2024-02-23T04:09:28  *** SFaulken_ is now known as SFaulken
2024-02-23T09:26:55  *** ldevulder_ is now known as ldevulder
2024-02-23T10:40:51  *** teepee_ is now known as teepee
2024-02-23T16:46:14  *** microchip_ is now known as OfficerPendejo
2024-02-23T19:31:51  *** teepee_ is now known as teepee
2024-02-23T19:48:06  <acidsys> is it only my account or does gitlab no longer tick "delete source branch" by default? in the repository settings we still have "Enable option by default" set for it yet recently I always need to manually tick it
2024-02-23T20:01:11  <cboltz> at least the form to create a MR has it enabled by default for me (I don't want to create a dummy MR just to test this)
2024-02-23T20:02:04  <acidsys> will have to test with the web form, I realize I always use the cli lately
2024-02-23T20:02:44  <cboltz> maybe the cli doesn't honor the web ui defaults?
2024-02-23T20:03:25  <acidsys> now that you say it, there indeed is a dedicated --remove-source-branch toggle
2024-02-23T20:03:34  <acidsys> I'll use that next time
2024-02-23T20:03:42  <acidsys> no short hand for it >(
2024-02-23T20:03:55  <cboltz> also not a config file to set it as default?
2024-02-23T20:05:28  <acidsys> does not seem like it. but whilst searching for it I found there is a way to set a default editor. so at least no more "nano not found" error
2024-02-23T20:05:54  <acidsys> alias it is
2024-02-23T20:06:18  <acidsys> oh, it supports built in aliases
2024-02-23T20:07:34  <acidsys> cool now I should have: glab mkmr => mr create -fy --remove-source-branch
2024-02-23T20:08:00  <acidsys> now what to contribute next to try it out
2024-02-23T20:43:20  <robin_listas> Hi, I suspect mail lists are down. No mails today on users, factory, de.
2024-02-23T20:45:17  <cboltz> indeed, even the daily progress.o.o ticket spam (which is also routed via a ML) is missing
2024-02-23T21:18:11  <cboltz> rebooting mailman.i.o.o (needed for some updates anyway, let's see if it also fixes mailman)
2024-02-23T21:50:32  <cboltz> I'm afraid the problem isn't that easy - when a mail arrives, mailman errors out with
2024-02-23T21:50:35  <cboltz> ModuleNotFoundError: No module named 'mailman_hyperkitty'
2024-02-23T21:50:54  <cboltz> Feb 23 21:47:03 2024 (1360) SHUNTING: [...]
2024-02-23T21:51:28  <cboltz> acidsys: any idea what could be wrong?
2024-02-23T21:53:05  <acidsys> yes
2024-02-23T21:53:11  <acidsys> mailman3 (lists.o.o):~ # head -n1 /usr/bin/mailman
2024-02-23T21:53:12  <acidsys> #!/usr/bin/python3.12
2024-02-23T21:53:16  <acidsys> i+ | python311-mailman-hyperkitty | package | 1.2.1-2.1 | noarch | repo-oss
2024-02-23T21:53:38  <cboltz> indeed, that explains it
2024-02-23T21:54:27  <acidsys> https://build.opensuse.org/package/view_file/devel:languages:python:mailman/python-HyperKitty/python-HyperKitty.spec?expand=1
2024-02-23T21:54:37  <acidsys> is this comment "# Newest python supported by mailman is Python 3.11" still accurate? should we try building against 3.12 ?
2024-02-23T21:56:16  <cboltz> osc blame says     66 (gladiac      2023-07-10 06:35:52    57) # Newest python supported by mailman is Python 3.11
2024-02-23T21:56:47  <acidsys> sounds timewise like just the latest available at the time
2024-02-23T22:00:16  <cboltz> since then, the .changes mentions an update to from 1.3.7 to 1.3.8, but I don't know if that also includes python 3.12 support
2024-02-23T22:00:41  <cboltz> OTOH, it's broken already, so testing with 3.12 shouldn't break too much ;-)
2024-02-23T22:01:05  <acidsys> yep. and I'd be surprised if there's anything major breaking with 3.12
2024-02-23T22:01:59  <cboltz> I'd hope so, but since upstream changelog mentiones fixes for (at least) 3.10 and 3.11, I wouldn't be too surprised if there are also issues with 3.12
2024-02-23T22:16:56  <acidsys> are you trying it or should I
2024-02-23T22:17:36  <cboltz> I'll let you try it
2024-02-23T22:40:51  <acidsys> ok, please test
2024-02-23T22:51:04  <cboltz> looks like it works again - a mail to the test ML went through
2024-02-23T22:51:22  <acidsys> ok, then only mailman-web left
2024-02-23T22:51:30  <cboltz> now the question is if the shunted mails get delivered automatically, or if they need manual intervention
2024-02-23T22:58:22  <cboltz> answering myself - it needs   mailman unshunt   (run as "mailman")
2024-02-23T22:58:29  <cboltz> now quite some mails are sent out
2024-02-23T23:00:17  <acidsys> cool!
2024-02-23T23:00:26  <acidsys> what is managing /etc/systemd/system/mailman-web.service ? I do not find it in Salt
2024-02-23T23:00:49  <acidsys> it needs to be amended to call gunicorn with the same python version as mailman-web instead of using the alternatives link
2024-02-23T23:02:18  <acidsys> but that is only a minor problem, I edited the override for now. more problematic: "ModuleNotFoundError: No module named 'settings'"
2024-02-23T23:03:24  <cboltz> no idea about mailman-web.service
2024-02-23T23:04:10  <cboltz> that "ModuleNotFound" might be related to the fact that openSUSE:infrastructure:mailman3/python-mailman-web fails to build (and is therefore probably outdated)
2024-02-23T23:05:26  <acidsys> ah that also ships a mailman-web.service :-)
2024-02-23T23:05:38  <acidsys> odd why we have one in /etc then
2024-02-23T23:06:52  <cboltz> that's indeed strange - is there a diff to the /usr/ version?
2024-02-23T23:07:03  <acidsys> no the /usr one does not exist at all
2024-02-23T23:07:12  <acidsys> there is no mailman-web from our repo installed
2024-02-23T23:07:51  <acidsys> I probably uninstalled it because I patched and installed python-mailman-web from home:crameleon:branches:devel:languages:python:mailman
2024-02-23T23:07:59  <acidsys> I forgot we had a different one
2024-02-23T23:08:32  <acidsys> (totally not confusing two packages with the same name and entirely different contents)
2024-02-23T23:17:45  <robin_listas> Thanks, mail lists seem to be back :-)
2024-02-23T23:26:55  <acidsys> :) web interface still takes a bit
2024-02-23T23:27:25  <acidsys> someone please contribute an openQA test for mailman/hyperkitty/postorius in factory :-(
2024-02-23T23:34:03  <acidsys> ok, hyperkitty and mailman-web work together again, next broken is Postorius, also has python311 pinned
2024-02-23T23:37:09  <acidsys> unfortunately, that one does not simply build with 3.12
2024-02-23T23:37:14  <acidsys> ImportError: cannot import name 'ThreadCollector' from 'debug_toolbar.utils' (/usr/lib/python3.12/site-packages/debug_toolbar/utils.py)
2024-02-23T23:37:52  <cboltz> I start to wonder if all these packages should require a small "mailman-python-version" package that defines a macro with the python version to use
2024-02-23T23:38:06  <acidsys> that sounds like a neat idea
2024-02-23T23:38:08  <cboltz> (won't fix build issues with new python, but at least keeps everything in sync)
2024-02-23T23:38:46  <acidsys> yeah and if one does not build it won't release the others