| *** spsurya has joined #zuul | 01:09 | |
| *** jamesmcarthur has joined #zuul | 02:37 | |
| *** bhavikdbavishi has joined #zuul | 02:41 | |
| *** bhavikdbavishi1 has joined #zuul | 02:44 | |
| *** bhavikdbavishi has quit IRC | 02:45 | |
| *** bhavikdbavishi1 is now known as bhavikdbavishi | 02:45 | |
| *** jamesmcarthur has quit IRC | 03:33 | |
| *** jamesmcarthur has joined #zuul | 03:47 | |
| *** jamesmcarthur has quit IRC | 04:00 | |
| *** jamesmcarthur has joined #zuul | 04:00 | |
| *** saneax has quit IRC | 04:09 | |
| *** jhesketh has quit IRC | 04:18 | |
| *** jhesketh has joined #zuul | 04:19 | |
| *** jamesmcarthur has quit IRC | 04:24 | |
| *** jamesmcarthur has joined #zuul | 04:26 | |
| *** jamesmcarthur has quit IRC | 04:37 | |
| *** jamesmcarthur has joined #zuul | 04:44 | |
| *** jamesmcarthur has quit IRC | 05:00 | |
| *** jamesmcarthur has joined #zuul | 05:05 | |
| *** raukadah is now known as chkumar|ruck | 05:23 | |
| *** jamesmcarthur has quit IRC | 05:38 | |
| *** jamesmcarthur has joined #zuul | 05:42 | |
| *** AJaeger has quit IRC | 05:42 | |
| *** jamesmcarthur has quit IRC | 05:44 | |
| *** AJaeger has joined #zuul | 05:45 | |
| *** themroc has joined #zuul | 05:58 | |
| openstackgerrit | Felix Schmidt proposed zuul/zuul master: Make direct-push configurable on project-level https://review.opendev.org/677109 | 06:02 |
|---|---|---|
| openstackgerrit | Felix Schmidt proposed zuul/zuul master: Implement push job in merger https://review.opendev.org/677110 | 06:02 |
| openstackgerrit | Felix Schmidt proposed zuul/zuul master: Push changes in GerritReporter if direct-push is enabled https://review.opendev.org/677111 | 06:02 |
| AJaeger | zuul experts, any idea why we see CONFIG_ERROR for periodic requirements jobs? See http://zuul.opendev.org/t/openstack/buildsets?result=CONFIG_ERROR&pipeline=periodic - you might need to look at the log files, I could not get any further information out of this. | 06:14 |
| *** saneax has joined #zuul | 06:27 | |
| *** threestrands has quit IRC | 06:34 | |
| *** threestrands has joined #zuul | 06:35 | |
| AJaeger | corvus: do we still need set-zuul-log-path-fact with Swift storage? | 06:56 |
| AJaeger | corvus: https://review.opendev.org/#/c/676756 looks fine, the tests don't show the URL anymore. | 06:56 |
| *** bhavikdbavishi has quit IRC | 07:19 | |
| *** bhavikdbavishi has joined #zuul | 07:19 | |
| *** threestrands has quit IRC | 07:19 | |
| *** themroc has quit IRC | 07:24 | |
| *** yolanda has quit IRC | 07:42 | |
| *** yolanda__ has joined #zuul | 07:43 | |
| *** jpena|off is now known as jpena | 07:44 | |
| *** hashar has joined #zuul | 07:56 | |
| arxcruz | zbr: https://review.rdoproject.org/r/#/c/21869/ ansible 2.8 | 08:23 |
| arxcruz | zbr: i added you as qe on https://tree.taiga.io/project/tripleo-ci-board/task/1256 | 08:24 |
| arxcruz | zbr: i did not add on the upstream base otherwise all upstream job would also move to ansible 2.8 which is not the intent | 08:24 |
| *** yolanda__ is now known as yolanda | 08:28 | |
| openstackgerrit | Felix Schmidt proposed zuul/zuul master: Implement push job in merger https://review.opendev.org/677110 | 08:32 |
| openstackgerrit | Felix Schmidt proposed zuul/zuul master: Push changes in GerritReporter if direct-push is enabled https://review.opendev.org/677111 | 08:32 |
| *** noorul has joined #zuul | 08:33 | |
| *** noorul has quit IRC | 08:34 | |
| zbr | arxcruz: why not switching tripleo-ci-base-ovb to 2.8 directly? | 08:49 |
| arxcruz | zbr: because the job is defined on tripleo-ci repo upstream | 08:50 |
| arxcruz | and doing so, we would move also upstream jobs | 08:50 |
| zbr | yeah. that is what I am asking, why not moving all | 08:51 |
| zbr | what prevents us from doing it? | 08:51 |
| arxcruz | zbr: we can discuss this today, my understanding was only rdo | 08:52 |
| arxcruz | zbr: also, it's in the commit message | 08:52 |
| arxcruz | Notice this could not be done in the upstream base because it would move | 08:52 |
| arxcruz | also the upstream jobs to ansible 2.8, which is not the intention right | 08:52 |
| arxcruz | now. | 08:52 |
| zbr | added comments | 08:52 |
| *** themroc has joined #zuul | 09:13 | |
| *** themroc has quit IRC | 09:16 | |
| *** themroc has joined #zuul | 09:16 | |
| *** hashar has quit IRC | 10:21 | |
| *** jpena is now known as jpena|lunch | 11:39 | |
| *** rlandy has joined #zuul | 11:47 | |
| *** rlandy is now known as rlandy|rover | 11:48 | |
| *** chkumar|ruck is now known as chkumar|rover | 11:49 | |
| *** rlandy|rover is now known as rlandy|ruck | 11:49 | |
| *** hashar has joined #zuul | 11:58 | |
| *** weshay_pto is now known as weshay | 12:00 | |
| *** rfolco has joined #zuul | 12:28 | |
| *** jpena|lunch is now known as jpena | 12:31 | |
| *** spsurya has quit IRC | 12:43 | |
| *** jeliu_ has joined #zuul | 13:57 | |
| *** sshnaidm is now known as sshnaidm|bbl | 14:16 | |
| *** yolanda has quit IRC | 14:31 | |
| *** SpamapS has quit IRC | 14:31 | |
| *** yolanda has joined #zuul | 14:36 | |
| *** noorul has joined #zuul | 14:48 | |
| *** saneax has quit IRC | 14:49 | |
| noorul | hi | 14:50 |
| noorul | I am trying to do docker pull insecure-ci-registry.opendev.org:5000/zuul/zuul-web:7a6c26a4237e4d77acb595487a8fc0b3_latest | 14:50 |
| noorul | I get following error | 14:50 |
| noorul | Error response from daemon: Get https://insecure-ci-registry.opendev.org:5000/v2/zuul/zuul-web/manifests/7a6c26a4237e4d77acb595487a8fc0b3_latest: no basic auth credentials | 14:50 |
| pabelanger | I don't think that registery is used by public is it? | 14:50 |
| pabelanger | just for testing | 14:50 |
| noorul | How do I get the containers for zuul locally? | 14:51 |
| pabelanger | they should be on docker hum | 14:51 |
| pabelanger | hub* | 14:51 |
| pabelanger | https://hub.docker.com/r/zuul/ | 14:51 |
| noorul | For a patch that is in review | 14:51 |
| pabelanger | oh | 14:51 |
| *** sgw has joined #zuul | 14:51 | |
| pabelanger | Will have to defer to somebody else | 14:51 |
| pabelanger | otherwise, you can download patchset and build locally | 14:52 |
| noorul | When I do docker build, it creates 4 images, but no tagging | 14:52 |
| pabelanger | I believe you need to manually tag | 14:53 |
| noorul | The problem is identifying which image belongs to which service | 14:53 |
| pabelanger | noorul: could you share what you are trying to do? | 14:54 |
| noorul | I am following this https://zuul-ci.org/docs/zuul/admin/quick-start.html | 14:54 |
| noorul | But I want to use https://review.opendev.org/#/c/677093 | 14:55 |
| panda | is it possible to ask zuul for meta information in a job ? e.g. "please tell me the initial zuul-inventory you would assemble while running job X" | 14:55 |
| pabelanger | noorul: you should be able to use git-review or gerrit links and download 677093 locally | 14:56 |
| pabelanger | eg: git-review -d 677093 | 14:56 |
| pabelanger | then docker-compose down / docker-compose up | 14:56 |
| pabelanger | and be running that commit | 14:56 |
| noorul | How does that work? | 14:57 |
| noorul | docker compose points to docker hub image | 14:57 |
| pabelanger | oh, I guess there is no option to force builds locally? | 14:57 |
| noorul | That is easy by running command "docker build ." | 14:58 |
| noorul | The problem is identifying which image is for which service so that I can tag | 14:58 |
| noorul | and use that in docker compose | 14:58 |
| noorul | Now I am running diocker image history and figuring out | 14:59 |
| pabelanger | okay | 14:59 |
| noorul | Shouldn't there be a better way? | 14:59 |
| *** SpamapS has joined #zuul | 14:59 | |
| pabelanger | not sure will defer to corvus | 14:59 |
| pabelanger | but, i don't think quickstart is setup to do development builds, more released builds from docker hub | 15:00 |
| noorul | I agree | 15:00 |
| pabelanger | maybe see what is needed to support that workflow | 15:00 |
| noorul | But my question is with respect to development | 15:00 |
| pabelanger | yah | 15:00 |
| pabelanger | noorul: https://opendev.org/zuul/zuul/src/branch/master/playbooks/quick-start is how we test on zuul.o.o | 15:01 |
| pabelanger | maybe you can follow along there for local | 15:01 |
| pabelanger | https://zuul-ci.org/docs/zuul-jobs/docker-image.html also gives some info about how it works | 15:02 |
| pabelanger | I believe use-buildset-registry might handle tagging part | 15:02 |
| corvus | the image names are all in the dockerfile | 15:04 |
| fungi | noorul: i think for the insecure-ci-registry images you'd need to disable authentication, though i haven't tried it myself. the idea was to be able to test with the built images for the proposed change series | 15:04 |
| fungi | since those won't appear on dockerhub until the changes get approved | 15:04 |
| corvus | just "docker build -l zuul-executor -t zuul-executor" to build the executor image | 15:04 |
| corvus | if you're building images locally, you don't need any registry stuff | 15:05 |
| corvus | just read the dockerfile | 15:05 |
| fungi | right, the suggestion of using the docker:// urls in the build result was in order to avoid having to rebuild them from source locally | 15:05 |
| fungi | but i don't know what sort of retention we actually have there, or if those might get deleted as soon as the buildset reports | 15:06 |
| noorul | corvus: what is -l option for? | 15:07 |
| noorul | corvus: First time hearing about that option | 15:07 |
| corvus | noorul: it's the label option for multi-stage builds: https://docs.docker.com/develop/develop-images/multistage-build/ | 15:07 |
| fungi | oh, and insecure-ci-registry actually seems to expect authentication now that i'm trying to fetch from it, so i guess that won't work | 15:07 |
| fungi | just gives a 401 Unauthorized otherwise | 15:08 |
| corvus | noorul: each of the images we build correspond 1:1 to a label in the dockerfile, so we build them all with a single label | 15:08 |
| clarkb | fungi: yes we manage a zuul secret that allow jobs to talk to it. It isn'ta public registry | 15:08 |
| corvus | noorul: but you have to run "docker build" for each one, specifying a label each time | 15:08 |
| fungi | ahh, too bad | 15:09 |
| fungi | i suppose listing those urls in the build info isn't terribly helpful in that case | 15:09 |
| noorul | Docker version 18.09.7, build 2d0083d | 15:10 |
| noorul | $ docker build -l zuul-executor -t zuul-executor | 15:10 |
| noorul | unknown shorthand flag: 'l' in -l | 15:10 |
| noorul | See 'docker build --help'. | 15:10 |
| corvus | noorul: it's --label | 15:10 |
| noorul | corvus: cool, that helps | 15:11 |
| noorul | corvus: I hope base image will be cached | 15:11 |
| corvus | noorul: yes, they're mostly the same image with only a different default command | 15:14 |
| corvus | should be fast | 15:14 |
| noorul | hmm | 15:14 |
| noorul | One more question, how do I get https://review.opendev.org/#/c/677093 locally re-based against master? | 15:15 |
| corvus | noorul: git review -d 677093 && git remote update && git rebase origin/master | 15:16 |
| fungi | you could use -x instead of -d while master is checked out | 15:16 |
| corvus | that's probably one of many ways | 15:16 |
| corvus | two of many ways :) | 15:16 |
| fungi | or what corvus said, yep | 15:16 |
| corvus | but there's a whole stack | 15:16 |
| noorul | will that get entire patch set? | 15:16 |
| corvus | so to use '-x' you'd check out master then use 'git review -x #' for each # in the stack | 15:16 |
| fungi | ahh, right, so you'd need to -x each dependency in order so rebase -i is easier | 15:16 |
| noorul | I wish I could use that image from insecure-ci-registry | 15:17 |
| *** saneax has joined #zuul | 15:17 | |
| *** jamesmcarthur has joined #zuul | 15:19 | |
| corvus | noorul: me too, but the docker registry doesn't have a way to require authentication for writes but not reads. hopefully someone will be able to improve on that someday. | 15:20 |
| noorul | corvus: Can we merge this? | 15:21 |
| fungi | i wonder if we could make those images separately downloadable as job artifacts? | 15:22 |
| fungi | so that you could at least fetch them with curl or wget or a browser | 15:22 |
| noorul | docker save | 15:22 |
| *** chkumar|rover is now known as raukadah | 15:24 | |
| *** noorul has quit IRC | 15:31 | |
| *** noorul has joined #zuul | 15:31 | |
| openstackgerrit | Tristan Cacqueray proposed zuul/zuul master: trigger: add job filter event https://review.opendev.org/639905 | 15:37 |
| *** themroc has quit IRC | 15:37 | |
| openstackgerrit | Tristan Cacqueray proposed zuul/zuul master: webtrigger: add initial driver and event https://review.opendev.org/555153 | 15:40 |
| openstackgerrit | Tristan Cacqueray proposed zuul/zuul master: webtrigger: add web route and rpclistener https://review.opendev.org/554839 | 15:46 |
| openstackgerrit | Tristan Cacqueray proposed zuul/zuul master: web: add build button to trigger job https://review.opendev.org/635716 | 15:46 |
| *** mattw4 has joined #zuul | 15:48 | |
| openstackgerrit | James E. Blair proposed zuul/zuul master: Render the logfile under the manifest https://review.opendev.org/676843 | 16:00 |
| *** sgw has quit IRC | 16:02 | |
| tristanC | panda: there is a topic to add an endpoint to query zuul job: https://review.opendev.org/#/q/topic:freeze_job . What type of meta information are you interested in? | 16:03 |
| tristanC | panda: here is an example freeze_job call and response: https://review.opendev.org/#/c/607078/31/tests/unit/test_web.py | 16:04 |
| *** jamesmcarthur has quit IRC | 16:11 | |
| *** jpena is now known as jpena|off | 16:11 | |
| *** mattw4 has quit IRC | 16:14 | |
| *** mattw4 has joined #zuul | 16:14 | |
| *** igordc has joined #zuul | 16:18 | |
| corvus | Shrews: is your autohold stuff ready to go? | 16:19 |
| Shrews | corvus: well, it's all in merge conflict now, but it *was* ready at one point | 16:20 |
| tobiash | oops | 16:21 |
| corvus | Shrews: the web auth stuff is merged now, so i think you've got priority there now; if you want to rebase i think we can get it in next | 16:23 |
| Shrews | corvus: Ok. I'm about to lunch and go to PT. Will get to that this afternoon. | 16:24 |
| corvus | Shrews: w00t, thx | 16:24 |
| *** mattw4 has quit IRC | 16:28 | |
| mordred | jeliu_: I'm about to go off to the dentist - but when I get back I can totally attempt to help with the percona stuff | 16:28 |
| *** hashar has quit IRC | 16:30 | |
| panda | tristanC: how come you always have only good answers for me ? | 16:33 |
| jeliu_ | mordred: Sounds good thank you! | 16:33 |
| panda | tristanC: that's exactly what I was looking for ... even if the state is in progress | 16:33 |
| noorul | Is bitbucket code in good shape to get merged? | 16:34 |
| panda | tristanC: we are probably going to push metatest more, and sometimes we have to test that our job is receiveing the correct setup | 16:34 |
| corvus | panda: what is metatest? | 16:38 |
| panda | corvus: the test to test all tests | 16:38 |
| corvus | noorul: last i looked there were some review comments that needed to be addressed | 16:38 |
| noorul | ofosos: hi | 16:39 |
| corvus | panda: is there any more information you can provide about that? | 16:40 |
| panda | corvus: similar to what I said before, for example. As we have long chains of parents, and we sometimes reparent, I want to run a job to check that a set of the jobs I defined is actually getting a specific set of variables | 16:40 |
| panda | corvus: or that a specfic pre play is included in the job. | 16:41 |
| corvus | panda: ah, so iiuc, essentially external job configuration validation | 16:41 |
| panda | corvus: yes | 16:41 |
| clarkb | does anyone know if the change object here https://opendev.org/openstack/project-config/src/branch/master/zuul.d/pipelines.yaml#L252 which comes from QueueItem.change # a ref will have a change.ref attribute when run in a timer triggered pipeline? Maybe if the job sets a branch override that ref attribute will be set? | 16:42 |
| panda | corvus: zuul is usually able to comment when a change is inconsistent, a job definition is missing, etc ... but then even if it becoms consistent it may not do what it's supposed to do. That's why I was looking for this type of information. | 16:45 |
| corvus | clarkb: yes, should always be there. will probably be refs/heads/branchname for a timer trigger | 16:47 |
| clarkb | cool I think that will address mark's comment about the stable smtp reporter lacking branch info | 16:49 |
| corvus | clarkb: .branch is not one of the attributes it has, but i bet we could add it if desired | 16:51 |
| fungi | ideally the reports could be updated to state what the branch name (or any similar parameter) is in the template, rather than relying on it being part of the url as a side effect of how logs are organized, right? | 16:51 |
| corvus | (ie, without the refs/heads) | 16:51 |
| corvus | fungi: i think that's what clarkb is working on | 16:51 |
| corvus | fungi: it would be in the subject | 16:51 |
| fungi | that's what it seemed like, yeah. cool | 16:52 |
| ofosos | noorul: hi | 16:54 |
| clarkb | corvus: I think .ref is probably close enough | 16:54 |
| clarkb | corvus: and least likely to break | 16:54 |
| noorul | ofosos: Did you get a chance to look into dependency support? | 16:54 |
| ofosos | No not yet, I have a patch for the broken pull url though | 16:55 |
| noorul | ofosos: oh, cool | 16:55 |
| *** jamesmcarthur has joined #zuul | 16:56 | |
| ofosos | I still have a stash of changes here, in principle it's just grepping through the PR descriptions. We have them in memory anyways | 16:57 |
| ofosos | Right now we're running some deployments, but I can push the change tomorrow | 16:58 |
| noorul | ofosos: I think we should work towards getting this merged | 16:58 |
| * corvus would love to have it merged :) | 16:58 | |
| ofosos | I'll so the dependencies on Wednesday | 16:58 |
| ofosos | I think that's the missing part to merge it | 16:58 |
| noorul | Hmm | 16:59 |
| ofosos | noorul: does the parallel Gating work with the new patch? | 16:59 |
| noorul | ofosos: I am in the process getting docker container | 16:59 |
| ofosos | Cool | 17:00 |
| ofosos | I'm still not certain why there is a test failure, I think I'll look into that tomorrow | 17:00 |
| noorul | It is pep8 failure | 17:01 |
| ofosos | No there's also a failure in the fictional tests | 17:01 |
| noorul | I forgot how gerrit works | 17:02 |
| noorul | https://review.opendev.org/#/c/677093/7 | 17:02 |
| noorul | That shows all the tests passed | 17:02 |
| noorul | except pep8 | 17:02 |
| ofosos | noorul: fuuu... that was the last change set I pushed before going to sleep :D | 17:03 |
| fungi | there was a functional test job timeout on 660683 | 17:04 |
| fungi | but that's possibly unrelated to the change, especially since it didn't manifest in the changes depending on 660683 | 17:04 |
| ofosos | great, I'll fix the pep8 and push the pull url right away. | 17:04 |
| ofosos | There was a problem, basically the branch name "foobar" was hardcoded in place of "master" in the tests. And in the process I cut out some useless code | 17:04 |
| fungi | oh, and 658835 was failing the functional job for some reason | 17:06 |
| fungi | 657837 timed out its last functional test job run too | 17:07 |
| openstackgerrit | Mark Meyer proposed zuul/zuul master: Change branch variable in PR https://review.opendev.org/677093 | 17:09 |
| openstackgerrit | Mark Meyer proposed zuul/zuul master: Change PR url https://review.opendev.org/677257 | 17:09 |
| openstackgerrit | James E. Blair proposed zuul/zuul master: Render the logfile under the manifest https://review.opendev.org/676843 | 17:13 |
| openstackgerrit | Mark Meyer proposed zuul/zuul master: Change PR url to point to the PR not the Repo https://review.opendev.org/677257 | 17:13 |
| openstackgerrit | Mark Meyer proposed zuul/zuul master: Change branch variable in PR https://review.opendev.org/677093 | 17:13 |
| noorul | At last done with pulling changes using git review -x | 17:14 |
| ofosos | fungI: problem with the change or infra? | 17:15 |
| *** jamesmcarthur has quit IRC | 17:15 | |
| fungi | ofosos: or with the job itself, i haven't looked at the log for the one with the failure | 17:17 |
| *** jamesmcarthur has joined #zuul | 17:17 | |
| fungi | i rechecked the ones with timeouts because it looked like previous runs had completed faster | 17:17 |
| * ofosos reminds himself to always look at the logs :( | 17:17 | |
| fungi | i'll take a quick look at them myself | 17:17 |
| fungi | good opportunity to test out opendev's shiny new log views | 17:18 |
| fungi | oh, right, that one was from when zuul was still posting incorrect log urls | 17:20 |
| *** jamesmcarthur has quit IRC | 17:22 | |
| fungi | er, nope, it's just really, really old | 17:22 |
| noorul | Maybe docker-compose should also include ara? | 17:22 |
| fungi | 2019-06-20 old | 17:22 |
| fungi | i'll recheck it for new results | 17:22 |
| clarkb | noorul: I thought it would because ara is listed as a dep for ansible | 17:30 |
| clarkb | well not directly but in zuul's ansible deps list | 17:30 |
| noorul | INFO:zuul.managed_ansible:Installing ansible 2.8, requirements: ['ansible>=2.8,<2.9', 'ara>=0.16.5,<1.0.0', 'openstacksdk', 'openshift', 'jmespath'], extra packages: [] | 17:31 |
| noorul | clarkb: Is that good enough? | 17:33 |
| clarkb | noorul: I think so, but I'm not 100% sure | 17:33 |
| clarkb | it is possible that we don't try to run ara out of that venv actually | 17:33 |
| *** mattw4 has joined #zuul | 17:33 | |
| clarkb | in which case maybe that won't work | 17:33 |
| tobiash | I think pabelanger had issues with that after the multi-ansible support, but afaik those have been sorted out | 17:35 |
| tobiash | pabelanger: is that correct? | 17:35 |
| pabelanger | you need to make sure ara is bindmount in to executor | 17:35 |
| clarkb | pabelanger: so we are not using ara from the venv then? | 17:35 |
| corvus | ara needs to be installed in 2 places -- inside the venv for the callback, and in the executor's environment for static generation | 17:35 |
| tobiash | it is automatically bindmounted as it's part of the ansible venv | 17:35 |
| clarkb | corvus: gotcha | 17:35 |
| tobiash | ah | 17:35 |
| tobiash | that was it | 17:35 |
| pabelanger | yah, we set ara_report_executable in base job | 17:36 |
| pabelanger | https://github.com/ansible/project-config/blob/master/zuul.d/jobs.yaml#L37 | 17:36 |
| pabelanger | because, likely not installed in right path, if in virtuelenv | 17:37 |
| clarkb | pabelanger: given that noorul could set it similarly in the base job and have it work without an additional ara install? | 17:38 |
| pabelanger | clarkb: yup | 17:38 |
| noorul | I have the configuration at the bottom | 17:38 |
| noorul | https://etherpad.openstack.org/p/rWL36RmF6W | 17:38 |
| pabelanger | ara-report role should fail if ara isn't found | 17:38 |
| noorul | It fails with ara not found | 17:38 |
| pabelanger | yah, so you need to set ara_report_exectable | 17:39 |
| pabelanger | in your base job | 17:39 |
| noorul | but to which ansible version? | 17:39 |
| pabelanger | we dynamically select it, based on version of ansible running | 17:40 |
| noorul | so, I just have to put this ara_report_executable: "/opt/venv/zuul-ansible/{{ ansible_version.full }}/bin/ara" | 17:40 |
| pabelanger | well, /opt/venv/zuul-ansible is likely different | 17:40 |
| pabelanger | that is the location of where ansible venv is | 17:40 |
| clarkb | ya it should be /usr/lib/zuul/ansible on the docker images I think | 17:41 |
| clarkb | noorul: you can docker exec an ls on that dir to confirm in your running container | 17:42 |
| * noorul is still waiting for container build to finish | 17:42 | |
| clarkb | /usr/lib/zuul/ansible/{{ ansible_version.full }}/bin/ara | 17:42 |
| pabelanger | ansible_version.full will give 2.8.4 for example | 17:42 |
| clarkb | oh then not .full | 17:43 |
| clarkb | we just want 2.8 | 17:43 |
| pabelanger | then I think you want {{ ansible_version.major }}.{{ ansible_version.minor }} | 17:44 |
| pabelanger | IIRC | 17:44 |
| noorul | executor_1 | 2019-08-19 17:46:15,627 DEBUG zuul.managed_ansible: Running pip: /var/lib/zuul/ansible-bin/2.8/bin/pip install ansible>=2.8,<2.9 ara>=0.16.5,<1.0.0 openstacksdk openshift jmespath | 17:46 |
| clarkb | hrm we must not run the command to install zuul's ansible venvs in the docker image build and instead do it at runtime? | 17:47 |
| clarkb | tobiash: ^ iirc the whole point of that split in behavior was to address container needs. Should we change that? | 17:47 |
| noorul | This is at runtime, when I ran docker-compose up | 17:49 |
| noorul | When I push config to changes to project-config master branch, will Zuul automatically get re-configured? | 17:49 |
| tobiash | clarkb: actually I'd favor installing it during image build time, that makes the container stateless | 17:50 |
| pabelanger | clarkb: also need to allow it for zuul.conf, so zuul bindmounts it into bwrap | 17:50 |
| tobiash | what could be improved though would be to install the ara executable into the PATH | 17:50 |
| *** bhavikdbavishi1 has joined #zuul | 17:52 | |
| pabelanger | https://github.com/ansible-network/windmill-config/blob/master/zuul/zuul.conf.j2#L43 is what we do, to for both ro_paths | 17:52 |
| pabelanger | I think we could drop untrusted, but haven't tested | 17:52 |
| noorul | ofosos: http://paste.openstack.org/show/760026/ | 17:53 |
| *** bhavikdbavishi has quit IRC | 17:53 | |
| *** bhavikdbavishi1 is now known as bhavikdbavishi | 17:53 | |
| *** jamesmcarthur has joined #zuul | 17:54 | |
| clarkb | pabelanger: I think that the regular ansible venv paths are already trusted otherwise you wouldn't be able to run ansible out of them | 17:54 |
| clarkb | tobiash: in that case maybe we could have the manage script install to that location by default? | 17:55 |
| clarkb | tobiash: seems there is far less benefit to containers if we are going to installthere anyway on startup | 17:55 |
| pabelanger | clarkb: IIRC, we do that because our /var/lib/ansible directly is a symlink to /opt, because we have versioned venvs for each ansible release. Over 2.8 and pip update | 17:55 |
| pabelanger | so we need both in executor | 17:56 |
| clarkb | pabelanger: ya so you may be a special case but in general you shouldn't need to do that | 17:56 |
| pabelanger | agree | 17:56 |
| *** jamesmcarthur has quit IRC | 17:59 | |
| *** jamesmcarthur has joined #zuul | 18:00 | |
| noorul | Any idea why I am getting http://paste.openstack.org/show/760026/ | 18:00 |
| *** jamesmcarthur has quit IRC | 18:05 | |
| tobiash | noorul: do you have a running executor? | 18:06 |
| ofosos | noorul: my deployments finished, I'm just spinning up a new deployment | 18:06 |
| ofosos | (a new deployment for zuul) | 18:06 |
| tobiash | this typically means that the merger/executor didn't exist or couldn't access the repo | 18:06 |
| fungi | i guess that could happen if the scheduler is started first? | 18:08 |
| noorul | http://paste.openstack.org/show/760027/ | 18:09 |
| fungi | looks like netifaces was installed from an sdist rather than a wheel, so wanted to build c extensions but "unable to execute 'gcc': No such file or directory" | 18:09 |
| noorul | Looks like an issue on master? | 18:10 |
| fungi | it's an issue with netifaces wheel builds | 18:11 |
| fungi | https://pypi.org/project/netifaces/#files | 18:11 |
| fungi | 0.10.9 has no provided manylinux1 build for cp37 | 18:12 |
| fungi | if you were using python 3.6 it would probably not hit that | 18:12 |
| fungi | i'm guessing we test against <3.7 | 18:13 |
| fungi | https://github.com/al45tair/netifaces/issues/33 | 18:13 |
| noorul | after restarting docker-compose, it looks ok | 18:13 |
| fungi | requested 8 months ago, no response from the netifaces maintainer | 18:13 |
| noorul | Now I am getting http://paste.openstack.org/show/760029/ | 18:14 |
| fungi | i wonder if netifaces is unmaintained lately | 18:14 |
| fungi | yeah, last commit on master was 2019-01-02 | 18:15 |
| corvus | wait let's back up | 18:15 |
| corvus | http://paste.openstack.org/show/760027/ looks like it's a docker container of a zuul-executor which is trying to.... build netifaces to install ansible? i can't quite tell, there isn't enough history in that logfile | 18:16 |
| corvus | but it's not supposed to do that. the docker images are supposed to have ansible venvs pre-installed | 18:16 |
| *** michael-beaver has joined #zuul | 18:17 | |
| openstackgerrit | Mark Meyer proposed zuul/zuul master: Change PR url to point to the PR not the Repo https://review.opendev.org/677257 | 18:18 |
| openstackgerrit | Mark Meyer proposed zuul/zuul master: Change branch variable in PR https://review.opendev.org/677093 | 18:18 |
| ofosos | all the typos, url is fixed nwo | 18:18 |
| corvus | noorul: so if that's the case, you should try to figure out how you got an executor images without ansible pre-installed | 18:18 |
| corvus | this is what an executor logfile is supposed to look like on startup: https://zuul.opendev.org/t/zuul/build/ffd54d55a14d4c9381225bc8ba1b1bfd/log/container_logs/executor.log | 18:19 |
| noorul | corvus: On restart it worked | 18:21 |
| *** noorul has quit IRC | 18:21 | |
| corvus | that's not reassuring :) | 18:21 |
| *** jamesmcarthur has joined #zuul | 18:23 | |
| corvus | here's a more polished/final version of the 'logfiles under manifest' change if anyone wants to try it out for ui feedback: https://storage.gra1.cloud.ovh.net/v1/AUTH_dcaab5e32b234d56b626f72581e3644c/logs_43/676843/4/check/zuul-build-dashboard/6a10611/npm/html/ | 18:32 |
| ofosos | Btw. what happens when I restart scheduler and executor but keep the build nodes running? Will the builds complete? | 18:33 |
| tobiash | corvus: awesome :) | 18:34 |
| fungi | ofosos: they will be lost as zuul keeps all its pipeline contents and queues in memory | 18:34 |
| corvus | ofosos: no, a scheduler restart terminates all builds and loses the entire state of the system. an executor restart will terminate the builds running there, but they may be restarted by the scheduler on another executor (or when the executor returns) | 18:35 |
| *** SpamapS has quit IRC | 18:35 | |
| fungi | the future plan to move that state information into zookeeper (primarily for purposes of being able to have high availability for the scheduler service) could in theory make it possible to restart services without losing builds | 18:36 |
| *** saneax has quit IRC | 18:47 | |
| *** sgw has joined #zuul | 18:51 | |
| ofosos | I think we have a problem with zuul-ci.org | 18:53 |
| ofosos | My cat jobs keep waiting for it. | 18:54 |
| ofosos | Nah, it's something different. I forgot to bake in the key to the image again :( | 18:55 |
| *** bhavikdbavishi has quit IRC | 19:01 | |
| *** SpamapS has joined #zuul | 19:01 | |
| corvus | ofosos: with support for sshkey, you should be able to bind-mount it in | 19:05 |
| *** jamesmcarthur has quit IRC | 19:08 | |
| ofosos | Hmm, I'm getting: 2019-08-19 19:09:59,160 DEBUG zuul.Merger: Unable to merge {'project': 'IAC/puppetmaster-bootstrap', 'connection': 'bitbucket', 'merge_mode': 2, 'ref': 'refs/heads/sys-7086-deploy', 'branch': 'SYS-6627', 'buildset_uuid': '62e1d2c9ab7445f2a29995c08edf770f', 'number': 39, 'patchset': '0d642695bed612b49ca543bb6e74bf0e64f4e9bf', 'oldrev': None, 'newrev': None} | 19:11 |
| ofosos | With the new patch applied | 19:12 |
| ofosos | git.exc.GitCommandError: Cmd('git') failed due to: exit code, Please tell me who you are. | 19:12 |
| ofosos | Branch and ref seem to be ok | 19:13 |
| ofosos | Welp | 19:14 |
| ofosos | 'ref' here is the branch I want to merge into 'branch' | 19:15 |
| ofosos | But why does it fail with git complaining about missing user config? | 19:15 |
| fungi | seems it wants to run `git merge ...` and needs to have some (arbitrary) user info configured to attribute the merge commit | 19:18 |
| *** sshnaidm|bbl is now known as sshnaidm | 19:19 | |
| ofosos | Ok, I put this in the zuul.conf, let's see what happens | 19:20 |
| fungi | i'm going to guess the new patch exposed that because previously the merges were all no-op or fast-forward | 19:20 |
| fungi | due to the problem you just fixed | 19:20 |
| ofosos | fungi: fingers crossed :) | 19:20 |
| openstackgerrit | Mark Meyer proposed zuul/zuul master: Change PR url to point to the PR not the Repo https://review.opendev.org/677257 | 19:28 |
| openstackgerrit | Mark Meyer proposed zuul/zuul master: Change branch variable in PR https://review.opendev.org/677093 | 19:28 |
| ofosos | Dang, I hate it when I can't change a simple string | 19:29 |
| openstackgerrit | David Shrewsbury proposed zuul/zuul master: Store autohold requests in zookeeper https://review.opendev.org/661114 | 19:31 |
| ofosos | fungi: looks good | 19:31 |
| openstackgerrit | Clark Boylan proposed zuul/zuul master: Include ref info in smtp reporter subjects https://review.opendev.org/677285 | 19:33 |
| clarkb | I havne't confirmed that the override to the opendev pipelines is working as expected but figured I should go ahead and get ^ up | 19:34 |
| openstackgerrit | David Shrewsbury proposed zuul/zuul master: Add caching of autohold requests https://review.opendev.org/663412 | 19:35 |
| openstackgerrit | David Shrewsbury proposed zuul/zuul master: Add autohold-info CLI command https://review.opendev.org/662487 | 19:35 |
| openstackgerrit | David Shrewsbury proposed zuul/zuul master: Record held node IDs with autohold request https://review.opendev.org/662498 | 19:35 |
| openstackgerrit | David Shrewsbury proposed zuul/zuul master: Auto-delete expired autohold requests https://review.opendev.org/663762 | 19:35 |
| openstackgerrit | David Shrewsbury proposed zuul/zuul master: Mark nodes as USED when deleting autohold https://review.opendev.org/664060 | 19:35 |
| *** saneax has joined #zuul | 19:36 | |
| Shrews | corvus: that tree is now rebased on master, but I need some input from mhu on what to do with the two new commands and how to integrate them with his recent admin changes | 19:36 |
| *** sanjayu_ has joined #zuul | 19:53 | |
| *** saneax has quit IRC | 19:56 | |
| *** sanjayu_ has quit IRC | 19:57 | |
| *** dkehn_ has quit IRC | 20:07 | |
| *** jamesmcarthur has joined #zuul | 20:28 | |
| jeliu_ | corvus, mordred: http://paste.openstack.org/show/760036/ !!! | 20:30 |
| corvus | jeliu_: \o/ | 20:32 |
| *** jamesmcarthur has quit IRC | 20:42 | |
| *** jamesmcarthur has joined #zuul | 20:50 | |
| *** mattw4 has quit IRC | 21:05 | |
| *** mattw4 has joined #zuul | 21:06 | |
| *** jamesmcarthur has quit IRC | 21:08 | |
| *** jamesmcarthur has joined #zuul | 21:10 | |
| *** jamesmcarthur has quit IRC | 21:10 | |
| *** jamesmcarthur has joined #zuul | 21:17 | |
| *** jamesmcarthur has quit IRC | 21:22 | |
| *** mattw4 has quit IRC | 21:23 | |
| *** mattw4 has joined #zuul | 21:24 | |
| *** sshnaidm is now known as sshnaidm|afk | 21:24 | |
| *** jamesmcarthur has joined #zuul | 21:28 | |
| *** jamesmcarthur has quit IRC | 21:29 | |
| *** jamesmcarthur has joined #zuul | 21:44 | |
| *** jamesmcarthur has quit IRC | 21:45 | |
| openstackgerrit | James E. Blair proposed zuul/zuul master: WIP: remove displayedFile from state https://review.opendev.org/677307 | 21:56 |
| *** jamesmcarthur has joined #zuul | 21:58 | |
| openstackgerrit | James E. Blair proposed zuul/zuul master: Add Tristan to Zuul Maintainers https://review.opendev.org/677308 | 22:00 |
| *** jamesmcarthur has quit IRC | 22:03 | |
| *** jeliu_ has quit IRC | 22:09 | |
| *** jeliu_ has joined #zuul | 22:10 | |
| *** jamesmcarthur has joined #zuul | 22:43 | |
| openstackgerrit | Jeff Liu proposed zuul/zuul-operator master: Add PerconaXDB Cluster to Zuul-Operator https://review.opendev.org/677315 | 22:49 |
| *** rlandy|ruck is now known as rlandy|ruck|bbl | 22:52 | |
| *** jeliu_ has quit IRC | 22:54 | |
| *** EmilienM|pto has quit IRC | 22:54 | |
| *** EmilienM has joined #zuul | 22:56 | |
| *** jamesmcarthur has quit IRC | 23:33 | |
| *** jamesmcarthur has joined #zuul | 23:56 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!