2025-09-02 09:49:45 <achill> algitbot: retry master
2025-09-02 15:14:59 <mio> algitbot: retry master
2025-09-02 15:16:27 <mio> algitbot: retry master
2025-09-02 15:20:01 <mio> algitbot: retry master
2025-09-02 15:28:57 <mio> algitbot: retry master
2025-09-02 15:47:25 <mio> algitbot: retry master
2025-09-02 15:52:04 <mio> algitbot: retry master
2025-09-03 12:28:30 <achill> algitbot: retry master
2025-09-03 15:32:01 <achill> cc PureTryOut <https://matrix.to/#/%40oftc_PureTryOut%3Apostmarketos.org>
2025-09-03 18:00:46 <PureTryOut> algitbot: retry master
2025-09-04 12:46:40 <ikke> Sooo many brittle tests....
2025-09-04 12:47:29 <ikke> revoke_test.go:1097: '-1' and '0' are not equal
2025-09-04 12:47:40 <ikke> Like that's helpful
2025-09-04 13:57:27 <achill> ikke: thanks for opening their respective issues!
2025-09-04 18:56:26 <ikke> mio: funny fix, just remove the failing assertion
2025-09-04 18:56:33 <ikke> (May be correct, but looks funny)
2025-09-05 23:37:00 <omni> ttps://github.com/vectordotdev/vector/pull/23645
2025-09-05 23:37:05 <omni> https://github.com/vectordotdev/vector/pull/23645
2025-09-06 13:23:21 <omni> !89740 (for anyone here who missed it)
2025-09-06 14:55:30 <achill> algitbot: retry master
2025-09-06 14:55:34 <achill> s/master/3.22-stable/
2025-09-06 14:55:51 <achill> algitbot: retry 3.22-stable
2025-09-06 14:56:49 <achill> lol?
2025-09-06 14:56:57 <achill> algitbot: retry master
2025-09-06 14:57:44 <ikke> dbus-example-objectmanager-visibility.h: No such file or directory
2025-09-06 14:57:56 <ikke> Is that a build system dependency issue?
2025-09-06 14:57:59 <achill> i guess some meson dependency issue
2025-09-06 14:58:00 <achill> yeah
2025-09-06 14:58:38 <achill> probably because of a meson upgrade, and the newer glib fixed it so let me search a commit
2025-09-06 15:02:20 <achill> ah now it built
2025-09-06 15:02:23 <achill> fine by me i guess
2025-09-06 15:07:41 <ikke> "this is fine" meme
2025-09-06 16:01:56 <achill> something is weird with build-edge-x86_64. it wants to build element-desktop with a old electron even tho the new one is in aports
2025-09-06 16:02:09 <achill> i dont know if it already built electron or the build order is just wrong
2025-09-06 16:02:12 <achill> 🤷
2025-09-06 16:02:16 <ikke> let me check
2025-09-06 16:03:29 <ikke> 36.8.1-r1 is latest version in the repos
2025-09-06 16:04:39 <achill> ahuh so element-desktop wants to build with electron-35.7.4-r0
2025-09-06 16:04:59 <ikke> achill: possibly a dependency holding it back?
2025-09-06 16:05:00 <achill> there is a constraint in the APKBUILD but it should only be ~35 which are satisfied by both versions
2025-09-06 16:05:09 <achill> hmm
2025-09-06 16:05:39 <ikke> achill: ~35 means >=35, <36
2025-09-06 16:05:58 <achill> yeah exactly, so that shouldnt be the problem
2025-09-06 16:06:42 <ikke> achill: maybe I'm missing something?
2025-09-06 16:06:55 <achill> ohhh the new electron version is 36 and not 35
2025-09-06 16:06:57 <ikke> yes
2025-09-06 16:09:07 <achill> i guess we'll see if it builds with the new electron 🙃
2025-09-06 16:12:53 <achill> algitbot: retry master
2025-09-06 16:13:52 <achill> now x86_64 doesnt respond to retries, huh?
2025-09-06 16:14:00 <ikke> achill: heh, I stopped the builder to check
2025-09-06 16:14:06 <achill> ah lol
2025-09-06 16:14:38 <achill> seems its building
2025-09-06 16:23:48 <PureTryOut> algitbot: retry master
2025-09-06 18:14:38 <achill> algitbot: retry 3.22-stable
2025-09-06 18:22:58 <achill> * algitbot: retry 3.22-stable
2025-09-06 18:25:55 <achill> * algitbot: retry 3.22-stable
2025-09-06 18:33:04 <achill> * algitbot: retry 3.22-stable
2025-09-06 19:08:51 <ikke> tailscale: https://github.com/tailscale/tailscale/issues/16966
2025-09-06 19:09:58 <ikke> aarch64 builder happens to run 6.12.42-0-lts
2025-09-06 19:25:30 <achill> * algitbot: retry 3.22-stable
2025-09-06 19:29:26 <ikke> kernel patch: https://patchwork.kernel.org/project/linux-fsdevel/patch/20250821105806.1453833-1-wangzijie1@honor.com/raw/
2025-09-06 19:43:49 <achill> * algitbot: retry 3.22-stable
2025-09-06 19:44:41 <achill> i guess we can just disable the test since its not related to tailscale
2025-09-06 20:30:00 <ikke> algitbot: retry 3.22-stable
2025-09-06 22:02:49 <achill> algitbot: retry master
2025-09-06 22:02:53 <achill> algitbot: retry 3.22-stable
2025-09-07 00:22:34 <mio> algitbot: retry 3.22-stable
2025-09-07 00:32:03 <mio> algitbot: retry 3.22-stable
2025-09-07 00:39:59 <mio> algitbot: retry 3.22-stable
2025-09-08 15:21:10 <ikke> algitbot: retry 3.22-stable
2025-09-08 17:58:17 <mio> algitbot: retry 3.22-stable
2025-09-08 19:19:03 <mio> algitbot: retry 3.22-stable
2025-09-08 19:19:16 <ikke> should we disable falcon?
2025-09-08 19:19:29 <ikke> Endless retrying does not make an illegal instruction legal
2025-09-08 19:19:36 <ikke> (disable on armv7)
2025-09-08 19:20:58 <achill> !89792
2025-09-08 19:21:53 <ikke> Still fails to build it seems
2025-09-08 19:23:22 <achill> yeah
2025-09-08 19:23:35 <achill> disabling seems fine to me
2025-09-08 19:24:53 <ikke> "Error: Failed to parse: 'typesystem_webenginecore.xml'"
2025-09-08 19:25:00 <ikke> Strange that that only happens on armv7
2025-09-08 19:25:16 <ikke> Can't find typesystem_webenginecore.xml
2025-09-08 19:26:12 <ikke> oh, apparently py3-pyside6 is not available for armv7
2025-09-08 19:29:13 <ikke> Or more precisely, does not provide that file
2025-09-08 19:35:17 <mio> possibly -r1 was not built with Qt6WebEngineQuick
2025-09-08 19:35:32 <mio> so it didn't include the .xml file
2025-09-08 19:36:08 <ikke> right, the builder has -r2
2025-09-08 19:36:10 <mio> maybe re-enabled on armv7 recently for qt6-qtwebengine, but since armv7 hasn't uploaded yet, ci doesn't have -r2
2025-09-08 19:36:13 <mio> yeah
2025-09-08 19:37:35 <ikke> So I suppose we can just merge that MR?
2025-09-08 19:37:59 <mio> guess so?
2025-09-09 08:12:54 <achill> ??
2025-09-09 09:27:31 <achill> algitbot: retry master
2025-09-09 15:37:12 <omni> algitbot: retry 3.20-stable
2025-09-09 15:37:33 <omni> is build-3-20-riscv64 stuck/broken?
2025-09-09 15:41:07 <ikke> network issue after reboot
2025-09-09 21:38:52 <mio> algitbot: retry 3.22-stable
2025-09-09 22:20:10 <andypost[m]> algitbot: retry 3.22-stable
2025-09-10 02:17:23 <mio> algitbot: retry 3.22-stable
2025-09-10 02:49:20 <mio> algitbot: retry 3.22-stable
2025-09-10 06:54:56 <achill> algitbot: retry master
2025-09-10 09:26:58 <omni> Revert "previous commit message" is not COMMITSTYLE.md compliant
2025-09-10 09:29:54 <achill> yeah good point
2025-09-10 09:30:24 <achill> was in a rush, i think there are certain advantages to the revert "..." messages but it needs discussion first :)
2025-09-10 09:34:43 <omni> it happens sometimes, but I think it's better if the commit message is amended, here would be "testing/ibus-typing-booster: downgrade to 2.27.73", and then the rest of the commit message body could be kept
2025-09-10 09:38:10 <omni> I know that I can be overly nitpicky, and what I often comment on is not that bad, but I think that, if we're too loose, git history could become messy and hard to search/follow
2025-09-10 09:39:40 <achill> i think the Revert ".." is useful if its a pure git revert without manual changes/conflict resolutions, and "downgrade" for stuff further than that like build fixes and stuff.  so that its visibile if its a pure revert or if it contains new changes
2025-09-10 10:36:48 <ikke> My preferences is "<pkg>: revert .."
2025-09-10 10:38:43 <omni> ah, yes, so "<pkg>: revert upgrade to .." for example
2025-09-10 10:39:38 <ikke> yup
2025-09-10 13:30:04 <andypost[m]> algitbot: retry master
2025-09-11 06:31:57 <achill> algitbot: retry 3.22-stable
2025-09-11 06:38:49 <achill> * algitbot: retry 3.22-stable
2025-09-11 15:42:32 <mio> algitbot: retry 3.22-stable
2025-09-13 09:26:23 <achill> algitbot: retry master
2025-09-13 10:35:14 <omni> algitbot: retry 3.22-stable
2025-09-14 02:05:31 <omni> wat
2025-09-14 05:12:31 <ikke>  That file certainly does exist
2025-09-14 13:47:10 <achill> algitbot: retry master
2025-09-14 15:41:38 <andypost[m]> algitbot: retry master
2025-09-14 18:53:16 <omni> is that hitting something that differ between ppc64le CI and package builder?
2025-09-14 18:53:28 <omni> passed here https://gitlab.alpinelinux.org/alpine/aports/-/jobs/2011952
2025-09-14 18:55:09 <ikke> omni: talking about racksdb?
2025-09-14 18:55:43 <omni> yes, and trying to look at it with crossed tired eyes
2025-09-14 18:57:41 <ikke> Both are power9 machines
2025-09-14 18:57:55 <ikke> One CI runner runs on the same host
2025-09-14 19:03:04 <omni> but why would it fail on the package builder? I haven't seen that in CI
2025-09-14 19:03:23 <omni> here's another successful CI pipeline https://gitlab.alpinelinux.org/alpine/aports/-/jobs/2012029
2025-09-14 19:17:19 <omni> !90114
2025-09-14 19:23:20 <omni> ikke: do you want to debug anything or should I merge that to deselect those tests?
2025-09-14 19:23:42 <ikke> I'd like to debug, but not sure where to look for
2025-09-14 19:24:25 <omni> are CI vs builders still different in setup?
2025-09-14 19:25:12 <omni> first thought was something in how /tmp is mounted, but that might be my tired eye looking at the wrong thing
2025-09-14 19:25:41 <omni> and.. I can't really say I have any idea where to look
2025-09-14 19:26:06 <ikke> docker vs lxc
2025-09-14 19:34:50 <omni> installing 112 packages vs upgrading 34 packages and installing 67
2025-09-14 19:34:59 <omni> but I'm not sure there's anything there
2025-09-14 19:35:11 <omni> they build on different base images?
2025-09-14 19:35:36 <ikke> Whatg is the opposte of --deselect?
2025-09-14 19:38:04 <ikke> Trying to run a single test
2025-09-14 19:38:15 <omni> ah, -k perhaps?
2025-09-14 19:38:21 <omni> pytest --help
2025-09-14 19:38:34 <ikke> -k takes a different syntax
2025-09-14 19:38:38 <omni> yeah...
2025-09-14 19:39:02 <ikke> I can do -k test_dump, but that select all tests called test_dump
2025-09-14 19:39:08 <ikke> TestRacksDBExec.test_dump selects nothing
2025-09-14 19:41:32 <ikke> ok, specify the filename separately
2025-09-14 19:41:34 <omni> what about
2025-09-14 19:41:37 <omni> pytest racksdb/tests/test_exec.py::TestRacksDBExec::test_dump
2025-09-14 19:41:38 <omni> etc
2025-09-14 19:43:37 <omni> guessing you could just run the same command without the --deselect flags and it would instead only run those specific tests
2025-09-14 19:44:05 <ikke> pytest -k test_dump ./racksdb/tests/test_exec.py worked
2025-09-14 19:45:28 <omni> k
2025-09-14 19:55:17 <ikke> looking with --pdb
2025-09-14 21:12:59 <omni> what was uploaded to testing?
2025-09-14 21:15:36 <omni> looked odd...
2025-09-16 09:53:04 <omni> ikke: do you still want me to hold off from merging !90114 ?
2025-09-16 10:07:02 <omni> hledger passed on aarch64 here https://gitlab.alpinelinux.org/strophy/aports/-/jobs/2014725
2025-09-16 10:07:29 <omni> I hope it will pass this time or the next, because I got to go
2025-09-16 13:03:00 <achill> algitbot: retry master
2025-09-16 14:53:09 <achill> algitbot: retry master
2025-09-16 15:28:50 <ptrc> of course it's oom
2025-09-16 15:28:52 <ptrc> bleh
2025-09-16 15:28:56 <ptrc> algitbot: retry master
2025-09-16 19:25:27 <achill> the glib failure is weird
2025-09-16 19:25:29 <achill> it worked on CI
2025-09-16 19:54:39 <ikke> potentially different hardware
2025-09-16 19:55:03 <achill> likely, but annoying
2025-09-16 19:56:00 <ikke> absolutely
2025-09-17 00:45:17 <omni> is build-3-22-riscv64 stuck on arti?
2025-09-17 00:55:48 <mio> omni: it would seem so
2025-09-17 00:58:34 <mio> 3.21 riscv64 also bulding the kernel for some time
2025-09-17 00:59:51 <mio> s/bulding/building/
2025-09-17 01:50:55 <mio> algitbot: retry master
2025-09-17 05:28:05 <ikke> algitbot: retry 3.18-stable
2025-09-17 07:12:31 <achill> algitbot: retry master
2025-09-17 22:40:17 <ptrc> sigh
2025-09-18 08:02:54 <achill> algitbot: retry master
2025-09-18 12:19:29 <ikke> checking racksdb again
2025-09-18 12:28:50 <omni> ok
2025-09-19 23:00:46 <omni> algitbot: retry master
2025-09-19 23:01:08 <omni> uhm, what's broken now?
2025-09-20 00:14:35 <andypost[m]> algitbot: retry master
2025-09-20 08:08:29 <achill> o/ first one
2025-09-20 11:40:24 <omni> algitbot: retry 3.22-stable
2025-09-21 13:12:02 <omni> hmm...
2025-09-21 13:14:39 <omni> =(
2025-09-21 13:16:18 <ikke> FAIL: scripts/unit.test
2025-09-21 13:16:23 <ikke> really clear..
2025-09-21 13:16:56 <ikke> and because the builder is already retrying, the log is gone..
2025-09-21 13:18:02 <ikke> stopped the builder now
2025-09-21 13:19:09 <ikke> ugh, multithreaded test log
2025-09-21 13:19:19 <ikke> at least, it looks like it
2025-09-21 13:24:03 <omni> ok, also !90496
2025-09-21 13:24:11 <omni> I missed the missing pkgrel bump
2025-09-21 13:24:18 <ikke> https://tpaste.us/yBdl 
2025-09-21 13:26:21 <omni> passed in CI https://gitlab.alpinelinux.org/sertonix/aports/-/jobs/2017622/raw
2025-09-21 13:26:39 <omni> (and that is where I saw that wolfssh needed a pkgrel bump)
2025-09-21 13:33:41 <omni> ikke: are you looking into it and should I hold off of merging other things to master?
2025-09-21 13:47:48 <ikke> I think I may have found it
2025-09-21 14:08:52 <omni> 👍
2025-09-21 14:22:24 <omni> ikke: what about x86_64?
2025-09-21 14:22:42 <ikke> I suppose I should start it again :P
2025-09-21 14:23:50 <ikke> I stopped it yesterday to debug the binsh issue
2025-09-21 14:24:13 <omni> oh, ah
2025-09-21 14:29:33 <ikke> for racksdb, I found where the exception comes from, but I still have to figure out what causes that problem
2025-09-21 14:45:06 <omni> but it is blocking community/ from being uploaded and testing/ to be built on that architecture?
2025-09-21 14:46:42 <omni> thinking it may give false positives in CI, if packages aren't at their expected version or build
2025-09-21 14:50:59 <omni> for php85 on riscv64, 837 tests were skipped in CI compared to 833 on the builder
2025-09-21 14:51:26 <omni> perhaps the one that is failing was one of those skipped in CI for some reason
2025-09-21 14:51:31 <omni> https://gitlab.alpinelinux.org/alpine/aports/-/jobs/2004894
2025-09-21 15:51:31 <ikke> omni: does the project check for CI being set?