help>
08025 <<< Sender: apfelsucht-bounces@company.at
08025 <<< Errors-To: apfelsucht-bounces@company.at
08025 <<<
08025 <<< ENDOFTEST
08025 <<<
08025 <<< .
08025 >>> 250 2.0.0 g9VDs9Nu008025 Message accepted for
delivery
08027 === CONNECT localhost
08025 <<< mail FROM:<apfelsucht-bounces@company.at>
size=1168
08025 >>> 250 2.1.0 <apfelsucht-bounces@company.at>...
Sender ok
08025 <<< rcpt TO:<al7tzliugfwefg@daemons.at>
08027 <<< 220 company.at LMTP company 20021003 ready
08027 >>> LHLO mx.company.at
Oct 31 14:54:09 tweak sm-mta[8025]: g9VDs9Nu008025:
from=<apfelsucht-bounces@company.at>, size=1168,
class=-30, nrcpts=1,
msgid=<C795F2CE-ECD9-11D6-B0B2-000393920DA6@company.at>,
proto=ESMTP, daemon=MTA, relay=localhost [127.0.0.1]
Oct 31 14:54:09 2002 (7608) Uncaught runner exception:
please run connect() first
Oct 31 14:54:09 2002 (7608) Traceback (most recent call
last):
  File "/opt/mailman/current/Mailman/Queue/Runner.py",
line 105, in _oneloop
    self._onefile(msg, msgdata)
  File "/opt/mailman/current/Mailman/Queue/Runner.py",
line 154, in _onefile
    keepqueued = self._dispose(mlist, msg, msgdata)
  File
"/opt/mailman/current/Mailman/Queue/OutgoingRunner.py",
line 61, in _dispose
    self._func(mlist, msg, msgdata)
  File
"/opt/mailman/current/Mailman/Handlers/SMTPDirect.py",
line 149, in process
    conn.quit()
  File
"/opt/mailman/current/Mailman/Handlers/SMTPDirect.py",
line 79, in quit
    self.__conn.quit()
  File "/usr/lib/python2.2/smtplib.py", line 671, in quit
    self.docmd("quit")
  File "/usr/lib/python2.2/smtplib.py", line 352, in docmd
    self.putcmd(cmd,args)
  File "/usr/lib/python2.2/smtplib.py", line 308, in putcmd
    self.send(str)
  File "/usr/lib/python2.2/smtplib.py", line 300, in send
    raise SMTPServerDisconnected('please run connect()
first')
SMTPServerDisconnected: please run connect() first

08027 <<< 250-company.at
08027 <<< 250-8BITMIME
08027 <<< 250-ENHANCEDSTATUSCODES
08027 <<< 250-PIPELINING
08027 <<< 250-SIZE
08027 <<< 250-AUTH EXTERNAL
08027 <<< 250 IGNOREQUOTA
08027 >>> MAIL From:<apfelsucht-bounces@company.at>
SIZE=1168
08027 <<< 250 2.1.0 ok
08027 >>> RCPT To:<kari>
08027 >>> DATA
08027 <<< 250 2.1.5 ok
08027 <<< 354 go ahead
08027 >>> Return-Path: <apfelsucht-bounces@company.at>
08027 >>> Received: from tweak.company.at (localhost
[127.0.0.1]) by mx.company.at (company.at/20021013)
with ESMTP id g9VDs9Nu008025 for
<recipient@subdomain.dhs.org>; Thu, 31 Oct 2002
14:54:09 +0100
08027 >>> Received: from company.at (fqdn.domain.net
[2.3.4.5]) by mx.company.at
08027 >>>       (company.at/20021013) with SMTP id
g9VDs6Nu008018 for <apfelsucht@company.at>;
08027 >>>       Thu, 31 Oct 2002 14:54:06 +0100
08027 >>> Date: Thu, 31 Oct 2002 15:05:15 +0100
08027 >>> Mime-Version: 1.0 (Apple Message framework v546)
08027 >>> Content-Type: text/plain; charset=US-ASCII;
format=flowed
08027 >>> From: SENDER REALNAME <SENDER@DOMAIN.NET>
08027 >>> To: apfelsucht@company.at
08027 >>> Content-Transfer-Encoding: 7bit
08027 >>> Message-Id:
<C795F2CE-ECD9-11D6-B0B2-000393920DA6@company.at>
08027 >>> X-Mailer: Apple Mail (2.546)
08027 >>> Subject: [Apfelsucht] test
08027 >>> X-BeenThere: apfelsucht@company.at
08027 >>> Precedence: list
08027 >>> List-Id: Appl'o'holics, MacLovers & Switchers
 <apfelsucht.company.at>
08027 >>> List-Post: <mailto:apfelsucht@company.at>
08027 >>> List-Subscribe:
<https://webmail.company.at/mailman/listinfo/apfelsucht>,
08027 >>>      
<mailto:apfelsucht-request@company.at?subject=subscribe>
08027 >>> List-Unsubscribe:
<https://webmail.company.at/mailman/listinfo/apfelsucht>,
08027 >>>      
<mailto:apfelsucht-request@company.at?subject=unsubscribe>
08027 >>> List-Archive: </pipermail>
08027 >>> List-Help:
<mailto:apfelsucht-request@company.at?subject=help>
08027 >>> Sender: apfelsucht-bounces@company.at
08027 >>> Errors-To: apfelsucht-bounces@company.at
08027 >>>
08027 >>> ENDOFTEST
08027 >>>
08027 >>> .
08027 <<< 250 2.1.5 Ok
08027 >>> QUIT
08027 <<< 221 2.0.0 bye
Oct 31 14:54:10 tweak sm-mta[8027]: g9VDs9Nu008025:
to=<recipient@subdomain.dhs.org>, delay=00:00:01,
xdelay=00:00:01, mailer=local,pri=85157,
relay=localhost, dsn=2.0.0, stat=Sent
---

PLEASE SEND HELP!




----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=100103&aid=631569&group_id=103

From noreply@sourceforge.net  Thu Oct 31 14:38:45 2002
From: noreply@sourceforge.net (noreply@sourceforge.net)
Date: Thu, 31 Oct 2002 06:38:45 -0800
Subject: [Mailman-Developers] 
 [ mailman-Bugs-631569 ] conn.quit() -> SMTPServerDisconnected
Message-ID: <E187GT7-0000JN-00@usw-sf-web3.sourceforge.net>

Bugs item #631569, was opened at 2002-10-31 15:33
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=100103&aid=631569&group_id=103

Category: mail delivery
Group: 2.1 beta
Status: Open
Resolution: None
>Priority: 6
Submitted By: Denzel D'aviagio (uplink2sf)
Assigned to: Nobody/Anonymous (nobody)
Summary: conn.quit() -> SMTPServerDisconnected

Initial Comment:
well this is driving me insane. had it since  2.1b3 on
RedHat 7.2 using python 2.2 (stock rpm install).

there's a few lists active but mail gets delivered
rarely - seems when sending mail,
Mailman/Handlers/SMTPDirect.py fails to conn.quit()
thus raising a
SMTPServerDisconnected: please run connect() first

...weird!

please find attached looots of logfiles that hopefully
make sense to you. I have too little knowledge of
python (<=0) to investigate more.

---
root [tweak] # $(which python2) -v
# /usr/lib/python2.2/site.pyc matches
/usr/lib/python2.2/site.py
import site # precompiled from /usr/lib/python2.2/site.pyc
# /usr/lib/python2.2/os.pyc matches
/usr/lib/python2.2/os.py
import os # precompiled from /usr/lib/python2.2/os.pyc
import posix # builtin
# /usr/lib/python2.2/posixpath.pyc matches
/usr/lib/python2.2/posixpath.py
import posixpath # precompiled from
/usr/lib/python2.2/posixpath.pyc
# /usr/lib/python2.2/stat.pyc matches
/usr/lib/python2.2/stat.py
import stat # precompiled from /usr/lib/python2.2/stat.pyc
# /usr/lib/python2.2/UserDict.pyc matches
/usr/lib/python2.2/UserDict.py
import UserDict # precompiled from
/usr/lib/python2.2/UserDict.pyc
Python 2.2 (#1, Apr 12 2002, 15:29:57)
[GCC 2.96 20000731 (Red Hat Linux 7.2 2.96-109)] on linux2
Type "help", "copyright", "credits" or "license" for
more information.
dlopen("/usr/lib/python2.2/lib-dynload/readline.so", 2);
import readline # dynamically loaded from
/usr/lib/python2.2/lib-dynload/readline.so
>>>
---
the following logfile mix is obfuscated a bit (email
and IP addresses)

08017 >>> 220 companyMail ESMTP
08017 <<< EHLO company.at
08017 >>> 250-mx.company.at Hello fqdn.domain.net
[2.3.4.5], pleased to meet you
08017 >>> 250-ENHANCEDSTATUSCODES
08017 >>> 250-PIPELINING
08017 >>> 250-8BITMIME
08017 >>> 250-SIZE
08017 >>> 250-DSN
08017 >>> 250-ETRN
08017 >>> 250-AUTH DIGEST-MD5 CRAM-MD5
08017 >>> 250-STARTTLS
08017 >>> 250-DELIVERBY
08017 >>> 250 HELP
08017 <<< STARTTLS
08017 >>> 220 2.0.0 Ready to start TLS
08018 >>> 220 companyMail ESMTP
08018 <<< HELO company.at
08018 >>> 250 mx.company.at Hello fqdn.domain.net
[2.3.4.5], pleased to meet you
08018 <<< MAIL FROM:<SENDER@DOMAIN.NET>
08018 >>> 250 2.1.0 <SENDER@DOMAIN.NET>... Sender ok
08018 <<< RCPT TO:<apfelsucht@company.at>
08018 >>> 250 2.1.5 <apfelsucht@company.at>... Recipient ok
08018 <<< DATA
08018 >>> 354 Enter mail, end with "." on a line by itself
08018 <<< Date: Thu, 31 Oct 2002 15:05:15 +0100
08018 <<< Mime-Version: 1.0 (Apple Message framework v546)
08018 <<< Content-Type: text/plain; charset=US-ASCII;
format=flowed
08018 <<< Subject: test
08018 <<< From: SENDER REALNAME <SENDER@DOMAIN.NET>
08018 <<< To: apfelsucht@company.at
08018 <<< Content-Transfer-Encoding: 7bit
08018 <<< Message-Id:
<C795F2CE-ECD9-11D6-B0B2-000393920DA6@company.at>
08018 <<< X-Mailer: Apple Mail (2.546)
08018 <<<
08018 <<< ENDOFTEST
08018 <<<
08018 <<< .
08018 >>> 250 2.0.0 g9VDs6Nu008018 Message accepted for
delivery
08018 <<< QUIT
08018 >>> 221 2.0.0 mx.company.at closing connection
08019 === EXEC smrsh -c
/opt/mailman/current/mail/mailman post apfelsucht
08019 >>> From SENDER@DOMAIN.NET Thu Oct 31 14:54:06 2002
08019 >>> Received: from company.at (fqdn.domain.net
[2.3.4.5]) by mx.company.at (company.at/20021013) with
SMTP id g9VDs6Nu008018 for <apfelsucht@company.at>;
Thu, 31 Oct 2002 14:54:06 +0100
08019 >>> Date: Thu, 31 Oct 2002 15:05:15 +0100
08019 >>> Mime-Version: 1.0 (Apple Message framework v546)
08019 >>> Content-Type: text/plain; charset=US-ASCII;
format=flowed
08019 >>> Subject: test
08019 >>> From: SENDER REALNAME <SENDER@DOMAIN.NET>
08019 >>> To: apfelsucht@company.at
08019 >>> Content-Transfer-Encoding: 7bit
08019 >>> Message-Id:
<C795F2CE-ECD9-11D6-B0B2-000393920DA6@company.at>
08019 >>> X-Mailer: Apple Mail (2.546)
08019 >>>
08019 >>> ENDOFTEST
08019 >>>
Oct 31 14:54:06 tweak sm-mta[8018]: g9VDs6Nu008018:
from=<SENDER@DOMAIN.NET>, size=377, class=0, nrcpts=1,
msgid=<C795F2CE-ECD9-11D6-B0B2-000393920DA6@company.at>,
proto=SMTP, daemon=MTA, relay=fqdn.domain.net [2.3.4.5]
08019 <<< [EOF]
Oct 31 14:54:07 tweak sm-mta[8019]: g9VDs6Nu008018:
to="|/opt/mailman/current/mail/mailman post
apfelsucht", ctladdr=<apfelsucht@company.at> (47/0),
delay=00:00:01, xdelay=00:00:01, mailer=prog,
pri=30568, dsn=2.0.0, stat=Sent
connect: ('localhost', 25)
connect: ('localhost', 25)
reply: '220 companyMail ESMTP\r\n'
reply: retcode (220); Msg: companyMail ESMTP
connect: companyMail ESMTP
send: 'ehlo tweak.company.at\r\n'
reply: '250-mx.company.at Hello localhost [127.0.0.1],
pleased to meet you\r\n'
reply: '250-ENHANCEDSTATUSCODES\r\n'
reply: '250-PIPELINING\r\n'
reply: '250-8BITMIME\r\n'
reply: '250-SIZE\r\n'
reply: '250-DSN\r\n'
reply: '250-ETRN\r\n'
reply: '250-AUTH DIGEST-MD5 CRAM-MD5\r\n'
reply: '250-STARTTLS\r\n'
reply: '250-DELIVERBY\r\n'
reply: '250 HELP\r\n'
reply: retcode (250); Msg: mx.company.at Hello
localhost [127.0.0.1], pleased to meet you
ENHANCEDSTATUSCODES
PIPELINING
8BITMIME
SIZE
DSN
ETRN
AUTH DIGEST-MD5 CRAM-MD5
STARTTLS
DELIVERBY
HELP
send: 'mail FROM:<apfelsucht-bounces@company.at>
size=1168\r\n'
reply: '250 2.1.0 <apfelsucht-bounces@company.at>...
Sender ok\r\n'
reply: retcode (250); Msg: 2.1.0
<apfelsucht-bounces@company.at>... Sender ok
send: 'rcpt TO:<recipient@subdomain.dhs.org>\r\n'
reply: '250 2.1.5 <recipient@subdomain.dhs.org>...
Recipient ok\r\n'
reply: retcode (250); Msg: 2.1.5
<recipient@subdomain.dhs.org>... Recipient ok
send: 'data\r\n'
reply: '354 Enter mail, end with "." on a line by
itself\r\n'
reply: retcode (354); Msg: Enter mail, end with "." on
a line by itself
data: (354, 'Enter mail, end with "." on a line by itself')
send: "Received: from company.at (fqdn.domain.net
[2.3.4.5]) by mx.company.at\r\n\t(company.at/20021013)
with SMTP id g9VDs6Nu008018 for
<apfelsucht@company.at>;\r\n\tThu, 31 Oct 2002 14:54:06
+0100\r\nDate: Thu, 31 Oct 2002 15:05:15
+0100\r\nMime-Version: 1.0 (Apple Message framework
v546)\r\nContent-Type: text/plain; charset=US-ASCII;
format=flowed\r\nFrom: SENDER REALNAME
<SENDER@DOMAIN.NET>\r\nTo:
apfelsucht@company.at\r\nContent-Transfer-Encoding:
7bit\r\nMessage-Id:
<C795F2CE-ECD9-11D6-B0B2-000393920DA6@company.at>\r\nX-Mailer:
Apple Mail (2.546)\r\nSubject: [Apfelsucht]
test\r\nX-BeenThere:
apfelsucht@company.at\r\nPrecedence: list\r\nList-Id:
Appl'o'holics, MacLovers & Switchers 
<apfelsucht.company.at>\r\nList-Post:
<mailto:apfelsucht@company.at>\r\nList-Subscribe:
<https://webmail.company.at/mailman/listinfo/apfelsucht>,\r\n\t<mailto:apfelsucht-request@company.at?subject=subscribe>\r\nList-Unsubscribe:
<https://webmail.company.at/mailman/listinfo/apfelsucht>,\r\n\t<mailto:apfelsucht-request@company.at?subject=unsubscribe>\r\nList-Archive:
</pipermail>\r\nList-Help:
<mailto:apfelsucht-request@company.at?subject=help>\r\nSender:
apfelsucht-bounces@company.at\r\nErrors-To:
apfelsucht-bounces@company.at\r\n\r\nENDOFTEST\r\n\r\n.\r\n"
reply: '250 2.0.0 g9VDs9Nu008025 Message accepted for
delivery\r\n'
reply: retcode (250); Msg: 2.0.0 g9VDs9Nu008025 Message
accepted for delivery
data: (250, '2.0.0 g9VDs9Nu008025 Message accepted for
delivery')
send: 'mail FROM:<apfelsucht-bounces@company.at>
size=1168\r\n'
reply: '250 2.1.0 <apfelsucht-bounces@company.at>...
Sender ok\r\n'
reply: retcode (250); Msg: 2.1.0
<apfelsucht-bounces@company.at>... Sender ok
send: 'rcpt TO:<al7tzliugfwefg@daemons.at>\r\n'
send: 'quit\r\n'
send: 'quit\r\n'
Oct 31 14:54:09 2002 (7608) All recipients refused:
please run connect() first
08025 >>> 220 companyMail ESMTP
08025 <<< ehlo tweak.company.at
08025 >>> 250-mx.company.at Hello localhost
[127.0.0.1], pleased to meet you
08025 >>> 250-ENHANCEDSTATUSCODES
08025 >>> 250-PIPELINING
08025 >>> 250-8BITMIME
08025 >>> 250-SIZE
08025 >>> 250-DSN
08025 >>> 250-ETRN
08025 >>> 250-AUTH DIGEST-MD5 CRAM-MD5
08025 >>> 250-STARTTLS
08025 >>> 250-DELIVERBY
08025 >>> 250 HELP
08025 <<< mail FROM:<apfelsucht-bounces@company.at>
size=1168
08025 >>> 250 2.1.0 <apfelsucht-bounces@company.at>...
Sender ok
08025 <<< rcpt TO:<recipient@subdomain.dhs.org>
08025 >>> 250 2.1.5 <recipient@subdomain.dhs.org>...
Recipient ok
08025 <<< data
08025 >>> 354 Enter mail, end with "." on a line by itself
08025 <<< Received: from company.at (fqdn.domain.net
[2.3.4.5]) by mx.company.at
08025 <<<       (company.at/20021013) with SMTP id
g9VDs6Nu008018 for <apfelsucht@company.at>;
08025 <<<       Thu, 31 Oct 2002 14:54:06 +0100
08025 <<< Date: Thu, 31 Oct 2002 15:05:15 +0100
08025 <<< Mime-Version: 1.0 (Apple Message framework v546)
08025 <<< Content-Type: text/plain; charset=US-ASCII;
format=flowed
08025 <<< From: SENDER REALNAME <SENDER@DOMAIN.NET>
08025 <<< To: apfelsucht@company.at
08025 <<< Content-Transfer-Encoding: 7bit
08025 <<< Message-Id:
<C795F2CE-ECD9-11D6-B0B2-000393920DA6@company.at>
08025 <<< X-Mailer: Apple Mail (2.546)
08025 <<< Subject: [Apfelsucht] test
08025 <<< X-BeenThere: apfelsucht@company.at
08025 <<< Precedence: list
08025 <<< List-Id: Appl'o'holics, MacLovers & Switchers
 <apfelsucht.company.at>
08025 <<< List-Post: <mailto:apfelsucht@company.at>
08025 <<< List-Subscribe:
<https://webmail.company.at/mailman/listinfo/apfelsucht>,
08025 <<<      
<mailto:apfelsucht-request@company.at?subject=subscribe>
08025 <<< List-Unsubscribe:
<https://webmail.company.at/mailman/listinfo/apfelsucht>,
08025 <<<      
<mailto:apfelsucht-request@company.at?subject=unsubscribe>
08025 <<< List-Archive: </pipermail>
08025 <<< List-Help:
<mailto:apfelsucht-request@company.at?subject=help>
08025 <<< Sender: apfelsucht-bounces@company.at
08025 <<< Errors-To: apfelsucht-bounces@company.at
08025 <<<
08025 <<< ENDOFTEST
08025 <<<
08025 <<< .
08025 >>> 250 2.0.0 g9VDs9Nu008025 Message accepted for
delivery
08027 === CONNECT localhost
08025 <<< mail FROM:<apfelsucht-bounces@company.at>
size=1168
08025 >>> 250 2.1.0 <apfelsucht-bounces@company.at>...
Sender ok
08025 <<< rcpt TO:<al7tzliugfwefg@daemons.at>
08027 <<< 220 company.at LMTP company 20021003 ready
08027 >>> LHLO mx.company.at
Oct 31 14:54:09 tweak sm-mta[8025]: g9VDs9Nu008025:
from=<apfelsucht-bounces@company.at>, size=1168,
class=-30, nrcpts=1,
msgid=<C795F2CE-ECD9-11D6-B0B2-000393920DA6@company.at>,
proto=ESMTP, daemon=MTA, relay=localhost [127.0.0.1]
Oct 31 14:54:09 2002 (7608) Uncaught runner exception:
please run connect() first
Oct 31 14:54:09 2002 (7608) Traceback (most recent call
last):
  File "/opt/mailman/current/Mailman/Queue/Runner.py",
line 105, in _oneloop
    self._onefile(msg, msgdata)
  File "/opt/mailman/current/Mailman/Queue/Runner.py",
line 154, in _onefile
    keepqueued = self._dispose(mlist, msg, msgdata)
  File
"/opt/mailman/current/Mailman/Queue/OutgoingRunner.py",
line 61, in _dispose
    self._func(mlist, msg, msgdata)
  File
"/opt/mailman/current/Mailman/Handlers/SMTPDirect.py",
line 149, in process
    conn.quit()
  File
"/opt/mailman/current/Mailman/Handlers/SMTPDirect.py",
line 79, in quit
    self.__conn.quit()
  File "/usr/lib/python2.2/smtplib.py", line 671, in quit
    self.docmd("quit")
  File "/usr/lib/python2.2/smtplib.py", line 352, in docmd
    self.putcmd(cmd,args)
  File "/usr/lib/python2.2/smtplib.py", line 308, in putcmd
    self.send(str)
  File "/usr/lib/python2.2/smtplib.py", line 300, in send
    raise SMTPServerDisconnected('please run connect()
first')
SMTPServerDisconnected: please run connect() first

08027 <<< 250-company.at
08027 <<< 250-8BITMIME
08027 <<< 250-ENHANCEDSTATUSCODES
08027 <<< 250-PIPELINING
08027 <<< 250-SIZE
08027 <<< 250-AUTH EXTERNAL
08027 <<< 250 IGNOREQUOTA
08027 >>> MAIL From:<apfelsucht-bounces@company.at>
SIZE=1168
08027 <<< 250 2.1.0 ok
08027 >>> RCPT To:<kari>
08027 >>> DATA
08027 <<< 250 2.1.5 ok
08027 <<< 354 go ahead
08027 >>> Return-Path: <apfelsucht-bounces@company.at>
08027 >>> Received: from tweak.company.at (localhost
[127.0.0.1]) by mx.company.at (company.at/20021013)
with ESMTP id g9VDs9Nu008025 for
<recipient@subdomain.dhs.org>; Thu, 31 Oct 2002
14:54:09 +0100
08027 >>> Received: from company.at (fqdn.domain.net
[2.3.4.5]) by mx.company.at
08027 >>>       (company.at/20021013) with SMTP id
g9VDs6Nu008018 for <apfelsucht@company.at>;
08027 >>>       Thu, 31 Oct 2002 14:54:06 +0100
08027 >>> Date: Thu, 31 Oct 2002 15:05:15 +0100
08027 >>> Mime-Version: 1.0 (Apple Message framework v546)
08027 >>> Content-Type: text/plain; charset=US-ASCII;
format=flowed
08027 >>> From: SENDER REALNAME <SENDER@DOMAIN.NET>
08027 >>> To: apfelsucht@company.at
08027 >>> Content-Transfer-Encoding: 7bit
08027 >>> Message-Id:
<C795F2CE-ECD9-11D6-B0B2-000393920DA6@company.at>
08027 >>> X-Mailer: Apple Mail (2.546)
08027 >>> Subject: [Apfelsucht] test
08027 >>> X-BeenThere: apfelsucht@company.at
08027 >>> Precedence: list
08027 >>> List-Id: Appl'o'holics, MacLovers & Switchers
 <apfelsucht.company.at>
08027 >>> List-Post: <mailto:apfelsucht@company.at>
08027 >>> List-Subscribe:
<https://webmail.company.at/mailman/listinfo/apfelsucht>,
08027 >>>      
<mailto:apfelsucht-request@company.at?subject=subscribe>
08027 >>> List-Unsubscribe:
<https://webmail.company.at/mailman/listinfo/apfelsucht>,
08027 >>>      
<mailto:apfelsucht-request@company.at?subject=unsubscribe>
08027 >>> List-Archive: </pipermail>
08027 >>> List-Help:
<mailto:apfelsucht-request@company.at?subject=help>
08027 >>> Sender: apfelsucht-bounces@company.at
08027 >>> Errors-To: apfelsucht-bounces@company.at
08027 >>>
08027 >>> ENDOFTEST
08027 >>>
08027 >>> .
08027 <<< 250 2.1.5 Ok
08027 >>> QUIT
08027 <<< 221 2.0.0 bye
Oct 31 14:54:10 tweak sm-mta[8027]: g9VDs9Nu008025:
to=<recipient@subdomain.dhs.org>, delay=00:00:01,
xdelay=00:00:01, mailer=local,pri=85157,
relay=localhost, dsn=2.0.0, stat=Sent
---

PLEASE SEND HELP!




----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=100103&aid=631569&group_id=103

From danny@terweij.nl  Thu Oct 31 14:48:08 2002
From: danny@terweij.nl (Danny Terweij)
Date: Thu, 31 Oct 2002 15:48:08 +0100
Subject: [Mailman-Developers] Much cpu/memory load
Message-ID: <01ab01c280ec$8739bb20$1e00a8c0@onsnet.org>

Hi,

Because my /qfiles/archive/*.* files are not processed by the
qrunner/ArchRunner my linux box gets overloaded.
I have now 55mb unproccessed archive files at the qfiles dir.
The pipermail archives are not updated.

When i do a mailmanctl stop then the Archrunner is still active. I did a
kill -9 PID and it stops.
Then i must remove all the lock files found where lockfiles are.

Then the cpu load is back to normal.

24290 ?        S      0:00 /usr/bin/python2 ./mailmanctl start
24291 ?        D     34:56 qrunner
/usr/local/mailman/bin/qrunner --runner=ArchRunner:0:1 -s
24292 ?        S      0:16 qrunner
/usr/local/mailman/bin/qrunner --runner=BounceRunner:0:1 -s
24293 ?        S      0:03 qrunner
/usr/local/mailman/bin/qrunner --runner=CommandRunner:0:1 -s
24294 ?        S      0:31 qrunner
/usr/local/mailman/bin/qrunner --runner=IncomingRunner:0:1 -s
24295 ?        S      0:03 qrunner
/usr/local/mailman/bin/qrunner --runner=NewsRunner:0:1 -s
24296 ?        S      0:29 qrunner
/usr/local/mailman/bin/qrunner --runner=OutgoingRunner:0:1 -s
24297 ?        S      0:07 qrunner
/usr/local/mailman/bin/qrunner --runner=VirginRunner:0:1 -s

I think that realy somthing is wrong with the ArchRunner. Looks like it is
in a loop or is doing nothing.
The /qfiles/archive/ dir is growing and growing, i must stop mailman very
often now otherwise de box gets overloaded.

The load was this morning on a average of 10 and all my memory was used
(20mb left on swap).
No error and messages at the logs dir :(

Please take a look Barry :) it is the latest cvs.
You may also take al look on my box if you want and have time...

Danny Terweij






From Dale@Newfield.org  Thu Oct 31 14:52:56 2002
From: Dale@Newfield.org (Dale Newfield)
Date: Thu, 31 Oct 2002 09:52:56 -0500 (EST)
Subject: [Mailman-Developers] CVS glitch.
In-Reply-To: <BC3B43CA-EC9D-11D6-9FAC-0003934516A8@plaidworks.com>
Message-ID: <Pine.OSX.4.44.0210310951000.20431-100000@core.newfield.org>

> > What happens when you do "cvs -q up -P -d -A" ?

> plaidworks.com 162# cvs -z3
> -d:pserver:anonymous@cvs.mailman.sourceforge.net:/cvsroot/mailman
> update mailman

Am I just being annoying if I point out that you didn't include all the
flags Barry suggested?  (Specifically "-A" says "make sure *nothing* is
sticky and go to the head of the main trunk.)

-Dale


From chuqui@plaidworks.com  Thu Oct 31 16:58:07 2002
From: chuqui@plaidworks.com (Chuq Von Rospach)
Date: Thu, 31 Oct 2002 08:58:07 -0800
Subject: [Mailman-Developers] errors in 2.1b4
Message-ID: <EDE7B5B2-ECF1-11D6-9FAC-0003934516A8@plaidworks.com>


I had some stuff in the shunt dir, so I ran unshunt. Most of it seemed 
to be weird spam with funky character sets in it.

Got various tracebacks and errors attempting to deal with it....


plaidworks.com 44# ../bin/unshunt

*** /Users/mailman/logs/error ***
Oct 31 08:53:11 2002 (393) Uncaught runner exception:
Oct 31 08:53:11 2002 (393) Traceback (most recent call last):
   File "/usr/local/mailman/Mailman/Queue/Runner.py", line 105, in 
_oneloop
     self._onefile(msg, msgdata)
   File "/usr/local/mailman/Mailman/Queue/Runner.py", line 154, in 
_onefile
     keepqueued = self._dispose(mlist, msg, msgdata)
   File "/usr/local/mailman/Mailman/Queue/IncomingRunner.py", line 129, 
in _dispose
     status = self._dopipeline(mlist, msg, msgdata, pipeline)
   File "/usr/local/mailman/Mailman/Queue/IncomingRunner.py", line 152, 
in _dopipeline
     sys.modules[modname].process(mlist, msg, msgdata)
   File "/usr/local/mailman/Mailman/Handlers/Hold.py", line 181, in 
process
     MessageTooBig(bodylen, mlist.max_message_size))
   File "/usr/local/mailman/Mailman/Handlers/Hold.py", line 207, in 
hold_for_approval
     id = mlist.HoldMessage(msg, reason, msgdata)
   File "/usr/local/mailman/Mailman/ListAdmin.py", line 202, in 
HoldMessage
     assert not self.__db.has_key(id)
AssertionError

Oct 31 08:53:13 2002 (395) Uncaught runner exception: sequence 
expected, NoneType found
Oct 31 08:53:13 2002 (395) Traceback (most recent call last):
   File "/usr/local/mailman/Mailman/Queue/Runner.py", line 105, in 
_oneloop
     self._onefile(msg, msgdata)
   File "/usr/local/mailman/Mailman/Queue/Runner.py", line 154, in 
_onefile
     keepqueued = self._dispose(mlist, msg, msgdata)
   File "/usr/local/mailman/Mailman/Queue/OutgoingRunner.py", line 61, 
in _dispose
     self._func(mlist, msg, msgdata)
   File "/usr/local/mailman/Mailman/Handlers/SMTPDirect.py", line 139, 
in process
     deliveryfunc(mlist, msg, msgdata, envsender, refused, conn)
   File "/usr/local/mailman/Mailman/Handlers/SMTPDirect.py", line 277, 
in verpdeliver
     d = {'bounces': bmailbox,
TypeError: sequence expected, NoneType found

Oct 31 08:53:13 2002 (393) Uncaught runner exception: ASCII decoding 
error: ordinal not in range(128)
Oct 31 08:53:13 2002 (393) Traceback (most recent call last):
   File "/usr/local/mailman/Mailman/Queue/Runner.py", line 105, in 
_oneloop
     self._onefile(msg, msgdata)
   File "/usr/local/mailman/Mailman/Queue/Runner.py", line 154, in 
_onefile
     keepqueued = self._dispose(mlist, msg, msgdata)
   File "/usr/local/mailman/Mailman/Queue/IncomingRunner.py", line 129, 
in _dispose
     status = self._dopipeline(mlist, msg, msgdata, pipeline)
   File "/usr/local/mailman/Mailman/Queue/IncomingRunner.py", line 152, 
in _dopipeline
     sys.modules[modname].process(mlist, msg, msgdata)
   File "/usr/local/mailman/Mailman/Handlers/Hold.py", line 181, in 
process
     MessageTooBig(bodylen, mlist.max_message_size))
   File "/usr/local/mailman/Mailman/Handlers/Hold.py", line 255, in 
hold_for_approval
     lang=lang)
   File "/usr/local/mailman/Mailman/Message.py", line 203, in __init__
     self['Subject'] = Header(subject, charset, header_name='Subject')
   File "/usr/local/mailman/pythonlib/email/Header.py", line 164, in 
__init__
     self.append(s, charset)
   File "/usr/local/mailman/pythonlib/email/Header.py", line 224, in 
append
     unicode(s, charset.get_output_charset())
UnicodeError: ASCII decoding error: ordinal not in range(128)

Not even sure this isn't the correct response to these beasts, but I 
thought I'd pass them along in case Barry wanted to tighten up some 
tests in the code...





-- 
Chuq Von Rospach, Architech
chuqui@plaidworks.com -- http://www.plaidworks.com/chuqui/blog/

Yes, I am an agent of Satan, but my duties
are largely ceremonial.


From andrew.clark@ucsb.edu  Thu Oct 31 18:16:38 2002
From: andrew.clark@ucsb.edu (Andrew D. Clark)
Date: Thu, 31 Oct 2002 10:16:38 -0800
Subject: [Mailman-Developers] re: Much cpu/memory load
In-Reply-To: <E187Ifr-00046V-03@mail.python.org>
References: <E187Ifr-00046V-03@mail.python.org>
Message-ID: <29390000.1036088198@modulus.oit.ucsb.edu>

I've had the same problem.  ArchRunner seems to have serious performance 
problems if you've got heavy list activity and are archiving to both 
pipermail and mbox.  I "fixed" it by setting ARCHIVE_TO_MBOX=1.  This will 
only archive to mbox, so your pipermail archives won't be updated.  You 
should be able to generate pipermail archives by hand later with bin/arch 
list_name.  It isn't an ideal solution, but at least your message will move 
out of the archive queue.

--
Andrew Clark
Campus Network Programmer
Office of Information Technology
University of California, Santa Barbara
andrew.clark@ucsb.edu (805) 893-5311


> Date: Thu, 31 Oct 2002 15:48:08 +0100
> From: "Danny Terweij" <danny@terweij.nl>
> To: <mailman-developers@python.org>
> Subject: [Mailman-Developers] Much cpu/memory load
> Message-ID: <01ab01c280ec$8739bb20$1e00a8c0@onsnet.org>
> Content-Type: text/plain;
> 	charset="iso-8859-1"
> MIME-Version: 1.0
> Content-Transfer-Encoding: 7bit
> Precedence: list
> Message: 1
>
> Hi,
>
> Because my /qfiles/archive/*.* files are not processed by the
> qrunner/ArchRunner my linux box gets overloaded.
> I have now 55mb unproccessed archive files at the qfiles dir.
> The pipermail archives are not updated.
>
> When i do a mailmanctl stop then the Archrunner is still active. I did a
> kill -9 PID and it stops.
> Then i must remove all the lock files found where lockfiles are.
>
> Then the cpu load is back to normal.
>
> 24290 ?        S      0:00 /usr/bin/python2 ./mailmanctl start
> 24291 ?        D     34:56 qrunner
> /usr/local/mailman/bin/qrunner --runner=ArchRunner:0:1 -s
> 24292 ?        S      0:16 qrunner
> /usr/local/mailman/bin/qrunner --runner=BounceRunner:0:1 -s
> 24293 ?        S      0:03 qrunner
> /usr/local/mailman/bin/qrunner --runner=CommandRunner:0:1 -s
> 24294 ?        S      0:31 qrunner
> /usr/local/mailman/bin/qrunner --runner=IncomingRunner:0:1 -s
> 24295 ?        S      0:03 qrunner
> /usr/local/mailman/bin/qrunner --runner=NewsRunner:0:1 -s
> 24296 ?        S      0:29 qrunner
> /usr/local/mailman/bin/qrunner --runner=OutgoingRunner:0:1 -s
> 24297 ?        S      0:07 qrunner
> /usr/local/mailman/bin/qrunner --runner=VirginRunner:0:1 -s
>
> I think that realy somthing is wrong with the ArchRunner. Looks like it is
> in a loop or is doing nothing.
> The /qfiles/archive/ dir is growing and growing, i must stop mailman very
> often now otherwise de box gets overloaded.
>
> The load was this morning on a average of 10 and all my memory was used
> (20mb left on swap).
> No error and messages at the logs dir :(
>
> Please take a look Barry :) it is the latest cvs.
> You may also take al look on my box if you want and have time...
>
> Danny Terweij




From rge1@egwpartners.com  Thu Oct 31 18:32:48 2002
From: rge1@egwpartners.com (Rick Ennis)
Date: Thu, 31 Oct 2002 13:32:48 -0500
Subject: [Mailman-Developers] Can't Approve/Reject/Discard from Admin
	WebInterface
References: <019301c28035$2e80f320$0202a8c0@anakin>
	<20021030181541.GC5220@cthulhu.gerg.ca>
Message-ID: <006801c2810b$e9d0bdf0$0202a8c0@anakin>

Thanks for the response Greg.  Actually I just figured it out.  But it looks
like I may be petitioning for a feature request or detail fix...

I was looking at the html source for one of the regular admin pages and I
noticed that the url for the form POST is a relative path
"../admin/somethingorother."  But the url for the page I was having trouble
with (the moderator requests one) is an absolute path starting with
http://servername/mailman/admindb/whatever.  This was the issue b/c I had
apache sending a redirect for any http://servername/mailman url to
https://servername/mailman (to run it over SSL).  The POST information was
getting lost in the redirect.

Does anyone else run mailman over https?  This worked great with the 2.0.x
releases.
Barry, can this be added as a configurable option?  Or could that path just
be made relative too?

Thanks,
Rick

----- Original Message -----
From: "Greg Ward" <gward@python.net>
To: <mailman-developers@python.org>
Sent: Wednesday, October 30, 2002 1:15 PM
Subject: Re: [Mailman-Developers] Can't Approve/Reject/Discard from Admin
WebInterface


> On 30 October 2002, Rick Ennis said:
> > Hi...  I have a quick question.
>
> Sorry, I don't have any quick answers -- just more questions.
>
> > I just started playing with 2.1 yesterday (snagged 2.1b4+ from cvs) and
have
> > been searching the mailing lists all morning.  What I can't seem to
figure
> > out is...  should the approve/reject/discard interface on the admin
screen
> > be working (in this beta)?   I click one of the radio buttons followed
by
> > "submit all data" but it just reloads the same page.  Logging out and
back
> > in shows no change in the db state.
>
> Which web browser, and on which OS?  Which web server, and on which OS?
> What does your web server log say?  Anything weird like an HTTP proxy
> server in the way?  Have you tried running a packet tracer (I like
> ethereal myself) to the traffic between web browser and server to make
> sure it looks sane?
>
> > None of my files in the mailman/logs
> > directory have been updated, nor my /var/log/messages file.
>
> AFAIK Mailman does not use syslog, so /var/log/* should be irrelevant.
>
> Do look in ~mailman/data though -- that's where it stores pending
> messages.  When you approve/reject/discard a message, a file should
> disappear from that directory (I think).
>
> > Is this not
> > implemented yet or did I just do something brain-dead during the
> > installation?
>
>         Greg



From claw@kanga.nu  Thu Oct 31 20:04:59 2002
From: claw@kanga.nu (J C Lawrence)
Date: Thu, 31 Oct 2002 12:04:59 -0800
Subject: [Mailman-Developers] CVS glitch. 
In-Reply-To: Message from barry@python.org (Barry A. Warsaw) 
	<15809.14420.439191.907147@gargle.gargle.HOWL> 
References: <15808.53941.660814.92001@gargle.gargle.HOWL>
	<BC3B43CA-EC9D-11D6-9FAC-0003934516A8@plaidworks.com>
	<15809.14420.439191.907147@gargle.gargle.HOWL> 
Message-ID: <22606.1036094699@kanga.nu>

On Thu, 31 Oct 2002 09:04:04 -0500 
Barry A Warsaw <barry@python.org> wrote:

JCL> <Elide rant on the gloriously broken state and quality of CVS>

> Sigh.  Greg Wilson said at a Python conference many years ago
> (paraphrasing): cvs is the worst tool all of us use on a daily basis.

Far be it from em to disagree.

> subversive-ly y'rs, -Barry

I've been there.  No thanks.

-- 
J C Lawrence                
---------(*)                Satan, oscillate my metallic sonatas. 
claw@kanga.nu               He lived as a devil, eh?		  
http://www.kanga.nu/~claw/  Evil is a name of a foeman, as I live.


From Dan.Mick@sun.com  Thu Oct 31 20:14:10 2002
From: Dan.Mick@sun.com (Dan Mick)
Date: Thu, 31 Oct 2002 12:14:10 -0800
Subject: [Mailman-Developers] CVS glitch.
References: <Pine.OSX.4.44.0210310951000.20431-100000@core.newfield.org>
Message-ID: <3DC18F12.8040704@sun.com>

Dale Newfield wrote:
>>>What happens when you do "cvs -q up -P -d -A" ?
> 
> 
>>plaidworks.com 162# cvs -z3
>>-d:pserver:anonymous@cvs.mailman.sourceforge.net:/cvsroot/mailman
>>update mailman
> 
> 
> Am I just being annoying if I point out that you didn't include all the
> flags Barry suggested?  (Specifically "-A" says "make sure *nothing* is
> sticky and go to the head of the main trunk.)

What he said; Chuq, can you soothe my mind about the panacea for CVS's
brokenness, and try those flags?

It is expected that cvs without -A might not get everything.  It is
unexpected that -A won't solve that.


From Dale@Newfield.org  Thu Oct 31 20:34:33 2002
From: Dale@Newfield.org (Dale Newfield)
Date: Thu, 31 Oct 2002 15:34:33 -0500 (EST)
Subject: [Mailman-Developers] CVS glitch.
In-Reply-To: <3DC18F12.8040704@sun.com>
Message-ID: <Pine.OSX.4.44.0210311510530.22199-100000@core.newfield.org>

On Thu, 31 Oct 2002, Dan Mick wrote:
> It is expected that cvs without -A might not get everything.  It is
> unexpected that -A won't solve that.

And just for the edification of the list, an explanation of the other two
flags Barry suggested:
"-P" says "give me new directories (and their contents) if any have been
           created."
"-d" says "remove unmodified files (and resulting empty directories) that
           are no longer relevant."

Order should not matter, and I never tracked it down as far as I should,
but let me mention that I've witnessed people having cvs problems when the
flags are ordered "-P -d" instead of "-d -P"

from cvs --help update :
	-A	Reset any sticky tags/date/kopts.
	-P	Prune empty directories.
	-d	Build directories, like checkout does.

-Dale


From danny@terweij.nl  Thu Oct 31 20:41:52 2002
From: danny@terweij.nl (Danny Terweij)
Date: Thu, 31 Oct 2002 21:41:52 +0100
Subject: [Mailman-Developers] re: Much cpu/memory load
References: <E187Ifr-00046V-03@mail.python.org>
	<29390000.1036088198@modulus.oit.ucsb.edu>
Message-ID: <000f01c2811d$f1e55b10$1e00a8c0@onsnet.org>

From: "Andrew D. Clark" <andrew.clark@ucsb.edu>

> I've had the same problem.  ArchRunner seems to have serious performance
> problems if you've got heavy list activity and are archiving to both
> pipermail and mbox.  I "fixed" it by setting ARCHIVE_TO_MBOX=1.  This will
> only archive to mbox, so your pipermail archives won't be updated.  You
> should be able to generate pipermail archives by hand later with bin/arch
> list_name.  It isn't an ideal solution, but at least your message will
move
> out of the archive queue.


Hmmzz. i am now trying to reboot my machine with

 9:33pm  up 1 day,  6:38,  7 users,  load average: 42.95, 41.97, 40.85

This is very bad! and this is happened after upgrade to b4(+)



> > Date: Thu, 31 Oct 2002 15:48:08 +0100
> > From: "Danny Terweij" <danny@terweij.nl>
> > To: <mailman-developers@python.org>
> > Subject: [Mailman-Developers] Much cpu/memory load
> > Message-ID: <01ab01c280ec$8739bb20$1e00a8c0@onsnet.org>
> > Content-Type: text/plain;
> > charset="iso-8859-1"
> > MIME-Version: 1.0
> > Content-Transfer-Encoding: 7bit
> > Precedence: list
> > Message: 1
> >
> > Hi,
> >
> > Because my /qfiles/archive/*.* files are not processed by the
> > qrunner/ArchRunner my linux box gets overloaded.
> > I have now 55mb unproccessed archive files at the qfiles dir.
> > The pipermail archives are not updated.
> >
> > When i do a mailmanctl stop then the Archrunner is still active. I did a
> > kill -9 PID and it stops.
> > Then i must remove all the lock files found where lockfiles are.
> >
> > Then the cpu load is back to normal.
> >
> > 24290 ?        S      0:00 /usr/bin/python2 ./mailmanctl start
> > 24291 ?        D     34:56 qrunner
> > /usr/local/mailman/bin/qrunner --runner=ArchRunner:0:1 -s
> > 24292 ?        S      0:16 qrunner
> > /usr/local/mailman/bin/qrunner --runner=BounceRunner:0:1 -s
> > 24293 ?        S      0:03 qrunner
> > /usr/local/mailman/bin/qrunner --runner=CommandRunner:0:1 -s
> > 24294 ?        S      0:31 qrunner
> > /usr/local/mailman/bin/qrunner --runner=IncomingRunner:0:1 -s
> > 24295 ?        S      0:03 qrunner
> > /usr/local/mailman/bin/qrunner --runner=NewsRunner:0:1 -s
> > 24296 ?        S      0:29 qrunner
> > /usr/local/mailman/bin/qrunner --runner=OutgoingRunner:0:1 -s
> > 24297 ?        S      0:07 qrunner
> > /usr/local/mailman/bin/qrunner --runner=VirginRunner:0:1 -s
> >
> > I think that realy somthing is wrong with the ArchRunner. Looks like it
is
> > in a loop or is doing nothing.
> > The /qfiles/archive/ dir is growing and growing, i must stop mailman
very
> > often now otherwise de box gets overloaded.
> >
> > The load was this morning on a average of 10 and all my memory was used
> > (20mb left on swap).
> > No error and messages at the logs dir :(
> >
> > Please take a look Barry :) it is the latest cvs.
> > You may also take al look on my box if you want and have time...
> >
> > Danny Terweij
>
>
>
>
> _______________________________________________
> Mailman-Developers mailing list
> Mailman-Developers@python.org
> http://mail.python.org/mailman/listinfo/mailman-developers


From barry@python.org  Thu Oct 31 21:20:14 2002
From: barry@python.org (Barry A. Warsaw)
Date: Thu, 31 Oct 2002 16:20:14 -0500
Subject: [Mailman-Developers] re: Much cpu/memory load
References: <E187Ifr-00046V-03@mail.python.org>
	<29390000.1036088198@modulus.oit.ucsb.edu>
	<000f01c2811d$f1e55b10$1e00a8c0@onsnet.org>
Message-ID: <15809.40590.896001.764415@gargle.gargle.HOWL>


A quick fix for the Archiver problems, until I can debug them more is
to add the following in ArchRunner.py, just under the "class
ArchRunner" line:

class ArchRunner(Runner):
    QDIR = mm_cfg.ARCHQUEUE_DIR
    SLEEPTIME = mm_cfg.minutes(10)

    def _dispose(self, mlist, msg, msgdata):

This at least makes ArchRunner only run once per 10 minutes.

-Barry

From noreply@sourceforge.net  Thu Oct 31 21:46:19 2002
From: noreply@sourceforge.net (noreply@sourceforge.net)
Date: Thu, 31 Oct 2002 13:46:19 -0800
Subject: [Mailman-Developers] 
 [ mailman-Bugs-631774 ] i18n unicode string interpolation
Message-ID: <E187N8t-0007WY-00@usw-sf-web5.sourceforge.net>

Bugs item #631774, was opened at 2002-10-31 22:46
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=100103&aid=631774&group_id=103

Category: None
Group: 2.1 beta
Status: Open
Resolution: None
Priority: 5
Submitted By: Simone Piunno (pioppo)
Assigned to: Nobody/Anonymous (nobody)
Summary: i18n unicode string interpolation

Initial Comment:
Hi, below you can find a report from one of the italian
translators (thanks Enea Zaffanella).

I believe somewhere between 2.1b3 and b4 a string has 
been converted into unicode and now string interpolation 
is not working anymore as expected.


Here u'' appears in html output (generated by admin.py):

======================================
  <table BORDER="2">
    <tr> 
     
<td><center><strong>Indirizzo/nome</strong></center></td>
      <td><center><strong>La tua
decisione</strong></center></td>

      <td><center><strong>Motivo del
rifiuto</strong></center></td>
    </tr>
    <tr>
**********************************************************************
      <td>u'zaffanella@example.it<br><em>Enea
Zaffanella</em>'</td>
**********************************************************************
      <td>
<table CELLSPACING="5">
  <tr>

    <td><input name="14" type="RADIO" value="0" CHECKED
>Rimando la
decisione</td>
    <td><input name="14" type="RADIO" value="4"
>Approvo</td>
    <td><input name="14" type="RADIO" value="2"
>Rigetta</td>
    <td><input name="14" type="RADIO" value="3"
>Scarta</td>
  </tr>
</table>
================================

And here (adminsubscribeack.txt) interpolation doesn't
happen:

================================
%(member)s è stato iscritto con successo a %(listname)s.
================================

They are 2 different cases: one is _(string) and the
other is Utils.Maketext(a template, dictionary of
strings), so this suggests even more that i18n.py has
been screwed up by 
unicode.

----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=100103&aid=631774&group_id=103

From danny@terweij.nl  Thu Oct 31 21:40:45 2002
From: danny@terweij.nl (Danny Terweij)
Date: Thu, 31 Oct 2002 22:40:45 +0100
Subject: [Mailman-Developers] re: Much cpu/memory load
References: 
	<E187Ifr-00046V-03@mail.python.org><29390000.1036088198@modulus.oit.ucsb.edu><000f01c2811d$f1e55b10$1e00a8c0@onsnet.org>
	<15809.40590.896001.764415@gargle.gargle.HOWL>
Message-ID: <005601c28126$2b9e9a30$1e00a8c0@onsnet.org>

From: "Barry A. Warsaw" <barry@python.org>

> A quick fix for the Archiver problems, until I can debug them more is
> to add the following in ArchRunner.py, just under the "class
> ArchRunner" line:

Thanx, i'll try that tomorrow. (Now here 22:39)

I'll let you know what then happens :)

Danny.



From e008993@cobalt.sbirmc.ac.uk  Wed Oct 30 20:29:44 2002
From: e008993@cobalt.sbirmc.ac.uk (Martin Whinnery <martin.whinnery@sbirmc.ac.uk>)
Date: Wed, 30 Oct 2002 20:29:44 +0000 (GMT)
Subject: [Mailman-Developers] Help programming MemberAdaptor...
In-Reply-To: <20021030180552.GA5220@cthulhu.gerg.ca>
Message-ID: <Pine.LNX.4.21.0210302018210.31723-100000@cobalt.sbirmc.ac.uk>

	

On Wed, 30 Oct 2002, Greg Ward wrote:

> > What I'd really like to do is have lists EITHER populate in the normal
> > file way OR populated from LDAP. I was thinking I might use an address
> > like
> 
> Err, define "populate".  Is this a one-time thing?  Or do you want the
> list membership to always come from an LDAP lookup?
> 

The latter. I help run a large educational campus, and we hold all
accounts and groups in LDAP. These groups change membership over time, and
I want to associate mailing lists with some of them. What I don't need is
the headache of remembering to keep the memberships in sync.

> If the former, I would just write a script that does the LDAP query and
> sets the list population from it.  Use bin/add_members as an example.
> 

If we absolutely must. It's what we do with DNS and DHCP. But I'd rather
not. Just more code to maintain.

> If the latter, why use Mailman at all?  Why not just use an MTA that can
> lookup aliases using LDAP, and then send mail to that alias?  I'm pretty
> sure both Exim and postfix can do that.
> 

'Cos mailman does the [approve|reject|defer|discard] thing. And
archives. And it deals with 'real' mailing lists (which we would die
rather than lose).

Martin Whinnery
*Assistant* Network Manager
South Birmingham College

'Just the assistant'


From e008993@cobalt.sbirmc.ac.uk  Wed Oct 30 20:56:11 2002
From: e008993@cobalt.sbirmc.ac.uk (Martin Whinnery <martin.whinnery@sbirmc.ac.uk>)
Date: Wed, 30 Oct 2002 20:56:11 +0000 (GMT)
Subject: [Mailman-Developers] More extend.py questions
In-Reply-To: <20021030181656.GD5220@cthulhu.gerg.ca>
Message-ID: <Pine.LNX.4.21.0210302030110.31723-100000@cobalt.sbirmc.ac.uk>



On Wed, 30 Oct 2002, Greg Ward wrote:

> On 30 October 2002, Martin Whinnery said:
> > Please indulge a python virgin. Barry has advised someone to 'overload
> > the Load & Save function' in extend.py . Would someone help me with a
> > little sample? It doesn't need to do anything useful, just replace the
> > original function.
> 
> Huh?  There's no file called extend.py in the Mailman source tree.  Just
> what are you trying to accomplish?
> 

And I quote (From NEWS 2.36)
"   
    - Membership Adaptors
        o Internally, mailing list memberships are accessed through a
          MemberAdaptor interface.  This would allow for integrating
          membership databases with external sources (e.g. Zope or
          LDAP), although the only MemberAdaptor currently implemented
          is a "classic" adaptor which stores the membership
          information on the MailList object.
...
    - List Extensibility
        o A framework has been added which can be used to specialize
          and extend specific mailing lists.  If there is a file
          called lists/<yourlist>/extend.py, it is execfile()'d after
          the MailList object is instantiated.  The file should
          contain a function extend() which will be called with the
          MailList instance...
"

Now after a little more digging (And I take your point about not hacking
the actual source, I don't know what I was thinking. Probably why I'm
'just the assistant') it appears that what I need to do is implement new
Load() and Save() functions ( orginally defined in MailList.py ), which
cause membership information to be read into the mlist structure from LDAP
rather than from config.pck, and which fail to write that information back
to config.pck.

I appreciate that this will leave membership options unable to be changed,
but that's what we want. A later version might be able to write that info
back to LDAP, but there's authentication issues and the matter of schema
definitions, and I don't need to go there just yet.

Here, look. This code is in file

/usr/local/mailman/lists/ldaplist/extend.py

# Extension mechanism for a list.
# This ought to cause a lists membership info to read from LDAP

from ldap import *
from Mailman.ldap_cfg import *
from Mailman.LDAPMemberships import *
## Note that the file LDAPMemberships.py is identical to
## OldStyleMembership.py except that names the class LDAPMemberships.
## All the methods are currently identical
 
class LDAPMailList:
    def __init__(self,oldfunc):
        self.oldfunc = oldfunc

    def Save(self):
        myldap = initialize(LDAP_URL)
        myldap.simple_bind("dummy","")
        return

    def Load(self):
        """Load method which ought to overload that defined in MailList"""
        myldap = initialize(LDAP_URL)
        myldap.simple_bind("","")
        # Populate membership
        filter = "(&(objectclass=" + LDAP_LIST_CLASS + ")(" +
LDAP_LIST_RDN_ATTR + "=" + mlist.real_name + "))"
	blah blah blah LDAP stuff blah blah blah
        return


def extend(mlist):
    mlist = LDAPMailList(mlist)
    mlist.__memberadaptor = LDAPMemberships(mlist)

----- snip ----------

Now, what I want is for the Load() and Save() functions defined in
MailList.py to be overridden for this list with my own versions.

I just don't understand Python/Mailman well enough.

I suspect that I'm trying to do one thing two ways, and that I should
forget the LDAPMemberships / MemberAdaptor stuff for the moment.

Any examples, flames, advice or witticisms will be revered in a darkened
room.


Thanks in advance (I like the desperate sound that makes when I say it)

Martin Whinnery
"Just the" Assistant Network Manager
South Birmingham College