| -@gerrit:opendev.org- Zuul merged on behalf of lotorev vitaly: [zuul/zuul] docs: Document regexp RE2 restricted syntax https://review.opendev.org/c/zuul/zuul/+/695991 | 00:11 | |
| -@gerrit:opendev.org- James E. Blair https://matrix.to/#/@jim:acmegating.com proposed: | 00:21 | |
| - [zuul/zuul-operator] Update doc theme settings to match Zuul https://review.opendev.org/c/zuul/zuul-operator/+/807798 | ||
| - [zuul/zuul-operator] Pin minikube to 1.22.0 https://review.opendev.org/c/zuul/zuul-operator/+/807805 | ||
| -@gerrit:opendev.org- James E. Blair https://matrix.to/#/@jim:acmegating.com proposed: [zuul/nodepool] WIP static node slots https://review.opendev.org/c/zuul/nodepool/+/807806 | 00:27 | |
| -@gerrit:opendev.org- Tobias Henkel proposed on behalf of Felix Edel: [zuul/zuul] Make reporting asynchronous https://review.opendev.org/c/zuul/zuul/+/691253 | 09:36 | |
| -@gerrit:opendev.org- Matthieu Huin https://matrix.to/#/@mhuin:matrix.org proposed: [zuul/zuul] Web UI: add checkbox, selects to filter toolbar https://review.opendev.org/c/zuul/zuul/+/729265 | 10:22 | |
| -@gerrit:opendev.org- Matthieu Huin https://matrix.to/#/@mhuin:matrix.org proposed: [zuul/zuul] Web UI: make more filters selectable in build, buildset searches https://review.opendev.org/c/zuul/zuul/+/793159 | 10:22 | |
| @gtema:matrix.org | does anybody has a suggestion how now (after security update) to be able to override `zuul.executor.log_root` from a test playbook in zuul-jobs? Using set_fact and passing `zuul` as var into the role does not work now | 12:48 |
|---|---|---|
| -@gerrit:opendev.org- Simon Westphahl proposed: [zuul/zuul] Checkout playbook branch for roles running on tag https://review.opendev.org/c/zuul/zuul/+/807868 | 12:59 | |
| -@gerrit:opendev.org- Matthieu Huin https://matrix.to/#/@mhuin:matrix.org proposed: [zuul/zuul] Web UI: add checkbox, selects to filter toolbar https://review.opendev.org/c/zuul/zuul/+/729265 | 13:09 | |
| @jim:acmegating.com | gtema: that should not be possible now (if you find a way, please open a private security bug); i imagine some new approach to testing that will need to be devised. | 13:19 |
| @gtema:matrix.org | that's exactly the point. Currently zuul-jobs-test-upload-logs-s3 job is trying exactly that | 13:20 |
| @gtema:matrix.org | so we either need to drop this job (at least for now) or find quickly alternative way | 13:20 |
| -@gerrit:opendev.org- Matthieu Huin https://matrix.to/#/@mhuin:matrix.org proposed: [zuul/zuul] Web UI: make more filters selectable in build, buildset searches https://review.opendev.org/c/zuul/zuul/+/793159 | 13:21 | |
| -@gerrit:opendev.org- Matthieu Huin https://matrix.to/#/@mhuin:matrix.org proposed: [zuul/zuul] UI: add pagination to builds, buildsets search https://review.opendev.org/c/zuul/zuul/+/793467 | 13:21 | |
| -@gerrit:opendev.org- Felix Edel proposed: [zuul/zuul] Improve some log messages in the executor/builds context https://review.opendev.org/c/zuul/zuul/+/807874 | 13:22 | |
| @gtema:matrix.org | corvus, are you fine with change disabling zuul-jobs-test-upload-logs-s3 so that we can finally proceed with https://review.opendev.org/c/zuul/zuul-jobs/+/776677? | 13:23 |
| @avass:vassast.org | gtema: if you disable the test wanna remove the `_undocumented_...` var as well ? :) | 13:24 |
| @gtema:matrix.org | avass, we can not now run this test at all, cause we have no possibility to override log_root | 13:24 |
| @gtema:matrix.org | it doesn't really matter whether it is with _undocumented_ or without | 13:24 |
| @avass:vassast.org | gtema: yeah I mean remove the variable from the role so it can't be used at all | 13:24 |
| @gtema:matrix.org | if we remove it at all we still can't fix the job | 13:25 |
| @gtema:matrix.org | or at least I see no possibility right now | 13:25 |
| @jim:acmegating.com | gtema: can you try not overriding the path; instead, make it exist on the test worker node and just let the role run with the real value? | 13:33 |
| @jim:acmegating.com | gtema: (instead of using tempfile for fake_zuul_logdir, just create zuul.executor.log_root, which will be something like /var/lib/zuul/builds/uuid/work/logs) | 13:34 |
| -@gerrit:opendev.org- Matthieu Huin https://matrix.to/#/@mhuin:matrix.org proposed: [zuul/zuul] Web UI: add checkbox, selects to filter toolbar https://review.opendev.org/c/zuul/zuul/+/729265 | 13:35 | |
| @gtema:matrix.org | that requires root privs, which are normally not there | 13:35 |
| @jim:acmegating.com | gtema: they normally are there; we typically only remove them for unit test jobs | 13:37 |
| @gtema:matrix.org | ah, right. | 13:37 |
| @gtema:matrix.org | ok, will try to invent something | 13:38 |
| -@gerrit:opendev.org- Artem Goncharov proposed: [zuul/zuul-jobs] [DNM] Test upload_logs_s3 role https://review.opendev.org/c/zuul/zuul-jobs/+/807132 | 13:54 | |
| @jim:acmegating.com | gtema: that's about what i was thinking; i left a suggestion inline | 14:52 |
| @jim:acmegating.com | and looking at the error, i think that's why the job is failing | 14:52 |
| -@gerrit:opendev.org- Simon Westphahl proposed: | 14:56 | |
| - [zuul/zuul] Add source interface for setting change attributes https://review.opendev.org/c/zuul/zuul/+/805836 | ||
| - [zuul/zuul] Reference change dependencies by key https://review.opendev.org/c/zuul/zuul/+/805844 | ||
| - [zuul/zuul] Implement ABC for caching changes in Zookeeper https://review.opendev.org/c/zuul/zuul/+/805835 | ||
| - [zuul/zuul] Cache Gerrit refs in Zookeeper https://review.opendev.org/c/zuul/zuul/+/805837 | ||
| - [zuul/zuul] Cache Pagure refs in Zookeeper https://review.opendev.org/c/zuul/zuul/+/806556 | ||
| - [zuul/zuul] Cache Gitlab refs in Zookeeper https://review.opendev.org/c/zuul/zuul/+/806557 | ||
| - [zuul/zuul] Cache Git refs (driver) in Zookeeper https://review.opendev.org/c/zuul/zuul/+/806755 | ||
| - [zuul/zuul] Periodically maintain connection caches https://review.opendev.org/c/zuul/zuul/+/806756 | ||
| - [zuul/zuul] Clean up dangling cache data nodes more often https://review.opendev.org/c/zuul/zuul/+/807102 | ||
| - [zuul/zuul] DNM: test correct change cache (de-)serialization https://review.opendev.org/c/zuul/zuul/+/807890 | ||
| -@gerrit:opendev.org- Simon Westphahl proposed: | 15:01 | |
| - [zuul/zuul] Add source interface for setting change attributes https://review.opendev.org/c/zuul/zuul/+/805836 | ||
| - [zuul/zuul] Reference change dependencies by key https://review.opendev.org/c/zuul/zuul/+/805844 | ||
| - [zuul/zuul] Implement ABC for caching changes in Zookeeper https://review.opendev.org/c/zuul/zuul/+/805835 | ||
| - [zuul/zuul] Cache Gerrit refs in Zookeeper https://review.opendev.org/c/zuul/zuul/+/805837 | ||
| - [zuul/zuul] Cache Github refs in Zookeeper https://review.opendev.org/c/zuul/zuul/+/805838 | ||
| - [zuul/zuul] Cache Pagure refs in Zookeeper https://review.opendev.org/c/zuul/zuul/+/806556 | ||
| - [zuul/zuul] Cache Gitlab refs in Zookeeper https://review.opendev.org/c/zuul/zuul/+/806557 | ||
| - [zuul/zuul] Cache Git refs (driver) in Zookeeper https://review.opendev.org/c/zuul/zuul/+/806755 | ||
| - [zuul/zuul] Periodically maintain connection caches https://review.opendev.org/c/zuul/zuul/+/806756 | ||
| - [zuul/zuul] Clean up dangling cache data nodes more often https://review.opendev.org/c/zuul/zuul/+/807102 | ||
| - [zuul/zuul] DNM: test correct change cache (de-)serialization https://review.opendev.org/c/zuul/zuul/+/807890 | ||
| @gtema:matrix.org | corvus, thanks. Will update now (was on call) | 15:33 |
| -@gerrit:opendev.org- Artem Goncharov proposed: [zuul/zuul-jobs] Fix upload_logs_s3 role https://review.opendev.org/c/zuul/zuul-jobs/+/807132 | 15:35 | |
| -@gerrit:opendev.org- Artem Goncharov proposed: [zuul/zuul-jobs] Fix zuul-jobs-test-upload-logs-s3 job https://review.opendev.org/c/zuul/zuul-jobs/+/807132 | 15:36 | |
| -@gerrit:opendev.org- Matthieu Huin https://matrix.to/#/@mhuin:matrix.org proposed: | 15:47 | |
| - [zuul/zuul] Web UI: add checkbox, selects to filter toolbar https://review.opendev.org/c/zuul/zuul/+/729265 | ||
| - [zuul/zuul] web: yarn update to patternfly 2021.11 https://review.opendev.org/c/zuul/zuul/+/807904 | ||
| -@gerrit:opendev.org- Matthieu Huin https://matrix.to/#/@mhuin:matrix.org proposed: [zuul/zuul] Web UI: make more filters selectable in build, buildset searches https://review.opendev.org/c/zuul/zuul/+/793159 | 15:47 | |
| -@gerrit:opendev.org- Matthieu Huin https://matrix.to/#/@mhuin:matrix.org proposed: [zuul/zuul] UI: add pagination to builds, buildsets search https://review.opendev.org/c/zuul/zuul/+/793467 | 15:48 | |
| -@gerrit:opendev.org- Artem Goncharov proposed on behalf of James E. Blair https://matrix.to/#/@jim:acmegating.com: [zuul/zuul-jobs] Revert "Revert "Update upload-logs roles to support endpoint override"" https://review.opendev.org/c/zuul/zuul-jobs/+/776677 | 16:38 | |
| -@gerrit:opendev.org- Artem Goncharov proposed on behalf of James E. Blair https://matrix.to/#/@jim:acmegating.com: [zuul/zuul-jobs] Revert "Revert "Update upload-logs roles to support endpoint override"" https://review.opendev.org/c/zuul/zuul-jobs/+/776677 | 16:38 | |
| @mordred:inaugust.com | tristanC: ^^ I like how the "on behalf of" feature works for new patches in addition to merges. I think that's really good and clear information | 16:47 |
| @jim:acmegating.com | ++ | 16:47 |
| @tristanc_:matrix.org | glad you find it useful, I'd be happy to further improve the gerrit's notices if you have more suggestions :-) | 16:51 |
| @clarkb:matrix.org | If there is a way to tell matrix to not try and preload the site preview that might be nice as gerrit doesn't support the bits that make that work. I have it disabled globally in my client which isn't exactly what I want either | 16:53 |
| @tristanc_:matrix.org | Clark: it seems like this is not supported according to https://github.com/matrix-org/synapse/issues/2497 | 16:54 |
| @clarkb:matrix.org | I also looked at what it would take to have gerrit support that but quickly got over my head. The issue is gerrit loads everything dynamically with js rather than supplying any info in the html. I'm sure its possible to have the gerrit server return more info in the html but not the sort of thing I'm good at doing | 16:54 |
| @clarkb:matrix.org | you basically get the change identifier and then the loaded js knows what to do with that once the browser has loaded the js and there are no hint tags as a result | 16:56 |
| @jim:acmegating.com | is there anything gerrit could return to say "don't try to preview this?" | 16:57 |
| @clarkb:matrix.org | corvus: I don't think so | 16:59 |
| @apevec:matrix.org | maybe Matrix identifies itself with some specific User-Agent: then have rewrite rule to return empty html :) | 17:00 |
| @jim:acmegating.com | apevec: possibly so, i think it's synapse that does the actual request.... | 17:00 |
| @jim:acmegating.com | could...erm....actually even have it return a useful url preview implemented as a cgi-bin script :) | 17:01 |
| @jim:acmegating.com | not saying that's a good idea. but it's an idea :) | 17:01 |
| @clarkb:matrix.org | The best thing is probably figuring out how to hijack gerrit's java webserver to respond with just a bit more detail :) | 17:02 |
| @clarkb:matrix.org | Though I wonder if upstream wouldn't want to do that as they seem to intentionally defer as much page load to js as possible | 17:03 |
| @mordred:inaugust.com | might be worth bringing up upstream - I imagine it's not just matrix - slack probably has the same issue yeah? | 17:03 |
| @clarkb:matrix.org | ya and twitter and facebook (those two apps actually basically define what the metadata should look like and everyone else has followed along) | 17:03 |
| @mordred:inaugust.com | ++ ... if there is a general mechanism for defining url previews it would probably make sense for gerrit to implement | 17:04 |
| @tristanc_:matrix.org | in the case of the gerritbot, i think we actually don't want a preview as the title is self explanatory. Thus the bot could explicitly disable preview as proposed in https://github.com/matrix-org/matrix-doc/pull/2385/files | 17:07 |
| @jim:acmegating.com | i agree, that looks like a good enhancement | 17:08 |
| @mordred:inaugust.com | ++ | 17:08 |
| @tristanc_:matrix.org | it seems like zuul-web also does not serve a preview compatible response | 17:14 |
| @jim:acmegating.com | does this error from the zuul-registry test job make any sense? https://303203e2c1792c8c7292-9add4719a9922a9385555a8552fc2366.ssl.cf1.rackcdn.com/807798/2/check/zuul-operator-functional-k8s/796fce8/docker/k8s_POD_test-registry-6rdf8_default_63ba35e1-7cdc-4c83-a402-3ec2b8afdf09_0.txt | 17:15 |
| @jim:acmegating.com | that's from https://zuul.opendev.org/t/zuul/build/e29afceff10e4399ad83dac7dcc664f4/console | 17:15 |
| @jim:acmegating.com | where we run a k8s batch job to run "podman login" to the registry | 17:15 |
| @jim:acmegating.com | the first link is the output of the pod running that. | 17:16 |
| @jim:acmegating.com | if there were some error with podman authenticating, i would expect an error message along those lines | 17:17 |
| @jim:acmegating.com | the describe says exit code 125 | 17:18 |
| @jim:acmegating.com | 'man podman' says 125 is "125 The error is with podman itself" | 17:18 |
| @jim:acmegating.com | though i also get 125 from just trying to connect to a host that doesn't exist; it seems like a bit of a catch all | 17:21 |
| @jim:acmegating.com | no errors logged in the registry pod itself | 17:23 |
| @jim:acmegating.com | we did change a lot of stuff with the registry recently:/ | 17:25 |
| @jim:acmegating.com | i'll set an autohold and see if i can catch it. | 17:25 |
| -@gerrit:opendev.org- Tristan Cacqueray https://matrix.to/#/@tristanc_:matrix.org proposed: [zuul/zuul-operator] Add debug log-level to the podman login test command https://review.opendev.org/c/zuul/zuul-operator/+/807931 | 17:39 | |
| @tristanc_:matrix.org | corvus: perhaps ^ may help debug? | 17:40 |
| @jim:acmegating.com | tristanC: ++ | 17:44 |
| -@gerrit:opendev.org- Zuul merged on behalf of James E. Blair https://matrix.to/#/@jim:acmegating.com: [zuul/zuul] Try harder to unlock failed build requests https://review.opendev.org/c/zuul/zuul/+/807221 | 18:34 | |
| @vlotorev:matrix.org | Hi, gentle ping for reviewing https://review.opendev.org/#/c/803269 | 18:46 |
| -@gerrit:opendev.org- Zuul merged on behalf of lotorev vitaly: [zuul/zuul] doc: Specify job.required-projects is extended by child jobs https://review.opendev.org/c/zuul/zuul/+/803269 | 20:20 | |
| @mhuin:matrix.org | Hey there, I'm adding results pagination for builds and buildsets in the web UI in https://review.opendev.org/c/zuul/zuul/+/793467 - I would like some advice on what to do regarding the builds and buildsets API | 20:28 |
| @mhuin:matrix.org | This change is API breaking, making the builds and buildsets endpoints return an object with the total results and the offset rather than just the list of results (which is limited to 50 reuslts by default) | 20:29 |
| @mhuin:matrix.org | But I could also just add an arbitrarily high limit in the web UI queries, and handle the pagination UI-side. This wouldn't break the API but I'm concerned by the performance impact on zuul-web | 20:30 |
| @jim:acmegating.com | Matthieu Huin: i think the api break is preferable | 20:31 |
| @mhuin:matrix.org | or I could make the limit argument API side optional, or accept limit=-1 as a way to query without a limit | 20:31 |
| @clarkb:matrix.org | I expect the primary user of that api is the ui as long as both update together we should avoid most issues | 20:32 |
| @jim:acmegating.com | Matthieu Huin: assuming we accept the api break as the way to go, there's one other decision: do we merge a monolithic change like you have there, or do we do the same thing you did for zuul-client in the zuul-web js: handle both in the js, then change the api, then remove the backwards compat in the js. the main reason to do that would be to be able to test the JS against deployed sites | 20:33 |
| @mhuin:matrix.org | Clark: yeah the PS handles changes both in the backend and the UI - and there's a depends-on on zuul-client to handle the change as well in the patch chain | 20:33 |
| @jim:acmegating.com | ideally we would do that, but if it's really hard, we can probably merge the monolithic change if you promised you tested it :) | 20:33 |
| @mhuin:matrix.org | corvus: pinky promise! but I'm okay with doing this by the book. I can't guarantee I haven't overlooked something | 20:34 |
| @jim:acmegating.com | (i guess the other reason to split the existing change is so that if we merge the monolithing one, we have to more or less immediately restart our zuuls in order to preview-test any other web changes) | 20:35 |
| @mhuin:matrix.org | I can introduce a temporary api endpoint with the breaking change, so that both are exposed | 20:35 |
| @jim:acmegating.com | * (i guess the other reason to split the existing change is that if we merge the monolithing one, we have to more or less immediately restart our zuuls in order to preview-test any other web changes) | 20:35 |
| @mhuin:matrix.org | or yeah, just handle both in the js | 20:37 |
| @mhuin:matrix.org | I'll get to it tomorrow, thanks for the wise words | 20:38 |
| @mhuin:matrix.org | in the meantime, the web OIDC auth patch chain is still up for reviews :) | 20:38 |
| @jim:acmegating.com | we restarted opendev's zuul on current master; if that looks good tomorrow, we can cut a release, and that'll be a good checkpoint for rollbacks for the next set of sos patches | 21:32 |
| @fungicide:matrix.org | Zuul-related question about testing zuul (nodepool really i guess?) with devstack using virtualbox: http://lists.openstack.org/pipermail/openstack-discuss/2021-September/024774.html | 21:38 |
| -@gerrit:opendev.org- James E. Blair https://matrix.to/#/@jim:acmegating.com proposed on behalf of Tristan Cacqueray https://matrix.to/#/@tristanc_:matrix.org: [zuul/zuul-operator] Add debug log-level to the podman login test command https://review.opendev.org/c/zuul/zuul-operator/+/807931 | 21:39 | |
| @jim:acmegating.com | fungi: that reads a lot like a devstack and/or virtualbox question; i don't think there's a zuul question there | 21:40 |
| @jim:acmegating.com | > <@gerrit:opendev.org> James E. Blair https://matrix.to/#/@jim:acmegating.com proposed on behalf of Tristan Cacqueray https://matrix.to/#/@tristanc_:matrix.org: [zuul/zuul-operator] Add debug log-level to the podman login test command https://review.opendev.org/c/zuul/zuul-operator/+/807931 | 21:41 |
| rebase to get the minikube pin | ||
| @fungicide:matrix.org | yes, i expect it will lead to virtualbox-related questions about trying to connect zuul through the host adapter driver as well, but i agree they're probably running into a virtualbox issue. i just didn't know if anyone here had tried something similar with zuu(nodepool)l+devstack+virtualbox or could suggest a saner way to test zuul/nodepool with devstack | 21:42 |
| @fungicide:matrix.org | probably they need to get familiar with kvm instead | 21:43 |
| @fungicide:matrix.org | anyway, hopefully someone on openstack-discuss can guide them in a direction which leads to a useful zuul test setup for them | 21:48 |
| -@gerrit:opendev.org- Zuul merged on behalf of Tobias Henkel: [zuul/zuul] Don't hide exception in _processCatJobs https://review.opendev.org/c/zuul/zuul/+/807421 | 22:28 | |
| @clarkb:matrix.org | I responded to that email | 23:01 |
Generated by irclog2html.py 2.17.2 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!