Mailman 3 January 2004 - Mailman-Developers

Filtering certain characters from email addresses?
by Matt Goebel 23 Jan '04

23 Jan '04

Hello, Searched the archives and didn't see any references to this. Would it be possible to add some character filtering for email addresses before they are added to a list? I've been having more and more problems with admins using the mass subscribe function where they are cutting and pasting from word and picking up null characters, and other odd things. Mailman will happily add the addresses but then can't do anything with them later on, forcing me to use a hex editor to remove the odd characters. Thanks, Mattias -- Matthew Goebel : goebel(a)emunix.emich.edu : Unix Jockey @ EMU : Hail Eris Neo-Student, Net Lurker, Donut consumer, and procrastinating Furry Fan. "Always with the negative waves, Moriarty" - Oddball "Comfort the troubled, and trouble the comfortable." - Dietrich Bonhoeffer

1 0

mailman setup
by Guillaume Rousse 23 Jan '04

23 Jan '04

Hello. I'm one of the guy managing the mailman package for mandrake-linux. I've currently trying to make setup a bit more standard and FHS compliant. I need developpers help however, first to make sure my changes won't cause trouble, second as I think some of these change should get merged upstream, as they would benefit more people. The first issue is file location. We are trying to achieve FHS compliance, so i had to move things a bit. I'm using the --with-var-prefix configure option to have variable files in /var/lib/mailman and non-variables files in /usr/lib/mailman. That is supported by mailman install. Then i move logs from /var/lib/mailman/logs to /var/log/mailman through a symlink. The config file location, however, is still puzzling me. Having a configuration file (mm_cfg.py) under /usr is non-standard, as /usr is supposed to be read-only. Moreover, it is a pain for us packagers, as we have to include everything but this file as normal files, then tag it as config file. I'd really think it should be moved in /etc instead, either by me using a symlink, but much better by an upstream change. The second issue are perms. The default install make everything owned by the uid/gid determined at configure stage (mail/mail for me), and make all directories setgid, even for non-variables files, which is quite really a non-standard practice. So i tried to clean it a bit. I made everything under /usr/lib/mailman owned by root.root, with standard perms (644 for files, 755 for directories) except the binaries in /usr/lib/mailman/cgi-bin and usr/lib/mailman/mail which are setgid, and owned by mail group. I tested, it seems to work. I also made top-level directory for variable files (/var/lib/mailman) owned by root/root with standard 755 perms, but i didn't tried yet further perm modifications there, as mailman need write permissions in subdirectories. You can have a look at the spec file in CVS if you want details (http://cvs.mandrakesoft.com/cgi-bin/cvsweb.cgi/SPECS/mailman/mailman.spec). Your comments on my changes are welcome, as well as what could get integrated into mailman. -- Those who hesitate under fire usually do not end up KIA, MIA or WIA -- Murphy's Military Laws n°101

3 4

embedded html problem with large lists
by ryan 23 Jan '04

23 Jan '04

No one has been able to answer this on mailman-users for over a week. I'm hoping one of you can help diagnose this problem. On one of my Mailman lists, I send out messages with embedded HTML. These messages contain inline images. I sent out 2 test messages to a pre-configured list. For the two tests, the list had only 6 staff members subscribed. The message looked great, everyone saw the content with the image correctly aligned in the body of the message, between the two blocks of text where it belonged. However, after I added the 6700+ actual email addresses to the SAME list, and sent out the SAME email, the image was removed from the body of the message and included as an attachment instead. This essentially made my pretty little email look like crap. Again, the same list (same configuration), and the same email were used for the successful test, and the unsuccessful real mail-out. I cannot stress enough that these were the *exact* same emails, for both the test with 6 users, and the later mailing with 6700+; the filesizes were the exact same, and the HTML was unchanged. So... my only theory is: at some cut-off point, Mailman decides that there are two many users on a list to embed html/images into every email. Perhaps because it is more efficient to store the html once and then reference it as an attachment when each member is emailed? Or could this be some behavior of my mailer daemon, Sendmail? Any thoughts? Thanks in advance, Ryan

2 1

BounceRunner.py patch
by SHIGENO Kazutaka 23 Jan '04

23 Jan '04

Hi, Please use if the following patch is required for Mailman. --- Mailman/Queue/BounceRunner.py- Sat Nov 22 07:45:04 2003 +++ Mailman/Queue/BounceRunner.py Thu Jan 8 19:15:56 2004 @@ -68,7 +68,7 @@ # All messages to list-owner(a)vdom.ain have their envelope sender set # to site-owner(a)dom.ain (no virtual domain). Is this a bounce for a # message to a list owner, coming to the site owner? - if msg.get('to', '') == Utils.get_site_email(extra='-owner'): + if msg.get('to', '') == Utils.get_site_email(extra='owner'): # Send it on to the site owners, but craft the envelope sender to # be the -loop detection address, so if /they/ bounce, we won't # get stuck in a bounce loop. -- shigeno

2 1

images embedded in html: inline vs. attached
by ryan 23 Jan '04

23 Jan '04

Hi, I've tried searching the list archives, and posting to the mailman-users list, but I can't seem to find an answer to my problem. Hopefully someone on this list can shed some light... A quick background: At the non-profit museum where I work, we use Mailman to inform our paying members (and other general members of the public who signup to our lists) about our current exhibitions and events. We like to use HTML formatted messages with in-body (inline?) images for this purpose. It looks nice, and conforms with the overall look of our website and other publications in print. I realize sending out bulk HTML mailings is probably not the most typical use of Mailman, but it works great for us, and allows people who don't want our emails to quickly remove themselves without requiring our intervention. Here's the problem: Today I sent out 2 test messages to a pre-configured list. For the two tests, the list had only 6 staff members subscribed. The message looked great, everyone saw the content with the image correctly aligned in the body of the message, between the two blocks of text where it belonged. However, after I added the 6700+ actual email addresses to the SAME list, and sent out the SAME email, the image was removed from the body of the message and included as an attachment instead. This essentially made our pretty little email look like crap. This behavior occurred on all the computers of the 6 staff members where I had previously tested, plus on additional accounts that I was able to confirm through phone calls, etc. Again, the same list (same configuration), and the same email were used for the successful test and the unsuccessful real mail-out. I cannot stress enough that these were the *exact* same emails, for both the test with 6 users, and the later mailing with 6700+; the filesizes were the exact same, and the HTML was unchanged. So... my only theory is: at some cut-off point, Mailman decides that there are two many users on a list to embed an image in every email. Perhaps because you thought it more efficient to store the image once and then reference it as an attachment when each member is emailed? Or could this be some behavior of my mailer daemon, Sendmail? Any thoughts? Thanks in advance, Ryan

2 1

Mailman/PostgreSQL
by Robby Russell 22 Jan '04

22 Jan '04

I am considering starting a new python program that would work very similar to mailman but involve a PostgreSQL back end. I haven't looked too far into the mailman code to determine just how much data it is currently storing, so I am not sure if an abstraction layer to pgsql is the best solution for what I am aiming to do. My goal is to create an application that works very similar to mailman with a database back end and contains reporting. At the moment, I am not sure whether or not mailman has any reporting features that would generate pretty pages with graphs, stats, etc. Is there a reporting feature for mailman that someone has built already? I want statistics of who posts the most on a list, how many people subscribed between dates, by month, year, how many people unsub'd during any specific period. (perhaps people unsub'd during a heated thread... i'd like to track that sort of information). So, as Mailman developers, do you see an 'easier' route using the existing infrastructure to handle this sort of reporting with a db backend? or should I start from scratch and use some of mailmans technology in the process? Anyone interested in assisting me in such a project? are there any projects that some of you may be aware that already exist that work like this? Cheers, Robby -- #------------------------------------------------------- # Robby Russell, | Sr. Administrator / Lead Programmer # Command Prompt, Inc. | http://www.commandprompt.com # rrussell(a)commandprompt.com | Telephone: (503) 667.4564 #-------------------------------------------------------

1 0

Redirect list to list?
by Michael Heydekamp 22 Jan '04

22 Jan '04

Hi, we want all messages sent by moderated members or non-members to our CVS list be automatically redirected to our Dev list (rather than really moderating or discarding them). Only a specific address (in fact the list address itself) and the non- moderated members shall be allowed to post to the list without being redirected. Is there any patch or other solution which does that job? Michael

3 4

List locks not getting relinquished
by Chris Boulter 22 Jan '04

22 Jan '04

Hi, I'm having some problems with lists not getting unlocked. I don't really know how to set about debugging this and I haven't found anything in this list's archive or the documentation to help. My Mailman installation is quite non-standard, so it may be hard to provide help, but I'd appreciate hints about how to debug the locking mechanism (for instance, can I tell which process created a lock?). I'm using Mailman 2.1.2. We have sync_members running in a loop from a daemon, syncing ~500 lists every few hours. Of these, half a dozen lists are failing (and my daemon has to forcibly kill the sync_members process when it detects a timeout). There seems to be no common feature to the lists which fail. When it happens, I can forcibly break the locks and restore correct behaviour by manually deleting Mailman/locks/*, but then the problem recurs eventually. However, once I've broken the locks, I can successfully run sync_members on the lists from the command line hundreds or thousands of times without failure, either making changes to the subscribers or not each time. I tried setting LIST_LOCK_DEBUGGING = On but didn't find the result enlightening (I can post some here if anyone thinks this is a good place to look). So maybe it's something about the daemon which doesn't relinquish a lock. What the daemon does is mostly to run sync_members but occasionally to run newlist or a Python script of my devising (to change a user's password on a list, or change the list owner). Possibly my own scripts are bad: I don't know much about Python, but I've checked them as carefully as I can. I've posted them here in case any kind individual could cast an eye over them (mostly kludged together from distributed Mailman code): http://chris.jellybaby.net/mailman/ The command being run by the daemon, and the stacktrace when I ran the command manually and hit ^C are below (nb I've modified sync_members and added an '--ignore-invalid' option, so the line numbers below aren't correct for the standard distro). /usr/local/mailman/bin/sync_members --ignore-invalid --welcome-msg=no --goodbye-msg=no --notifyadmin=no --file /var/tmp/mlmsync124897.tmp latin.america ^CTraceback (most recent call last): File "/usr/local/mailman/bin/sync_members", line 301, in ? main() File "/usr/local/mailman/bin/sync_members", line 234, in main mlist = MailList.MailList(listname) File "/usr/local/mailman/Mailman/MailList.py", line 122, in __init__ self.Lock() File "/usr/local/mailman/Mailman/MailList.py", line 155, in Lock self.__lock.lock(timeout) File "/usr/local/mailman/Mailman/LockFile.py", line 312, in lock self.__sleep() File "/usr/local/mailman/Mailman/LockFile.py", line 496, in __sleep time.sleep(interval) KeyboardInterrupt Many thanks, Chris

2 3

Bug in Mailman 2.1.4 {Subscription problem}
by Martin Maechler 22 Jan '04

22 Jan '04

Dear Mailman developers, A user ("ADBONT@...", cc'ed) who wanted to subscribe to one of our mailing lists received a >> Bug in Mailman version 2.1.4 >> >> We're sorry, we hit a bug! and sent me the full message, I've found the same message in mailman's data/logs/error file, and append it below. Note that this is quite an active mailing list, ---> https://www.stat.math.ethz.ch/mailman/listinfo/bioconductor/ and many people do subscribe and unsubscribe successfully both through the web and the email interface. Though -- I've got reports more than once that subscription through the web interface produced a "you are subscribed ...." confirmation (web) and welcome email and the person was *not* subscribed after all. That is really another problem, but maybe related... Note that our setup is a bit non-standard maybe: Web server ("hubble") is Redhat9 Mail server ("hypatia") is Solaris 8 -- mailman installed on both Many thanks for mailman, and for your consideration of this report. Martin Maechler <maechler(a)stat.math.ethz.ch> http://stat.ethz.ch/~maechler/ Seminar fuer Statistik, ETH-Zentrum LEO C16 Leonhardstr. 27 ETH (Federal Inst. Technology) 8092 Zurich SWITZERLAND phone: x-41-1-632-3408 fax: ...-1228 <>< Here the excerpt from <mailman>/data/logs/error : Jan 21 11:03:34 2004 admin(31707): @@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@ admin(31707): [----- Mailman Version: 2.1.4 -----] admin(31707): [----- Traceback ------] admin(31707): Traceback (most recent call last): admin(31707): File "/sfs/s/solaris/8/app/mailman-sfs/2.1.4/scripts/driver", line 87, in run_main admin(31707): main() admin(31707): File "/sfs/s/solaris/8/app/mailman-sfs/2.1.4/Mailman/Cgi/subscribe.py", line 69, in main admin(31707): language = cgidata.getvalue('language') admin(31707): File "/scratch/local/app/python/2.3/lib/python2.3/cgi.py", line 560, in getvalue admin(31707): if key in self: admin(31707): File "/scratch/local/app/python/2.3/lib/python2.3/cgi.py", line 611, in __contains__ admin(31707): raise TypeError, "not indexable" admin(31707): TypeError: not indexable admin(31707): [----- Python Information -----] admin(31707): sys.version = 2.3 (#1, Oct 16 2003, 05:04:34) [GCC 3.2.2 20030222 (Red Hat Linux 3.2.2-5)] admin(31707): sys.executable = /scratch/local/bin/python admin(31707): sys.prefix = /scratch/local/app/python/current admin(31707): sys.exec_prefix = /scratch/local/app/python/current admin(31707): sys.path = /scratch/local/app/python/current admin(31707): sys.platform = linux2 admin(31707): [----- Environment Variables -----] admin(31707): force_response_1_0: 1 admin(31707): SSL_SERVER_I_DN_OU: Department of Mathematics admin(31707): SSL_CIPHER_EXPORT: false admin(31707): SSL_SERVER_S_DN_Email: webmaster(a)math.ethz.ch admin(31707): SERVER_SOFTWARE: Apache/2.0.40 (Red Hat Linux) admin(31707): SSL_VERSION_INTERFACE: mod_ssl/2.0.40 admin(31707): SCRIPT_NAME: /mailman/subscribe admin(31707): SSL_SERVER_A_KEY: rsaEncryption admin(31707): SSL_SERVER_S_DN_ST: Switzerland admin(31707): SERVER_SIGNATURE: <address>Apache/2.0.40 Server at www.stat.math.ethz.ch Port 443</address> admin(31707): admin(31707): REQUEST_METHOD: POST admin(31707): PATH_INFO: /bioconductor admin(31707): SERVER_PROTOCOL: HTTP/1.1 admin(31707): SSL_SERVER_S_DN: /C=CH/ST=Switzerland/L=Zurich/O=ETH Zurich/OU=Seminar for Statistics/CN=www.stat.math.ethz.ch/emailAddress=webmaster(a)math.ethz.ch admin(31707): SSL_CIPHER: RC4-MD5 admin(31707): SSL_SERVER_V_START: Feb 7 16:20:08 2003 GMT admin(31707): SSL_SESSION_ID: 319B2196DF311D397B6A4EBE58D8F8307D04AE1C58665697A02A9A59325182B9 admin(31707): CONTENT_LENGTH: 96 admin(31707): SSL_CLIENT_VERIFY: NONE admin(31707): SSL_SERVER_I_DN_ST: Switzerland admin(31707): SSL_SERVER_S_DN_O: ETH Zurich admin(31707): HTTP_REFERER: https://www.stat.math.ethz.ch/mailman/listinfo/bioconductor, https://www.stat.math.ethz.ch/mailman/listinfo/bioconductor admin(31707): SERVER_NAME: www.stat.math.ethz.ch admin(31707): REMOTE_ADDR: 148.177.129.213 admin(31707): SSL_CIPHER_ALGKEYSIZE: 128 admin(31707): SSL_SERVER_I_DN: /C=CH/ST=Switzerland/L=Zurich/O=ETH Zurich/OU=Department of Mathematics/CN=ISG D-MATH/emailAddress=webmaster(a)math.ethz.ch admin(31707): SSL_SERVER_I_DN_C: CH admin(31707): SSL_SERVER_M_VERSION: 1 admin(31707): SSL_SERVER_I_DN_O: ETH Zurich admin(31707): SERVER_ADDR: 129.132.148.136 admin(31707): DOCUMENT_ROOT: /var/www/vhosts/sfs/htdocs admin(31707): SSL_SERVER_S_DN_OU: Seminar for Statistics admin(31707): SERVER_PORT: 443 admin(31707): SSL_SERVER_I_DN_Email: webmaster(a)math.ethz.ch admin(31707): SSL_VERSION_LIBRARY: OpenSSL/0.9.7a admin(31707): PYTHONPATH: /sfs/s/solaris/8/app/mailman-sfs/2.1.4 admin(31707): SCRIPT_FILENAME: /var/www/vhosts/sfs/mailman/subscribe admin(31707): SERVER_ADMIN: webmaster(a)math.ethz.ch admin(31707): SCRIPT_URI: https://www.stat.math.ethz.ch/mailman/subscribe/bioconductor admin(31707): SSL_SERVER_A_SIG: md5WithRSAEncryption admin(31707): SSL_SERVER_S_DN_L: Zurich admin(31707): HTTP_HOST: www.stat.math.ethz.ch admin(31707): SCRIPT_URL: /mailman/subscribe/bioconductor admin(31707): HTTPS: on admin(31707): SSL_SERVER_I_DN_L: Zurich admin(31707): HTTP_CONNECTION: Keep-Alive admin(31707): PATH_TRANSLATED: /var/www/vhosts/sfs/htdocs/bioconductor admin(31707): HTTP_CACHE_CONTROL: no-cache admin(31707): SSL_SERVER_S_DN_C: CH admin(31707): REQUEST_URI: /mailman/subscribe/bioconductor admin(31707): HTTP_ACCEPT: image/gif, image/x-xbitmap, image/jpeg, image/pjpeg, application/vnd.ms-powerpoint, application/vnd.ms-excel, application/msword, application/x-shockwave-flash, */* admin(31707): SSL_SERVER_M_SERIAL: 07 admin(31707): SSL_SERVER_S_DN_CN: www.stat.math.ethz.ch admin(31707): HTTP_USER_AGENT: Mozilla/4.0 (compatible; MSIE 5.5; Windows NT 5.0; JP Rev04) admin(31707): nokeepalive: 1 admin(31707): GATEWAY_INTERFACE: CGI/1.1 admin(31707): SSL_PROTOCOL: SSLv3 admin(31707): REMOTE_PORT: 19699 admin(31707): HTTP_ACCEPT_LANGUAGE: en-au, en-au admin(31707): HTTP_ACCEPT_ENCODING: gzip, deflate, gzip, deflate admin(31707): ssl_unclean_shutdown: 1 admin(31707): SSL_SERVER_V_END: Feb 4 16:20:08 2013 GMT admin(31707): CONTENT_TYPE: application/x-www-form-urlencoded, application/x-www-form-urlencoded admin(31707): SSL_SERVER_I_DN_CN: ISG D-MATH admin(31707): QUERY_STRING: admin(31707): SSL_CIPHER_USEKEYSIZE: 128 admin(31707): downgrade_1_0: 1 admin(31707): UNIQUE_ID: TLeriIGElIIAAHklESEAAAAI

2 1

Shunt
by Juan Enrique Gómez 22 Jan '04

22 Jan '04

Hi! could any one explain me what's intended for the shunt directory in mailman?.. just curiosity... Thanks in advance, -- Saludos, -------------------------------------------------------------------------- |Juan Enrique Gomez Perez |Metropoli2000 Networks, S.L. | Phone: +34 914250023 Fax: +34 914250136 | Mobile: +34 615 922 820 | email: juan.enrique.gomez(a)metropoli2000.com -------------------------------------------------------------------------- PGP Fingerprint: 6B39 3A2B A17B 1E8E CFFD FC14 678E 0A22 BD80 C486 Public Key: http://pgp.mit.edu:11371/pks/lookup?op=get&search=0xBD80C486 If this message hasn't a correct signature please notify it to juanen(a)metropoli2000.com. To get the public key please use the above url. Thanks for your help. --------------------------------------------------------------------------

2 1