Thursday, 2026-02-05

-@gerrit:opendev.org- Ghanshyam proposed: [opendev/irc-meetings] 975725: Shifting OpenStack RBAC meeting to 18 UTC https://review.opendev.org/c/opendev/irc-meetings/+/97572501:54
-@gerrit:opendev.org- Tim Burke proposed: [zuul/zuul-jobs] 975737: Ignore empty tags for container jobs https://review.opendev.org/c/zuul/zuul-jobs/+/97573702:23
-@gerrit:opendev.org- Zuul merged on behalf of Ghanshyam: [opendev/irc-meetings] 975725: Shifting OpenStack RBAC meeting to 18 UTC https://review.opendev.org/c/opendev/irc-meetings/+/97572508:17
@fungicide:matrix.orgdebian's ci team reveals some of the efforts they've taken to shield their systems from llm training crawlers: https://lists.debian.org/debian-devel-announce/2026/02/msg00001.html14:55
@fungicide:matrix.orgpip 26.0.1 today fixing a subtle bug that could in theory impact some of our users (i believe we use the affected options together in some places): https://pip.pypa.io/en/stable/news/15:24
@clarkb:matrix.orgthe fail2ban approach seems similar to what we are experimenting with (basically use a heuristic to identify bad traffic then block them entirely)15:45
@clarkb:matrix.orgI suspect the use of `--pre` is not very common though15:45
@clarkb:matrix.orgfungi: I've just approved https://review.opendev.org/c/opendev/system-config/+/975696 so that hopefully we can see the 1700 ish backup succeed15:47
@fungicide:matrix.orgthanks!15:54
@fungicide:matrix.orgheading out for lunch/errands, back in a while16:02
-@gerrit:opendev.org- Zuul merged on behalf of Jeremy Stanley https://matrix.to/#/@fungicide:matrix.org: [opendev/system-config] 975696: Backup keycloak database over IPv6 loopback https://review.opendev.org/c/opendev/system-config/+/97569616:08
-@gerrit:opendev.org- Clark Boylan proposed: [opendev/system-config] 975816: Use ::1 for keycloak db backups https://review.opendev.org/c/opendev/system-config/+/97581616:41
@clarkb:matrix.orgfungi: there were actually two locations that needed to be updated ^ I was confused for a mintue why the script on the host hadn't updated despite the deployment being successful. That is why16:41
@clarkb:matrix.orgfwiw I skimmed through RETRY status jobs this morning and haven't found any that look like cpu count problems. There are a lot that fail due to missing python3.13-dev packages on arm64 which surprised me.17:24
@clarkb:matrix.orgI'm guessing that error condition is simply really infrequent?17:24
@fungicide:matrix.orgoh, i missed that too18:12
@fungicide:matrix.orgsorry about that, approved now18:13
@fungicide:matrix.orgthe debian ci announcement led into a discussion on the debian-devel ml here: https://lists.debian.org/debian-devel/2026/02/msg00048.html18:23
@fungicide:matrix.orgnotably they tried anubis but didn't have much luck with it18:24
-@gerrit:opendev.org- Zuul merged on behalf of Clark Boylan: [opendev/system-config] 975816: Use ::1 for keycloak db backups https://review.opendev.org/c/opendev/system-config/+/97581618:32
@clarkb:matrix.orgfungi: that is really good info re anubis since I've been somewhat skeptical of how useful it would be18:33
@clarkb:matrix.orgI'm happy to maybe be on the right track but even more happy that we can learn from others experiences without doing a lot of heavy lifting :)18:33
@fungicide:matrix.orgalso the idea that fighting runaway utilization with proof-of-work is not the most environmentally friendly approach18:34
@jim:acmegating.comyeah it's getting warm in here18:35
-@gerrit:opendev.org- Clark Boylan proposed: [opendev/system-config] 945143: DNM testing trixie build of lodgeit https://review.opendev.org/c/opendev/system-config/+/94514319:28
@clarkb:matrix.orgThat change will both check that lodgeit on trixie runs successfully in our system-config-run-paste job and if we recheck it tomorrow should double check the zuul-registry pruning fix didn't delete the manifest list for the lodgeit image early19:28
@clarkb:matrix.orglooks like lodgeit is all happy (we could land the change today if we want then I'd need a different test for the registry pruning tomorrow)21:03
@clarkb:matrix.orgthe sun is out and rain is coming this weekend so I'm going to pop out for a bike ride in a bit. I'll plan to check on registry pruning later today and reviews on the gerrit 3.11.8 change would be great if we think upgrading gerrit to the latest bugfix tomorrow is a good idea.21:25
@fungicide:matrix.orgyeah, i'll take a look, i agree tomorrow makes sense for gerrit21:26
@fungicide:matrix.org+2 on the gerrit change, but might be good to get corvus to weigh in on your comment about depends-on with the upstream gerrit repo since i'm also unsure23:23
@fungicide:matrix.orgthat's https://review.opendev.org/c/opendev/system-config/+/975176 for reference23:24
@clarkb:matrix.orgthanks!23:53

Generated by irclog2html.py 4.0.0 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!