August 2025 - Mailman-users - lists.mailman3.org

db migration from mailman3.2.2 to mailman3.3.9 (sqlite3 to postgres16)
by Peter Thornton Aug. 13, 2025

Aug. 13, 2025

Hello, I know this question comes up here periodically, but I am having some trouble performing a data migration from an older version of mailman3 to a more recent one. We have a legacy environment with: mailman3.2.2 postorius 1.2.4 hyperkitty 1.2.2 there are two sqlite3 dbs for mailman core and mailman web. I've staged a new environment using the docker-mailman project with the following: mailman3.3.9 postorius 1.3.10 hyperkitty 1.3.8 there is a single postgres db here for both core and web. I've tried migration a few different ways, using pgloader, and also with some manual python scripts. Data is definitely not my forte, but I am struggling to get a working instance. I am running into several issues such as foreignkeyviolations, datatype mismatches, unique PK errors, etc.. Does anyone out there have a prescriptive process for this migration, or a suggestion for workflow/tooling that has worked in a similar upgrade process? Thanks! -Peter

2 1

Website 404 error
by peter＠pi4cc.nl Aug. 11, 2025

Aug. 11, 2025

Hello I'm running Debian 12 with the latest version and installed mailman3 via apt install-mailman3-full. Everything works, I'm using MySQL as the database. There's something strange, and I can't find a source/solution. When I first access the website domain.nl/mailman3, it's redirected to domain.nl/mailman/postorius/lists/, resulting in a 404 error (3 after mailman is missing). When I accessed it a second time with the same link (domain.nl/mailman3), I was redirected to domain.nl/mailman3/postorius/lists and the welcome page appeared. This is strange... I tested it on a mobile device with Chrome and Firefox, and on a desktop with Chrome and Edge. All the results were the same. Is there a specific file to look add? Regards, Peter

3 2

Migration of archives mailman2 to mailman3: python: can't open file '/opt/mailman/manage.py': [Errno 2] No such file or directory
by Wikinaut Aug. 10, 2025

Aug. 10, 2025

Your current *) documentation says Migrate the list archives from Mailman 2 to Mailman 3 by running the following command: $ python manage.py hyperkitty_import -l foo-list(a)example.com $var_prefix/archives/private/foo-list.mbox/foo-list.mbox (venv) /opt/mailman $ python manage.py python: can't open file '/opt/mailman/manage.py': [Errno 2] No such file or directory What did I wrong or what is that part of your documentation missing? *) https://docs.mailman3.org/en/latest/migration.html

3 6

I wanted to upgrade the mailman3 documentation, but cannot see the promised link "edit on github"
by Wikinaut Aug. 10, 2025

Aug. 10, 2025

https://docs.mailman3.org/en/latest/documentation.html#edit-on-gitlab Please can you fix that?

1 0

Vote irregularities
by Ron Aug. 10, 2025

Aug. 10, 2025

I've been playing with the voting system on a couple of lists (this one included). There are some weird behaviours I've noticed: * votes are frequently not immediately reflected - a page reload is required * some votes display as: 0 0 You dislike it (cancel) Here we see I've voted on a message, but the counts are still zero up and zero down. I just down-voted one of my posts as a test, and nothing changed in the page to reflect the vote. Often it takes multiple clicks, sometimes a page reload. Browser console shows voting generates a 200 response to a POST method, but the JSON confuses me: like 0 dislike 0 html '\n\n <form method="post" class="likeform"\n action="/archives/list/talk(a)lists.domain.tld/message/IK6WH4IQPZXOG3CIFV4CJGOKVFARGXA5/vote">\n <input type="hidden" name="csrfmiddlewaretoken" value="***">\n \n\t\t\n\t\t\t<a class="youlike vote "\n\t\t\thref="#like" data-vote="1" aria-label="Like thread">\n\t\t\t\t\t<i class="fa fa-thumbs-o-up"></i> 0\n\t\t\t</a>\n\t\t\t<a class="youdislike vote"\n\t\t\thref="#dislike" data-vote="-1" aria-label="Dislike thread">\n\t\t\t\t<i class="fa fa-thumbs-o-down"></i> 0\n\t\t\t</a>\n\t\t\n \n\n </form>\n' The uwsgi.log file entry for that vote is: [pid: 4010303|app: 0|req: 552908/575956] 127.0.0.1 () {66 vars in 1374 bytes} [Sun Aug 10 02:28:21 2025] POST /archives/list/talk(a)lists.domain.tld/message/IK6WH4IQPZXOG3CIFV4CJGOKVFARGXA5/vote => generated 696 bytes in 376 msecs (HTTP/1.1 200) 9 headers in 394 bytes (1 switches on core 1) Following the Permalink beside the voting options loads the specific message with no votes showing. Ctrl+F5 shows no votes. Re-casting vote and reloading shows no votes. Clicking "Back to thread" shows my message and: > 0 0 You dislike it (cancel) I don't think I've ever seen reliable voting on any MM3 site, including this one (I don't think I've seen zero / zero You {like|dislike} it (cancel) here, but multiple vote attempts to see the results is not unusual). Searching the archive shows something from ~6-7 years ago where caching was modified, seems relevant but nothing since. Am I alone in this, or is the feature so unused that nobody else noticed this?

2 1

Problem upgrading from 2.1 to 3, dockerized
by jesper.holck＠ibsgaarden.dk Aug. 9, 2025

Aug. 9, 2025

I recently followed the instructions here https://github.com/g4rf/dockerized-mailcow-mailman, and was able to create a new Mailman 3, Debian-based installation. I then followed the instructions here https://docs.mailman3.org/en/latest/migration.html in order to migrate from my old, Centos-based installation, to the new, Debian-based installation. It seemed to work well, but unfortunately a problem showed up. Whenever someone posted to one of the lists on the new installation, the messages were rejected by Postfix: postfix-mailcow-1 | Aug 5 20:55:48 ef2b760c2e61 postfix/cleanup[8681]: B7B856099EC: message-id=<A8946B8E-2A47-4701-9548-4544981FAC5E(a)gmail.com> postfix-mailcow-1 | Aug 5 20:55:49 ef2b760c2e61 postfix/cleanup[8681]: B7B856099EC: milter-reject: END-OF-MESSAGE from unknown[172.22.1.1]: 5.7.1 This message does not meet our delivery requirements; from=<mylist-bounces(a)myserver.dk> to=<mylistuser1(a)gmail.com> proto=ESMTP helo=<[172.19.199.4]> postfix-mailcow-1 | Aug 5 20:55:50 ef2b760c2e61 postfix/cleanup[8681]: 216D36099EC: message-id=<A8946B8E-2A47-4701-9548-4544981FAC5E(a)gmail.com> postfix-mailcow-1 | Aug 5 20:55:50 ef2b760c2e61 postfix/cleanup[8681]: 216D36099EC: milter-reject: END-OF-MESSAGE from unknown[172.22.1.1]: 5.7.1 This message does not meet our delivery requirements; from=<mylist-bounces(a)myserver.dk> to=<mylistuser2(a)mail.dk> proto=ESMTP helo=<[172.19.199.4]> So, for each of the existing lists, I had to delete the old list and create a new one - with the same name and the same users. After this, everything seems to work as expected. Thanks for all the work you all has put into this, Jesper

2 1

mailman addmember/syncmember commands print undecoded UTF-8 [ADD] or [DEL] mail addresses in its messages like "[ADD] =?utf-8?q?John_=....."
by Wikinaut Aug. 8, 2025

Aug. 8, 2025

My post here is merely about the information output when running mailman addmembers/syncmembers where I discovered strange looking lines. I noticed that addmember and syncmember output looks like [ADD] =?utf-8?q?John_=28Foo=29_Doe_=5Bcomment1=5D_=5Bcomment2=5D?= <john.foo.doe(a)example.com> whereas the entries in the Web-GUI look fine. I triple-checked the textfile from which the names stem from, yes, the file is UTF8. Is this behaviour of addmember/syncmember command output commonly known - or is this a small cosmetic bug? PS. At some moment I had, probably from my test and legay data file, also these codings in about fifty entries of my users lists, but deleted that users with a small script, so that user list is now clean. My post here is merely about the information output when running mailman addmembers/syncmembers.

3 9

clearing pending db
by Morris Jones Aug. 8, 2025

Aug. 8, 2025

I have an admittedly out-of-date installation of Mailman 3.3 (3.3.2 Tom Sawyer) but it's been working without issues for a long time, so I haven't had much incentive to bring it up to date. Sorry! A couple of my larger mailing lists (around 500 members) are currently very slow when making changes through the UI (maybe that's my raison-to-update). They also both have very large lists of members "pending confirmation." My suspicion is that the large pending list is slowing down other operations. But pendingdb.evict() doesn't seem to clear the list. For example, on one of my lists: >>> from mailman.interfaces.pending import IPendings >>> >>> from zope.component import getUtility >>> >>> pendingdb = getUtility(IPendings) >>> pendingdb.count() 5040 >>> pendingdb.evict() >>> pendingdb.count() 5040 >>> I have several lists, but none of them are extremely active, so I don't have a problem with wiping the pending db for all of them. And I also might be totally wrong in my guess. Mojo -- Morris Jones, Monrovia, CA

3 3

Support question, not a bug: Is it possible and how, programmatically to change all settings from domain1 to domain2?
by Wikinaut Aug. 8, 2025

Aug. 8, 2025

@Mark & Stephen, now, where here everything appears to run smoothly on a new server, I want to change the testmail.domain (domain1) settings to mail.domain (domain2) settings. It would be a great help, if this can be done by python script respectively a mailman3 shell script. Is there any on your desk, or can you help me? I already defined in mailman/admin (django admin) that new target domain. I guess, at the end, there might also be a very changes in mailman.cfg and settings.py necessary.

1 0

Main daemon process sticks around after forking
by Jérôme Charaoui Aug. 8, 2025

Aug. 8, 2025

Hello, I'm running a (rather large) mailman3 instance on Debian 13 (trixie) using the distribution packages. To manage the mailman3 daemon, the package installs this systemd service unit: https://sources.debian.org/src/mailman3/3.3.10-2/debian/mailman3.service/ It has been working fine until recently: after a reboot, we noticed the mailman3 unit consistently failing to start due to a timeout error. I figured out that the timeout was related to the service unit type (Type=forking) and the fact that the main/parent process was forking subprocesses (and running normally), writing a PID file, but it wasn't exiting, so systemd identifies this as an error and aborts the service. Since switching to Type=simple, it has been working fine. I'm wondering if anyone else has seen such behavior and how I could troubleshoot it. Thanks! -- Jérôme

2 3