| -@gerrit:opendev.org- Anil Shashikumar Belur proposed: | 00:39 | |
| - [opendev/system-config] 1002420: Build hound on python3.14 https://review.opendev.org/c/opendev/system-config/+/1002420 | ||
| - [opendev/system-config] 1002421: Build haproxy and zk statsd containers on python3.14 https://review.opendev.org/c/opendev/system-config/+/1002421 | ||
| - [opendev/system-config] 1002422: Use asyncio.run() to start the matrix-eavesdrop bot https://review.opendev.org/c/opendev/system-config/+/1002422 | ||
| - [opendev/system-config] 1002423: Build irc and matrix bots on python3.14 https://review.opendev.org/c/opendev/system-config/+/1002423 | ||
| - [opendev/system-config] 1002424: Build gerrit-base on python3.14 https://review.opendev.org/c/opendev/system-config/+/1002424 | ||
| - [opendev/system-config] 1002425: Default the python base images to python3.14 https://review.opendev.org/c/opendev/system-config/+/1002425 | ||
| -@gerrit:opendev.org- Anil Shashikumar Belur proposed: [opendev/system-config] 1002426: Restart gitea when its config files change https://review.opendev.org/c/opendev/system-config/+/1002426 | 00:41 | |
| -@gerrit:opendev.org- Monty Taylor https://matrix.to/#/@mordred:inaugust.com proposed: [opendev/project-config] 1002429: Re-enqueue vouched changes on trivial rebases https://review.opendev.org/c/opendev/project-config/+/1002429 | 02:34 | |
| @mordred:waterwanders.com | [@jim:acmegating.com](https://matrix.to/#/@jim:acmegating.com) ^^ I think like that, right? | 02:41 |
|---|---|---|
| @yaron.bar:matrix.org | thanks fungi and Jens Harbott | 05:15 |
| sorry for the noise earlier, I assumed a CLA was needed. please disregard. Registered SSH and that resolved it | ||
| @harbott.osism.tech:regio.chat | infra-root: debian-security volume is full, I'm trying to bump quota a bit, but we also are > 93% usage of afs1.dfw | 09:11 |
| `Error 122 creating lock file '/afs/.openstack.org/mirror/debian-security/db/lockfile': Disk quota exceeded!` | ||
| @harbott.osism.tech:regio.chat | ok, bumped the quota to 550G, rm-ed some size 0 files and rerunning reprepro-mirror-update now | 11:26 |
| @harbott.osism.tech:regio.chat | hmm, the removed files are marked missing now, running wget for them | 11:34 |
| @harbott.osism.tech:regio.chat | `Released volume mirror.debian-security successfully` | 13:08 |
| @fungicide:matrix.org | Yaron Bar: where did the need for the cla come up? was it in actual documentation or were you asking an llm and following hallucinated directions? | 13:16 |
| @fungicide:matrix.org | thanks for fixing up the mirror volume Jens Harbott, the recent flood of security fixes is leading to a lot more package churn in distros, so i agree we need to look at freeing or making space in afs somehow | 13:18 |
| -@gerrit:opendev.org- Stephen Finucane proposed wip: [zuul/zuul-jobs] 1002494: Support doc dependency-groups https://review.opendev.org/c/zuul/zuul-jobs/+/1002494 | 13:39 | |
| -@gerrit:opendev.org- Stephen Finucane proposed: [zuul/zuul-jobs] 1002504: Simplify installation of doc dependencies https://review.opendev.org/c/zuul/zuul-jobs/+/1002504 | 14:21 | |
| -@gerrit:opendev.org- Stephen Finucane proposed wip: [zuul/zuul-jobs] 1002494: Support doc dependency-groups https://review.opendev.org/c/zuul/zuul-jobs/+/1002494 | 14:21 | |
| @clarkb:matrix.org | fungi: Jens Harbott: should we be thinking about cleaning up bullseye as soon as it is eol ish on the 31st? | 15:33 |
| @clarkb:matrix.org | That should give us back ~1/3 of those volumes? I'm not sure how well used bullseye is in zuul though | 15:33 |
| @jim:acmegating.com | looks like not a lot, but non-zero: https://graphite.opendev.org/?width=586&height=308&target=sumSeries(stats_counts.zuul.provider.*.image.debian-bullseye.launch.ready) | 15:40 |
| @jim:acmegating.com | looks like it's mostly used by starlingx | 15:42 |
| @jim:acmegating.com | and a very small number of jobs in openstack/requirements, openstack/horizon on "unmaintained" branches | 15:43 |
| @jim:acmegating.com | (and of course zuul-jobs, but that doesn't count) | 15:43 |
| @fungicide:matrix.org | it won't be quite 1/3 i don't think, owing to the recent increase in package churn | 15:44 |
| @fungicide:matrix.org | also it's debian lts which is expiring for bullseye this month, extended lts carries on until 2031-06-30 | 15:44 |
| @fungicide:matrix.org | https://wiki.debian.org/LTS/Extended | 15:45 |
| @jim:acmegating.com | 256 out of 270 requests for bullseye nodes since utc midnight were for starlingx | 15:47 |
| @clarkb:matrix.org | fungi: ELTS is a paid service though not somethign we would have access to | 15:57 |
| @clarkb:matrix.org | considering that starlingx is already saying they can't do anything due to their release removing bullseye that quickly is likely to be problematic for them | 15:57 |
| @fungicide:matrix.org | debian elts is a paid service for commercial use but afaik doesn't require a sign-up and essentially works on the honor system expecting companies to pay for licenses if they rely on it | 15:59 |
| @fungicide:matrix.org | though it's a good point that i think the elts package repositories are separate and we don't mirror them, so not actually relevant to the situation after all | 16:00 |
| @clarkb:matrix.org | fungi: their documetnation says you need to pay them to get the username and password to the apt repos | 16:01 |
| @clarkb:matrix.org | maybe they would give them to us for free if we asked nicely. We probably wouldn't be able to mirror them though | 16:02 |
| @fungicide:matrix.org | aha, so the free for community and individual use requires registration after all | 16:03 |
| @fungicide:matrix.org | and i agree they wouldn't want those packages mirrored publicly for the same reasons rhel doesn't allow it | 16:04 |
| @clarkb:matrix.org | I'm headed back to the optometrist in a bit (ran out of time last time to find new glasses) but afterwards I'm going to try and catch up on reviews | 16:20 |
| @fungicide:matrix.org | https://mirror.sjc3.raxflex.opendev.org/ is up and serving afs content again, so i've taken it out of the emergency list to get its cert updated | 16:52 |
| -@gerrit:opendev.org- Jeremy Stanley https://matrix.to/#/@fungicide:matrix.org proposed: [opendev/system-config] 1002534: Clean up old static servers from cacti hosts list https://review.opendev.org/c/opendev/system-config/+/1002534 | 16:56 | |
| @clarkb:matrix.org | fungi: any idea what happened with account auth yet? | 17:06 |
| @fungicide:matrix.org | no, i'm still digging into that | 17:07 |
| @fungicide:matrix.org | since it repurposes the rax classic "api key" for a password, i'm wondering if that may have been generated with some sort of expiration | 17:08 |
| @clarkb:matrix.org | Oh maybe? And if we set up one account before the other it's possible that there is a delta between expirations | 17:10 |
| @fungicide:matrix.org | but i don't see where that would be set if so, and the rackspace portal does let me view the api key and it matches what we've got set as the password in clouds.yaml | 17:11 |
| -@gerrit:opendev.org- Clark Boylan proposed: [opendev/system-config] 1002179: Add support for shared anubis secret keys in gitea https://review.opendev.org/c/opendev/system-config/+/1002179 | 18:45 | |
| @clarkb:matrix.org | infra-root the stack ending at https://review.opendev.org/c/opendev/system-config/+/1002425 is the python3.14 update chagnes for system-config container images. The linked last change updates the base python image Dockerfiles to set the default version. We are always explicit when building the images in CI, but that update has a side effect of rebuilding the images for us. Do we think that landing this last chagne first is worthwhile to avoid needing to rebuild images for python updates later? Otherwise the first few changes should be pretty safe to approve today and I'm happy to monitor if someone wants to take a look. The irc bot and gerrit image updates are the two that we'll need to plan to avoid unwanted downtimes | 18:58 |
| @clarkb:matrix.org | and then https://review.opendev.org/c/opendev/system-config/+/1002426/ is an improvement to restart Gitea when docker-compose or anubis config files update. This way we won't have to manually restart anubis when we update the botpolicy (this was an issue in the past) | 18:59 |
| @clarkb:matrix.org | also happy to monitor that one if someone else can take a look at it | 18:59 |
| -@gerrit:opendev.org- Clark Boylan proposed: [opendev/system-config] 1002179: Add support for shared anubis secret keys in gitea https://review.opendev.org/c/opendev/system-config/+/1002179 | 20:45 | |
| @clarkb:matrix.org | https://review.opendev.org/c/opendev/system-config/+/1002200 is another straightforward review to add to the list if anyone else has a moment (this one drops a couple of node exporter collectors to remove some metrics) | 21:06 |
| @clarkb:matrix.org | ok I think the shared anubis key is working now: https://review.opendev.org/c/opendev/system-config/+/1002179 | 22:36 |
| -@gerrit:opendev.org- Zuul merged on behalf of Clark Boylan: [opendev/system-config] 1002200: Disable dmi and entropy collectors in node-exporter https://review.opendev.org/c/opendev/system-config/+/1002200 | 22:36 | |
| @clarkb:matrix.org | the testing doesn't actually confirm that two different anubis instances accept a cookie from one client since we only run one backend in CI. But it doesn't explode so in theory it is using that secret for the signing | 22:37 |
Generated by irclog2html.py 4.1.0 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!