Today I began maintaining a mailman list, and noticed
a small misspelling. I hope this is the appropriate
place to post patches...
Thanks,
Michael
csf at moscow.com
Index: mailman/Mailman/Cgi/admin.py
===================================================================
RCS file: /projects/cvsroot/mailman/Mailman/Cgi/admin.py,v
retrieving revision 1.36
diff -u -r1.36 admin.py
--- admin.py 1999/03/15 19:03:55 1.36
+++ admin.py 1999/03/16 01:08:10
@@ -299,7 +299,7 @@
doc.AddItem(form)
form.AddItem("Make your changes below, and then submit them using the"
- " bottom at the bottom. (You can change your password"
+ " button at the bottom. (You can change your password"
" there, too.)")
form.AddItem(FormatOptionsSection(category, lst))
From Dionysos at Dionysia.org Tue Mar 16 06:45:29 1999
From: Dionysos at Dionysia.org (Dan Delaney)
Date: Tue, 16 Mar 1999 00:45:29 -0500 (EST)
Subject: [Mailman-Users] Large lists and Postfix
Message-ID:
Hi all,
Is anyone using Mailman for lists with thousands of subscribers?
If so, how well does it work and what kind of load does it put on
the system? (My list has 200 subscriber with an average of 50
messages/day.)
Also, how well does it work with Postfix?
Thanks.
--Dan
-----------------------------------------------------------------------
Daniel G. Delaney The Louisville Times Chorus
Dionysos at Dionysia.org www.LouisvilleTimes.org
www.Dionysia.org/~dionysos/ Dionysia Design
ICQ Number: 8171285 www.Dionysia.com/design/
-----------------------------------------------------------------------
"Only two things are infinite: the universe and stupidity--
and I'm not sure about the former."
--Albert Einstein
From bruce at hams.com Tue Mar 16 06:35:49 1999
From: bruce at hams.com (bruce at hams.com)
Date: 16 Mar 1999 05:35:49 -0000
Subject: [Mailman-Users] Large lists and Postfix
Message-ID: <19990316053549.12149.qmail@hams.com>
The load is almost entirely dependent on the mail delivery agent and your
network connection, not Mailman. Linux, qmail and smartlist are able to get
more than 100,000 messages a day delivered using a '486. This is a reliable,
repeatable, every day figure. It should be at least that good for postfix.
Thanks
Bruce
From Dionysos at Dionysia.org Tue Mar 16 07:09:55 1999
From: Dionysos at Dionysia.org (Dan Delaney)
Date: Tue, 16 Mar 1999 01:09:55 -0500 (EST)
Subject: [Mailman-Users] Large lists and Postfix
In-Reply-To: <19990316053549.12149.qmail@hams.com>
Message-ID:
On 16 Mar 1999 bruce at hams.com wrote:
> The load is almost entirely dependent on the mail delivery agent and your
> network connection, not Mailman. Linux, qmail and smartlist are able to get
> more than 100,000 messages a day delivered using a '486. This is a reliable,
> repeatable, every day figure. It should be at least that good for postfix.
Thanks. So, considering the claims made about ezmlm for qmail, does
that have anything to do with ezmlm, or is it qmail that's making
mlm so fast? Would exmlm do just as will using Postfix (if it
could), and would qmail work just as well with Mailman instead of
ezmlm?
--Dan
-----------------------------------------------------------------------
Daniel G. Delaney The Louisville Times Chorus
Dionysos at Dionysia.org www.LouisvilleTimes.org
www.Dionysia.org/~dionysos/ Dionysia Design
ICQ Number: 8171285 www.Dionysia.com/design/
-----------------------------------------------------------------------
"Only two things are infinite: the universe and stupidity--
and I'm not sure about the former."
--Albert Einstein
From bruce at hams.com Tue Mar 16 07:17:57 1999
From: bruce at hams.com (bruce at hams.com)
Date: 16 Mar 1999 06:17:57 -0000
Subject: [Mailman-Users] Large lists and Postfix
Message-ID: <19990316061757.13450.qmail@hams.com>
From: Dan Delaney
> Thanks. So, considering the claims made about ezmlm for qmail, does
> that have anything to do with ezmlm, or is it qmail that's making
> mlm so fast?
Ezmlm is a very lightweight mailing list manager, and it adds essentially
no overhead to that of qmail. Mailman would add some more overhead, but
still not enough to be significant. However, the main overhead of mailing
lists is their overhead on _your_time_ to manage them, not the computer.
Mailman is very good about reducing _that_ overhead, much better than ezmlm.
Ezmlm's main advantage is that it uses the qmail "percent hack" to get
reliable information about bounces - the "envelope from" address is
different for each recepient, and you can tell who a bouncing message
was addressed to by the address the bounce comes to. For example,
mail to me would have an envelope From something like this:
balloons-owner-bruce%hams.com at lists.hams.com
If my address bounces, the bounce mail goes to that address, a qmail default
rule catches it, and ezmlm decodes bruce%hams.com and realizes that I've
had a bounce. It doesn't have to look at the bounce message at all, just
the address it goes to.
When I get sick of watching undecoded bounces go by, I'll add that "percent
hack" feature to Mailman.
Thanks
Bruce
From ds-list-mailman at sws5.ctd.ornl.gov Tue Mar 16 14:25:52 1999
From: ds-list-mailman at sws5.ctd.ornl.gov (Dave Sill)
Date: Tue, 16 Mar 1999 08:25:52 -0500 (EST)
Subject: [Mailman-Users] Large lists and Postfix
In-Reply-To: <19990316061757.13450.qmail@hams.com>
References: <19990316061757.13450.qmail@hams.com>
Message-ID: <14062.23520.982260.884101@sws5.ctd.ornl.gov>
bruce at hams.com wrote:
>However, the main overhead of mailing lists is their overhead on
>_your_time_ to manage them, not the computer. Mailman is very good
>about reducing _that_ overhead, much better than ezmlm.
I'm a long-time majordomo user who's evaluating alternatives including
ezmlm, mailman, and listar. I'd appreciate an explanation of how
mailman is better at reducing the list manager's overhead than ezmlm.
>Ezmlm's main advantage is that it uses the qmail "percent hack" to get
Actually, it's called "VERP" (Variable Envelope Return Path), not
"percent hack". The "percenthack" qmail configuration variable turns
on user%host at relay style relaying. But your explanation is, otherwise,
accurate.
>If my address bounces, the bounce mail goes to that address, a qmail default
>rule catches it, and ezmlm decodes bruce%hams.com and realizes that I've
>had a bounce. It doesn't have to look at the bounce message at all, just
>the address it goes to.
Right, and the neat thing about this from the user's point of view is
that ezmlm keeps track of which messages each recipient bounces so if
the problem is temporary, it tells them and they can retrieve the
messages from the archive.
>When I get sick of watching undecoded bounces go by, I'll add that "percent
>hack" feature to Mailman.
You really need VERP support in the MTA, otherwise you'll have to
inject a copy of each message to each recipient. Right now, qmail is
the only MTA I've aware of that does VERP. Venema wants to add it as
an option to Postfix, though.
-Dave
From bruce at hams.com Tue Mar 16 17:08:54 1999
From: bruce at hams.com (bruce at hams.com)
Date: 16 Mar 1999 16:08:54 -0000
Subject: [Mailman-Users] Large lists and Postfix
Message-ID: <19990316160854.27636.qmail@hams.com>
From: Dave Sill
> I'm a long-time majordomo user who's evaluating alternatives including
> ezmlm, mailman, and listar. I'd appreciate an explanation of how
> mailman is better at reducing the list manager's overhead than ezmlm.
Simply because the web interface is excellent, and it allows the user
to do most of their own management easily, and the confirmation system
avoids problems with malicious subscription. Users don't in general
want to bother the manager if they can solve their own problem with a
button click. Yes, I know other list servers have these features as
optional packages, but in Mailman they are well-integrated.
Mailman seems to have the best features. I really like the way that
Mailman figures out what host URL it's being called from and presents the
appropriate list choices for that URL. I have the same mailman instance
running on several virtual web sites, and it does not display the lists
of one virtual site when called with another virtual site's host name.
Ezmlm is too minimal. I don't know about listar. Majordomo and SmartList
are fine, but their source code is more difficult for team-development
becuase of their language choices. Mailman's development is superscedeing
these other programs because of that.
> Actually, it's called "VERP" (Variable Envelope Return Path), not
> "percent hack". The "percenthack" qmail configuration variable turns
> on user%host at relay style relaying. But your explanation is, otherwise,
> accurate.
I missed this usage when I was working on qmail, or it didn't exist then.
I stand corrected.
> You really need VERP support in the MTA, otherwise you'll have to
> inject a copy of each message to each recipient. Right now, qmail is
> the only MTA I've aware of that does VERP. Venema wants to add it as
> an option to Postfix, though.
Right. What Mailman needs is a way to pass in the VERP from a bounce in
an MDA-independent fashion. Something like "~mailman/mail/wrapper bounce
", where address is the decoded VERP. Given that, one could use
a front-end script like my qmail-to-mailman.py to handle the MDA-dependent
portion.
Thanks
Bruce
From th at nextel.no Tue Mar 16 18:37:18 1999
From: th at nextel.no (Tor Houghton)
Date: Tue, 16 Mar 1999 18:37:18 +0100 (MET)
Subject: [Mailman-Users] Problem with "bin/newlist"
Message-ID:
Hi,
Got the 1.09b source down and installed it. I couldn't see any reference
to this particular error (although i found one for mmpasswd):
redlance:mailman {25} bin/newlist mylist th at nextel.no mypasswd
Traceback (innermost last):
File "bin/newlist", line 38, in ?
from Mailman.pythonlib import getpass
ImportError: No module named Mailman.pythonlib
What's wrong?
Regards,
Tor Houghton.
From castro at usmatrix.net Tue Mar 16 18:37:15 1999
From: castro at usmatrix.net (Edgard Castro)
Date: Tue, 16 Mar 1999 12:37:15 -0500
Subject: [Mailman-Users] Sugestions
Message-ID: <19990316123714.A22397@usmatrix.net>
Hello,
I'm currently using MailMan with Postfix and I noticed some
'missing' features (or maybe my fault :P).
Those are: Disable sorting of recipients when injecting
the messages. This will help using less
memory when using MTAs like Postfix or QMail.
Disable or hardlimit (globally) the number
of concurrent SMTP connections that MailMan
will open to inject the messages. This will
be helpfull to let a user maintain it's our
list without hogging the system down. (You
know... Stupid users who set this to like
100!).
Are these implemented? Sorry if it's already. :P
e.
--
Edgard Castro | 305-273-5003 -o)
Chief Network Engineer - U.S. Matrix Internet, Inc. /\\
Linux Counter User #29078 -- http://counter.li.org _\_V
From ds-list-mailman at sws5.ctd.ornl.gov Tue Mar 16 18:40:34 1999
From: ds-list-mailman at sws5.ctd.ornl.gov (Dave Sill)
Date: Tue, 16 Mar 1999 12:40:34 -0500 (EST)
Subject: [Mailman-Users] Large lists and Postfix
In-Reply-To: <19990316160854.27636.qmail@hams.com>
References: <19990316160854.27636.qmail@hams.com>
Message-ID: <14062.38802.920407.329536@sws5.ctd.ornl.gov>
bruce at hams.com wrote:
>Simply because the web interface is excellent, and it allows the user
>to do most of their own management easily,
I agree that mailman's web interface is excellent. And given the
choice between an MLM with an e-mail only interface, and one with a
good web interface, too, I'd usually prefer the latter. Unfortunately,
it's not always practical for the list server to be co-located with a
web server.
For example, we have two list servers: one for internal use, and one
for internal+external use. The internal server happens to also be our
main internal web server. (Yay!) But implementing new CGI's on this
system is an exercise in frustration. First you have to request an
account on the development system. Then you have to implement and test
the CGI on the development system using a CGI wrapper. If you need
something python installed on both the development and production
systems, that's additional work and delay. (Boo!)
The public list server is also an FTP and netnews server, but it's not
a web server. It's already got its hands full, so I'm not keen to run
a web server there too, but even if I did, I'd have to essentially get
a license to run it. More work, paperwork, and delay.
So for my immediate need, the web interface is not important. The MLM
has to "win" on the basis of its e-mail interface alone.
>and the confirmation system
>avoids problems with malicious subscription.
Majordomo (optionally) and ezmlm (automatically) do this, too. I'm
sure listar does, too. This is a must-have feature, these days.
>Users don't in general
>want to bother the manager if they can solve their own problem with a
>button click. Yes, I know other list servers have these features as
>optional packages, but in Mailman they are well-integrated.
My qmail+majordomo+bounceman set-up requires *very* little owner
intervention. The major remaining problem area is people who can't
remember their subscription address and therefore can't unsubscribe.
Does Mailman have any better way of handling this? That doesn't
require the use of the web interface?
>Mailman seems to have the best features. I really like the way that
>Mailman figures out what host URL it's being called from and presents the
>appropriate list choices for that URL. I have the same mailman instance
>running on several virtual web sites, and it does not display the lists
>of one virtual site when called with another virtual site's host name.
Virtual list servers aren't a plus me at the moment.
>Ezmlm is too minimal.
Are you talking about ezmlm, or ezmlm-idx? Plain old ezmlm is pretty
minimal, but the idx package makes it pretty complete--and practically
everyone running ezmlm is using ezmlm-idx. It has no web interface,
but I like the way it does some things, like moderated lists,
especially with multiple moderators.
-Dave
From th at nextel.no Tue Mar 16 18:49:53 1999
From: th at nextel.no (Tor Houghton)
Date: Tue, 16 Mar 1999 18:49:53 +0100 (MET)
Subject: [Mailman-Users] "No module named pythonlib"
Message-ID:
The Cron daemon reports the following for the mailman user:
Traceback (innermost last):
File "/home/mailman/cron/gate_news", line 27, in ?
from Mailman import MailList
File "/home/mailman/Mailman/MailList.py", line 33, in ?
File "/home/mailman/Mailman/ListAdmin.py", line 28, in ?
File "/home/mailman/Mailman/Message.py", line 31, in ?
ImportError: No module named pythonlib
(Had I noticed this earlier, I would have included it in my previous
mail.)
Cheerio,
Tor Houghton.
From bruce at hams.com Tue Mar 16 18:33:53 1999
From: bruce at hams.com (bruce at hams.com)
Date: 16 Mar 1999 17:33:53 -0000
Subject: [Mailman-Users] Large lists and Postfix
Message-ID: <19990316173353.30279.qmail@hams.com>
From: Dave Sill
> I agree that mailman's web interface is excellent. And given the
> choice between an MLM with an e-mail only interface, and one with a
> good web interface, too, I'd usually prefer the latter. Unfortunately,
> it's not always practical for the list server to be co-located with a
> web server.
You have a point, but I think the best way to deal with this would be to
have Mailman become able to make use of a _remote_MDA_. The admin forms
could simply ask what host to use as the SMTP server for each list. That
would distribute the entire load of list delivery onto one or more
separate MDA hosts, leaving the server upon which Mailman resides with the
load of WWW service, bounce processing, and list management. The actual
subscriber lists should be on the same machine that performs web service,
so that web subscription management will work interactively. Mail delivery
can take place elsewhere. This leaves you with the task of hosting a
virtual domain for list reception and a web server on the same machine,
not much in terms of load but still possibly a pain to get through the
administration of your company. Note however that any unprivileged 486
that you sit on the outside of your firewall would be equal to the task.
> Are you talking about ezmlm, or ezmlm-idx?
I don't know about ezmlm-idx. Smartlist does multiple moderators too, I don't
know if it handles them as well.
Thanks
Bruce
From th at nextel.no Tue Mar 16 19:20:19 1999
From: th at nextel.no (Tor Houghton)
Date: Tue, 16 Mar 1999 19:20:19 +0100 (MET)
Subject: [Mailman-Users] mailman-archive bug?
Message-ID:
Hi (again),
I was browsing the mailman-users archive, and I sorted the March entries
by Author.
I clicked on the following:
[Mailman-Users] Problem with "bin/newlist" Tor Houghton
(http://www.python.org/pipermail/mailman-users/1999-March/000762.html)
But was given an entry by Edgard Castro titled: "[Mailman-Users]
Sugestions" (Tue, 16 Mar 1999 12:37:15 -0500).
Bug or feature?
Tor.
From th at nextel.no Tue Mar 16 19:27:28 1999
From: th at nextel.no (Tor Houghton)
Date: Tue, 16 Mar 1999 19:27:28 +0100 (MET)
Subject: [Mailman-Users] newlist traceback
Message-ID:
I created a symbolic link in /home/mailman/bin/Mailman to point to
/home/mailman/Mailman, and in so doing, I got a little further with the
newlist command. It no longer complains about a missing pythonlib.
However, I now get:
Initial friends password:
Traceback (innermost last):
File "/home/mailman/bin/newlist", line 141, in ?
raise SystemExit(main(sys.argv))
File "/home/mailman/bin/newlist", line 91, in main
newlist.Create(list_name, owner_mail, pw)
File "/home/mailman/Mailman/MailList.py", line 658, in Create
self.Lock()
File "/home/mailman/Mailman/MailList.py", line 1189, in Lock
self._lock_file.lock('w|', 1)
File "/usr/local/lib/python1.5/posixfile.py", line 190, in lock
flock = fcntl.fcntl(self._file_.fileno(), cmd, flock)
IOError: (22, 'Invalid argument')
I am running (as you can see), Python 1.5 on an OpenBSD machine (2.3).
Sorry for this influx of messages, but I am completely new to Python (and
Mailman).
Regards (again),
Tor.
From ptb at home.ptb.org Tue Mar 16 19:29:59 1999
From: ptb at home.ptb.org (Paul Barnett)
Date: Tue, 16 Mar 1999 12:29:59 -0600
Subject: [Mailman-Users] Mass subscription of users -- random password?
Message-ID: <014601be6fdb$165b5d80$0a01a882@barnettpc>
I looked back through the archives, but haven't found anything about this:
I did an "mass subscribe" through the /admin/../members web page for a
single user (which happened to be me), and started wondering what my
password was. Since I didn't specify sending the welcome message, I hadn't
received any sort of response.
So, I went through the web interface, and asked it to send me my password.
I got:
g`OM
I don't know if this was random characters, or an intentionally generated
random password. It did work (and allowed me to change it to something
else), but I was wondering if there was an algorithm to come up with a
random password... if so, I'd suggest a combination of lower case, upper
case, numbers, and MAYBE well known punctuation marks, but only if they not
the beginning or end characters.
That particular character between the "g" and "O" above is potentially very
confusing to a user that doesn't know there is a difference between ` and '
From gstein at lyra.org Tue Mar 16 21:36:00 1999
From: gstein at lyra.org (Greg Stein)
Date: Tue, 16 Mar 1999 12:36:00 -0800
Subject: [Mailman-Users] Large lists and Postfix
References: <19990316173353.30279.qmail@hams.com>
Message-ID: <36EEC0B0.4986C70C@lyra.org>
bruce at hams.com wrote:
>
> From: Dave Sill
> > I agree that mailman's web interface is excellent. And given the
> > choice between an MLM with an e-mail only interface, and one with a
> > good web interface, too, I'd usually prefer the latter. Unfortunately,
> > it's not always practical for the list server to be co-located with a
> > web server.
>
> You have a point, but I think the best way to deal with this would be to
> have Mailman become able to make use of a _remote_MDA_. The admin forms
> could simply ask what host to use as the SMTP server for each list. That
> would distribute the entire load of list delivery onto one or more
> separate MDA hosts, leaving the server upon which Mailman resides with the
> load of WWW service, bounce processing, and list management. The actual
> subscriber lists should be on the same machine that performs web service,
> so that web subscription management will work interactively. Mail delivery
> can take place elsewhere. This leaves you with the task of hosting a
> virtual domain for list reception and a web server on the same machine,
> not much in terms of load but still possibly a pain to get through the
> administration of your company. Note however that any unprivileged 486
> that you sit on the outside of your firewall would be equal to the task.
There is an "SMTPHOST" configuration option. That would allow you to
shove your outbound delivery over to another machine.
The webserver still needs to run an MTA for incoming mail.
In my setup, I have a mail machine (ns1) and mailing-list machine
(cartman). Inbound mail is forwarded from ns1 over to cartman. cartman
delivers outbound mail, but I could reconfig the system to point at ns1
for outbound.
Note that this doesn't entirely meet Dave's requirement, though: you
still have a web server on the same box as the MLM. This just shoves
mail delivery elsewhere.
Cheers,
-g
--
Greg Stein, http://www.lyra.org/
From bruce at hams.com Tue Mar 16 21:26:13 1999
From: bruce at hams.com (bruce at hams.com)
Date: 16 Mar 1999 20:26:13 -0000
Subject: [Mailman-Users] Large lists and Postfix
Message-ID: <19990316202613.1156.qmail@hams.com>
From: Greg Stein
> There is an "SMTPHOST" configuration option. That would allow you to
> shove your outbound delivery over to another machine.
Good. I'd still like to have it graphicaly configurable for each list,
so that it could be used for "scaling". Then, the limiting factor would
become how many incoming list mails and bounces the MLM machine could
process and archive. I suspect you could run 10 MDA machines flat out
before the MLM machine would have a problem.
> The webserver still needs to run an MTA for incoming mail.
Right.
> Note that this doesn't entirely meet Dave's requirement, though: you
> still have a web server on the same box as the MLM. This just shoves
> mail delivery elsewhere.
You are correct. Dave gets to suffer for his workplace's policies. It's not
clear that accomodating them should be a design goal for Mailman.
Thanks
Bruce
From John at list.org Tue Mar 16 22:30:19 1999
From: John at list.org (John Viega)
Date: Tue, 16 Mar 1999 13:30:19 -0800
Subject: [Mailman-Users] Large lists and Postfix
In-Reply-To: <36EEC0B0.4986C70C@lyra.org>; from Greg Stein on Tue, Mar 16, 1999 at 12:36:00PM -0800
References: <19990316173353.30279.qmail@hams.com> <36EEC0B0.4986C70C@lyra.org>
Message-ID: <19990316133019.A25923@viega.org>
I've set up mailman installs where the MLM was on a separate machine
from the MTA and the web server. Of course, it was an environment
where all the machines used the same OS and were NFS'd together. You
could do a little hacking w/ rsh, etc to get things to work in a less
homogenous environment.
On Tue, Mar 16, 1999 at 12:36:00PM -0800, Greg Stein wrote:
> bruce at hams.com wrote:
> >
> > From: Dave Sill
> > > I agree that mailman's web interface is excellent. And given the
> > > choice between an MLM with an e-mail only interface, and one with a
> > > good web interface, too, I'd usually prefer the latter. Unfortunately,
> > > it's not always practical for the list server to be co-located with a
> > > web server.
> >
> > You have a point, but I think the best way to deal with this would be to
> > have Mailman become able to make use of a _remote_MDA_. The admin forms
> > could simply ask what host to use as the SMTP server for each list. That
> > would distribute the entire load of list delivery onto one or more
> > separate MDA hosts, leaving the server upon which Mailman resides with the
> > load of WWW service, bounce processing, and list management. The actual
> > subscriber lists should be on the same machine that performs web service,
> > so that web subscription management will work interactively. Mail delivery
> > can take place elsewhere. This leaves you with the task of hosting a
> > virtual domain for list reception and a web server on the same machine,
> > not much in terms of load but still possibly a pain to get through the
> > administration of your company. Note however that any unprivileged 486
> > that you sit on the outside of your firewall would be equal to the task.
>
> There is an "SMTPHOST" configuration option. That would allow you to
> shove your outbound delivery over to another machine.
>
> The webserver still needs to run an MTA for incoming mail.
>
> In my setup, I have a mail machine (ns1) and mailing-list machine
> (cartman). Inbound mail is forwarded from ns1 over to cartman. cartman
> delivers outbound mail, but I could reconfig the system to point at ns1
> for outbound.
>
> Note that this doesn't entirely meet Dave's requirement, though: you
> still have a web server on the same box as the MLM. This just shoves
> mail delivery elsewhere.
>
> Cheers,
> -g
>
> --
> Greg Stein, http://www.lyra.org/
>
> ------------------------------------------------------
> Mailman-Users maillist - Mailman-Users at python.org
> http://www.python.org/mailman/listinfo/mailman-users
From lindsey at ncsa.uiuc.edu Tue Mar 16 22:42:39 1999
From: lindsey at ncsa.uiuc.edu (Christopher Lindsey)
Date: Tue, 16 Mar 1999 15:42:39 -0600 (CST)
Subject: [Mailman-Users] Large lists and Postfix
In-Reply-To: <36EEC0B0.4986C70C@lyra.org> from "Greg Stein" at Mar 16, 99 12:36:00 pm
Message-ID: <199903162142.PAA23932@ferret.ncsa.uiuc.edu>
A non-text attachment was scrubbed...
Name: not available
Type: text
Size: 2334 bytes
Desc: not available
Url : http://mail.python.org/pipermail/mailman-users/attachments/19990316/87c21c6f/attachment.asc
From udall at brassica.agronomy.wisc.edu Tue Mar 16 23:58:05 1999
From: udall at brassica.agronomy.wisc.edu (Joshua Udall)
Date: Tue, 16 Mar 1999 16:58:05 -0600
Subject: [Mailman-Users] make update
Message-ID: <199903162312.RAA123860@mail1.doit.wisc.edu>
I'm updating to b9 but I have a few questions. If I don't do 'make
update' rather 'make install' will it successfully overwrite? I'd like
to change some options that were used in the configure process.
I'm not sure I got them correct.
BTW, in the UPGRADING text file - which directs to give the 'make
update' command - it'd be nice to mention that one should run
configure first.
Josh
From ayu1 at nycap.rr.com Wed Mar 17 01:20:19 1999
From: ayu1 at nycap.rr.com (Alex Yu)
Date: Tue, 16 Mar 1999 19:20:19 -0500
Subject: [Mailman-Users] make update
In-Reply-To: <199903162312.RAA123860@mail1.doit.wisc.edu>
Message-ID: <4.1.19990316192001.0097a4d0@pop1.rpi.edu>
At 04:58 PM 1999/3/16 -0600, Joshua Udall wrote:
>I'm updating to b9 but I have a few questions. If I don't do 'make
>update' rather 'make install' will it successfully overwrite? I'd like
Read the Makefile file and you should be able to figure out.
Alex
From soneill at cen.com Wed Mar 17 05:33:49 1999
From: soneill at cen.com (Sean O'Neill)
Date: Tue, 16 Mar 1999 23:33:49 -0500
Subject: [Mailman-Users] [Q]: Problem with trying to create first mailing-list
Message-ID: <000201be702f$5b5dcb80$d933f2cc@sean>
I just installed Mailman on my system and got Apache working correctly but
I'm have a problem getting the bin/newlist python script to work correctly.
I get the following error when trying to run it:
$ /home/mailman/bin/newlist
Traceback (innermost last):
File "/home/mailman/bin/newlist", line 38, in ?
from Mailman.pythonlib import getpass
ImportError: No module named Mailman.pythonlib
I've confirmed that there is a Mailman directory with a pythonlib
subdirectory which contains two files: getpass.py and getpass.pyc. Doing a
file on both of them tells me that the getpass.pyc is a "compiled" python
script.
I'm new at python so be gentle. Hopefully this is something fairly easy.
I've looked at the Installation instructions multiple times and I don't
think I missed anything.
Any help appreciated.
---------------------------------
Sean O'Neill
AppNet, Inc.
Century Computing Division
301-953-3330
soneill at cen.com
soneill at centurycomputing.com
From ray at bigfoot.everett.org Wed Mar 17 05:35:44 1999
From: ray at bigfoot.everett.org (Ray Everett-Church)
Date: Tue, 16 Mar 1999 23:35:44 -0500 (EST)
Subject: [Mailman-Users] archive addresses not disguised
Message-ID:
Running 1.0b9 and all seems well, except that one of my testing lists has
been set up with obscure_addresses turned on. But the addresses are clear
as a bell. Suggestions?
Thanks,
-Ray
From lindsey at ncsa.uiuc.edu Wed Mar 17 16:30:27 1999
From: lindsey at ncsa.uiuc.edu (Christopher Lindsey)
Date: Wed, 17 Mar 1999 09:30:27 -0600 (CST)
Subject: [Mailman-Users] Sugestions
In-Reply-To: <19990316123714.A22397@usmatrix.net> from "Edgard Castro" at Mar 16, 99 12:37:15 pm
Message-ID: <199903171530.JAA29183@ferret.ncsa.uiuc.edu>
A non-text attachment was scrubbed...
Name: not available
Type: text
Size: 1605 bytes
Desc: not available
Url : http://mail.python.org/pipermail/mailman-users/attachments/19990317/e18aa82c/attachment.pot
From de5 at sws5.ctd.ornl.gov Wed Mar 17 16:45:53 1999
From: de5 at sws5.ctd.ornl.gov (Dave Sill)
Date: Wed, 17 Mar 1999 10:45:53 -0500 (EST)
Subject: [Mailman-Users] Large lists and Postfix
In-Reply-To: <199903162142.PAA23932@ferret.ncsa.uiuc.edu>
References: <36EEC0B0.4986C70C@lyra.org>
<199903162142.PAA23932@ferret.ncsa.uiuc.edu>
Message-ID: <14063.52785.784928.754102@sws5.ctd.ornl.gov>
Christopher Lindsey wrote:
>
>Regarding Dave's comments about confirmations... majordomo 1.94.4 (don't
>know about the almost-alpha 2.0) has an easily spoofable confirmation
>key.
Oh? How so? Don't you have to know the $cookie_seed? Or are you
assuming everyone uses the default? I've never had a problem with
faked confirmations.
>Regarding Dave's comments about "The major remaining problem area is
>people who can't remember their subscription address and therefore
>can't unsubscribe..." Look at the Received: headers. :) Seriously, there's
>no other user-specific key added to the messages.
Actually, qmail's VERP helps a lot here--if the user's MTA or MDA puts
the envelope return path in a header field. There are users on systems
that strip some useful headers and fail to include useful
information. I pity them, but it's really not my problem.
-Dave
From bruce at hams.com Wed Mar 17 16:43:06 1999
From: bruce at hams.com (bruce at hams.com)
Date: 17 Mar 1999 15:43:06 -0000
Subject: [Mailman-Users] Sugestions
Message-ID: <19990317154306.29772.qmail@hams.com>
> From: Christopher Lindsey
> > Those are: Disable sorting of recipients when injecting
> > the messages. This will help using less
> > memory when using MTAs like Postfix or QMail.
>
> I don't think this has been addressed as an option yet.
Can someone please explain to me how this is going to have a serious effect
on memory use? If the problem is that some software is going to try to sort
an already-sorted list, I would not suggest worrying about it.
Thanks
Bruce
From ds-list-mailman at sws5.ctd.ornl.gov Wed Mar 17 17:33:47 1999
From: ds-list-mailman at sws5.ctd.ornl.gov (Dave Sill)
Date: Wed, 17 Mar 1999 11:33:47 -0500 (EST)
Subject: [Mailman-Users] Sugestions
In-Reply-To: <19990317154306.29772.qmail@hams.com>
References: <19990317154306.29772.qmail@hams.com>
Message-ID: <14063.55659.453664.28194@sws5.ctd.ornl.gov>
bruce at hams.com wrote:
>> From: Christopher Lindsey
>> >
>> > Those are: Disable sorting of recipients when injecting
>> > the messages. This will help using less
>> > memory when using MTAs like Postfix or QMail.
>>
>> I don't think this has been addressed as an option yet.
>
>Can someone please explain to me how this is going to have a serious effect
>on memory use? If the problem is that some software is going to try to sort
>an already-sorted list, I would not suggest worrying about it.
I can't answer your question, but I can give another reason not to
sort lists. I sort my busiest, most important lists in order of
average delivery time. Since qmail delivers everything in order,
faster recipients get their messages faster and the slow pokes don't
hold them up.
-Dave
From bruce at hams.com Wed Mar 17 17:16:17 1999
From: bruce at hams.com (bruce at hams.com)
Date: 17 Mar 1999 16:16:17 -0000
Subject: [Mailman-Users] Sugestions
Message-ID: <19990317161617.30616.qmail@hams.com>
From: Dave Sill
> I sort my busiest, most important lists in order of
> average delivery time.
MTA should keep track of this information, by MX host. Where are you getting
it, from a log file?
Thanks
Bruce
From lindsey at ncsa.uiuc.edu Wed Mar 17 17:47:50 1999
From: lindsey at ncsa.uiuc.edu (Christopher Lindsey)
Date: Wed, 17 Mar 1999 10:47:50 -0600 (CST)
Subject: [Mailman-Users] Large lists and Postfix
In-Reply-To: <14063.52785.784928.754102@sws5.ctd.ornl.gov> from "Dave Sill" at Mar 17, 99 10:45:53 am
Message-ID: <199903171647.KAA29665@ferret.ncsa.uiuc.edu>
A non-text attachment was scrubbed...
Name: not available
Type: text
Size: 936 bytes
Desc: not available
Url : http://mail.python.org/pipermail/mailman-users/attachments/19990317/01708e86/attachment.asc
From ds-list-mailman at sws5.ctd.ornl.gov Wed Mar 17 18:17:05 1999
From: ds-list-mailman at sws5.ctd.ornl.gov (Dave Sill)
Date: Wed, 17 Mar 1999 12:17:05 -0500 (EST)
Subject: [Mailman-Users] Sugestions
In-Reply-To: <19990317161617.30616.qmail@hams.com>
References: <19990317161617.30616.qmail@hams.com>
Message-ID: <14063.58257.899890.854152@sws5.ctd.ornl.gov>
bruce at hams.com wrote:
>From: Dave Sill
>>
>> I sort my busiest, most important lists in order of
>> average delivery time.
>
>MTA should keep track of this information, by MX host. Where are you getting
>it, from a log file?
The qmailanalog package processes qmail log files and produces various
statistics. It does delivery time by recipient, though, not by MX.
Here's a sample:
Recipients
One line per recipient. Information on each line:
* sbytes is the number of bytes successfully delivered to this recipient.
* mess is the number of messages sent to this recipient (success plus failure).
* tries is the number of delivery attempts (success, failure, deferral).
* xdelay is the total xdelay incurred by this recipient.
sbytes mess tries xdelay recipient
25849 30 30 18.03 local.alpha-osf-managers-approval at sws1.ctd.ornl.gov
197935 74 74 385.09 local.alpha-osf-managers-decbounce at sws1.ctd.ornl.gov
698068 162 162 71.37 local.alpha-osf-managers-owner- at sws1.ctd.ornl.gov
...
197935 74 74 890.78 remote.BISSxx at AESIR.MIT.EDU
197935 74 74 1901.97 remote.BIVIxx at nebeng.otis.utc.com
197935 74 74 968.25 remote.Bxx at OREGON.UOREGON.EDU
197935 74 74 1162.04 remote.BOICHxx at OSMIUM.llnl.gov
197935 74 74 1044.84 remote.BOLSxx at frango.hs.washington.edu
197935 74 74 901.39 remote.BRIAxx at tiger.hsc.edu
3859 1 1 7.86 remote.BRYAN.SKOKxx at em.doe.gov
197935 74 74 457.39 remote.Bxx at novo.dk
197935 74 76 1161.35 remote.B_Oxx at VENUS.TWU.EDU
...
And an example of why MX isn't granular enough:
197935 74 74 1071.19 remote.elynxx at MIT.EDU
197935 74 74 301.06 remote.gyouxx at MIT.EDU
-Dave
From castro at usmatrix.net Wed Mar 17 18:18:31 1999
From: castro at usmatrix.net (Edgard Castro)
Date: Wed, 17 Mar 1999 12:18:31 -0500
Subject: [Mailman-Users] Sugestions
In-Reply-To: <19990317154306.29772.qmail@hams.com>; from bruce@hams.com on Wed, Mar 17, 1999 at 03:43:06PM -0000
References: <19990317154306.29772.qmail@hams.com>
Message-ID: <19990317121831.A10588@usmatrix.net>
On Wed, Mar 17, 1999 at 03:43:06PM -0000, bruce at hams.com wrote:
> Those are: Disable sorting of recipients when injecting
> the messages. This will help using less
> memory when using MTAs like Postfix or QMail.
>
> I don't think this has been addressed as an option yet.
>
> Can someone please explain to me how this is going to have a serious effect
> on memory use? If the problem is that some software is going to try to sort
> an already-sorted list, I would not suggest worrying about it.
The problem is not sort a already-sorted list, but SORT at all.
Postfix and QMail for example, have a better way to delivery
the messages without the need to sort them by domain or any
other kind of kludge. Sorting by domain is GREAT if you use
sendmail, but with other MTAs, is just a waste of resources.
e.
--
Edgard Castro | 305-273-5003 -o)
Chief Network Engineer - U.S. Matrix Internet, Inc. /\\
Linux Counter User #29078 -- http://counter.li.org _\_V
From bruce at hams.com Wed Mar 17 18:32:03 1999
From: bruce at hams.com (bruce at hams.com)
Date: 17 Mar 1999 17:32:03 -0000
Subject: [Mailman-Users] Sugestions
Message-ID: <19990317173203.32527.qmail@hams.com>
> And an example of why MX isn't granular enough:
>
> 197935 74 74 1071.19 remote.elynxx at MIT.EDU
> 197935 74 74 301.06 remote.gyouxx at MIT.EDU
I don't understand why an MX would have different delays per _recepient_.
In general it accepts the mail, hangs up the SMTP connection, and _then_
tries to deliver it.
Thanks
Bruce
From ds-list-mailman at sws5.ctd.ornl.gov Wed Mar 17 19:21:29 1999
From: ds-list-mailman at sws5.ctd.ornl.gov (Dave Sill)
Date: Wed, 17 Mar 1999 13:21:29 -0500 (EST)
Subject: [Mailman-Users] Sugestions
In-Reply-To: <19990317173203.32527.qmail@hams.com>
References: <19990317173203.32527.qmail@hams.com>
Message-ID: <14063.62121.92069.529395@sws5.ctd.ornl.gov>
bruce at hams.com wrote:
>> And an example of why MX isn't granular enough:
>>
>> 197935 74 74 1071.19 remote.elynxx at MIT.EDU
>> 197935 74 74 301.06 remote.gyouxx at MIT.EDU
>
>I don't understand why an MX would have different delays per _recepient_.
>In general it accepts the mail, hangs up the SMTP connection, and _then_
>tries to deliver it.
In general, perhaps... In this specific case, that's obviously not
true. These statistics reflect 74 approximately simultaneous
deliveries of the same messages to two addresses at the same
MX. Perhaps mit.edu's mailer doesn't just queue the message and
declare victory. Maybe it waits for the next delivery to succeed. Or
maybe they do per-user spam filtering and elynxx has lots of filters
and gyouxx doesn't. Who knows? The point is that you have to look at
recipients, not MX's to optimize list order for an MTA that doesn't
group by MX.
-Dave
From listadm at metaphor.unh.edu Wed Mar 17 19:25:15 1999
From: listadm at metaphor.unh.edu (The List Server Administrator at UNH)
Date: Wed, 17 Mar 1999 13:25:15 -0500 (EST)
Subject: [Mailman-Users] Feature questions.
Message-ID:
Dear Mailman Users,
I am currently investigating migrating our mailing list services from
CREN ListProc to Mailman. I have a few specific questions which I was
unable to answer by looking at the FAQ or the list archives. Perhaps
someone on this list can help.
o What are the restrictions for the name of the list. I.e. how long
can the list name be and what characters are allowed -- or if it is
a shorter list, disallowed. :-)
BTW I am already aware that Mailman maps the name to lowercase -- a
Good Thing(tm) I would think.
o Is there a built-in user address alias facility? Is it available
server wide? On a list-by-list basis?
For example, we currently have 5 systems that all share the same
usernames and directories -- that is, a user may log into any of
these systems and get their home environment. Now on our current
ListProc server we are able to use a server-wide re-write rule such
that any mail from:
wfc at alberti.unh.edu
wfc at christa.unh.edu
wfc at hopper.unh.edu
etc.
will be changed to the generic:
wfc at cisunix.unh.edu
(There are historical reasons why this isn't already the return
address from these nodes, but in any case...)
This aliasing is done by the ListProc list server very early on in
*all* mail processing, so even if a person subscribes themselves
using one of these specific node names, the re-written address gets
save to the subscriber file instead.
Now this re-writing could be handled at the MTA level, but given
that Mailman also allows subscriptions via a web page, I would
think that this re-write needs to happen within the server's
arena instead to maintain consistency.
So, does this aliasing feature currently exist in Mailman? Or,
given Mailman's architecture, would it seem like a reasonable hack
for a programming literate site-admin to implement?
TIA for your time and help.
Cordially,
The List Server Admin
list.admin at unh.edu
(currently Bill Costa)
From sbolduc at uni-global.com Wed Mar 17 20:48:29 1999
From: sbolduc at uni-global.com (Sylvain Bolduc)
Date: Wed, 17 Mar 1999 14:48:29 -0500
Subject: [Mailman-Users] Mailman 1.0b9, password requirement.
Message-ID: <001c01be70af$2184eda0$989b2fce@sly.uniconseil.com>
Greetings,
Is there a way for users to subscribe via email or web but NOT
to have to specify a password.
Why it's seems absolutely necessary ? For security reasons ?
Some other mailling list program doesn't require it, they are just looking
for the Sender Address that match the one into the database.
For flexibility, it could be good to have and option into Mailman
that let us to enable/disable it.
Or if we leave that field empty, by default have one autogenerated in that case.
(Just alike the Mass Subscribe Members options)
Best regards
Sylvain Bolduc
UniGlobal
1801, McGill College, suite 1010
Montreal (Quebec) H3A 2N4
Tel: (514) 840-1158 poste 351
Fax: (514) 840-1166
sbolduc at uni-global.com
http://www.uni-global.com
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.python.org/pipermail/mailman-users/attachments/19990317/893c200f/attachment.htm
From wang at allmalt.cs.uwm.edu Wed Mar 17 21:42:34 1999
From: wang at allmalt.cs.uwm.edu (Dr. Yongge Wang)
Date: Wed, 17 Mar 1999 14:42:34 -0600 (CST)
Subject: [Mailman-Users] New Bie
Message-ID: <199903172042.OAA18631@allmalt.cs.uwm.edu>
Hi, I have just installed mailman lists.
when I create a new test list: wangtest
then it told me to add the following lines
to /etc/aliases:
## wangtest mailing list
## created: 17-Mar-1999 root
wangtest: "|/usr/mailman/mail/wrapper post wangtest"
wangtest-admin: "|/usr/mailman/mail/wrapper mailowner wangtest"
wangtest-request: "|/usr/mailman/mail/wrapper mailcmd wangtest"
owner-wangtest: wangtest-admin
wangtest-owner: wangtest-admin
And I have added these lines.
Then when I try to subscribe to this list, it goes on and
ask me to confirm.
Then I confirmed by sending an email to:
wangtest-request at ...
And wrong things happens. it says that
|/usr/mailman/mail/wrapper mailcmd wangtest has
permament error. Indeed, when I send an email to
wangtest-request at ...
with help subject line. The same error appears.
Could any one tell me what is wrong here?
Perhpas the reason is that i must replace the
"mailcmd" and "mailowner" with some other thing??
what will be that??
Thank you very much for your help!!
Another question: When I create a separate
mailman.aliases in mailman directory, and
put the line OA/...mailman
in sendmail.cf. I will get some wrong things
since when I run newaliases, then it will complain that
cannot find : mailman.aliases.db
and also then my mail system will be wrong.
How could i generate the database: mailman.aliases.db
Thank you very much for your help in advance
Sincerely,
Yongge Wang
From ayu1 at nycap.rr.com Thu Mar 18 01:33:24 1999
From: ayu1 at nycap.rr.com (Alex Yu)
Date: Wed, 17 Mar 1999 19:33:24 -0500
Subject: [Mailman-Users] Mailman 1.0b9, password requirement.
In-Reply-To: <001c01be70af$2184eda0$989b2fce@sly.uniconseil.com>
Message-ID: <4.1.19990317193243.0095e680@pop1.rpi.edu>
At 02:48 PM 1999/3/17 -0500, Sylvain Bolduc wrote:
>
> Some other mailling list program doesn't require it, they are just looking
> for the Sender Address that match the one into the database.
I can spoof your email address very easily. Wanna try it?
Alex
From paz at apriori.net Thu Mar 18 03:06:40 1999
From: paz at apriori.net (paz)
Date: Wed, 17 Mar 1999 21:06:40 -0500 (EST)
Subject: [Mailman-Users] upgrade attempt - error message
Message-ID:
I decided to have a go at upgrading from 0.9 to 1.0b9 tonight. I backed up
the system, and then I read the mail on the mailman-users list for
whatever I could find on upgrading. The instructions seemed to be:
./configure && make
[disable MTA]
make install
make update
[enable MTA]
What I did:
[shut off PPP, effectively disabling mail deliveries to mailman]
su - mailman
cd /home/mailman/mailman-1.0b9 [directory where I'd unzipped new dist]
./configure && make
Results:
=====================
mailman:~/mailman-1.0b9/mailman-1.0b9[128] ./configure && make
loading cache ./config.cache
checking for --with-python...
checking for python... /usr/local/bin/python
checking Python interpreter... /usr/local/bin/python
checking for a BSD compatible install... /usr/bin/install -c
checking whether make sets ${MAKE}... yes
checking for --without-gcc... no
checking for gcc... gcc
checking whether the C compiler (gcc ) works... yes
checking whether the C compiler (gcc ) is a cross-compiler... no
checking whether we are using GNU C... yes
checking whether gcc accepts -g... yes
checking whether #! works in shell scripts... yes
checking for mailman UID... mailman
checking for mailman GID... 1502
checking permissions on /home/mailman... configure: error:
***** Installation directory /home/mailman is not configured properly!
***** Set-gid bit must be set for directory: /home/mailman
mailman:~/mailman-1.0b9/mailman-1.0b9[129]
=====================
Looked at mailman home directory:
=====================
drwxrwxr-x 16 mailman mailman 1024 Mar 17 20:36 .
drwxr-xr-x 11 root wheel 512 Mar 3 20:49 ..
-rw-r--r-- 1 mailman mailman 9347 Feb 5 13:45 .addressbook
-rw-r--r-- 1 mailman mailman 17686 Feb 5 13:46 .addressbook.lu
-rw-r--r-- 1 mailman mailman 260 Feb 2 19:18 .alias
-rw-r--r-- 1 mailman mailman 907 Feb 2 19:19 .cshrc
-rw-r--r-- 1 mailman mailman 509 Feb 2 19:17 .cshrc.orig
-rw------- 1 mailman mailman 2218 Feb 15 09:25 .history
-rw-r--r-- 1 mailman mailman 585 Feb 2 19:17 .login
-rw-r--r-- 1 mailman mailman 561 Feb 2 19:17 .login.orig
-rw-r--r-- 1 mailman mailman 139 Feb 2 13:01 .login_conf
-rw------- 1 mailman mailman 351 Feb 2 13:01 .mail_aliases
-rw-r--r-- 1 mailman mailman 313 Feb 2 13:01 .mailrc
-rw-r--r-- 1 mailman mailman 10963 Feb 6 04:34 .pinerc
-rw-r--r-- 1 mailman mailman 749 Feb 2 13:01 .profile
-rw------- 1 mailman mailman 257 Feb 2 13:01 .rhosts
-rw-r--r-- 1 mailman mailman 832 Feb 2 13:01 .shrc
drwxrwsr-x 6 root mailman 1536 Feb 3 05:44 Mailman
drwxrwsr-x 4 root mailman 512 Feb 2 13:31 archives
drwxrwsr-x 2 root mailman 512 Feb 3 02:26 bin
drwxrwsr-x 2 root mailman 512 Feb 3 02:27 cgi-bin
drwxrwsr-x 2 root mailman 512 Feb 3 02:26 cron
drwxrwsr-x 2 root mailman 512 Mar 17 12:00 data
drwxrwsr-x 2 root mailman 512 Feb 3 02:26 filters
drwxrwsr-x 6 root mailman 512 Feb 20 18:04 lists
drwxrwsr-x 2 root mailman 512 Mar 17 20:42 locks
drwxrwsr-x 2 root mailman 512 Mar 4 01:43 logs
drwxrwsr-x 2 root mailman 512 Feb 5 13:50 mail
drwxr-xr-x 3 mailman mailman 512 Mar 17 20:38 mailman-1.0b9
drwxrwsr-x 2 root mailman 512 Feb 3 02:26 scripts
drwxrwsr-x 2 root mailman 1024 Feb 18 19:21 templates
=====================
The only weird thing is that it looks like I had installed mailman as
root. Shall I chown -R mailman * from the mailman home directory and try
this again, or will this hose things up?
From jwm at plain.co.nz Thu Mar 18 03:33:24 1999
From: jwm at plain.co.nz (John Morton)
Date: Thu, 18 Mar 1999 15:33:24 +1300 (NZDT)
Subject: [Mailman-Users] Who's collecting bounce messages?
Message-ID: <14064.25984.206896.881028@antares.plain.co.nz>
Who out there is collecting bounce messages to improve Mailman's
handling of them? I've got a few produced by PMDF (some sort of vms
MTA) that aren't being picked up properly.
John.
From che at debian.org Thu Mar 18 03:43:50 1999
From: che at debian.org (Ben Gertzfield)
Date: 17 Mar 1999 18:43:50 -0800
Subject: [Mailman-Users] Ignoring administrative requests from certain addresses
Message-ID:
Is it possible, with Mailman, to have the system automatically
send any subscribe/unsubscribe/etc. requests from, say, .*@hotmail\.com
to /dev/null?
I was trying to convince a friend of mine to switch and he wanted
this functionality. I know mailman can send *normal* list messages
from certain addresses to the bitbucket, but how about subscribes
and whatnot?
Ben
--
Brought to you by the letters I and F and the number 3.
"Elate means having wings."
Debian GNU/Linux maintainer of Gimp and GTK+ -- http://www.debian.org/
I'm on FurryMUCK as Che, and EFNet/Open Projects IRC as Che_Fox.
From clark.evans at manhattanproject.com Thu Mar 18 06:32:36 1999
From: clark.evans at manhattanproject.com (Clark Evans)
Date: Thu, 18 Mar 1999 05:32:36 +0000
Subject: [Mailman-Users] Sendmail virtual hosting
References:
Message-ID: <36F08FF4.3C2E0C7@manhattanproject.com>
How do I get mailman to work with sendmail
virtual hosting.
is /etc/aliases still used?
Can I put an entry in /etc/virtual that referenses
an alias in /etc/aliases ?
Thanks!
Clark
From jwm at plain.co.nz Thu Mar 18 06:48:25 1999
From: jwm at plain.co.nz (John Morton)
Date: Thu, 18 Mar 1999 18:48:25 +1300 (NZDT)
Subject: [Mailman-Users] Sendmail virtual hosting
In-Reply-To: <36F08FF4.3C2E0C7@manhattanproject.com>
References:
<36F08FF4.3C2E0C7@manhattanproject.com>
Message-ID: <14064.37706.164901.420044@antares.plain.co.nz>
Clark Evans writes:
> How do I get mailman to work with sendmail
> virtual hosting.
>
> is /etc/aliases still used?
>
> Can I put an entry in /etc/virtual that referenses
> an alias in /etc/aliases ?
That's what I do, and it works fine. Though I guess you could cut out
the aliases file and just have the line 'name at domain command' , but
I've never tried it.
John.
From clark.evans at manhattanproject.com Thu Mar 18 08:23:17 1999
From: clark.evans at manhattanproject.com (Clark Evans)
Date: Thu, 18 Mar 1999 07:23:17 +0000
Subject: [Mailman-Users] Sendmail virtual hosting
References:
<36F08FF4.3C2E0C7@manhattanproject.com> <14064.37706.164901.420044@antares.plain.co.nz>
Message-ID: <36F0A9E5.FE917969@manhattanproject.com>
Clark Evans writes:
> How do I get mailman to work with sendmail
> virtual hosting. Is /etc/aliases still used?
> Can I put an entry in /etc/virtual that referenses
> an alias in /etc/aliases ?
John Morton replied:
> That's what I do, and it works fine. Though I guess you could cut out
> the aliases file and just have the line 'name at domain command' , but
> I've never tried it.
I had to explain to my ISP why so many entries in these
files. Since the explanation may be useful to someone,
here is a copy of it.
:) Clark
---------------------------------------------------------
Let's take a sample list and work through it so that it
is clearly understandable.
Take the announcement list "announce".
Let's say someone subscribes. To do this, they go to the
web page, fill out a form, and press "OK". It sends e-mail
to them asking them to 'confirm'. The return address for this
confirmation is 'announce-request at jos.org'. In their mailer,
they reply to the message and press send. The message goes
to jos.org, destined for announce-request.
The goal is for the reply message to somehow get piped
to '/sites/jos/mailman/mail/wrapper mailcmd announce'.
This is a program with two arguments, one of them stating
that it's a request, and the other argument stating what
list the request is for.
To have mail re-directed to a program, an entry like
test-list: "|/dir/program-to-run"
must exist in the /etc/mail/aliases file.
Therefore, an entry, say
announce-request: "|/sites/jos/mailman/mail/wrapper mailcmd announce"
needs to be in /etc/mail/aliases. For the message to
be delivered. Now. There is a problem, first
announce-request may not be unique to the server,
second the virtual mapping
@jos.org warpii at spin.de
Will catch the message to announce-request at jos.org
and send it to Markus before it get's to the
alias file! Bad.
THUS, we need an entry in the /etc/mail/virtual file
to make sure that the message for this e-mail address
makes it to the alias file so that it can be re-directed
to the computer program *wheu*
SO, to kill two birds with one stone, we re-name the
"announce-request" in the /etc/mail/aliases file and make
it jos specific, hence, list-jos-announce-request.
THEN, we add "announce-request at jos.org" to the
/etc/mail/virtual file to point to list-jos-announce-request.
/etc/mail/aliases:
list-jos-announce-request: "|/sites/jos/mailman/mail/wrapper mailcmd announce"
/etc/mail/virtual:
announce-request at jos.org list-jos-announce-request
THEREFORE, mail sent to announce-request at jos.org gets re-directed
to the program /sites/jos/mailman/mail/wrapper with the
arguments "mailcmd announce".
Now, once this program get's the mail, it sends out a 'Hurray!'
thank you for joining message. This message has a reply-to
address of 'announce-admin at jos.org' -- just in case the user
has complaints. If the person replys to this message, it goes
to "|/sites/jos/mailman/mail/wrapper mailowner announce" to be
put in a basket for the administator for the list (who ever it
is at the time ) to answer questions about the list.
Let's assume our user is happy and wants to send a message
to the list. He sends it to "announce at jos.org", which goes to
/sites/jos/mailman/mail/wrapper with the arguments "post announce"
So, we have the same pattern once again.
THUS, for each list, we have three entries in the /etc/mail/aliases :
The first for 'requests' (automated)
The second one for 'admin' (Stored for the administrator's review)
The third one is for 'posts' (Sent out to the list at large)
THEN, for each entry in /etc/mail/aliases we have one in /etc/mail/virtual
This is necessary to re-direct from xxx at jos.org to list-jos-xxxx
ALSO, other common ways for an end user to access the list administrator
is by a 'owner', so there are two more /etc/mail/virtual entries,
'owner-announce' and 'announce-owner'. This is for backward compatibility
with the many mailers out there that have an 'owner' for the list.
THEREFORE, for each list, there is (3) /etc/mail/aliases entries,
and (5) /etc/mail/virtual entries.
Multiply this by the number of lists (11) and you have a large
number of entries. However, I don't think that it's really all
that big of a deal.
Hope this help explains the request. Please consider the entries.
Thank you!
Clark Evans
-------------------------------------------------------------
Additional entries for /etc/mail/virtual
-------------------------------------------------------------
announce at jos.org list-jos-announce
announce-admin at jos.org list-jos-announce-admin
announce-request at jos.org list-jos-announce-request
owner-announce at jos.org list-jos-announce-admin
announce-owner at jos.org list-jos-announce-admin
-------------------------------------------------------------
Additional Entries for /etc/mail/aliases -------------------------------------------------------------
list-jos-announce: "|/sites/jos/mailman/mail/wrapper post announce"
list-jos-announce-admin: "|/sites/jos/mailman/mail/wrapper mailowner announce"
list-jos-announce-request: "|/sites/jos/mailman/mail/wrapper mailcmd announce"
From nathan at robotics.net Thu Mar 18 19:58:32 1999
From: nathan at robotics.net (Nathan Stratton)
Date: Thu, 18 Mar 1999 13:58:32 -0500 (EST)
Subject: [Mailman-Users] wrapper error
Message-ID:
I upgraded from 1.06 to 1.09 and everything went well, or so I thought.
When uses now send mail to my lists I get a mail wrapper error. I think it
has something to do with permission, but I am not sure.
-rwxr-sr-x 1 root mailman 21075 Mar 14 22:05 wrapper
I also tried chaning the owner to mailman.mailman and I get the same
error.
This is what is sent to postmaster:
>From MAILER-DAEMON at ginger.robotics.net Thu Mar 18 13:51:19 1999
Date: Thu, 18 Mar 1999 10:58:12 -0500 (EST)
From: Mail Delivery Subsystem
To: nathan at ginger.robotics.net
Subject: Returned mail: unknown mailer error 2
The original message was received at Thu, 18 Mar 1999 10:58:09 -0500 (EST)
from c697754-a.baden1.pa.home.com [24.1.42.229]
----- The following addresses had permanent fatal errors -----
"|/home/mailman/mail/wrapper post cleclist"
(expanded from: )
----- Transcript of session follows -----
554 "|/home/mailman/mail/wrapper post cleclist"... unknown mailer error 2
--
Check out the new CLEC mailing list at http://www.robotics.net/clec
><>
Nathan Stratton Telecom & ISP Consulting
http://www.robotics.net nathan at robotics.net
From rbelk at noc.umsmed.edu Thu Mar 18 20:40:25 1999
From: rbelk at noc.umsmed.edu (Randy Belk)
Date: Thu, 18 Mar 1999 13:40:25 -0600
Subject: [Mailman-Users] Proxy Problems and Wish List !
Message-ID:
There seems to be an issue if you access mailman's pages using a web browser that has a "proxy" entry set and that proxy also does caching. Can the automated Web Pages insert the "no cache" tag when the page is generated?
From listadm at metaphor.unh.edu Thu Mar 18 21:28:04 1999
From: listadm at metaphor.unh.edu (The List Server Administrator at UNH)
Date: Thu, 18 Mar 1999 15:28:04 -0500 (EST)
Subject: [Mailman-Users] Sending a human message to the LISTNAME-request address.
Message-ID:
In investigating the differences between Mailman and ListProc, I
noticed that the use of the `standard' list aliases are not the same.
Specifically Mailman uses the "-request" address for contacting the
server, while ListProc uses this address for contacting the list's
human admin.
Now I've seen different list servers go different ways on this issue.
For example I think Majordomo takes the same approach as Mailman where
this particular alias is concerned.
But ignoring this issue, look at error message I received from the
server when I sent my `human message' to the "-request" address
(see below). Don't you think it would make sense for the server
to include some sort of information on how to get help? If I'm a
typical user, all I'll know from this reply is that my message
apparently went to a robot, not a person. But know I don't know
who I should contact or what I should do for help.
While we're talking about error messages, a question. Are the
various text strings used for generating error messages and
diagnostics sent to users kept separate from the code? I'll
briefly mention that I think such a feature is vital. Separating
the displayed text from the code allows for much easier
internationalization as well as making it easier for different
sites to customize the messages as necessary to reflect local
policies. Having the error and diagnostic messages hardwired into
the server code is probably my number one frustration with
ListProc. In that particular case, the supplied error messages are
rarely helpful, but there is little that I, as a site admin, can
do about it. Of course because the error messages are so bad, I
end up handling many more user questions than I probably should
have to. (But at least each error message contains a blurb on who
to contact for help.)
Just my $0.02 USD.
Cordially,
The List Server Admin
list.admin at unh.edu
(currently Bill Costa)
---------- Forwarded message ----------
Date: Thu, 18 Mar 1999 14:52:25 -0500 (EST)
From: mailman-users-request at python.org
To: listadm at metaphor.unh.edu
Subject: Mailman results for Mailman-Users
**** Subject line ignored: This is a test to see what the server will do...
>>>> Please excuse this test. It appears that Mailman uses the "-request"
**** please: Command UNKNOWN.
>>>> version of the list address differently than other servers I have seen.
**** version: Command UNKNOWN.
>>>> I want to see what will happen if I send a note addressed to a human
**** i: Command UNKNOWN.
>>>> to this address.
**** to: Command UNKNOWN.
>>>> Cordially,
**** cordially,: Command UNKNOWN.
>>>> The List Server Admin
**** the: Command UNKNOWN.
>>>> list.admin at unh.edu
**** list.admin at unh.edu: Command UNKNOWN.
>>>> (currently Bill Costa)
**** (currently: Command UNKNOWN.
From Dionysos at Dionysia.org Thu Mar 18 22:05:41 1999
From: Dionysos at Dionysia.org (Dan Delaney)
Date: Thu, 18 Mar 1999 16:05:41 -0500 (EST)
Subject: [Mailman-Users] Sending a human message to the LISTNAME-request
address.
In-Reply-To:
Message-ID:
On Thu, 18 Mar 1999, The List Server Administrator at UNH wrote:
> Don't you think it would make sense for the server
> to include some sort of information on how to get help?
I have to agree whole heartedly on that one. One of the things I
really like about Petidomo is that if it receives a message with NO
commands it actually replies with a copy of the help file with a
subject of something like "Your request was indeciferable." Now
THAT's friendly. And there's no need for an "end" command on
Petidomo because it simply and quietly ignores lines which are not
commands. If I send it a command and leave my sig in the message,
all I get in reply is the status or result of my command. It does't
tell me that it couldn't understand EVERY SINGLE LINE of my message
(as with Bill's example).
> Separating the displayed text from the code allows for much
> easier internationalization as well as making it easier for
> different sites to customize the messages as necessary to
> reflect local policies.
That is an excellent idea.
--Dan
-----------------------------------------------------------------------
Daniel G. Delaney The Louisville Times Chorus
Dionysos at Dionysia.org www.LouisvilleTimes.org
www.Dionysia.org/~dionysos/ Dionysia Design
ICQ Number: 8171285 www.Dionysia.com/design/
-----------------------------------------------------------------------
"Only two things are infinite: the universe and stupidity--
and I'm not sure about the former."
--Albert Einstein
From gstein at lyra.org Thu Mar 18 22:09:51 1999
From: gstein at lyra.org (Greg Stein)
Date: Thu, 18 Mar 1999 13:09:51 -0800
Subject: [Mailman-Users] Sending a human message to the LISTNAME-request address.
References:
Message-ID: <36F16B9F.7F2B7A15@lyra.org>
Dan Delaney wrote:
>
> On Thu, 18 Mar 1999, The List Server Administrator at UNH wrote:
> > Don't you think it would make sense for the server
> > to include some sort of information on how to get help?
>
> I have to agree whole heartedly on that one. One of the things I
> really like about Petidomo is that if it receives a message with NO
> commands it actually replies with a copy of the help file with a
> subject of something like "Your request was indeciferable." Now
> THAT's friendly. And there's no need for an "end" command on
> Petidomo because it simply and quietly ignores lines which are not
> commands. If I send it a command and leave my sig in the message,
> all I get in reply is the status or result of my command. It does't
> tell me that it couldn't understand EVERY SINGLE LINE of my message
> (as with Bill's example).
Interesting idea, and probably simple to do. Simply set a flag if any
command was recognized. If no command was found, then forget about the
line-by-line response of unknown commands and return the help.
> > Separating the displayed text from the code allows for much
> > easier internationalization as well as making it easier for
> > different sites to customize the messages as necessary to
> > reflect local policies.
>
> That is an excellent idea.
Yah, all great and excellent and whatnot. Somebody has to do it. Either
of you could certainly volunteer to submit the patches to do this in the
post-1.0 release.
Most people don't have the time or inclination to do i18n work in the
FIRST release of a tool. I'm thankful that I have the Mailman package to
begin with. If I had to wait for i18n, then I'd be waiting another year,
if ever. I'd probably be running something else instead and never switch
to Mailman.
Cheers,
-g
--
Greg Stein, http://www.lyra.org/
From clark.evans at manhattanproject.com Thu Mar 18 22:21:35 1999
From: clark.evans at manhattanproject.com (Clark Evans)
Date: Thu, 18 Mar 1999 21:21:35 +0000
Subject: [Mailman-Users] Use "From" not "Sender"
References: <19990131002210.25497.qmail@www06.netaddress.usa.net>
Message-ID: <36F16E5F.8930BC0D@manhattanproject.com>
Hello all. For subscribe messages and such, mailman
uses the "Sender:" rather than "From:" this is bad.
Netscape and many other mailers don't let you
set the "Sender:" and mine is _always_ wrong.
My user name is "clark" on my linux box at home,
I'm using "mail.mindspring.com" for my SMTP server.
Somehow... god know how, I get "Sender: clark at mindspring.com"
and there is _no_ way I can change that (that I know of).
Now. I'm _not_ clark at mindspring.com , yet when I
subscribe and stuff, it sends errors to this poor bloke.
Thus, I respectfully request that Mailman use "From:"
if it is there.
Thank you!
Clark Evans
P.S. I you don't believe me, check the headers
for this message.
From roger at infomed.sld.cu Thu Mar 18 23:34:05 1999
From: roger at infomed.sld.cu (Roger Pe?a Escobio)
Date: Thu, 18 Mar 1999 17:34:05 -0500
Subject: [Mailman-Users] Sending a human message to the LISTNAME-request address.
Message-ID: <01be718f$6e406ad0$1c7001c4@ntserver.sld.cu>
-----Original Message-----
From: The List Server Administrator at UNH
To: mailman-users at python.org
Date: Thursday, March 18, 1999 3:53 PM
Subject: [Mailman-Users] Sending a human message to the LISTNAME-request
address.
> In investigating the differences between Mailman and ListProc, I
> noticed that the use of the `standard' list aliases are not the same.
> Specifically Mailman uses the "-request" address for contacting the
> server, while ListProc uses this address for contacting the list's
> human admin.
>
ListProc uses only one -request address, I think that this is better than
have a lot of -request address (like mailman) all doing (basically) the
same., after all , the list manager is mailman, not the list in question.
> Now I've seen different list servers go different ways on this issue.
> For example I think Majordomo takes the same approach as Mailman where
> this particular alias is concerned.
>
> But ignoring this issue, look at error message I received from the
> server when I sent my `human message' to the "-request" address
> (see below). Don't you think it would make sense for the server
> to include some sort of information on how to get help? If I'm a
> typical user, all I'll know from this reply is that my message
> apparently went to a robot, not a person. But know I don't know
> who I should contact or what I should do for help.
*****-owner at xxxx.xxxx.xxx
I'm looking one interesting things, how could someone look all the list
running on mailman if he dont subscribe to any of them?
sending e-mail to where??
suppose that he dont know nothing about mailman and he dont have direct
access to internet, so he has only one chance: e-mail command
From bruce at hams.com Thu Mar 18 23:41:21 1999
From: bruce at hams.com (bruce at hams.com)
Date: 18 Mar 1999 22:41:21 -0000
Subject: [Mailman-Users] Sending a human message to the LISTNAME-request address.
Message-ID: <19990318224121.11544.qmail@hams.com>
From: "Chuck Swiger"
> ListProc uses only one -request address, I think that this is better than
> have a lot of -request address (like mailman) all doing (basically) the
> same., after all , the list manager is mailman, not the list in question.
You can run Mailman with only one -request address, just edit your aliases
that way. The -admin address is useful if you have different administrators
for different lists.
Thanks
Bruce
From castro at usmatrix.net Fri Mar 19 00:39:20 1999
From: castro at usmatrix.net (Edgard Castro)
Date: Thu, 18 Mar 1999 18:39:20 -0500
Subject: [Mailman-Users] Use "From" not "Sender"
In-Reply-To: <36F16E5F.8930BC0D@manhattanproject.com>; from Clark Evans on Thu, Mar 18, 1999 at 09:21:35PM +0000
References: <19990131002210.25497.qmail@www06.netaddress.usa.net> <36F16E5F.8930BC0D@manhattanproject.com>
Message-ID: <19990318183920.A3035@usmatrix.net>
On Thu, Mar 18, 1999 at 09:21:35PM +0000, Clark Evans wrote:
> Thus, I respectfully request that Mailman use "From:"
> if it is there.
Or, at least, make it an option. I have somewhat
the same situation here. I subscribe to lists using
my email address (castro at usmatrix.net) but I usually
use my linux box to send email, so my Sender goes
with the name of the box and then Mailman does not
recognize me as myself (oh well). :)
e.
--
Edgard Castro
From clark.evans at manhattanproject.com Fri Mar 19 01:30:26 1999
From: clark.evans at manhattanproject.com (Clark Evans)
Date: Fri, 19 Mar 1999 00:30:26 +0000
Subject: [Mailman-Users] Sending a human message to the LISTNAME-request
address.
References: <19990318224121.11544.qmail@hams.com>
Message-ID: <36F19AA2.7334D90D@manhattanproject.com>
bruce at hams.com wrote:
>
> From: "Chuck Swiger"
> > ListProc uses only one -request address, I think that this is better than
> > have a lot of -request address (like mailman) all doing (basically) the
> > same., after all , the list manager is mailman, not the list in question.
>
> You can run Mailman with only one -request address, just edit your aliases
> that way. The -admin address is useful if you have different administrators
> for different lists.
>
How do you do this?
one: "|/mailman/mail/wrapper mailcmd one"
two: "|/mailman/mail/wrapper mailcmd two"
Seems that there is a 1-1 relationship here.
:) Clark
From clark.evans at manhattanproject.com Fri Mar 19 01:32:13 1999
From: clark.evans at manhattanproject.com (Clark Evans)
Date: Fri, 19 Mar 1999 00:32:13 +0000
Subject: [Mailman-Users] Sending a human message to the LISTNAME-request
address.
References: <19990318224121.11544.qmail@hams.com>
Message-ID: <36F19B0D.20E1BA48@manhattanproject.com>
bruce at hams.com wrote:
>
> From: "Chuck Swiger"
> > ListProc uses only one -request address, I think that this is better than
> > have a lot of -request address (like mailman) all doing (basically) the
> > same., after all , the list manager is mailman, not the list in question.
>
> You can run Mailman with only one -request address, just edit your aliases
> that way. The -admin address is useful if you have different administrators
> for different lists.
On a related topic, with majordomo you can unsubscribe from
all of the lists on a server by sending:
"unsubscribe * my-email at my-domain"
Can people do this with Mailman?
:) Clark
From bruce at hams.com Fri Mar 19 01:55:41 1999
From: bruce at hams.com (bruce at hams.com)
Date: 19 Mar 1999 00:55:41 -0000
Subject: [Mailman-Users] Sending a human message to the LISTNAME-request address.
Message-ID: <19990319005541.14995.qmail@hams.com>
> How do you do this?
>
> one: "|/mailman/mail/wrapper mailcmd one"
> two: "|/mailman/mail/wrapper mailcmd two"
OK, one request address _per_list_.
If you are lucky enough to be running qmail, you can use my virtual domain
script and run Mailman with _no_aliases_whatsoever_.
Thanks
Bruce
From roger at infomed.sld.cu Fri Mar 19 02:44:12 1999
From: roger at infomed.sld.cu (Roger Pe?a Escobio)
Date: Thu, 18 Mar 1999 20:44:12 -0500
Subject: [Mailman-Users] Sending a human message to the LISTNAME-request address.
Message-ID: <01be71a9$fd12ec00$1c7001c4@ntserver.sld.cu>
>bruce at hams.com wrote:
>>
>> From: "Chuck Swiger"
>> > ListProc uses only one -request address, I think that this is better
than
>> > have a lot of -request address (like mailman) all doing (basically) the
>> > same., after all , the list manager is mailman, not the list in
question.
>>
>> You can run Mailman with only one -request address, just edit your
aliases
>> that way. The -admin address is useful if you have different
administrators
>> for different lists.
>>
>
>How do you do this?
>
>one: "|/mailman/mail/wrapper mailcmd one"
>two: "|/mailman/mail/wrapper mailcmd two"
>
>Seems that there is a 1-1 relationship here.
the point is that you need to know what lists are running on mailman for
obtein some help or exec any other commands, i think that this wrong.
roger
From paz at apriori.net Fri Mar 19 13:50:23 1999
From: paz at apriori.net (paz)
Date: Fri, 19 Mar 1999 07:50:23 -0500 (EST)
Subject: [Mailman-Users] no mail
Message-ID:
First - Thanks to the kind souls who replied with the helpful advice to my
previous posting - a chown -R and a chmod g+s /home/mailman allowed the
upgrade to work. I fixed one blarf in the install with "--with-cgi-gid",
and the web pages came back.
The only problem now seems to be no mail! I went back throught the
installation messages, and it seems to be using the correct group for
mail; but in my test last evening, the message I "sent out" never got
mailed anywhere.
Hints & comments welcome (besides the obvious "it will be apparent from
reading the makefile/INSTALL/UPGRADE file...").
cheers -
-- Philip.
philip zimmermann paz at apriori.net
www.apriori.net ayer, ma usa
From roger at infomed.sld.cu Fri Mar 19 14:30:59 1999
From: roger at infomed.sld.cu (Roger Pe?a Escobio)
Date: Fri, 19 Mar 1999 08:30:59 -0500
Subject: [Mailman-Users] no mail
Message-ID: <01be720c$b99d7270$1c7001c4@ntserver.sld.cu>
>The only problem now seems to be no mail! I went back throught the
>installation messages, and it seems to be using the correct group for
>mail; but in my test last evening, the message I "sent out" never got
>mailed anywhere.
>
>Hints & comments welcome (besides the obvious "it will be apparent from
>reading the makefile/INSTALL/UPGRADE file...").
first of all look if you have the alias of the list in the aliases file,
look in the maillog file maybe you find some king of trace that allow you to
figure out what is hapen, another thing that you to check is permission of
the wrapper programs, it must be sgid by mailman
>
>cheers -
>-- Philip.
>
>philip zimmermann paz at apriori.net
>www.apriori.net ayer, ma usa
>
>
>
>------------------------------------------------------
>Mailman-Users maillist - Mailman-Users at python.org
>http://www.python.org/mailman/listinfo/mailman-users
From boris at infoplease.com Fri Mar 19 21:08:56 1999
From: boris at infoplease.com (Boris Goldowsky)
Date: Fri, 19 Mar 1999 15:08:56 -0500
Subject: [Mailman-Users] help setting up mailman
Message-ID: <81435F08832BD111844A006097BA974822926F@banana.infoplease.com>
I'm trying to set up mailman to run on a Solaris 2.5.1 system with sendmail
8.9.3. The configure, build, and creation of a test mailing list worked
without errors.
However, when I send any message that tries to use the wrapper function, I
get the following error in syslog:
Mar 19 14:43:51 spaceheater sendmail[18681]: OAA18681:
to="|/opt/mailman/mail/wrapper mailowner mmtest",
delay=00:00:00, xdelay=00:00:00, mailer=prog,
stat=Operating system error
(line breaks added for clarity) There is no other indication that I can
find of what the error might be. There is nothing in mailman/logs/error,
for instance.
This does not appear to be a wrong-GID issue, since when I manually execute
mail/wrapper I see a very different error message, which includes the
expected and actual group ids.
Can anyone explain this error to me?
Thanks,
Bng
From listadm at metaphor.unh.edu Fri Mar 19 21:50:24 1999
From: listadm at metaphor.unh.edu (The List Server Administrator at UNH)
Date: Fri, 19 Mar 1999 15:50:24 -0500 (EST)
Subject: [Mailman-Users] Sending a human message to the LISTNAME-request
address.
In-Reply-To: <01be718f$6e406ad0$1c7001c4@ntserver.sld.cu>
Message-ID:
A couple of preliminaries:
o I'm not sure if a discussion of this nature -- i.e. design
philosophy -- is best done here or on the developers' list.
I worry that end-users may be arguing these fine points but
the developers are blissfully unaware of these proceedings.
o If you have never read it, check out the "Mailing List Management
Software FAQ" which you can find at places like:
http://www.faqs.org/faqs/mail/list-admin/software-faq/
ftp://ftp.uu.net/usenet/news.answers/mail/list-admin/software-faq
It contains lots of good information about list servers in
general and has reviews of what is currently out there.
Unfortunately I have not found a copy that has been updated
any later than 1995.
o In some respects I feel like I should not get too excited about
this issue of how you communicate with the server via e-mail,
given Mailman's web orientation. But I think it is an issue
that should be delt with sooner than later with a firm
understanding as to why various decisions where made.
Now...
Thanks for all the replies regarding list aliases and error
messages. Here are my thoughts on some of the issues raised.
Chuck Swiger posted (in part):
CS> ListProc uses only one -request address, I think that this is better than
CS> have a lot of -request address (like mailman) all doing (basically) the
CS> same., after all , the list manager is mailman, not the list in question.
Hmmmm, I think I know what you're saying here, but let's establish
some definitions. Looking at some other mailers and what they do
for addresses we have:
LISTSERV ListProc Majordomo Smartlist
TO CONTACT -------------- -------------- -------------- -------------
server listserv listproc majordomo *-request
list owner *-request *-request ? ?
delivery err owner-* owner-* ? ?
all owners all-request n/a ? ?
site admin owner-listserv n/a ? ?
LISTSERV, ListProc, and Majordomo all have the idea of a central
server address for contacting the server. Smartlist and Mailman
seem to instead use the model that there is (what appears to be) a
server dedicated to each list. An advantage to having server
commands go to the *-request address is that by virtue of the fact
that you are mentioning the list, you have established a context
within which to interpret commands. But with Mailman, the only
thing that will seem to pull everything together, i.e. display a
set of lists available from the server, will be the web page. Or
am I missing something?
I would rather see the *-request address used for contacting a
human since that is how it is most widely used. If the desire is
to have a model of a `separate server for every list', I would
recommend using *-server instead. As it turns out LISTSERV also
uses this address format as well, and I don't believe it
contradicts any other existing use for this form of address.
I had previously posted:
BC> > [speaking as the user] But know [SIC] I don't know
BC> > who I should contact or what I should do for help.
to which Roger Escobio posted:
RE> *****-owner at xxxx.xxxx.xxx
Sure as a list admin -- I know that. You know that. But how does
the *user* know that? There is nothing in the e-mail message sent
back that tells the user who to go to for help or make suggestions
what to do next:
From mailman-users-request at python.org Fri Mar 19 08:35:04 1999
Date: Thu, 18 Mar 1999 14:52:25 -0500 (EST)
From: mailman-users-request at python.org
To: listadm at metaphor.unh.edu
Subject: Mailman results for Mailman-Users
**** Subject line ignored: This is a test...
>>>> Please excuse this test.
**** please: Command UNKNOWN.
The only addresses I see here are my own (listadm) and the list's
server address (mailman-users-request). That's it. There isn't
even a hint that I could send back a HELP command to the server
address. No mention of how to contact a human. Nothing.
Here is an extract from a Majordomo response to a similar problem:
From Majordomo at pop.psu.edu Fri Mar 19 14:44:00 1999
Date: Fri, 19 Mar 1999 12:23:16 -0500 (EST)
From: Majordomo at pop.psu.edu
To: List.Admin at unh.edu
Subject: Majordomo results
--
>>>> get file majordomo-faq
**** get: unknown list 'file'.
>>>>
**** Help for Majordomo at pop.psu.edu:
This help message is being sent to you from the Majordomo
mailing list management system at Majordomo at pop.psu.edu.
This is version 1.94.4 of Majordomo.
If you're familiar with mail servers, an advanced user's
summary of Majordomo's commands appears at the end of this
message.
etc.
I believe it was Roger Escobio who also wrote:
RE> I'm looking one interesting things, how could someone look all the
RE> list running on mailman if he dont subscribe to any of them? sending
RE> e-mail to where??
This is a good point. While Mailman seems to be following the
Smartlist model for aliases, most servers seem to offer a central
address for contacting the server without the context of any given
list. Perhaps it would be trivial to do the same thing with
Mailman -- but is that indeed part of the current installation
standard?
In any case, going back to Chuck's original comment, I tend to
agree that I like LISTSERV/ListProc model better, where there is a
single address for the server, and all list addresses are really
unique for that list. But I don't know if we're in a position to
change the current Mailman approach radically given the current
stage of development. But it would be nice to have that approach
and philosophy spelled out. I think one of the worst things about
ListProc, reflected by its chaotic command structure, is that grew
by virtue of creeping featurism. To me ListProc doesn't look like
it was designed, it just grew like a weed.
Bruce take on the multiple aliases:
B> You can run Mailman with only one -request address, just edit your aliases
B> that way. The -admin address is useful if you have different administrators
B> for different lists.
We might be getting to an important point -- what type of model is
Mailman aimed at? Are we talking about an installation where the
list server admin is managing all of the lists, or a situation
where the lists are all `owned' by different (often non-technical)
people who are managing their own lists. In our particular case,
we have over 400 lists on our server, so we are definitely the
latter. So you do need a unique address for contacting that
list's manager. But in the case of ListProc it is using the
"*-request" alias for that, and establishes a single address for the
server itself.
Clark Evans wrote regarding the
idea of having a single address for the server:
CE> How do you do this?
CE>
CE> one: "|/mailman/mail/wrapper mailcmd one"
CE> two: "|/mailman/mail/wrapper mailcmd two"
This brings me back to my question of -- does Mailman have a
central server address, for contacting the server without respect
to a given list. I said that I thought establishing such an
address would be trivial, i.e.:
mailman: ""|/mailman/mail/wrapper mailcmd mailman"
Where mailman is a phantom list. But now we have established a
list context which may cause problems depending upon the command
given. What is needed is a list contextless way of contacting the
server.
Clark also followed up:
CE> On a related topic, with majordomo you can unsubscribe from
CE> all of the lists on a server by sending:
CE>
CE> "unsubscribe * my-email at my-domain"
CE>
CE> Can people do this with Mailman?
Precisely! ListProc has a similar feature.
Finally Bruce followed up to Clark's posting:
CE> How do you do this? (one address for the server)
B> If you are lucky enough to be running qmail, you can use my virtual domain
B> script and run Mailman with _no_aliases_whatsoever_.
I think this misses the point. The issue here is the address
model presented to the end-user, not how this address model is
implemented.
In summary, there doesn't appear to be much standardization on
this issue of list and server addresses. From the Norm Aleks
Mailing List Management Software FAQ:
"A standardized addressing system would be very useful, there's
no doubt. Everyone has their own opinion on what the best
choice would be, and I have mine, but most of all I would like
to see the MLMs' writers agree on *some* standard. Just my
opinion -- but it certainly would help the end users."
I think there is a lot to be said for following the mostly widely
used conventions which I believe would be
*-request at somewhere.org
for contacting the list owner. But more important, I think, is the
establishment of the model Mailman is going to follow, articulate
that model, and then design the software accordingly.
Sorry about the length.
Cordially,
The List Server Admin
list.admin at unh.edu
(currently Bill Costa)
From Dionysos at Dionysia.org Fri Mar 19 22:09:12 1999
From: Dionysos at Dionysia.org (Dan Delaney)
Date: Fri, 19 Mar 1999 16:09:12 -0500 (EST)
Subject: [Mailman-Users] Sending a human message to the LISTNAME-request
address.
In-Reply-To:
Message-ID:
Bill wrote:
> I like LISTSERV/ListProc model better, where there is a
> single address for the server, and all list addresses are really
> unique for that list. But I don't know if we're in a position to
I like that approach as well. But some people like it the other way.
That's one of the nice things about majordomo, it goes both ways (so
to speak :-). If I have a majordomo list called "mylist", the
someone can subscribe one of two ways. They can send a message to
"majordomo at mydomain.com" with "subscribe myemail at anotherdomain.com"
in the message, or they can send a message to
"mylist-request at mydomain.com" with just "subscribe" in the message.
That's awefully nice. Mailman, however, limits the user to
the latter method, making "mailman at mydomain.com" go to the Mailman
administrator. I think it would make more sense to just stick with
"mailman-owner at mydomain.com", or even add
"mailman-admin at mydomain.com" and make "mailman at mydomain.com" an
address for sending commands to the user. If this is supposed to
replace majordomo, that's what a lot of majordomo users are used
to.
As far as "mylist-request at mydomain.com" goes, I don't see a
problem with that list going to the server. I think "mylist-owner"
is meant for messages to a human. You are sending a "request" to the
server when sending to "mylist-request", and sending a message to
the owner of the list at "mylist-owner".
--Dan
____________________________________________________________________________
Daniel G. Delaney * Dionysos at Dionysia.org * www.Dionysia.org/~dionysos
PGP Public Key: http://Dionysia.org/~dionysos/pgp5.html
"Only two things are infinite: the universe and stupidity--
and I'm not sure about the former."
--Albert Einstein
From roger at infomed.sld.cu Fri Mar 19 21:54:55 1999
From: roger at infomed.sld.cu (=?iso-8859-1?Q?Roger_Pe=F1a_Escobio?=)
Date: Fri, 19 Mar 1999 15:54:55 -0500
Subject: [Mailman-Users] help setting up mailman
Message-ID: <01be724a$be1f9600$1c7001c4@ntserver.sld.cu>
-----Original Message-----
From: Boris Goldowsky
To: 'mailman-users at python.org'
Date: Friday, March 19, 1999 3:33 PM
Subject: [Mailman-Users] help setting up mailman
>I'm trying to set up mailman to run on a Solaris 2.5.1 system with sendmail
>8.9.3. The configure, build, and creation of a test mailing list worked
>without errors.
>
>However, when I send any message that tries to use the wrapper function, I
>get the following error in syslog:
>
>Mar 19 14:43:51 spaceheater sendmail[18681]: OAA18681:
> to="|/opt/mailman/mail/wrapper mailowner mmtest",
> delay=00:00:00, xdelay=00:00:00, mailer=prog,
> stat=Operating system error
>
make a link to the mailman wrapper programs in the /etc/smrsh dir
#ln -s /home/mailman/mail/wrapper wrapper
maybe your sendmail need this for run the wrapper programs
>Thanks,
>Bng
>
>
>------------------------------------------------------
>Mailman-Users maillist - Mailman-Users at python.org
>http://www.python.org/mailman/listinfo/mailman-users
From jwblist at olympus.net Fri Mar 19 22:35:51 1999
From: jwblist at olympus.net (John W Baxter)
Date: Fri, 19 Mar 1999 13:35:51 -0800
Subject: [Mailman-Users] Sending a human message to the
LISTNAME-request address.
In-Reply-To:
References: <01be718f$6e406ad0$1c7001c4@ntserver.sld.cu>
Message-ID:
At 15:50 -0500 3/19/99, The List Server Administrator at UNH wrote:
> I would rather see the *-request address used for contacting a
> human since that is how it is most widely used.
With Majordomo, *-request is also widely used as a portal to the
server. The list setup instructions discuss both the "send the
request to Majordomo@" method and the "send the request to
*-request@" method. (We have both set up, but only talk about
*-request.)
I don't know how the Majordomo population as a whole sets up their
lists...I do know that all of our not-very-many lists use *-request@
to reach the server, and *-owner and owner-* to reach a human. [And
we can't change now, since our users are trained that way, although
when we switch to Mailman (not before 1.0 non-beta) we will retrain
them to the Web.]
--John
--
John Baxter jwblist at olympus.net Port Ludlow, WA, USA
Give a man a fish and you feed him for a day. Teach him to fish,
and you get rid of him for the weekend.
From listadm at metaphor.unh.edu Fri Mar 19 22:56:52 1999
From: listadm at metaphor.unh.edu (The List Server Administrator at UNH)
Date: Fri, 19 Mar 1999 16:56:52 -0500 (EST)
Subject: [Mailman-Users] Site configurable server messages. (was Sending a human message...)
In-Reply-To: <199903191705.MAA15108@python.org>
Message-ID:
BC = Bill Costa
DD = Dan Delaney
GS = Greg Stein
BC> > > Separating the displayed text from the code allows for much
BC> > > easier internationalization as well as making it easier for
BC> > > different sites to customize the messages as necessary to
BC> > > reflect local policies.
DD> > That is an excellent idea.
GS> Yah, all great and excellent and whatnot. Somebody has to do it. Either
GS> of you could certainly volunteer to submit the patches to do this in the
GS> post-1.0 release.
I'll go on record that if we adopt Mailman to replace ListProc at
this site, I would be more than happy to volunteer to do this.
What will it take to come to a consensus that this would indeed
be a Good-Thing(tm) to do?
Cordially,
The List Server Admin
list.admin at unh.edu
(currently Bill Costa)
From boris at infoplease.com Fri Mar 19 23:14:27 1999
From: boris at infoplease.com (Boris Goldowsky)
Date: Fri, 19 Mar 1999 17:14:27 -0500
Subject: [Mailman-Users] help setting up mailman
Message-ID: <81435F08832BD111844A006097BA9748229283@banana.infoplease.com>
Roger Pe?a Escobio diagnosed my problem correctly. Copying the mail/wrapper
into /usr/adm/sm.bin solved that problem.
However, now I get different errors when I try to subscribe. I entered my
email address on the web form to subscribe, received the confirmation
message, replied to it, and the answering email I get is included below.
I did try setting /opt/mailman/scripts/* to be executable (they were
installed with mode 664), but that didn't help. Any clues?
Bng
The original message was received at Fri, 19 Mar 1999 16:23:47 -0500 (EST)
from banana.infoplease.com [208.222.166.11]
----- The following addresses had permanent fatal errors -----
"|/usr/adm/sm.bin/mailman-wrapper mailcmd mmtest"
(expanded from: )
----- Transcript of session follows -----
Traceback (innermost last):
File "/opt/mailman/scripts/mailcmd", line 52, in ?
list.ParseMailCommands()
File "/opt/mailman/Mailman/MailCommandHandler.py", line 162, in
ParseMailCommands
self._cmd_dispatch[cmd](args, line, mail)
File "/opt/mailman/Mailman/MailCommandHandler.py", line 512, in
ProcessConfirmCmd
self.ProcessConfirmation(cookie)
File "/opt/mailman/Mailman/MailList.py", line 874, in ProcessConfirmation
got = Pending().confirmed(cookie)
File "/opt/mailman/Mailman/Pending.py", line 80, in confirmed
db = self.__load()
File "/opt/mailman/Mailman/Pending.py", line 94, in __load
self.__assert_lock(self.db_lock_timeout)
File "/opt/mailman/Mailman/Pending.py", line 120, in __assert_lock
self.__lock.lock(timeout)
File "/opt/mailman/Mailman/flock.py", line 119, in lock
os.link(self.lockfile, self.tmpfname)
os.error: (13, 'Permission denied')
554 "|/usr/adm/sm.bin/mailman-wrapper mailcmd mmtest"... unknown mailer
error 1
-----Original Message-----
From: Roger Pe?a Escobio [mailto:roger at infomed.sld.cu]
Sent: Friday, March 19, 1999 3:55 PM
To: mailman-u; Boris Goldowsky
Subject: Re: [Mailman-Users] help setting up mailman
-----Original Message-----
From: Boris Goldowsky
To: 'mailman-users at python.org'
Date: Friday, March 19, 1999 3:33 PM
Subject: [Mailman-Users] help setting up mailman
>I'm trying to set up mailman to run on a Solaris 2.5.1 system with sendmail
>8.9.3. The configure, build, and creation of a test mailing list worked
>without errors.
>
>However, when I send any message that tries to use the wrapper function, I
>get the following error in syslog:
>
>Mar 19 14:43:51 spaceheater sendmail[18681]: OAA18681:
> to="|/opt/mailman/mail/wrapper mailowner mmtest",
> delay=00:00:00, xdelay=00:00:00, mailer=prog,
> stat=Operating system error
>
make a link to the mailman wrapper programs in the /etc/smrsh dir
#ln -s /home/mailman/mail/wrapper wrapper
maybe your sendmail need this for run the wrapper programs
>Thanks,
>Bng
>
>
>------------------------------------------------------
>Mailman-Users maillist - Mailman-Users at python.org
>http://www.python.org/mailman/listinfo/mailman-users
From claw at varesearch.com Fri Mar 19 15:20:18 1999
From: claw at varesearch.com (J C Lawrence)
Date: Fri, 19 Mar 1999 06:20:18 -0800
Subject: [Mailman-Users] Sugestions
In-Reply-To: Message from bruce@hams.com
of "17 Mar 1999 17:32:03 GMT." <19990317173203.32527.qmail@hams.com>
Message-ID:
On 17 Mar 1999 17:32:03 -0000
bruce wrote:
>> And an example of why MX isn't granular enough:
>>
>> 197935 74 74 1071.19 remote.elynxx at MIT.EDU 197935 74 74 301.06
>> remote.gyouxx at MIT.EDU
> I don't understand why an MX would have different delays per
> _recepient_. In general it accepts the mail, hangs up the SMTP
> connection, and _then_ tries to deliver it.
Authentication (which can be done during acceptance of a message)
can be variously slow, especially if they are running thru NIS maps,
a curious IMAP setup, or some other baroqueness.
--
J C Lawrence Internet: claw at kanga.nu
---------(*) Internet: claw at varesearch.com
...Honorary Member of Clan McFud -- Teamer's Avenging Monolith...
From alann at ihs.com Fri Mar 19 23:23:10 1999
From: alann at ihs.com (Alan Neiman)
Date: Fri, 19 Mar 1999 15:23:10 -0700 (MST)
Subject: [Mailman-Users] question
Message-ID: <199903192224.PAA00382@dns1.ihs.com>
I would like to have our mailing lists 100% controlled by email requests. Is
there a way to change the welcome email to still give instructions, but not to
refrence the fact they can change things via the web page?
Also, is there a way to turn off users passwords, so they don't need one to
unsubscribe?
thanks
Alan Neiman
alan.neiman at ihs.com
Unix System Administrator
Information Handling Services
--------------------------------------------------------------
** 1997 & 1998 World Champions **
17 & 2 -- Simply The Best -- The Broncos -- The Denver Broncos
** 1997 & 1998 World Champions **
--------------------------------------------------------------
From gstein at lyra.org Sat Mar 20 00:23:48 1999
From: gstein at lyra.org (Greg Stein)
Date: Fri, 19 Mar 1999 15:23:48 -0800
Subject: [Mailman-Users] question
References: <199903192224.PAA00382@dns1.ihs.com>
Message-ID: <36F2DC84.33E179BF@lyra.org>
Alan Neiman wrote:
>
> I would like to have our mailing lists 100% controlled by email requests. Is
> there a way to change the welcome email to still give instructions, but not to
> refrence the fact they can change things via the web page?
Edit $prefix/templates/subscribeack.txt
> Also, is there a way to turn off users passwords, so they don't need one to
> unsubscribe?
Nope.
-g
--
Greg Stein, http://www.lyra.org/
From clark.evans at manhattanproject.com Sat Mar 20 01:24:32 1999
From: clark.evans at manhattanproject.com (Clark Evans)
Date: Sat, 20 Mar 1999 00:24:32 +0000
Subject: [Mailman-Users] User limit?
References: <199901290308.TAA67170@under.engr.sgi.com>
Message-ID: <36F2EAC0.DA5627B8@manhattanproject.com>
Is there a limit as to the number of subscribers?
Is 5000 too much?
Clark
From bwarsaw at cnri.reston.va.us Sat Mar 20 06:07:13 1999
From: bwarsaw at cnri.reston.va.us (Barry A. Warsaw)
Date: Sat, 20 Mar 1999 00:07:13 -0500 (EST)
Subject: [Mailman-Users] Jitterbug for Mailman
Message-ID: <14067.11521.211373.182229@anthem.cnri.reston.va.us>
Hey folks, sorry I've been so unresponsive lately.
Anyway, I have installed Jitterbug on python.org for managing bug
reports on Mailman. Hopefully this will work better than trying to
use my inbox :-). If Jitterbug works out well, I plan on creating
interfaces for other python.org projects.
To check out the Jitterbug reporting interface for Mailman, please see
http://www.python.org/mailman-bugs
Let me know if you have any problems.
-Barry
From bwarsaw at cnri.reston.va.us Sat Mar 20 06:42:18 1999
From: bwarsaw at cnri.reston.va.us (Barry A. Warsaw)
Date: Sat, 20 Mar 1999 00:42:18 -0500 (EST)
Subject: [Mailman-Users] Sending a human message to the LISTNAME-request address.
References:
Message-ID: <14067.13626.253250.31672@anthem.cnri.reston.va.us>
Let me try to follow up to this thread. Apologies if I've missed out
on part of the discussion, and for the random nature of the responses
below.
- I agree that mailman@ ought to be a context-independent way of
contacting the server. It's more inline with what Majordomo does,
which is what I think John modeled Mailman after most closely.
However mailcmd would have to be fixed so it could handle such
context-independent messages.
- mailman-owner@, owner-mailman@, and mailman-admin@ should go to a
human, probably the site administrator. This is more in line with
the standard addresses generated for real lists.
- list-request@ goes to the server. I think it's too late to change
this default, although if you wanted to, you should be able to
change the aliases to use a different standard (I haven't tried
this, so I can't say whether this actually works or not).
- mailcmd should be fixed so that it sends a much more helpful
response when the message contains errors. Greg's right that it
probably isn't a huge amount of work, but given the Web-centric view
of Mailman, it's not surprising this hasn't had as much attention as
it should.
- I18N has been discussed many times. Much of the text that a human
eventually sees is factored out into files, but not all of it. I'd
like to make setting up the I18N architecture the big feature for
the 1.1 release.
- Design questions are probably best discussed on mailman-developers.
- To answer Bill's question about the `admin model' Mailman supports.
Here at python.org, we have a large number of lists, managed by many
independent remote admins. While this is an important design
decision, Mailman could do better. Harald has recently talked about
some plans for introducing real user objects into Mailman, which
would allow an admin to manage several lists more easily (it would
also allow users to manage their multiple subscriptions much more
easily). Post 1.0
Hope that catches me up with this thread. I'll try to add some of
these things to the new Jitterbug database.
-Barry
From udall at brassica.agronomy.wisc.edu Sun Mar 21 07:48:10 1999
From: udall at brassica.agronomy.wisc.edu (Joshua)
Date: Sun, 21 Mar 1999 00:48:10 -0600
Subject: [Mailman-Users] Aliases not working
Message-ID: <199903200640.AAA137022@mail1.doit.wisc.edu>
Mailmen -
I just upgraded to b9. Running Sparc/Linux on a SS4 with RH 5.2.
Got things working ok - almost.
I created a test list 'test3' and the admin, info work ok. I can't tell if
list_members does cause I haven't been able to subscribe to the
list.
these email addresses:
test3-request at brassica.agronomy.wisc.edu
test3 at brassica.agronomy.wisc.edu
return:
From: Mail Delivery Subsystem
Subject: Returned mail: User unknown
However, the confimation of request was formatted correctly:
From: test3-request at brassica.agronomy.wisc.edu
Date sent: Sat, 20 Mar 1999 00:26:21 -0600
Subject: Test3 -- confirmation of subscription -- request 357936
To: jaudall at students.wisc.edu
Send reply to: test3-request at brassica.agronomy.wisc.edu
I didn't think there needed to be 'hard' aliases for these addresses -
Though I don't know how mailman does it - but, the method is not
working.
Note:
The aliases put into /etc/sendmail.cf work fine. (i.e. mailman-
owner at brassica.agronomy.wisc.edu)
Any suggestions? I could use some help.
Josh
From lindsey at ncsa.uiuc.edu Sat Mar 20 08:11:12 1999
From: lindsey at ncsa.uiuc.edu (Christopher Lindsey)
Date: Sat, 20 Mar 1999 01:11:12 -0600 (CST)
Subject: [Mailman-Users] Aliases not working
In-Reply-To: <199903200640.AAA137022@mail1.doit.wisc.edu> from "Joshua" at Mar 21, 99 00:48:10 am
Message-ID: <199903200711.BAA15377@ferret.ncsa.uiuc.edu>
A non-text attachment was scrubbed...
Name: not available
Type: text
Size: 549 bytes
Desc: not available
Url : http://mail.python.org/pipermail/mailman-users/attachments/19990320/1f4e684d/attachment.asc
From jeffw at fatman.pro-libertate.com Sat Mar 20 18:40:02 1999
From: jeffw at fatman.pro-libertate.com (Jeff Wallace)
Date: Sat, 20 Mar 1999 11:40:02 -0600
Subject: [Mailman-Users] Pipermail
Message-ID: <36F3DD71.4E7B38AF@fatman.pro-libertate.com>
Hi,
I'm converting mailman from majordomo, can someone point me in the
direction of some documentation on moving my majordomo archives over to
use pipermail within mailman?
-jeffw
From paz at apriori.net Sat Mar 20 20:56:16 1999
From: paz at apriori.net (paz)
Date: Sat, 20 Mar 1999 14:56:16 -0500 (EST)
Subject: [Mailman-Users] Thanks -
In-Reply-To: <199903200711.BAA15377@ferret.ncsa.uiuc.edu>
Message-ID:
My upgrade from 0.9 to 1.0b9 seems to have succeeded!
./configure wanted the additional --with-mail-gid=1 so that wrappers
worked (seems like something should have seen mail = gid 6 and daemon =
gid 1, but I got lucky with this attempt) and --with-cgi-gid=65534.
Web pages are working, mailing postings works.
Root got a funny message a couple of hours ago, which I don't understand:
========================
Traceback (innermost last):
File "/home/mailman/cron/gate_news", line 119, in ?
main()
File "/home/mailman/cron/gate_news", line 58, in main
mlist = MailList.MailList(name, lock=0)
File "/home/mailman/Mailman/MailList.py", line 59, in __init__
self.Load()
File "/home/mailman/Mailman/MailList.py", line 712, in Load
raise mm_cfg.MMBadListError, 'Failed to unmarshal config info'
AttributeError: MMBadListError
========================
If anyone can heads or tails of that, particularly if it's of any real
concern to me, I'd appreciate knowing.
Thanks to everybody who responded with their helpful advice! (You know who
your are...)
cheers -
-- Philip.
From roger at infomed.sld.cu Sat Mar 20 21:40:13 1999
From: roger at infomed.sld.cu (=?iso-8859-1?Q?Roger_Pe=F1a_Escobio?=)
Date: Sat, 20 Mar 1999 15:40:13 -0500
Subject: [Mailman-Users] Sending a human message to the LISTNAME-requestaddress.
Message-ID: <01be7311$db0811d0$1c7001c4@ntserver.sld.cu>
>On Fri, 19 Mar 1999 15:50:24 -0500 (EST), The List Server Administrator at
>UNH wrote:
>> Chuck Swiger posted (in part):
>>
>CS> ListProc uses only one -request address, I think that this is better
than
>CS> have a lot of -request address (like mailman) all doing (basically) the
>CS> same., after all , the list manager is mailman, not the list in
question.
>
>Not guilty! I didn't say that.... :-)
sorry , i am the guilty :-)))))))))))))
i said that
>
>-Chuck
>
> Charles Swiger | chuck at codefab.com | Yeah, yeah-- disclaim away.
> ----------------+-------------------+----------------------------
> You have come to the end of your journey. Survival is everything.
From roger at infomed.sld.cu Sat Mar 20 21:47:42 1999
From: roger at infomed.sld.cu (Roger Pe?a Escobio)
Date: Sat, 20 Mar 1999 15:47:42 -0500
Subject: [Mailman-Users] Sending a human message to the LISTNAME-requestaddress.
Message-ID: <01be7312$e68f0210$1c7001c4@ntserver.sld.cu>
-----Original Message-----
From: Dan Delaney
To: mailman-users at python.org
Date: Friday, March 19, 1999 6:52 PM
Subject: Re: [Mailman-Users] Sending a human message to the
LISTNAME-requestaddress.
>Bill wrote:
>> I like LISTSERV/ListProc model better, where there is a
>> single address for the server, and all list addresses are really
>> unique for that list. But I don't know if we're in a position to
>
>I like that approach as well. But some people like it the other way.
>That's one of the nice things about majordomo, it goes both ways (so
>to speak :-). If I have a majordomo list called "mylist", the
>someone can subscribe one of two ways. They can send a message to
>"majordomo at mydomain.com" with "subscribe myemail at anotherdomain.com"
>in the message, or they can send a message to
>"mylist-request at mydomain.com" with just "subscribe" in the message.
>That's awefully nice. Mailman, however, limits the user to
>the latter method, making "mailman at mydomain.com" go to the Mailman
>administrator. I think it would make more sense to just stick with
>"mailman-owner at mydomain.com", or even add
>"mailman-admin at mydomain.com" and make "mailman at mydomain.com" an
>address for sending commands to the user. If this is supposed to
>replace majordomo, that's what a lot of majordomo users are used
>to.
> As far as "mylist-request at mydomain.com" goes, I don't see a
>problem with that list going to the server. I think "mylist-owner"
>is meant for messages to a human. You are sending a "request" to the
>server when sending to "mylist-request", and sending a message to
>the owner of the list at "mylist-owner".
I AGREE WITH YOU, completely, in all that you say
>
> --Dan
thank Dan, you say what i wan to say :-)
roger
From bwarsaw at cnri.reston.va.us Sat Mar 20 22:34:29 1999
From: bwarsaw at cnri.reston.va.us (Barry A. Warsaw)
Date: Sat, 20 Mar 1999 16:34:29 -0500 (EST)
Subject: [Mailman-Users] private archiving problems
References: <199902140622.AAA02426@ferret.ncsa.uiuc.edu>
Message-ID: <14068.5221.550758.71233@anthem.cnri.reston.va.us>
Just remember that archives are only private because they go through a
cgi script that does the authorization. Public archives are vended
directly by the Web server. The latter is so that the most common
case is fast. However without a lot of file system reorganization on
the switch b/w private and public archiving, when you access a private
archive, you are invoking a cgi.
From bwarsaw at cnri.reston.va.us Sat Mar 20 22:58:02 1999
From: bwarsaw at cnri.reston.va.us (Barry A. Warsaw)
Date: Sat, 20 Mar 1999 16:58:02 -0500 (EST)
Subject: [Mailman-Users] Mass subscription of users -- random password?
References: <014601be6fdb$165b5d80$0a01a882@barnettpc>
Message-ID: <14068.6634.463607.118657@anthem.cnri.reston.va.us>
>>>>> "PB" == Paul Barnett writes:
PB> if so, I'd suggest a combination of lower case, upper case,
PB> numbers, and MAYBE well known punctuation marks, but only if
PB> they not the beginning or end characters.
I've changed this so it will only generate upper and lower case
letters. This should still be random enough for our purposes. I left
out numbers because some fonts make l/1 and 0/O hard to distinguish.
Thanks,
-Barry
From bwarsaw at cnri.reston.va.us Sat Mar 20 23:14:15 1999
From: bwarsaw at cnri.reston.va.us (Barry A. Warsaw)
Date: Sat, 20 Mar 1999 17:14:15 -0500 (EST)
Subject: [Mailman-Users] No passwords for members?
References: <199902150452.WAA09633@ferret.ncsa.uiuc.edu>
Message-ID: <14068.7607.573658.724404@anthem.cnri.reston.va.us>
>>>>> "CL" == Christopher Lindsey writes:
CL> I could've sworn that this was addressed on the list already,
CL> but I can't seem to find the message about it... So here I go
CL> (again?)
CL> Anyhow, I recently converted a SmartList list to Mailman via
CL> the add_members script, but now I'm getting this error message
CL> for all of the users who try accessing their password:
CL> Mailman noticed in .MailUserPassword() that:
| User: 'xxxxxxxxxx at aol.com'
| List: rose-list
CL> lacks a password. Please notify the Mailman system manager
CL> at this site!
CL> So what can I do about it? There's no option to
CL> default/recreate a password for the user... Do I need to go
CL> through and manually subscribe each person ...
CL> The other interesting that I noticed is that some of the
CL> addresses had the 'plain' option unchecked, regardless of
CL> whether or not they were receiving mail in digest form.
The user probably has a password, so you shouldn't need to recreate
one. There was a bug in 1.0b9 that caused this error message to occur
when the case-preserved user's address had upper case letters in it.
This will be fixed in the next release. A patch against 1.0b9 is
attached.
-Barry
Index: Deliverer.py
===================================================================
RCS file: /projects/cvsroot/mailman/Mailman/Deliverer.py,v
retrieving revision 1.50
retrieving revision 1.53
diff -c -r1.50 -r1.53
*** Deliverer.py 1999/01/13 23:53:16 1.50
--- Deliverer.py 1999/03/09 02:24:03 1.53
***************
*** 247,253 ****
def MailUserPassword(self, user):
listfullname = '%s@%s' % (self.real_name, self.host_name)
ok = 1
! if self.passwords.has_key(user):
recipient = self.GetMemberAdminEmail(user)
subj = '%s maillist reminder\n' % listfullname
# get the text from the template
--- 245,255 ----
def MailUserPassword(self, user):
listfullname = '%s@%s' % (self.real_name, self.host_name)
ok = 1
! # find the case-preserved version of the user's address
! cpuser = self.members.get(self.FindUser(user))
! if type(cpuser) == type(''):
! user = cpuser
! if user and self.passwords.has_key(user):
recipient = self.GetMemberAdminEmail(user)
subj = '%s maillist reminder\n' % listfullname
# get the text from the template
From bwarsaw at cnri.reston.va.us Sat Mar 20 23:49:07 1999
From: bwarsaw at cnri.reston.va.us (Barry A. Warsaw)
Date: Sat, 20 Mar 1999 17:49:07 -0500 (EST)
Subject: [Mailman-Users] Pipermail
References: <36F3DD71.4E7B38AF@fatman.pro-libertate.com>
Message-ID: <14068.9699.261845.583999@anthem.cnri.reston.va.us>
>>>>> "JW" == Jeff Wallace writes:
JW> I'm converting mailman from majordomo, can someone point me in
JW> the direction of some documentation on moving my majordomo
JW> archives over to use pipermail within mailman?
http://www.python.org/pipermail/mailman-users/1999-March/000636.html
-Barry
From bwarsaw at cnri.reston.va.us Sun Mar 21 21:30:31 1999
From: bwarsaw at cnri.reston.va.us (Barry A. Warsaw)
Date: Sun, 21 Mar 1999 15:30:31 -0500 (EST)
Subject: [Mailman-Users] Thanks -
References: <199903200711.BAA15377@ferret.ncsa.uiuc.edu>
Message-ID: <14069.22248.24172.474001@anthem.cnri.reston.va.us>
>>>>> "paz" == writes:
paz> If anyone can heads or tails of that, particularly if it's of
paz> any real concern to me, I'd appreciate knowing.
Looks like your config.db file is corrupt. Here's a script I just
added to the system that can give a little more information about
what's going on.
-Barry
-------------------- snip snip --------------------
#! /usr/bin/env python
"""Check the raw config.db for a mailing list.
Also check the config.db.last if the live file is corrupted.
Usage: %(program)s listname
"""
import sys
import os
import marshal
import paths
import Mailman.mm_cfg
program = sys.argv[0]
def testfile(filename):
try:
fp = open(filename)
except IOError, (code, msg):
print filename, 'cannot be opened:\n\t', msg
return 1
else:
try:
d = marshal.load(fp)
except (EOFError, ValueError, TypeError), msg:
print filename, 'is corrupted:\n\t', msg
return 1
else:
print filename, 'is fine'
return 0
def main():
if len(sys.argv) == 2:
listname = sys.argv[1]
else:
print __doc__ % globals()
sys.exit(1)
listpath = os.path.join(Mailman.mm_cfg.LIST_DATA_DIR, listname)
configdb = os.path.join(listpath, 'config.db')
lastdb = os.path.join(listpath, 'config.db.last')
origbad = testfile(configdb)
backupbad = testfile(lastdb)
if origbad and not backupbad:
print """
***** ALERT *****
The original database file is corrupt, but the backup seems fine.
Consider copying
%(lastdb)s
to
%(configdb)s
however, you may lose some data.""" % locals()
elif origbad and backupbad:
print """
***** ALERT *****
Both the original database file and the backup seem
corrupted. You will probably need to recover both
%(configdb)s
and
%(lastdb)s
from a system backup, or remove the list `%(listname)s' and
re-create it from scratch.""" % locals()
if __name__ == '__main__':
main()
From rpyne at kinfolk.org Mon Mar 22 00:36:54 1999
From: rpyne at kinfolk.org (Richard B. Pyne)
Date: Sun, 21 Mar 1999 16:36:54 -0700
Subject: [Mailman-Users] Pipermail
In-Reply-To: <14068.9699.261845.583999@anthem.cnri.reston.va.us>
Message-ID: <199903212336.QAA07627@ns.kinfolk.org>
I, too, was looking for this information. I found the message
referenced:
-------------------------------------------------
> messages from when it was still under majordomo. Is there any way
for me
> to have these archived?
"`$prefix/bin/arch listname mbox_filename' is your friend!" (tm) :-)
you could also merge $prefix/archives/private/.mbox/.mbox
with his mbox to preserve it for future use.
------------------------------------------------------
Could someone please put this in english?
On 20 Mar 99, at 17:49, Barry A. Warsaw wrote:
>
> >>>>> "JW" == Jeff Wallace writes:
>
> JW> I'm converting mailman from majordomo, can someone point me in
> JW> the direction of some documentation on moving my majordomo
> JW> archives over to use pipermail within mailman?
>
> http://www.python.org/pipermail/mailman-users/1999-March/000636.html
>
> -Barry
>
> ------------------------------------------------------
> Mailman-Users maillist - Mailman-Users at python.org
> http://www.python.org/mailman/listinfo/mailman-users
------------------------------
Richard B. Pyne, KB7RMU
rpyne at kinfolk.org
http://pyne.kinfolk.org/rbp2
From listadm at metaphor.unh.edu Mon Mar 22 20:32:47 1999
From: listadm at metaphor.unh.edu (The List Server Administrator at UNH)
Date: Mon, 22 Mar 1999 14:32:47 -0500 (EST)
Subject: [Mailman-Users] (no subject)
Message-ID:
John W Baxter posted to the mailman-users
list (in part):
> And we can't change now, since our [Majordomo] users are trained that
> way [to use *-owner for reaching a human], although when we switch to
> Mailman (not before 1.0 non-beta) we will retrain them to the Web.
Our problem is that our ListProc subscribers (11,732 distinct e-mail
addresses spread across 398 lists, managed by 202 distinct owners)
are already trained to use *-request to contact the human. What I am
going to have to hope is that given the wonderful web interface of
mailman, few if any will complain about the change. :-)
Cordially,
The List Server Admin
list.admin at unh.edu
(currently Bill Costa)
From bwarsaw at cnri.reston.va.us Tue Mar 23 00:34:27 1999
From: bwarsaw at cnri.reston.va.us (Barry A. Warsaw)
Date: Mon, 22 Mar 1999 18:34:27 -0500 (EST)
Subject: [Mailman-Users] New Mailing List
Message-ID: <14070.54147.17042.254891@anthem.cnri.reston.va.us>
There have been a number of requests to start an announce-only list
for Mailman. I've now done this
http://www.python.org/mailman/listinfo/mailman-announce
The function of the lists are now better explained on the www.list.org
site. Let me know if you have any problems.
-Barry
From roger at infomed.sld.cu Tue Mar 23 03:27:13 1999
From: roger at infomed.sld.cu (=?iso-8859-1?Q?Roger_Pe=F1a_Escobio?=)
Date: Mon, 22 Mar 1999 21:27:13 -0500
Subject: [Mailman-Users] patch to fix the alias-wrapper.c
Message-ID: <01be74d4$a9981e00$1c7001c4@ntserver.sld.cu>
Hi...
few days ago we discuses about aliases file, well... this my patch of
mailman-1.0b9 to fix alias-wrapper, also fix newlist.
this patch also changes the rmlist file, including an rmaliases programs
(rmalias-wrapper.c, this programs edit the aliases file when you eliminate a
list ).
Of course, you are free to apply the patch or not, but if will do it look
inside first, see what it will do and if you agree, ?? great !!! lets do it
:-)
this line are for the brave souls :-) who like apply the patch
1-move the patch to the source dir , for example , /usr/local/src if mailman
source is in /usr/local/src/mailman-1.0b9
2-rename the alias-wrapper.c source to addalias-wrapper.c , it is, mv
./mailman-1.0b9/src/alias-wrapper.c ./mailman-1.0b9/src/addalias-wrapper.c
3-exec the command : patch -p0 < mailman-1.0b9-patch-r.patch
4- if you dont get any estrange line or question every thing was fine, if
not ...... check if you really have the mailman-1.0b9 version or if you are
in the correct dir.
5- everything is ok ? great, now put the rmalias-wrapper.c into the src dir,
it is ./mailman-1.0b9/src/
6-now you have to check some "little" thing, first of all , my aliases file
for mailman is /etc/mailman/aliases, you can change that editing the
addalias-wrapper.c and rmalias-wrapper.c, is really simple, take a look, I
dont know if mailman user can edit the /etc/aliases file ( I think he cant)
but in the other way I prefer to have two aliases file one for system alias
and another for mailman alias, if I convince you about two alias, you will
have to change your sendmail.cf and create the /etc/mailman/ dir :
6.1-creating the /etc/mailman/ dir : #mkdir /etc/mailman ; chgrp mailman
/etc/mailman ; chmod 755 /etc/mailman
6.2-changing the sendmail.cf: do this really depend on your OS , in
RedHat you have to do this:
cd /usr/lib/sendmail-cf/cf/
edit your *.mc , if you never do this before ( if you have the
sendmail.cf from distribution) your *.mc is redhat.mc, put this line
define(`ALIAS_FILE',`/etc/aliases,/etc/mailman/aliases')
exec the command : #m4 ../m4/cf.m4 redhat.mc > myhost.cf
make a copy of your old sendmail.cf : #cp /etc/sendmail.cf
/etc/sendmail.cf.old
copy the new *.cf file to /etc/sendmail.cf: #cp myhost.cf
/etc/sendmail.cf
restart sendmail: #/etc/rc.d/init.d/sendmail restart
and thats all !!!! are you still with me ???? :-)
6.3-check if you have /var/tmp dir , addalias-wrapper.c and
rmalias-wrapper.c need it temporally, redhat already have but other OS dont
7-the only thing that rest is run autoconf INSIDE mailman source , it is :
#cd /usr/local/mailman-1.0b9 ; autoconf
that is for make configure from configure.in
8-well, if you do all above you really have my mailman ready to install, so
install it :-))
a final words.... maybe you think something like this "should I do all this
thing instead of edit the aliases file manually?? are you crazy!!!" but you
only need to do this only once, and then when you create or eliminate a list
all will be do automatically.
well a hope that this patch help someone .
Roger
-------------- next part --------------
A non-text attachment was scrubbed...
Name: rmalias-wrapper.c
Type: application/octet-stream
Size: 3227 bytes
Desc: not available
Url : http://mail.python.org/pipermail/mailman-users/attachments/19990322/9fba647b/attachment.obj
-------------- next part --------------
A non-text attachment was scrubbed...
Name: mailman-1.0b9-patch-r.patch
Type: application/octet-stream
Size: 9105 bytes
Desc: not available
Url : http://mail.python.org/pipermail/mailman-users/attachments/19990322/9fba647b/attachment-0001.obj
From gconnor at nekodojo.org Tue Mar 23 07:44:21 1999
From: gconnor at nekodojo.org (Greg Connor)
Date: Mon, 22 Mar 1999 22:44:21 -0800
Subject: [Mailman-Users] Making two "connected" lists
In-Reply-To: <14070.54147.17042.254891@anthem.cnri.reston.va.us>
Message-ID: <4.1.19990322222108.009fcee0@pop.nekodojo.org>
I would like to make two "connected" lists... The idea is that some users
want to get "announcements only" and other users want "announcements and
discussion".
One quick and dirty way to do this would be to make a combined alias, like
foo: "|/home/mailman/mail/wrapper post foo"
foo-announce: "|/home/mailman/mail/wrapper post foo-announce", foo
which would deliver "foo-announce" to both foo-announce and foo.
Ideally, people would subscribe only to one or the other (since they will
get the "announce" messages either way).
This might help avoid some confusion with the two separate lists I have now
(under a different MLM). Currently people have to sign up for both in
order to get all messages, and have to unsub both when leaving... and some
people still send "announce" material to both, so most people get two
copies of the announcement. I would prefer one for announce-only and
another for announce+discussion. (Eventually, we might want to make
replies go to the discussion list, even if replying to an announcement, but
that's not so important right now -- I have replies going to the sender only)
All fine up to this point... However, the issue I still haven't resolved
is how this would work with the "Restrict posting privilege to list
members" option. The announce list would have the "announce-only" users,
but if the discussion list members send to the announce list, they might
get "You are not subscribed". For now I can turn off the members-only
limitation on the first list, but eventually I want to find away around
this, and let members of either list post to the first one.
I just wanted to ask if someone had done this before, and whether there is
a feature to help me do what I want to do. (I noticed there is a couple of
features for "umbrella" lists, but I'm not sure if this is what it is for,
or whether umbrella lists work with the members-only option.) Write back
if you have information that may help. Also, if there is interest I can
summarize the results to anyone interested (or back to the list if there is
enough interest) once I get it all working how I like it.
Thanks to Barry and other mailman developers for a great package!
gregc
From ptb at home.ptb.org Tue Mar 23 15:14:22 1999
From: ptb at home.ptb.org (Paul Barnett)
Date: Tue, 23 Mar 1999 08:14:22 -0600
Subject: [Mailman-Users] List adminstration and moderation
Message-ID: <004801be7537$75675570$0201a8c0@ptb.org>
Since I'm a new user (Mailman 1.0b9), I'm not sure if the "problem" I'm
describing is actually a feature, but I'll submit it for consideration:
When I first enter mailing list administration:
http://x/mailman/admin/test
I get a demand for "Administrative Authentication". That's expected, but if
I immediately click on "Tend to pending administrative requests":
http://x/mailman/admindb/test
I get a second demand for "Administrative Authentication". I thought this
might be so that a list manager could specify a different password for
moderation and thus let other people perform that function without giving
them access to everything else, but I can't find the ability to specify that
option.
Is there another reason it is demanding a password a second time?
From ptb at home.ptb.org Tue Mar 23 15:48:32 1999
From: ptb at home.ptb.org (Paul Barnett)
Date: Tue, 23 Mar 1999 08:48:32 -0600
Subject: [Mailman-Users] How does Mailman use identd?
Message-ID: <008401be753c$39325500$0201a8c0@ptb.org>
I just subscribed to the mailman-announce list.
About that time, my firewall denied eight requests to port 113 (identd) from
parrot.python.org
It obviously failed, but I still got the subscription confirmation.
How does it use the information, if it manages to get it?
From Dionysos at Dionysia.org Tue Mar 23 15:51:09 1999
From: Dionysos at Dionysia.org (Dan Delaney)
Date: Tue, 23 Mar 1999 09:51:09 -0500 (EST)
Subject: [Mailman-Users] How about this for the error message?
Message-ID:
This is what Mailman returns for by sig if I don't put an "end"
command":
>>>> --Dan
**** --dan: Command UNKNOWN.
>>>> ____________________________________________________________________________
**** ____________________________________________________________________________: Command UNKNOWN.
>>>> Daniel G. Delaney * Dionysos at Dionysia.org * www.Dionysia.org/~dionysos
**** daniel: Command UNKNOWN.
>>>> PGP Public Key: http://Dionysia.org/~dionysos/pgp5.html
**** pgp: Command UNKNOWN.
>>>> "Only two things are infinite: the universe and stupidity--
**** "only: Command UNKNOWN.
>>>> and I'm not sure about the former."
**** and: Command UNKNOWN.
>>>> --Albert Einstein
**** --albert: Command UNKNOWN.
That's pretty darn obtrusive, and a little daunting to average Joe
Shmoe computer user who gets nervous when the computer starts
yelling at him about errors. So, how about the following instead?
Unrecognized line: --Dan
Unrecognized line: _______________________________
Unrecognized line: Daniel G. Delaney * Dionysos@
Unrecognized line: PGP Public Key: http://Dionysia
Unrecognized line: "Only two things are infinite:
Unrecognized line: and I'm not sure about the form
Unrecognized line: --Albert Einstein
That's so much easier to look at, and it conveys the message--that
the mail server didn't recognize those lines. Note that it should
only returns the first 40 or so characters of the line instead of
the whole thing.
Here's another idea. Petidomo will STOP parsing the message for
commands if it gets FOUR unrecognized lines in a row. I think that's
a pretty darn good idea. You could even simply put a little message
after the fourth "Unrecognized line" message that says
"Ignoring remainder of message."
What's everyone think? Should this idea go to the dev list instead?
If so, could someone who's also on that list please forward it over
there?
Cheers.
--Dan
____________________________________________________________________________
Daniel G. Delaney * Dionysos at Dionysia.org * www.Dionysia.org/~dionysos
PGP Public Key: http://Dionysia.org/~dionysos/pgp5.html
"Only two things are infinite: the universe and stupidity--
and I'm not sure about the former."
--Albert Einstein
From petrilli at amber.org Tue Mar 23 16:09:48 1999
From: petrilli at amber.org (Christopher Petrilli)
Date: Tue, 23 Mar 1999 10:09:48 -0500
Subject: [Mailman-Users] How does Mailman use identd?
In-Reply-To: <008401be753c$39325500$0201a8c0@ptb.org>; from Paul Barnett on Tue, Mar 23, 1999 at 08:48:32AM -0600
References: <008401be753c$39325500$0201a8c0@ptb.org>
Message-ID: <19990323100948.A8268@amber.org>
On Tue, Mar 23, 1999 at 08:48:32AM -0600, Paul Barnett wrote:
> I just subscribed to the mailman-announce list.
Welcome :-) Wilkommen :-)
> About that time, my firewall denied eight requests to port 113 (identd) from
> parrot.python.org
Sounds like an MTA using identd to figure out who is sending mail, this
is done a lot to try and track spam... it's not that out of the
ordinary.
> It obviously failed, but I still got the subscription confirmation.
Yup, it's not a "stop" but just an "FYI" piece of information.
> How does it use the information, if it manages to get it?
I believe Gudio uses it to target his new beta mind-altering
Pythonicmindbenderwidget contraption... but I could be wrong :-)
Chris
--
| Christopher Petrilli ``Television is bubble-gum for
| petrilli at amber.org the mind.''-Frank Lloyd Wright
From petrilli at amber.org Tue Mar 23 16:16:22 1999
From: petrilli at amber.org (Christopher Petrilli)
Date: Tue, 23 Mar 1999 10:16:22 -0500
Subject: [Mailman-Users] How about this for the error message?
In-Reply-To: ; from Dan Delaney on Tue, Mar 23, 1999 at 09:51:09AM -0500
References:
Message-ID: <19990323101622.B8268@amber.org>
On Tue, Mar 23, 1999 at 09:51:09AM -0500, Dan Delaney wrote:
> This is what Mailman returns for by sig if I don't put an "end"
> command":
>
> [ Big ugly stuff deleted ]
>
> That's pretty darn obtrusive, and a little daunting to average Joe
> Shmoe computer user who gets nervous when the computer starts
> yelling at him about errors. So, how about the following instead?
I agree that it's obtrusive, but "oh by the way", the "defacto-standard"
(which I believe Mailman follows, at least I've never gotten a complaint
about it) is that signatures start with a double dash line, then
carriage-return... al'la "--\n(.*)" Which means that it's getting
confused where the "start" of your signature is... just so you know
what's happening.
> Unrecognized line: --Dan
> Unrecognized line: _______________________________
> Unrecognized line: Daniel G. Delaney * Dionysos@
> Unrecognized line: PGP Public Key: http://Dionysia
> Unrecognized line: "Only two things are infinite:
> Unrecognized line: and I'm not sure about the form
> Unrecognized line: --Albert Einstein
This looks better to my untrained eyes :-)
Chris
--
| Christopher Petrilli ``Television is bubble-gum for
| petrilli at amber.org the mind.''-Frank Lloyd Wright
From Dionysos at Dionysia.org Tue Mar 23 16:37:09 1999
From: Dionysos at Dionysia.org (Dan Delaney)
Date: Tue, 23 Mar 1999 10:37:09 -0500 (EST)
Subject: [Mailman-Users] How about this for the error message?
In-Reply-To: <19990323101622.B8268@amber.org>
Message-ID:
On Tue, 23 Mar 1999, Christopher Petrilli wrote:
> I agree that it's obtrusive, but "oh by the way", the "defacto-standard"
> (which I believe Mailman follows, at least I've never gotten a complaint
> about it) is that signatures start with a double dash line, then
> carriage-return... al'la "--\n(.*)" Which means that it's getting
> confused where the "start" of your signature is... just so you know
> what's happening.
Point taken. But the sig was just an example. The example used a
week or so ago was when someone actually sends a message meant for a
human to the -request address. Now THAT would generate an obtrusive
reply! In that case, the "ignore after four unrecognized lines" idea
would be awefully helpful.
--
____________________________________________________________________________
Daniel G. Delaney * Dionysos at Dionysia.org * www.Dionysia.org/~dionysos
PGP Public Key: http://Dionysia.org/~dionysos/pgp5.html
"Only two things are infinite: the universe and stupidity--
and I'm not sure about the former."
--Albert Einstein
From bwarsaw at cnri.reston.va.us Tue Mar 23 16:54:14 1999
From: bwarsaw at cnri.reston.va.us (Barry A. Warsaw)
Date: Tue, 23 Mar 1999 10:54:14 -0500 (EST)
Subject: [Mailman-Users] How about this for the error message?
References:
<19990323101622.B8268@amber.org>
Message-ID: <14071.47398.987773.73798@anthem.cnri.reston.va.us>
>>>>> "CP" == Christopher Petrilli writes:
CP> I agree that it's obtrusive, but "oh by the way", the
CP> "defacto-standard" (which I believe Mailman follows, at least
CP> I've never gotten a complaint about it) is that signatures
CP> start with a double dash line, then carriage-return... al'la
CP> "--\n(.*)" Which means that it's getting confused where the
CP> "start" of your signature is... just so you know what's
CP> happening.
I think the `standard' is actually double-dash-space-newline
"^-- \n" but from my quick scan of MailCommandHandler.py it shouldn't
matter either way.
-Barry
From bwarsaw at cnri.reston.va.us Tue Mar 23 17:03:44 1999
From: bwarsaw at cnri.reston.va.us (Barry A. Warsaw)
Date: Tue, 23 Mar 1999 11:03:44 -0500 (EST)
Subject: [Mailman-Users] How about this for the error message?
References:
<19990323101622.B8268@amber.org>
<14071.47398.987773.73798@anthem.cnri.reston.va.us>
Message-ID: <14071.47968.457627.678502@anthem.cnri.reston.va.us>
Me> I think the `standard' is actually double-dash-space-newline
Me> "^-- \n" but from my quick scan of MailCommandHandler.py it
Me> shouldn't matter either way.
Except that you must have *some* whitespace following the double dash.
-Barry
From listadm at metaphor.unh.edu Tue Mar 23 17:35:20 1999
From: listadm at metaphor.unh.edu (The List Server Administrator at UNH)
Date: Tue, 23 Mar 1999 11:35:20 -0500 (EST)
Subject: [Mailman-Users] How about this for the error message?
In-Reply-To: <19990323101622.B8268@amber.org>
Message-ID:
Christopher Petrilli recently posted (in part):
> I agree that it's obtrusive, but "oh by the way", the "defacto-standard"
> (which I believe Mailman follows, at least I've never gotten a complaint
> about it) is that signatures start with a double dash line, then
> carriage-return... al'la "--\n(.*)" Which means that it's getting
> confused where the "start" of your signature is... just so you know
> what's happening.
Actually I think the standard is that it is two hyphens followed
by a space followed by the end of the line: "-- \n"
I remember reading this years ago, but I don't remember where.
However I did an Internet search and found at least one site that
says this very thing:
http://www.helpdesk.demon.net/faq/welcome.html
But perhaps this is a very old standard that is quite frankly a bit
too subtle to ask today's average subscriber to adhere to.
Cordially,
The List Server Admin
list.admin at unh.edu
(currently Bill Costa)
From tomas at euronetics.se Tue Mar 23 17:48:03 1999
From: tomas at euronetics.se (Tomas Fasth)
Date: Tue, 23 Mar 1999 17:48:03 +0100
Subject: [Mailman-Users] How about this for the error message?
References:
<19990323101622.B8268@amber.org>
<14071.47398.987773.73798@anthem.cnri.reston.va.us> <14071.47968.457627.678502@anthem.cnri.reston.va.us>
Message-ID: <36F7C5C3.9C204DE@euronetics.se>
"Barry A. Warsaw" wrote:
>
> Me> I think the `standard' is actually double-dash-space-newline
> Me> "^-- \n" but from my quick scan of MailCommandHandler.py it
> Me> shouldn't matter either way.
>
> Except that you must have *some* whitespace following the double dash.
I thought the "standard" was double-dash-whitespace-anything.
-- Tomas
From bwarsaw at cnri.reston.va.us Tue Mar 23 17:55:11 1999
From: bwarsaw at cnri.reston.va.us (Barry A. Warsaw)
Date: Tue, 23 Mar 1999 11:55:11 -0500 (EST)
Subject: [Mailman-Users] How about this for the error message?
References: <19990323101622.B8268@amber.org>