| -@gerrit:opendev.org- Tobias Urdin proposed: [zuul/zuul] 877587: web: add dark mode and theme selection https://review.opendev.org/c/zuul/zuul/+/877587 | 11:24 | |
| -@gerrit:opendev.org- James E. Blair https://matrix.to/#/@jim:acmegating.com proposed: [zuul/nodepool] 881237: Release the thread lock when we delete a node https://review.opendev.org/c/zuul/nodepool/+/881237 | 15:35 | |
| @jim:acmegating.com | Clark: fungi ^ this should fix the observed malbehavior in opendev | 15:35 |
|---|---|---|
| -@gerrit:opendev.org- James E. Blair https://matrix.to/#/@jim:acmegating.com proposed: [zuul/nodepool] 881238: Reduce statemachine driver logging https://review.opendev.org/c/zuul/nodepool/+/881238 | 15:37 | |
| @jim:acmegating.com | i have finished copying all the images to quay.io, so i think we can make the switch in repo now | 15:46 |
| @jim:acmegating.com | * i have finished copying all the images to quay.io, so i think we can make the switch in the publishing jobs now | 15:47 |
| -@gerrit:opendev.org- James E. Blair https://matrix.to/#/@jim:acmegating.com proposed: [zuul/nodepool] 881238: Reduce statemachine driver logging https://review.opendev.org/c/zuul/nodepool/+/881238 | 15:52 | |
| @jim:acmegating.com | Clark: fungi ^ can you rereview that pls? | 15:52 |
| @fungicide:matrix.org | i like the term "malbehavior," it seems perfectly cromulent | 15:53 |
| @fungicide:matrix.org | 881238,2 lgtm. what was the problem with using f-strings? | 15:55 |
| @jim:acmegating.com | it's not ideal for logging -- if you screw up a log message with %s and arguments, then the logging handler is able to still produce a coherent error message. but if you mess up f strings, the error isn't so great. so trying to remember to keep using %s there | 15:57 |
| @jim:acmegating.com | not important enough to go yank stuff out that's working, but whenever we touch it, probably a good idea | 15:57 |
| @fungicide:matrix.org | makes sense. so nothing inherently wrong with f-strings, they're just not robust enough for some logging edge cases | 16:12 |
| @jim:acmegating.com | yeah, i'll use 'em everywhere else | 16:13 |
| -@gerrit:opendev.org- James E. Blair https://matrix.to/#/@jim:acmegating.com proposed: [zuul/zuul] 881243: Publish container images to quay.io https://review.opendev.org/c/zuul/zuul/+/881243 | 16:24 | |
| -@gerrit:opendev.org- James E. Blair https://matrix.to/#/@jim:acmegating.com proposed: [zuul/zuul] 881243: Publish container images to quay.io https://review.opendev.org/c/zuul/zuul/+/881243 | 16:29 | |
| -@gerrit:opendev.org- James E. Blair https://matrix.to/#/@jim:acmegating.com proposed: [zuul/nodepool] 881244: Publish container images to quay.io https://review.opendev.org/c/zuul/nodepool/+/881244 | 16:31 | |
| -@gerrit:opendev.org- Zuul merged on behalf of James E. Blair https://matrix.to/#/@jim:acmegating.com: [zuul/nodepool] 881237: Release the thread lock when we delete a node https://review.opendev.org/c/zuul/nodepool/+/881237 | 16:49 | |
| -@gerrit:opendev.org- James E. Blair https://matrix.to/#/@jim:acmegating.com proposed: [zuul/zuul-operator] 881245: Publish container images to quay.io https://review.opendev.org/c/zuul/zuul-operator/+/881245 | 16:56 | |
| -@gerrit:opendev.org- James E. Blair https://matrix.to/#/@jim:acmegating.com proposed: [zuul/zuul-registry] 881247: Publish container images to quay.io https://review.opendev.org/c/zuul/zuul-registry/+/881247 | 17:03 | |
| -@gerrit:opendev.org- James E. Blair https://matrix.to/#/@jim:acmegating.com proposed: [zuul/zuul-preview] 881249: Publish container images to quay.io https://review.opendev.org/c/zuul/zuul-preview/+/881249 | 17:06 | |
| -@gerrit:opendev.org- James E. Blair https://matrix.to/#/@jim:acmegating.com proposed: [zuul/zuul-storage-proxy] 881250: Publish container images to quay.io https://review.opendev.org/c/zuul/zuul-storage-proxy/+/881250 | 17:10 | |
| -@gerrit:opendev.org- James E. Blair https://matrix.to/#/@jim:acmegating.com proposed: [zuul/gcp-authdaemon] 881251: Publish container images to quay.io https://review.opendev.org/c/zuul/gcp-authdaemon/+/881251 | 17:13 | |
| @jim:acmegating.com | That's the lot of them. | 17:14 |
| @clarkb:matrix.org | corvus: zuul-registry failed because it appears to still be trying to use docker jobs for some reason | 17:18 |
| @clarkb:matrix.org | or docker playbooks/roles anyway | 17:18 |
| @clarkb:matrix.org | oh it explicitly runs that playbook out of the local repo | 17:19 |
| @clarkb:matrix.org | should just be a matter of udpating the job I think | 17:19 |
| -@gerrit:opendev.org- James E. Blair https://matrix.to/#/@jim:acmegating.com proposed: [zuul/zuul-jobs] 881252: Move containerfile setting in container build https://review.opendev.org/c/zuul/zuul-jobs/+/881252 | 17:20 | |
| @jim:acmegating.com | Clark: thx, i'll look. ^ that is why zuul-operator failed | 17:20 |
| -@gerrit:opendev.org- James E. Blair https://matrix.to/#/@jim:acmegating.com proposed: [zuul/zuul-registry] 881247: Publish container images to quay.io https://review.opendev.org/c/zuul/zuul-registry/+/881247 | 17:23 | |
| @clarkb:matrix.org | corvus: storage proxy was similar to the registry | 17:24 |
| -@gerrit:opendev.org- James E. Blair https://matrix.to/#/@jim:acmegating.com proposed: [zuul/zuul-storage-proxy] 881250: Publish container images to quay.io https://review.opendev.org/c/zuul/zuul-storage-proxy/+/881250 | 17:24 | |
| @jim:acmegating.com | interesting. i wonder if we'd write those jobs that way now... | 17:24 |
| -@gerrit:opendev.org- James E. Blair https://matrix.to/#/@jim:acmegating.com proposed: [zuul/zuul-operator] 881245: Publish container images to quay.io https://review.opendev.org/c/zuul/zuul-operator/+/881245 | 17:25 | |
| -@gerrit:opendev.org- Zuul merged on behalf of James E. Blair https://matrix.to/#/@jim:acmegating.com: [zuul/nodepool] 881238: Reduce statemachine driver logging https://review.opendev.org/c/zuul/nodepool/+/881238 | 17:36 | |
| @jim:acmegating.com | Clark: ianw https://review.opendev.org/881244 is showing a possible issue related to the buildx/multi-arch path that i'm not in a position to track down, if you or anyone else wants to look into it | 19:28 |
| @jim:acmegating.com | i will recheck just to be sure, but i think that's likely a legit error | 19:28 |
| @clarkb:matrix.org | corvus: could that be related to the skopeo version thing? maybe skopeo isn't new enough there to understand the manifest? | 19:38 |
| @jim:acmegating.com | Clark: no idea -- but i don't think we changed push-to-intermediate-registry | 19:39 |
| @clarkb:matrix.org | corvus: the skopeo that failed is on localhost which I believe is up to date | 19:44 |
| @clarkb:matrix.org | oh it is failing when talking to the local registry | 19:45 |
| @clarkb:matrix.org | corvus: I think I see the issue (haven't look at code yet to know how/why though) compare the paths in https://zuul.opendev.org/t/zuul/build/8ce8d32ce1e54064ae6e5eaab13cf47f/log/job-output.txt#7996 to https://zuul.opendev.org/t/zuul/build/8ce8d32ce1e54064ae6e5eaab13cf47f/log/job-output.txt#10672 the buildset registry which we are trying to pull from has the image without the quay.io prefix but then when we try to send it to the intermediate registry it does | 20:01 |
| @clarkb:matrix.org | Pretty sure that is a valid 404 as a result | 20:01 |
| @jim:acmegating.com | yeah, compare to zuul pushing to the buildset registry; it has quay.io: https://zuul.opendev.org/t/zuul/build/20690c6ade924b05badee1f3987f867f/log/job-output.txt#10080 | 20:06 |
| @jim:acmegating.com | so seems like we need to narrow it down to the buildx temporary registry->buildset registry area | 20:07 |
| @clarkb:matrix.org | ya I think I see it in the code. | 20:07 |
| @clarkb:matrix.org | in build-container-image push.yaml we use zj_image.repository as a prefix but we don't in buildx.yaml | 20:08 |
| @clarkb:matrix.org | let me work on a patch that adds that in | 20:08 |
| @jim:acmegating.com | kk, thx | 20:08 |
| @clarkb:matrix.org | I think I introduced this issue since we didn't want the registry name in some places but do in others? | 20:09 |
| -@gerrit:opendev.org- Clark Boylan proposed: [zuul/zuul-jobs] 881277: Use full image url in container buildx path https://review.opendev.org/c/zuul/zuul-jobs/+/881277 | 20:16 | |
| @clarkb:matrix.org | corvus: ^ I think that may fix it | 20:17 |
| @jim:acmegating.com | cool; i think the next challenge is the py27 job failing in zuul-jobs | 20:19 |
| @jim:acmegating.com | https://zuul.opendev.org/t/zuul/build/953e65f109664c5ea80ff6e3cd9943ac | 20:19 |
| @clarkb:matrix.org | corvus: the python27 issue is one that swift ran into as well. New(er) virtualenv execs a python script under what it thinks is a detected python interpreter for some reason and now that script includes f strings so python2.7 explodes | 20:31 |
| @clarkb:matrix.org | swift fixed this by running their tox job on focal I think to get python3.8 which used older virtualenv? I can take a look at that for zuul jobs too to see if something makes sense | 20:31 |
| -@gerrit:opendev.org- Clark Boylan proposed: [zuul/zuul-jobs] 881279: Pin virtualenv in tox environments https://review.opendev.org/c/zuul/zuul-jobs/+/881279 | 20:40 | |
| @clarkb:matrix.org | I think that might work | 20:40 |
| @jim:acmegating.com | Clark: can you change that comment to say "necessary for python 2.7"? | 20:42 |
| @jim:acmegating.com | so we know when to remove | 20:42 |
| -@gerrit:opendev.org- Clark Boylan proposed: [zuul/zuul-jobs] 881279: Pin virtualenv in tox environments https://review.opendev.org/c/zuul/zuul-jobs/+/881279 | 20:43 | |
| @jim:acmegating.com | i approved that; the gate will tell us if it works | 20:45 |
| @clarkb:matrix.org | the py27 job did pass | 20:56 |
| -@gerrit:opendev.org- Zuul merged on behalf of Clark Boylan: [zuul/zuul-jobs] 881279: Pin virtualenv in tox environments https://review.opendev.org/c/zuul/zuul-jobs/+/881279 | 20:57 | |
| @jim:acmegating.com | i'm going to approve 881252 then | 21:23 |
| @jim:acmegating.com | rechecking the nodepool change | 21:23 |
| -@gerrit:opendev.org- Clark Boylan proposed: [zuul/zuul-jobs] 877834: Add ensure-quay-repo role https://review.opendev.org/c/zuul/zuul-jobs/+/877834 | 21:31 | |
| -@gerrit:opendev.org- Zuul merged on behalf of James E. Blair https://matrix.to/#/@jim:acmegating.com: [zuul/zuul-jobs] 881252: Move containerfile setting in container build https://review.opendev.org/c/zuul/zuul-jobs/+/881252 | 21:34 | |
| @clarkb:matrix.org | the ensure-quay-repo role update is just a rebase. I'm trying to update it to mimic some of the other work we've done recently more closely | 21:38 |
| -@gerrit:opendev.org- Clark Boylan proposed: [zuul/zuul-jobs] 877834: Add ensure-quay-repo role https://review.opendev.org/c/zuul/zuul-jobs/+/877834 | 22:05 | |
| @clarkb:matrix.org | corvus: ianw ^ thats a bit of a rewrite and I haven't manually tested it yet. BUt I realized as I was sketching out the system-config job updates that it would be nice to have the role work like this. I'll get the sketch of the system-config change up soon too | 22:05 |
| -@gerrit:opendev.org- Clark Boylan proposed: [zuul/zuul-jobs] 877834: Add ensure-quay-repo role https://review.opendev.org/c/zuul/zuul-jobs/+/877834 | 22:08 | |
| -@gerrit:opendev.org- Clark Boylan proposed: [zuul/zuul-jobs] 877834: Add ensure-quay-repo role https://review.opendev.org/c/zuul/zuul-jobs/+/877834 | 22:10 | |
| @clarkb:matrix.org | apologies I'm failing at git | 22:10 |
| -@gerrit:opendev.org- Clark Boylan proposed: [zuul/zuul-jobs] 881277: Use full image url in container buildx path https://review.opendev.org/c/zuul/zuul-jobs/+/881277 | 22:26 | |
| @clarkb:matrix.org | that fixed a merge conflict with the containerfile fix | 22:27 |
| -@gerrit:opendev.org- James E. Blair https://matrix.to/#/@jim:acmegating.com proposed: [zuul/nodepool] 881287: Add Azure gallery image support https://review.opendev.org/c/zuul/nodepool/+/881287 | 22:40 | |
| @clarkb:matrix.org | corvus: I'm not sure how to deal with this failure https://zuul.opendev.org/t/zuul/build/310b4777d09549d580796a9ed4d0ae1e in the test suite for that change | 22:52 |
| @clarkb:matrix.org | I guess maybe the port doesn't matter and I can just drop it? | 22:52 |
| @clarkb:matrix.org | oh no except that is what defines where you push to in the end | 22:52 |
| @clarkb:matrix.org | can we/should we url encode there? I' mwondering why the non buildx promotion testing doesn't seem to have this problem | 22:56 |
| @clarkb:matrix.org | oh it could be because we push everything into the multiarch registry to collate things together I think | 22:57 |
| @clarkb:matrix.org | hrm no this fails before that in the buildx command | 22:57 |
| @clarkb:matrix.org | I think this works in the single arch case because we don't cover the buildset registry case there which would push to the buildset registry using an invalid tag specifier | 23:18 |
| -@gerrit:opendev.org- Clark Boylan proposed: [zuul/zuul-jobs] 881277: Use full image url in container buildx path https://review.opendev.org/c/zuul/zuul-jobs/+/881277 | 23:33 | |
| @clarkb:matrix.org | I don't expect ^ that to work, but if it does that may be a shorter term workaround. Basically we rely on localhost having the docker like registry rather than a hostname:port combo that can't be used in the urls | 23:33 |
| @clarkb:matrix.org | But we should probably come up with some scheme to support 127.0.0.1:5100 like hosts. Maybe replace : with _. I think the regular non multiarch side may need cleaning up too. | 23:34 |
| @clarkb:matrix.org | Wow that last patchset looks like it may actually work. I'm not sure I'm a fan but if it works and gets us moving forward we can figure out the existing issue with host:port in buildset registry paths later I guess | 23:40 |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!