Mailman 3 March 2001 - Mailman-Developers

mailman and the web
by Jerry Asher 20 Mar '01

20 Mar '01

I would like to unify mailman with an ACS bboard. The Open ACS (http://openacs.org) is a toolkit developing interactive community websites based on PostgreSQL and Tcl. ACS Bboards are just what they sound like, a discussion forum in which visitors to a site can post questions, or post answers to the questions that others have asked. Currently the ACS system is such that visitors can request that follow up answers be sent to them, which the ACS implements by creating hundreds of individual emails and mailing them out. The "From" header is set to a bogus address (bboard-robot) and the Reply-To: header is set to the author of the new answer. Vacation messages get sent back to the author of the post. This means that if you reply to a message, the answer IS NOT posted on the bulletin board. I would like to "unify" the ACS bulletin board code with "some" mailing list manager. The benefits to the ACS toolkit I perceive would include: 1 detecting/catching vacation messages 2 detecting/catching/(and maybe unsubscribing) bounces 3 ability to post directly to the bulletin from email (and have the posts end up in the right threads) 4 but I am worried about mail loops. Can someone tell me "how well" Mailman performs at detecting/catching vacation and bounce messages? I know little about mailing lists, or Mailman, or mail headers. But with a few pointers, I am willing to learn. I wonder what the issues involved in this unification are? Are the four issues I am interested/worried significant enough that it makes sense to want to delegate this to Mailman? Or would it be easier/satisfactory to "just change the headers of the outgoing emails, and create some sort of procmail script to process new messages and post them into the right thread? If you feel that unification would be useful for my purposes, can someone tell me the right "places" in Mailman to get started? If I want to "inject" a message into Mailman for sending out to others, how would I do that? If I wanted to "capture" the incoming message for processing, where do I look for that? And how might I implement threading? Thank you, Jerry Asher ===================================================== Jerry Asher jerry(a)hollyjerry.org 1678 Shattuck Avenue Suite 161 Tel: (510) 549-2980 Berkeley, CA 94709 Fax: (877) 311-8688

9 20

Re: [Mailman-Developers] mailman and the web
by J C Lawrence 20 Mar '01

20 Mar '01

On Sun, 18 Mar 2001 22:35:50 -0500 Jay R Ashworth <jra(a)baylink.com> wrote: > On Sun, Mar 18, 2001 at 10:17:04PM -0500, Barry A. Warsaw wrote: >> In Mailman, this message would be handed off to the local MTA as >> one copy, with 1000 recipients. Then it would be up to the MTA >> to do as efficient delivery as is possible. Mailman has some >> optional chunking parameters so that it can be tuned to your MTA, >> but in general, it tries to hand off the message with the fewest >> dialog with the MTA as possible, and then it gets out of the way. > And for good reason. From an adventure last week, on a P200: > Time to add 1000 recipients to a list with welcome mails: 70-80 > minutes. > Time to add 1000 recipients to a list *without* welcome mails: > roughly 2 minutes. Translation your MTA is misconfigured to do some sort of extra processing (likely domain verification) on list server messages as they are delivered. Bad Idea (and a FAQ). A local test: On a dual PII-33 with 512M RAM running Postfix under Debian/Woody (kanga.nu), I created a test list and then subscribed 1,000 addresses to it via `~/bin/add_members -n members -w y`. Total execution time for the add_members command: 38 seconds. The member list in question was artificial (a set of trash +-addresses on localhost I have set to deliver to /dev/null but which are unique as far as postfix and Mailman are concerned). Qrunner on the same machine, when delivering a single message to one of my lists which has almost exactly 1,000 members (slightly over) takes just over 3.5 seconds with SMTP_MAX_RCPTS set to 5. I presume the extra time for add_members is due to poor lock management, but have not checked. -- J C Lawrence claw(a)kanga.nu ---------(*) http://www.kanga.nu/~claw/ --=| A man is as sane as he is dangerous to his environment |=--

1 0

python 2.0 & mailman
by Ricardo Kustner 18 Mar '01

18 Mar '01

Hi, After almost a year of succesfull mailman usage, my server is starting to run into performance trouble again (probably due to an enormous increase in webtraffic and list subscribers). So I'm looking into ways of increasing performance again. I was wondering if python 2.0 has significant performance increase compared to 1.5.2? I'm sure I can sqeaze some more performance out of this Pentium-1 box :) ps: my message should be much less messy now... I'm using joe as my mutt editor... I remember the keybindings from QEdit from those good old DOS days so I think I can manage pps: I didn't realize I was amongst true hackers who still use emacs on here; I'm honored ! ;) -- Regards, Ricardo

1 0

german templates - final
by hypnose＠t-online.de 17 Mar '01

17 Mar '01

Hi. For all those who are interested, I have the templates of 2.1 translated to German language and usable in www. It´s a windows zip-archive of 18k. Those who want them, please reply to hypnose(a)t-online.de with subject "german templates - please send" guenter guenter wessling (hypnose(a)t-online.de)

1 0

Here's a little issue with the news gatewaying: If the n
by room_maildev＠bbs.pixel.citadel.org 16 Mar '01

16 Mar '01

Here's a little issue with the news gatewaying: If the news server you're trying to get to is down, or is slow and having problems mailman creates a lock file and doesn't unlock. The next time (every 5 minutes in my case) the gateway attempts to run the lock files are in place and it evidently fails. Is there anything I can do to clear those lock files manually, or, more importantly, is there anyway that mailman can be configured to time out, and remove its own lockfiles/ thanks.

5 15

Re: [Mailman-Developers] Problems with "make install"
by Magnus Solvang 15 Mar '01

15 Mar '01

Someone wrote in email: | Dunno. I've used Python 1.5.2 and Python 2.0, and not seen this; | maybe 1.6.1 was broken. | | Python itself is pretty easy to build from source. I did build Python from source just before installing Mailman, and the installation was flawless. The machine in question is an old Red Hat installation, with several old versions of Python on it, but as long as I set the PYTHONPATH to the new version and set --with-python to point at the new python-binary, this should be enough, right? - M (who hopes he can move to mailman-users soon :)

2 1

Problems with "make install"
by Magnus Solvang 15 Mar '01

15 Mar '01

I'm trying to setup Mailman on a Linux machine, and have installed Python 1.6.1 for this purpose. In other words, I'm new to both. Anyway, "configure" (and "make") looks good, but "make install" fails. Can anybody guide me through this problem? I have searched for clues via google.com, but have not come up with anything similar (at least not in this part of the process). I guess this is a problem with my Python-installation, but I would appreciate any hints on how to fix this. - M mailman-2.0.1]$ export PYTHONPATH=/usr/local/lib/python1.6 mailman-2.0.1]$ ./configure --with-python=/usr/local/bin/python1.6 \ --with-mail-uid=mailman --with-cgi-uid=nobody \ --with-python=/usr/local/bin/python1.6 [...] mailman-2.0.1]# make install [...] Listing /home/mailman/Mailman/pythonlib ... Compiling /home/mailman/Mailman/pythonlib/StringIO.py ... Compiling /home/mailman/Mailman/pythonlib/__init__.py ... Compiling /home/mailman/Mailman/pythonlib/cgi.py ... Compiling /home/mailman/Mailman/pythonlib/nntplib.py ... Compiling /home/mailman/Mailman/pythonlib/rfc822.py ... Compiling /home/mailman/Mailman/pythonlib/smtplib.py ... Compiling /home/mailman/Mailman/pythonlib/tempfile.py ... Compiling /home/mailman/Mailman/versions.py ... Traceback (innermost last): File "bin/update", line 25, in ? import os File "/usr/local/lib/python1.6/os.py", line 38, in ? import posixpath File "/usr/local/lib/python1.6/posixpath.py", line 319 _varprog = re.compile(r'\$(\w+|\{[^}]*\})') ^ SyntaxError: invalid syntax

2 1

PHP wrappers?
by Harold Paulson 15 Mar '01

15 Mar '01

Hey, I was trying to wrap several of the Mailman CLI tools (list_lists and such) in a little PHP so I could easily integrate them into a larger web site. Using PHP's System() function and the backtick operator to exec the Mailman tools never seems to result in any output, but no errors either. However, I can drop a script of my own (a little Perl hack or whatever) in the ~mailman/bin directory with the same ownership and permissions, and it seems to exec just fine. Any idea what I'm overlooking? - H -- Harold Paulson Sierra Web Design haroldp(a)sierraweb.com http://www.sierraweb.com VOICE: 775.833.9500 FAX: 810.314.1517

4 5

Wiki maintenance: bad link
by Dan Mick 14 Mar '01

14 Mar '01

Browsing the Wiki for the first time in a while, I see there's content there. I was interested in Future of Pipermail, Bill Bumgarner's patch; so I looked at the patch and was mystified; then I tried to go to the description and it was a bad link. Seeing no "comments/bugs to this mailto link" link on the page: http://www.zope.org/Members/bwarsaw/MailmanDesignNotes/FutureOfPipermail and no "owner" listed on the page above it: http://www.zope.org/Members/bwarsaw/MailmanDesignNotes/RecentChanges ...I appeal here for help.

1 0

DRAFT Datasource Interface v0.1
by Chris Ryan 14 Mar '01

14 Mar '01

All, I've written up a templatish python code of the inteface with some comments and the like. I've placed everything on the Mailman ZWiki page http://www.zope.org/Members/bwarsaw/MailmanDesignNotes/CentralizedUserDatab… at Ricardo's sugestion so that you may all read it over and comment/add/substract ?delete? :). There were a few places that I left empty in the interface but I commented about it these were The Admin/Request/Postings Response classes and the bounce_info stuff which will be located in the BounceData class. This was due to my lack of overall understanding of what these are at this time. Feel Free to fill them in following the overall flow of the rest. Once we get something that is more or less acceptable my next step will be to impliment the interface to use the existing method of data storage. Then I would start modifying the code to use the interface. Once I've got the interface implimented we should have a concreate definition and anyone should be able to write a datasource driver using the interface to access what ever source they wish to use. Chris Ryan chris(a)greatbridge.com

3 3