| *** mattw4 has quit IRC | 00:06 | |
| *** jamesmcarthur has quit IRC | 00:06 | |
| *** jamesmcarthur has joined #zuul | 00:40 | |
| SpamapS | Anybody know what's going on with the Zuul Kubernetes operator? | 00:45 |
|---|---|---|
| SpamapS | Getting tired of maintaining my spruce/yaml/bash/Makefile monstrosity.. wondering if I should divert that effort to working on the operator. | 00:46 |
| ianw | re testinfra, filed https://storyboard.openstack.org/#!/story/2006498 as i can't see what's up | 00:51 |
| clarkb | SpamapS: I think you want jeliu | 00:52 |
| *** jamesmcarthur has quit IRC | 01:06 | |
| *** jamesmcarthur has joined #zuul | 01:06 | |
| SpamapS | jeliu, clarkb: are there patches up? Storyboard tasks, anything I can use to help/track/etc. ? | 01:10 |
| *** mattw4 has joined #zuul | 01:11 | |
| clarkb | SpamapS: https://review.opendev.org/#/q/project:zuul/zuul-operator | 01:15 |
| *** jamesmcarthur has quit IRC | 01:21 | |
| *** spsurya has joined #zuul | 01:28 | |
| *** jamesmcarthur has joined #zuul | 02:05 | |
| *** jamesmcarthur has quit IRC | 02:18 | |
| *** jamesmcarthur has joined #zuul | 02:20 | |
| *** jamesmcarthur has quit IRC | 02:26 | |
| *** jamesmcarthur_ has joined #zuul | 02:26 | |
| *** jamesmcarthur_ has quit IRC | 02:31 | |
| *** jamesmcarthur has joined #zuul | 02:34 | |
| *** jamesmcarthur has quit IRC | 03:34 | |
| *** jamesmcarthur has joined #zuul | 03:37 | |
| *** rfolco has quit IRC | 03:38 | |
| *** jamesmcarthur has quit IRC | 03:42 | |
| *** jamesmcarthur has joined #zuul | 04:05 | |
| *** igordc has joined #zuul | 04:10 | |
| *** jamesmcarthur has quit IRC | 04:12 | |
| *** igordc has quit IRC | 04:23 | |
| *** jank has joined #zuul | 04:48 | |
| *** jamesmcarthur has joined #zuul | 05:09 | |
| *** jamesmcarthur has quit IRC | 05:13 | |
| SpamapS | clarkb:ty | 05:21 |
| *** pcaruana has joined #zuul | 05:25 | |
| flaper87 | tristanC: do you have a custom executor image? The base one doesn't have kubectl/oc installed. | 05:33 |
| *** swest has joined #zuul | 05:44 | |
| *** jamesmcarthur has joined #zuul | 05:49 | |
| *** jamesmcarthur has quit IRC | 05:54 | |
| *** mattw4 has quit IRC | 06:00 | |
| *** raukadah is now known as chandankumar | 06:03 | |
| *** jank has quit IRC | 06:17 | |
| *** jank has joined #zuul | 06:19 | |
| *** bjackman_ has joined #zuul | 06:20 | |
| *** bjackman_ has quit IRC | 06:46 | |
| *** jamesmcarthur has joined #zuul | 06:51 | |
| *** bjackman_ has joined #zuul | 06:54 | |
| *** jamesmcarthur has quit IRC | 06:55 | |
| *** jank has quit IRC | 07:00 | |
| *** threestrands has quit IRC | 07:10 | |
| *** saneax has joined #zuul | 07:21 | |
| *** jamesmcarthur has joined #zuul | 07:22 | |
| *** jpena|off is now known as jpena | 07:41 | |
| *** jamesmcarthur has quit IRC | 07:42 | |
| *** jank has joined #zuul | 07:49 | |
| *** pcaruana has quit IRC | 08:08 | |
| *** pcaruana has joined #zuul | 08:13 | |
| *** badboy has joined #zuul | 08:15 | |
| badboy | how do I change the logging level for Zuul and it's components? | 08:16 |
| badboy | under [logger_zuul] change the level=DEBUG to level=INFO? | 08:16 |
| *** tosky has joined #zuul | 08:49 | |
| *** NBorg has joined #zuul | 09:05 | |
| *** rfolco has joined #zuul | 09:07 | |
| *** NBorg has quit IRC | 11:08 | |
| *** jpena is now known as jpena|lunch | 11:36 | |
| tristanC | flaper87: we use origin-clients provided by centos-release-openshift-origin39 | 11:41 |
| *** johnsom has quit IRC | 11:53 | |
| *** jpena|lunch has quit IRC | 11:54 | |
| *** johnsom has joined #zuul | 11:55 | |
| *** jpena has joined #zuul | 11:56 | |
| *** bjackman_ has quit IRC | 12:04 | |
| *** EmilienM is now known as EvilienM | 12:24 | |
| *** spsurya has quit IRC | 12:26 | |
| *** bjackman has joined #zuul | 12:37 | |
| *** badboy has quit IRC | 12:45 | |
| *** gtema_ has joined #zuul | 12:51 | |
| corvus | SpamapS, clarkb, tristanC: i've just got word that jeliu's internship has ended and he won't have very much time to work on the operator; i think diverting attention/resources toward that would be prudent. i think he was most recently working on getting the percona operator working. the spec is here: https://zuul-ci.org/docs/zuul/developer/specs/kubernetes-operator.html | 13:01 |
| *** gtema_ has quit IRC | 13:01 | |
| *** bjackman has quit IRC | 13:01 | |
| *** jank has quit IRC | 13:05 | |
| *** avass has joined #zuul | 13:06 | |
| *** jamesmcarthur has joined #zuul | 13:07 | |
| avass | corvus: Is there a reason why mirror-workspace-git-repos runs git reset --hard but not git clean -xdf? | 13:09 |
| corvus | avass: i don't know; are you running into a case where that's needed? maybe you're reusing workspaces, but we just haven't done that much upstream so it was omitted... | 13:12 |
| avass | corvus: Yep, since we're on static nodes the workspace are gonna be reused a lot | 13:13 |
| avass | corvus: encountered a bug because some untracked files didn't get removed from a previous run | 13:13 |
| corvus | avass: it seems reasonable to propose a patch to add that, but also, i just woke up, so i may still be missing things :) | 13:14 |
| avass | corvus: Added that locally for now. Just wanted to make sure I didn't miss something | 13:15 |
| corvus | avass: zuul's merger (which is designed to deal with all kinds of repo states) does a hard reset followed by a clean -xdf, so that sounds like another vote in favor of adding it :) | 13:15 |
| avass | Yep :) | 13:16 |
| *** jank has joined #zuul | 13:22 | |
| openstackgerrit | Albin Vass proposed zuul/zuul-jobs master: Clean non-bare remote repos https://review.opendev.org/680689 | 13:42 |
| pabelanger | SpamapS: corvus: clarkb: tristanC: re operator, I can make some time to work on it, won't be 100% but I can start getting involved there. But if others want to help drive it, happy to see that too | 13:46 |
| corvus | i think there's enough work there for several folks to help out in bits, and the spec should keep us on the same page. maybe if someone wants to claim tasks like "get pxc working", "get zk working", "write nodepool config sharding script", and "update operator crd to match spec" we could start a storyboard | 13:49 |
| pabelanger | Yay, breaking it out into parts would work, like we did in zuulv3 spec. Then figure out all the humans who want to do the parts | 13:52 |
| pabelanger | I can maybe look to creating for this weekend | 13:52 |
| *** jank has quit IRC | 13:52 | |
| corvus | pabelanger: do you have any idea why opendev uses "use-cached-repos" rather than "prepare-workspace-git" ? | 13:56 |
| corvus | avass: your change looks fine, but in looking at it, i noticed a hole in our testing, so i'm going to take a minute and see about closing that up | 13:58 |
| fungi | is that because we preinstall caches of our git repositories during image building? | 13:58 |
| fungi | is prepare-workspace-git supposed to also accommodate that? | 13:58 |
| pabelanger | corvus: no, feels like we just didn't do the migration after tobiash added it | 13:58 |
| pabelanger | we do use prepare-workspace-git for zuul.a.c | 13:58 |
| corvus | fungi: i it is supposed to do that; it looks like a slightly more advanced version of use-cached-repos | 13:58 |
| fungi | pabelanger: aha, that does sound like something we'd do ;) | 13:58 |
| avass | corvus: alright | 13:58 |
| corvus | okay, so i think we should: 1) migrate opendev to that role; 2) switch the zuul-jobs testing to that role; 3) recheck avass's change, then merge | 13:59 |
| pabelanger | +1 | 13:59 |
| fungi | sgtm | 13:59 |
| corvus | because it sure looks to me like zuul-jobs is testing use-cached-repos which is crazy since that's not even in that repo :) | 13:59 |
| fungi | i concur, based on http://codesearch.openstack.org/?q=use-cached-repos | 14:02 |
| *** hashar has joined #zuul | 14:05 | |
| *** jamesmcarthur has quit IRC | 14:07 | |
| openstackgerrit | James E. Blair proposed zuul/zuul-jobs master: Update the base-roles test to use prepare-workspace-git https://review.opendev.org/680703 | 14:15 |
| *** jamesmcarthur has joined #zuul | 14:15 | |
| corvus | i think #1 and #2 can happen independently, so i'm going to go ahead and rebase avass's change on that ^ which should get it tested | 14:15 |
| openstackgerrit | James E. Blair proposed zuul/zuul-jobs master: Clean non-bare remote repos https://review.opendev.org/680689 | 14:16 |
| *** avass has quit IRC | 14:22 | |
| openstackgerrit | Tristan Cacqueray proposed zuul/zuul-base-jobs master: Add support for kubectl connection https://review.opendev.org/680711 | 14:31 |
| openstackgerrit | Tristan Cacqueray proposed zuul/zuul-jobs master: Add prepare-workspace-openshift role https://review.opendev.org/631402 | 14:33 |
| openstackgerrit | Tristan Cacqueray proposed zuul/zuul-jobs master: Add remove-zuul-sshkey https://review.opendev.org/680712 | 14:33 |
| openstackgerrit | Tristan Cacqueray proposed zuul/zuul-base-jobs master: Add support for kubectl connection https://review.opendev.org/680711 | 14:33 |
| tristanC | flaper87: here are some changes to make the base jobs work with kubectl connection: https://review.opendev.org/680711 | 14:34 |
| corvus | tristanC: is prepare-workspace-openshift openshift specific or could it be k8s? | 14:36 |
| tristanC | corvus: it may works with k8s, but we are using the oc client that provides a convenient rsync wrapper | 14:37 |
| corvus | does oc work with plain k8s or does it require os? (i know the openshift python module works with plain k8s) | 14:39 |
| tristanC | corvus: it may very well work, i haven't tested it | 14:39 |
| corvus | we should really get that k8s up in opendev for stuff like this :) | 14:39 |
| tristanC | corvus: we are actually testing this service: https://pagure.io/software-factory/k1s | 14:40 |
| *** mattw4 has joined #zuul | 14:42 | |
| corvus | tristanC: podman on the backend? | 14:43 |
| tristanC | corvus: yes, podman | 14:43 |
| corvus | neat, i think i understand :) | 14:44 |
| tristanC | a bit of a hack, but we are looking forward swapping the runc driver by this new service :-) | 14:44 |
| corvus | i like that plan. it's a good hack :) | 14:45 |
| *** michael-beaver has joined #zuul | 14:49 | |
| *** mattw4 has quit IRC | 14:49 | |
| *** mattw4 has joined #zuul | 14:49 | |
| *** chandankumar is now known as raukadah | 14:54 | |
| *** mattw4 has quit IRC | 14:57 | |
| *** saneax has quit IRC | 15:09 | |
| corvus | i think i have a lead on the task bug from yesterday -- i think it's because that last task was a handler | 15:10 |
| openstackgerrit | James E. Blair proposed zuul/zuul master: Record handler tasks in json job output https://review.opendev.org/680726 | 15:16 |
| corvus | pabelanger, clarkb: ^ mystery from yesterday solved | 15:16 |
| *** noorul has joined #zuul | 15:33 | |
| noorul | ofosos: hi | 15:33 |
| *** Diabelko has quit IRC | 15:36 | |
| *** Diabelko has joined #zuul | 15:36 | |
| *** mattw4 has joined #zuul | 15:51 | |
| clarkb | corvus: have a moment for quick rereview of https://review.opendev.org/#/c/679670/ the faq page? | 15:53 |
| *** hashar has quit IRC | 15:54 | |
| corvus | clarkb: +3 thx | 15:54 |
| openstackgerrit | Merged zuul/zuul-website master: Add Zuul FAQ page https://review.opendev.org/679670 | 15:57 |
| noorul | Is zuul for opendev deployed and configured using gitops? If so can someone point me to the code? | 15:59 |
| corvus | noorul: yes, but don't do what we do :) we're moving to containers and will abandon our current system. 1 sec and i'll get links | 16:00 |
| corvus | noorul: https://docs.openstack.org/infra/system-config/zuul.html | 16:00 |
| noorul | corvus: By containers do you mean k8s operator? | 16:01 |
| corvus | noorul: eventually, but we'd be happy just moving to docker | 16:01 |
| corvus | it's all puppet now | 16:01 |
| noorul | corvus: What will you recommend for production? | 16:03 |
| corvus | you have to keep in mind that when we deployed zuul initially, ansible itself was only 3 months old, and k8s hadn't been invented yet. :) | 16:03 |
| corvus | noorul: if i were setting up a new deployment today, i would use docker with upstream latest images (running on master) based on the docker-compose file. if you have a k8s, it should be fairly straightforward to translate that to k8s crd. | 16:04 |
| clarkb | ya I'm not sure I would deploy a k8s just to run a zuul; however, if you plan to run tests in k8s or have k8s for other applications it wouldn't be a bad choice | 16:05 |
| corvus | noorul: if you don't want to use containers at all, the windmill project is a fine choice for deploying directly on vms | 16:05 |
| corvus | https://opendev.org/windmill/ | 16:06 |
| clarkb | corvus: probably ^ is a good FAQ page entry too? | 16:07 |
| clarkb | "Can zuul run in k8s?" YES "Does zuul have docker images?" YES | 16:07 |
| corvus | clarkb: ++ | 16:07 |
| openstackgerrit | Clark Boylan proposed zuul/zuul-website master: Add faq entry about kubernetes and docker images https://review.opendev.org/680742 | 16:14 |
| corvus | AJaeger_: fyi, i was just reading the pdf thread -- not sure if you're aware of this option: https://zuul-ci.org/docs/zuul/user/config.html#attr-project.vars | 16:21 |
| noorul | How will the user know via review system that corresponding job has started running? | 16:21 |
| corvus | noorul: depends on the review system and the zuul pipeline configuration for the 'start' action | 16:22 |
| corvus | noorul: https://zuul-ci.org/docs/zuul/user/config.html#attr-pipeline.start | 16:22 |
| noorul | corvus: I see | 16:23 |
| corvus | noorul: in opendev gerrit we do nothing in the check pipeline, but we leave a comment and clear the vote on gate. in github, folks usually set the status. | 16:23 |
| noorul | corvus: I think on stash also we need to set the status | 16:24 |
| noorul | corvus: Otherwise users will be clueless | 16:24 |
| noorul | corvus: s/stash/bitbucket | 16:24 |
| corvus | noorul: seems reasonable; i think there are some doc changes to add sample pipelines for the different drivers. would be good to add that to the bitbucket stack | 16:25 |
| clarkb | corvus: I believe the gerrit one merged but I think there wa sstill some questions on the github one? | 16:26 |
| * clarkb finds it | 16:26 | |
| clarkb | https://review.opendev.org/#/c/672712/ is the github one | 16:27 |
| clarkb | looks like pabelanger has been reviewing it | 16:27 |
| noorul | Any idea when bitbucket patches will land in master? | 16:28 |
| clarkb | noorul: the current review comments need to be addressed at least | 16:29 |
| pabelanger | clarkb: yah, github pipelines a little more tricky :( | 16:31 |
| noorul | Unable to reach out to ofosos :( | 16:34 |
| *** pcaruana has quit IRC | 16:37 | |
| noorul | Do we test the zuul containers in the pipeline? | 16:41 |
| clarkb | noorul: the zuul quickstart job tests the containers | 16:43 |
| *** jpena is now known as jpena|off | 16:48 | |
| *** noorul has quit IRC | 16:50 | |
| *** pcaruana has joined #zuul | 16:51 | |
| *** tosky has quit IRC | 16:58 | |
| ofosos | noorul: what's up? | 17:05 |
| ofosos | I'll have a look at the comments on the weekend. | 17:05 |
| ofosos | Have a nice weekend everybody. | 17:05 |
| ofosos | Off for a Schnitzel and beers. | 17:06 |
| clarkb | ofosos: enjoy! | 17:06 |
| *** hashar has joined #zuul | 17:15 | |
| clarkb | hrm something wrong with the website promotion job? maybe that is what you meant about adding it to the artifacts? | 17:31 |
| clarkb | if we are promoting we need that artifact ya? | 17:31 |
| corvus | clarkb: that is not what i meant, but now that you mention it, it sounds very plausible | 17:32 |
| clarkb | zuul-website runs a post job | 17:32 |
| clarkb | so maybe a different problem | 17:32 |
| corvus | clarkb: is http://zuul.opendev.org/t/zuul/build/acbb494ef30f4bb6a3fba109928f738d the right build? | 17:33 |
| clarkb | ya that ref is correct | 17:34 |
| clarkb | maybe the afs vos release isn't working | 17:34 |
| corvus | clarkb: yeah, i see the files in afs, must not have released | 17:35 |
| clarkb | the project.zuul volume is not locked for a release | 17:39 |
| clarkb | ow I have to remember where we run the releasees from | 17:39 |
| corvus | clarkb: i think i've got something | 17:40 |
| corvus | clarkb: they run on afsdb01 | 17:40 |
| corvus | the log shows it getting up to starlingx but no further | 17:40 |
| corvus | clarkb: vos examine -localauth project.starlingx.readonly is hanging for me; let's switch to -infra | 17:41 |
| clarkb | k | 17:41 |
| *** hashar has quit IRC | 17:42 | |
| *** jamesmcarthur has quit IRC | 17:45 | |
| *** jangutter has quit IRC | 17:54 | |
| *** hashar has joined #zuul | 18:11 | |
| corvus | faq is published! | 18:25 |
| corvus | https://zuul-ci.org/docs/faq.html | 18:26 |
| Shrews | tobiash: clarkb: ooh, while trying help tobias with a test for https://review.opendev.org/671704 , i found an actual bug in the new code that we all missed. yay tests! :) | 18:49 |
| clarkb | the log call ? | 18:51 |
| Shrews | yeah. i left a comment | 18:51 |
| *** igordc has joined #zuul | 18:53 | |
| clarkb | corvus: https://review.opendev.org/#/c/680498/1 failed tests because we test the heuristics that were removed | 18:57 |
| clarkb | corvus: and that is parent for the handler fix | 18:57 |
| clarkb | corvus: I can push up a new patchset if you want | 18:57 |
| corvus | clarkb: derp. if you have it handy, please do; i'm warming up lunch | 19:04 |
| clarkb | k | 19:04 |
| openstackgerrit | Clark Boylan proposed zuul/zuul master: Web: rely on new attributes when determining task failure https://review.opendev.org/680498 | 19:06 |
| openstackgerrit | Clark Boylan proposed zuul/zuul master: Record handler tasks in json job output https://review.opendev.org/680726 | 19:06 |
| clarkb | done | 19:07 |
| clarkb | I need to sort out lunc too and will then be running errands | 19:09 |
| openstackgerrit | David Shrewsbury proposed zuul/nodepool master: Fix node failures when at volume quota https://review.opendev.org/671704 | 19:15 |
| Shrews | clarkb: tobiash: that adds the test for tobias, but we might want to wait until he returns from vacation to validate it?? i don't think it's urgent | 19:17 |
| *** hashar has quit IRC | 19:46 | |
| *** hashar has joined #zuul | 19:47 | |
| openstackgerrit | Jeremy Stanley proposed zuul/zuul-website master: Update to page titles and Users https://review.opendev.org/680459 | 20:22 |
| *** rfolco has quit IRC | 20:50 | |
| *** hashar has quit IRC | 21:14 | |
| *** hashar has joined #zuul | 21:21 | |
| corvus | \o/ here's a milestone: 2019-09-06 14:46:42,836 DEBUG zuul.Pipeline.zuultest.check: [e: c4e8b6ad59f844c3b523a03ab663dd86] Event <GerritTriggerEvent pending-check localhost/test 1,1> for change <Change 0x7fd6412d0470 test 1,1> matched <GerritEventFilter types: pending-check uuid: zuul:check ignore_deletes: True> in pipeline <IndependentPipelineManager check> | 21:47 |
| *** EvilienM is now known as EmilienM | 21:47 | |
| corvus | that's zuul using the checks api to enqueue an item in a pipeline | 21:48 |
| corvus | just got that working in local testing | 21:48 |
| fungi | that's what i was going to guess from the keywords present | 21:48 |
| fungi | awesome! | 21:48 |
| corvus | this is the pipeline config for that: http://paste.openstack.org/show/772113/ | 21:49 |
| corvus | paladox: ^ | 21:49 |
| paladox | \o/ | 21:49 |
| paladox | yay! | 21:50 |
| openstackgerrit | James E. Blair proposed zuul/zuul master: WIP: Add support for the Gerrit checks plugin https://review.opendev.org/680778 | 21:50 |
| corvus | that's my very un-finished work-in-progress code with bunches of comments and stuff, just to checkpoint that since i'm going to eod soon | 21:51 |
| * fungi is excited | 21:53 | |
| fungi | have a good weekend! | 21:53 |
| *** hashar has quit IRC | 21:55 | |
| *** rfolco has joined #zuul | 22:38 | |
| *** michael-beaver has quit IRC | 22:38 | |
| *** mattw4 has quit IRC | 23:05 | |
| clarkb | https://review.opendev.org/#/c/680498/2 passes tests now. I +2'd it | 23:07 |
| clarkb | if anyone else wants to review that it will unstick another fix in the console rendering | 23:08 |
| paladox | i'm very excited when zuul takes over the ci on gerrit-review! | 23:10 |
| paladox | will be easier to tell your builds from the ui :) | 23:10 |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!