| *** jamesmcarthur has quit IRC | 00:20 | |
| *** jamesmcarthur has joined #zuul | 00:33 | |
| *** rlandy|bbl is now known as rlandy | 01:45 | |
| *** CrayZee has quit IRC | 01:52 | |
| *** smyers has quit IRC | 01:52 | |
| *** mgoddard has quit IRC | 01:52 | |
| *** decimuscorvinus has quit IRC | 01:52 | |
| *** tristanC has quit IRC | 01:52 | |
| *** andreaf has quit IRC | 01:52 | |
| *** corvus has quit IRC | 01:53 | |
| *** CrayZee has joined #zuul | 01:54 | |
| *** smyers has joined #zuul | 01:54 | |
| *** mgoddard has joined #zuul | 01:54 | |
| *** decimuscorvinus has joined #zuul | 01:54 | |
| *** tristanC has joined #zuul | 01:54 | |
| *** andreaf has joined #zuul | 01:54 | |
| *** corvus has joined #zuul | 01:54 | |
| openstackgerrit | Ian Wienand proposed zuul/nodepool master: Update to Python 3.9 https://review.opendev.org/c/zuul/nodepool/+/789655 | 02:05 |
|---|---|---|
| openstackgerrit | Ian Wienand proposed zuul/nodepool master: Update to Python 3.9 https://review.opendev.org/c/zuul/nodepool/+/789655 | 02:08 |
| openstackgerrit | Ian Wienand proposed zuul/zuul master: Update to Python 3.9 https://review.opendev.org/c/zuul/zuul/+/789654 | 02:10 |
| *** josefwells has quit IRC | 02:24 | |
| openstackgerrit | Ian Wienand proposed zuul/zuul master: Run jobs automatically on playbook changes https://review.opendev.org/c/zuul/zuul/+/755988 | 02:30 |
| openstackgerrit | Ian Wienand proposed zuul/zuul master: Add always-files option to jobs https://review.opendev.org/c/zuul/zuul/+/747614 | 02:31 |
| *** evrardjp has quit IRC | 02:33 | |
| *** evrardjp has joined #zuul | 02:33 | |
| openstackgerrit | James E. Blair proposed zuul/zuul master: Support serializing encrypted secret objects https://review.opendev.org/c/zuul/zuul/+/789985 | 02:42 |
| openstackgerrit | James E. Blair proposed zuul/zuul master: Decrypt secrets on the executors https://review.opendev.org/c/zuul/zuul/+/789986 | 02:42 |
| ianw | corvus: i'd like to make sure you're ok with https://review.opendev.org/c/zuul/nodepool/+/756643 to add podman into the nodepool containers. this will be necessary for using the containerfile element which i can beat into final shape (it works, in dib testing) | 02:44 |
| corvus | ianw: that change is wonderful, thank you for picking that up. containerfile will be a dream come true. :) | 02:54 |
| corvus | ianw: +2 with comment, feel free to +w at will | 02:54 |
| corvus | swest: it may be worth your time to take a quick look at 789985 and 756643 just for the general idea -- that gives us an option to use yaml serialization for some or all of the objects we move into ZK | 03:00 |
| ianw | corvus: thanks will review after lunch :) | 03:02 |
| *** openstackgerrit has quit IRC | 03:31 | |
| *** jamesmcarthur has quit IRC | 03:45 | |
| *** jamesmcarthur has joined #zuul | 03:46 | |
| *** jamesmcarthur has quit IRC | 03:51 | |
| *** ykarel has joined #zuul | 04:10 | |
| *** jamesmcarthur has joined #zuul | 04:16 | |
| *** jfoufas1 has joined #zuul | 04:31 | |
| *** jamesmcarthur has quit IRC | 04:49 | |
| *** jamesmcarthur has joined #zuul | 05:03 | |
| *** saneax has joined #zuul | 05:06 | |
| *** jamesmcarthur has quit IRC | 05:09 | |
| swest | corvus: I can't find 756643. did you mean 789986? | 05:20 |
| *** CrayZee has quit IRC | 05:20 | |
| *** snapiri has joined #zuul | 05:21 | |
| *** jamesmcarthur has joined #zuul | 05:21 | |
| *** jamesmcarthur has quit IRC | 05:28 | |
| *** jamesmcarthur has joined #zuul | 05:43 | |
| *** vishalmanchanda has joined #zuul | 05:57 | |
| *** jcapitao has joined #zuul | 06:42 | |
| *** reiterative has quit IRC | 06:49 | |
| *** reiterative has joined #zuul | 06:49 | |
| *** openstackgerrit has joined #zuul | 06:53 | |
| openstackgerrit | Merged zuul/zuul master: Fix resolveBranch working on outdated data https://review.opendev.org/c/zuul/zuul/+/789253 | 06:53 |
| openstackgerrit | Merged zuul/zuul master: Fix race in test for cancel of starting builds https://review.opendev.org/c/zuul/zuul/+/789551 | 06:53 |
| *** jpena|off is now known as jpena | 07:01 | |
| *** snapiri has quit IRC | 07:08 | |
| *** rpittau|afk is now known as rpittau | 07:14 | |
| *** nils has joined #zuul | 07:22 | |
| openstackgerrit | Ian Wienand proposed zuul/nodepool master: Update to Python 3.9 https://review.opendev.org/c/zuul/nodepool/+/789655 | 07:24 |
| openstackgerrit | Ian Wienand proposed zuul/nodepool master: bindep.txt: Update libffi for focal install https://review.opendev.org/c/zuul/nodepool/+/790004 | 07:24 |
| openstackgerrit | Merged zuul/zuul-jobs master: collect-container-logs: don't copy on failure https://review.opendev.org/c/zuul/zuul-jobs/+/787019 | 07:35 |
| *** jamesmcarthur has quit IRC | 07:43 | |
| *** tosky has joined #zuul | 07:47 | |
| *** snapiri has joined #zuul | 07:49 | |
| *** sam_wan has joined #zuul | 07:55 | |
| *** jamesmcarthur has joined #zuul | 07:59 | |
| openstackgerrit | Simon Westphahl proposed zuul/zuul master: Cache unparsed config files in Zookeeper https://review.opendev.org/c/zuul/zuul/+/771455 | 08:05 |
| *** ykarel is now known as ykarel|lunch | 08:05 | |
| *** jamesmcarthur has quit IRC | 09:02 | |
| *** holser has joined #zuul | 09:21 | |
| *** patrick34 has quit IRC | 09:32 | |
| openstackgerrit | Matthieu Huin proposed zuul/zuul-client master: Add change-status subcommand https://review.opendev.org/c/zuul/zuul-client/+/759838 | 10:01 |
| *** ykarel|lunch is now known as ykarel | 10:02 | |
| *** jamesmcarthur has joined #zuul | 10:04 | |
| openstackgerrit | Matthieu Huin proposed zuul/zuul master: Web UI: add Autoholds, Autohold page https://review.opendev.org/c/zuul/zuul/+/768199 | 10:04 |
| sshnaidm | what does happen when same pattern is in "files" and "irrelevant_files"? | 10:11 |
| sshnaidm | not really same, but something like "^setup.*$" and "setup.cfg" and setup.cfg file is changed | 10:12 |
| sshnaidm | who will win? | 10:12 |
| avass | sshnaidm: irrelevant files does it looks like | 10:18 |
| sshnaidm | avass, thanks | 10:19 |
| avass | sshnaidm: https://opendev.org/zuul/zuul/src/branch/master/zuul/model.py#L1753 | 10:19 |
| sshnaidm | yeah, see now.. | 10:19 |
| sshnaidm | and irrelevant files are inherited, right? Like all other parameters | 10:20 |
| avass | yeah but files and irrelevant-files get overriden if the child sets them | 10:22 |
| openstackgerrit | Simon Westphahl proposed zuul/zuul master: Add pipeline event queue lengths to status JSON https://review.opendev.org/c/zuul/zuul/+/790038 | 10:58 |
| openstackgerrit | Simon Westphahl proposed zuul/zuul master: Add metrics for pipeline event queues https://review.opendev.org/c/zuul/zuul/+/790039 | 10:58 |
| *** jcapitao is now known as jcapitao_lunch | 11:19 | |
| *** jpena is now known as jpena|lunch | 11:32 | |
| *** rlandy has joined #zuul | 11:46 | |
| openstackgerrit | Benjamin Schanzel proposed zuul/zuul-jobs master: Allow Specifying Log File Retention in s3 Uploader https://review.opendev.org/c/zuul/zuul-jobs/+/790050 | 12:00 |
| *** jcapitao_lunch is now known as jcapitao | 12:01 | |
| *** Shrews has joined #zuul | 12:03 | |
| openstackgerrit | Benjamin Schanzel proposed zuul/zuul-jobs master: Allow Specifying Log File Retention in s3 Uploader https://review.opendev.org/c/zuul/zuul-jobs/+/790050 | 12:05 |
| sshnaidm | avass, are "files:" also overriden if child job sets them? | 12:06 |
| *** jamesmcarthur has quit IRC | 12:06 | |
| *** jamesmcarthur has joined #zuul | 12:06 | |
| sshnaidm | because I can't understand then why this job is running: https://review.opendev.org/c/openstack/ansible-collections-openstack/+/789818/4/.zuul.yaml#322 | 12:07 |
| avass | sshnaidm: yeah they are | 12:07 |
| sshnaidm | the only changed file is .zuul.yaml there | 12:07 |
| avass | sshnaidm: but jobs also run their configuration updates, that also includes updates to parents in the same repo :) | 12:08 |
| *** hashar has joined #zuul | 12:08 | |
| avass | sshnaidm: that can be configured with: https://zuul-ci.org/docs/zuul/reference/job_def.html#attr-job.match-on-config-updates | 12:09 |
| sshnaidm | avass, that's interesting, thanks! | 12:10 |
| avass | np :) | 12:11 |
| openstackgerrit | Simon Westphahl proposed zuul/zuul master: Add metrics for pipeline event queues https://review.opendev.org/c/zuul/zuul/+/790039 | 12:19 |
| *** josefwells has joined #zuul | 12:22 | |
| openstackgerrit | Felix Edel proposed zuul/zuul master: Lock/unlock nodes on executor server https://review.opendev.org/c/zuul/zuul/+/774610 | 12:23 |
| *** jpena|lunch is now known as jpena | 12:26 | |
| *** ykarel_ has joined #zuul | 12:30 | |
| *** ykarel has quit IRC | 12:32 | |
| openstackgerrit | Benjamin Schanzel proposed zuul/zuul-jobs master: Allow Specifying Log File Retention in s3 Uploader https://review.opendev.org/c/zuul/zuul-jobs/+/790050 | 12:34 |
| *** patrick34 has joined #zuul | 12:37 | |
| *** sam_wan has quit IRC | 12:59 | |
| *** sduthil has quit IRC | 12:59 | |
| *** sduthil has joined #zuul | 13:00 | |
| *** ykarel_ is now known as ykarel | 13:02 | |
| avass | bschanzel: ++ | 13:02 |
| *** dustinc has joined #zuul | 13:07 | |
| avass | corvus: I think we had a misunderstanding with the public bucket config for s3 uploads. swift etc makes the bucket private/public while the s3 public config sets the acl on the logs uploaded (which overrides the bucket configuration). | 13:07 |
| patrick34 | Morning ! | 13:21 |
| openstackgerrit | Simon Westphahl proposed zuul/zuul master: Add metrics for pipeline event queues https://review.opendev.org/c/zuul/zuul/+/790039 | 13:22 |
| patrick34 | Following yesterday's discussion about this https://storyboard.openstack.org/#!/story/2008819 It's mission-critical for me. So let me know if I can help in any way to make it work :) | 13:22 |
| corvus | swest: yes, that's what i meant, sorry | 13:39 |
| *** Shrews_ has joined #zuul | 13:53 | |
| *** Shrews has quit IRC | 13:55 | |
| tobiash | patrick34: the git driver has no real change support, did you try enqueue-ref/dequeue-ref? | 14:04 |
| tobiash | enqueue/dequeue act on changes (aka gerrit, github, etc) | 14:05 |
| fungi | tobiash: there is no dequeue-ref | 14:05 |
| patrick34 | There's no dequeue-ref in the cli of zuul 3.19.1 | 14:05 |
| patrick34 | Enqueue ref I need to try it I'll let you know if it works | 14:06 |
| tobiash | ah yes, dequeue can do both, use the --ref instead of --change | 14:06 |
| fungi | tobiash: https://storyboard.openstack.org/#!/story/2008819 shows the traceback from trying to use --ref with dequeue and the git driver | 14:06 |
| mhu | IIRC the problem was that a merge was impossible | 14:07 |
| patrick34 | tobiash exactmy, but dequeue --ref does not work as I expected. Also hum yeah I will add more details to my storyboard | 14:07 |
| mhu | but it was for enqueue | 14:08 |
| patrick34 | just tried with enqueue-ref, same result https://gist.github.com/plaurin84/db92f3aabbc21ed75b463fef49bb4011 | 14:18 |
| patrick34 | oh wait it works If I use the actual ref id like "b47e0357d35ca6e86a2d217d99ddee3ae8c74efe" | 14:19 |
| *** patrick34 has quit IRC | 14:22 | |
| *** patrick34 has joined #zuul | 14:23 | |
| patrick34 | I basically expected enqueue-ref to work with either refs/heads/master or the commit id, but only commit id works. | 14:24 |
| patrick34 | and I expect dequeue --ref to work the same, but does not work at all for me. I will update the story with this information soon | 14:24 |
| openstackgerrit | Matthieu Huin proposed zuul/zuul master: Web UI: add Autoholds, Autohold page https://review.opendev.org/c/zuul/zuul/+/768199 | 14:24 |
| mhu | patrick34, this might require a patch to the doc to explain when to use revs or refs | 14:25 |
| mhu | (I'll admit I'm not too clear on that) | 14:25 |
| *** ykarel_ has joined #zuul | 14:27 | |
| *** ykarel has quit IRC | 14:30 | |
| openstackgerrit | Merged zuul/nodepool master: Install podman in container for container-based builds https://review.opendev.org/c/zuul/nodepool/+/756643 | 14:34 |
| patrick34 | mhu let me know how I can help make this a reality :) So I don't get my dev team roll back into jenkins '=D | 14:40 |
| mhu | patrick34, if you're familiar with gerrit's workflow you're more than welcome to contribute to zuul-client's documentation | 14:41 |
| patrick34 | yes I've been managing my own gerrit server for years :) | 14:41 |
| patrick34 | Well before fixing the doc I guess we should fix the actual issue right? I might have issues fixing at that level tough... | 14:42 |
| patrick34 | I'm devops so I only speak yaml haha | 14:42 |
| mhu | oh my bad, I read a bit too fast - I assumed specifying the commit id was the fix | 14:43 |
| patrick34 | well what I need is both enqueue-ref with refs/heads/master style, and a working dequeue for git driver | 14:43 |
| mhu | IIRC zuul-client assumes a commit id of 0000...000 if not specified for ref-related actions. I guess it doesn't work in your specific case | 14:44 |
| patrick34 | so, dequeue is only built for the gerrit driver? | 14:45 |
| mhu | I'm not sure, but since gerrit is the "historical" review system for zuul, it might have more solid support | 14:48 |
| corvus | patrick34: how did the item get into the queue in the first place? | 14:48 |
| corvus | patrick34: did you 'zuul enqueue-ref' or was it a pipeline trigger? | 14:49 |
| patrick34 | corvus I use time trigger in pipelines.yaml | 14:49 |
| patrick34 | mhu agreed, I know zuul was built with gerrit in mind at first. | 14:51 |
| corvus | then i don't think there's a way to fix this without a change to zuul, since it's enqueing a branch head with a null oldrev and it can't dequeue that. one of those two things would need to change. | 14:52 |
| patrick34 | corvus okay. Would it be considered a valid 'git driver' bug? How can I help make you make this happen? :) | 14:53 |
| corvus | i'd definitely consider it either a bug or missing feature; we aim for driver parity. | 14:55 |
| *** vishalmanchanda has quit IRC | 14:56 | |
| *** hashar has quit IRC | 15:00 | |
| *** ykarel_ is now known as ykarel|away | 15:01 | |
| *** hashar has joined #zuul | 15:09 | |
| openstackgerrit | Matthieu Huin proposed zuul/zuul master: Example Docker compose: keycloak integration https://review.opendev.org/c/zuul/zuul/+/769943 | 15:11 |
| *** jfoufas1 has quit IRC | 15:20 | |
| *** ykarel|away has quit IRC | 15:25 | |
| openstackgerrit | Matthieu Huin proposed zuul/zuul master: Example Docker compose: keycloak integration https://review.opendev.org/c/zuul/zuul/+/769943 | 15:33 |
| *** rpittau is now known as rpittau|afk | 15:42 | |
| *** sshnaidm is now known as sshnaidm|afk | 15:44 | |
| *** jpena is now known as jpena|away | 16:07 | |
| *** mgoddard has quit IRC | 16:13 | |
| *** mgoddard has joined #zuul | 16:15 | |
| *** jcapitao has quit IRC | 16:22 | |
| *** nils has quit IRC | 16:23 | |
| *** jpena|away is now known as jpena|off | 16:47 | |
| *** hashar has quit IRC | 16:59 | |
| avass | mhu: do we need to support ubuntu 16 with quickstart at this point? It's on extended security maintanence now and I don't think zuul runs any tests on xenial. | 17:00 |
| avass | I suppose if we want make it easier for people running xenial to use quickstart we should keep it. | 17:00 |
| avass | corvus: ^ ? | 17:00 |
| avass | mhu: ( I saw that you bumped the docker-compose version) | 17:01 |
| *** patrick34 has quit IRC | 17:05 | |
| mordred | I don't think we should be encouraging anyone to start anything new on xenial at this point | 17:07 |
| clarkb | we also dropped python3.5 support which is what xenial has on it | 17:08 |
| avass | I suppose it makes sense to drop docker-compose 2 support if needed then. worst case it's easy enough to install a newer version with pip | 17:10 |
| fungi | technically opendev is still running its scheduler on ubuntu 16.04 lts | 17:15 |
| fungi | but in containers with newer python | 17:15 |
| fungi | (using the zuul container images) | 17:15 |
| fungi | though we expect to be upgrading the os "soon" | 17:16 |
| mordred | yah - and we're neither using distro docker-compose nor using the quickstart compose file - but yeah | 17:18 |
| *** jamesmcarthur has quit IRC | 17:22 | |
| *** jamesmcarthur has joined #zuul | 17:23 | |
| corvus | is there anything in docker-compose.yaml that *requires* > 2? | 17:26 |
| clarkb | for gerrit we did it to support better controls over shutdown process iirc | 17:27 |
| clarkb | signals and wait times iirc | 17:27 |
| *** jamesmcarthur has quit IRC | 17:27 | |
| clarkb | perhaps something similar got into the docker compose for gerrit in quickstart? | 17:27 |
| corvus | because that's just a minimum requirement. if we're not using any compose features > 2, we should just leave it as is. raising it only serves to make it harder for people to use the quickstart if they happen to be on a platform with an older docker-compose. | 17:27 |
| avass | corvus: I don't think so. I just saw the comments about xenial | 17:29 |
| avass | (and apparently mhu needed version 2.3) | 17:30 |
| corvus | i wonder why. | 17:30 |
| mordred | maybe it would be good to understand why | 17:30 |
| mordred | yeah | 17:30 |
| mhu | avass: I remember I needed to bump the version, but I can't remember why, I'll need to look into it | 17:33 |
| mhu | or just try to set up the compose with a lower version and see what happens | 17:33 |
| corvus | mhu: are you proposing that change as something for reviewers to use to evaluate the integration, or as a permanent part of the quickstart? | 17:34 |
| *** smyers_ has joined #zuul | 17:34 | |
| mhu | however patch https://review.opendev.org/c/zuul/zuul/+/769943 is more meant as a way for people to test the admin GUI changes, it doesn't necessarily need to be merged | 17:34 |
| corvus | ok just my question :) | 17:35 |
| corvus | then the docker-compose detail isn't that important.... | 17:35 |
| mhu | corvus, my intent was evaluation purposes, but we could also work more on it later on if we want to offer an all-in-one compose that includes auth | 17:36 |
| corvus | mhu: i feel like having an example like that might be helpful for users, but i worry about making the initial quickstart too big and complicated for users who really want a "quick" intro to zuul | 17:36 |
| mhu | corvus, yeah I understand completely | 17:36 |
| corvus | mhu: guillaumec has a stack where he's adding tutorials onto the quickstart, this might make sense as part of that | 17:36 |
| mhu | I'd be okay with this compose living somewhere else | 17:36 |
| corvus | mhu: basically "so you got a quickstart up and running, here's how to add auth with keycloak" | 17:36 |
| *** smyers has quit IRC | 17:37 | |
| *** smyers_ is now known as smyers | 17:37 | |
| corvus | mhu: that might be tricky depending on how much of gerrit needs to change for it, so an "alternate" compose may be necessary, but it's a thought | 17:37 |
| corvus | we were talking about something similar for a 'github' tutorial the other day too | 17:37 |
| avass | also installing docker-compose on xenial got me version 1.18 which is suppose to support up to compose file format 3.5 with docker engine 17.06.0+ | 17:38 |
| *** hamalq has joined #zuul | 17:38 | |
| avass | and with "apt-get install docker.io" i got docker version 18.09.7 | 17:38 |
| corvus | avass: wow, i guess they upgraded? | 17:38 |
| avass | corvus: yeah they must have | 17:39 |
| *** hamalq has quit IRC | 17:39 | |
| mhu | avass: on fedora, which I am using, podman didn't support compose until recently | 17:39 |
| *** hamalq has joined #zuul | 17:40 | |
| corvus | anyway, i don't want us to be limited; if we need to use something we can bump it. but there was a period where peopre were writing compose 3 files while 2 was the latest thing offered by distros and that was user unfriendly. so i guess i'd ask just keep in mind that we're aiming for keeping the barrier to entry low :) | 17:41 |
| avass | corvus: oh no, my fault. docker-compose version 1.8, docker engine 18. mixed up the numbers :) | 17:41 |
| *** vishalmanchanda has joined #zuul | 17:43 | |
| mhu | for now anyway, it's really for testing purposes. But since the doc dir is doc/source/exampleS we could have a separate "auth-all-in-one" compose for the advanced testers | 17:43 |
| mhu | That makes me think, I have this change for autoholds pages on the GUI, but it's part of the admin GUI patch chain: https://review.opendev.org/c/zuul/zuul/+/768199 - Should I make one without the admin parts so that it could be merged faster? ie just having the autoholds listing without the admin management | 17:52 |
| avass | mhu: autohold listings would be nice to have. :) | 17:57 |
| avass | but "zuul autohold-list" works for the moment | 17:58 |
| openstackgerrit | Ade Lee proposed zuul/zuul-jobs master: Add role to enable FIPS on a node https://review.opendev.org/c/zuul/zuul-jobs/+/788778 | 17:59 |
| *** jamesmcarthur has joined #zuul | 18:17 | |
| *** Shrews_ is now known as Shrews | 18:23 | |
| josefwells | Hey zuulers! I'm building some general roles that would be used by lots of my repositories. I see the recommendation not to put these in a "config-project", but how do I tell project_1/2/3 in its zuul.yaml/zuul.d* to use jobs defined in "job-project", my not "config-project" project that just hods roles/playbooks?? | 18:30 |
| avass | josefwells: you just use them, as long as the jobs are defined in the same tenant they're available to all projects :) | 18:31 |
| josefwells | avass, before I realized the recommendation, I just built the roles into my config-project, but when I list the job zuul complains.. mabye Role vs. Job? I get "Job make-test not defined" | 18:33 |
| josefwells | avass, maybe since config-project is privileged, I'm not allowed to reach in and run those? | 18:33 |
| fungi | jobs aren't roles, if that's the confusion | 18:33 |
| fungi | jobs are tied to playbooks in the same project where they're defined, though those playbooks can use roles from other projects as long as they indicate they're doing so | 18:34 |
| avass | josefwells: yeah, if you want to use roles you need to configure your jobs to use them from a different repo | 18:34 |
| josefwells | fungi, I'm certainly confused about the difference, practically roles allow default/replaceable variables, but beyond that it seems hazy | 18:34 |
| avass | josefwells: https://zuul-ci.org/docs/zuul/reference/job_def.html#attr-job.roles | 18:35 |
| fungi | also we define what "job" means here: https://zuul-ci.org/docs/zuul/reference/glossary.html#term-job | 18:35 |
| avass | ^ that should allow you to use roles from a different project in your playbooks. note that the roles directory need to be in the root of the repository | 18:36 |
| josefwells | jobs use roles, roles don't use jobs | 18:36 |
| avass | jobs use playbooks that use roles :) | 18:36 |
| fungi | jobs are zuul's unit of reuse, roles are ansible's unit of reuse. jobs run playbooks which can consist of plays containing tasks, or roles which encapsulate a set of tasks for reusability | 18:37 |
| *** jamesmcarthur has quit IRC | 18:37 | |
| josefwells | gotcha, I'm trying to avoid needing playbooks in a bunch of repositories that are just duplicates of eachother | 18:37 |
| *** jamesmcarthur has joined #zuul | 18:37 | |
| fungi | if you define a job in one project, you can add it to pipelines for other projects | 18:38 |
| josefwells | or really even a zuul.yaml that needs to list the components of a job. Ok, I see, let me give it a shot | 18:38 |
| fungi | whether or not the project containing it is a config project or an untrusted project | 18:38 |
| fungi | jobs in untrusted projects are flexible in that you get speculative execution of proposed changes to them. on the other hand, only jobs defined in config projects get privileges to perform certain trusted actions | 18:39 |
| josefwells | all jobs inherit from the base job, which at least in the guides does stuff like "prepare-workspace". So if you specify multiple jobs, and each of them inherit from the base job, does that mean that each job will go through its own individual pre/run/post as defined by its parent? | 18:41 |
| *** jamesmcarthur has quit IRC | 18:42 | |
| avass | if you specify multiple jobs in your check/gate etc pipelines yeah. They will land on different hosts and execute in parallel | 18:42 |
| josefwells | avass, ok, good stuff, thanks for the info | 18:44 |
| avass | you can also setup dependencies between two jobs if for example one job builds an artifact and another job tests it. | 18:45 |
| avass | but in that case there would need to be some kind of storage to pass the artifact between the jobs | 18:46 |
| fungi | note that pre-run and post-run playbooks are inherited, but run playbooks are overridden/replaced | 18:46 |
| josefwells | avass, I saw some of that, but my crazy workspace setup is going to cause some issues, might need to change around how I do stuff. for dependencies like that, I assume they fall under the same zuul.build? | 18:46 |
| avass | josefwells: same buildset but different builds | 18:46 |
| avass | (if that's what you mean) | 18:47 |
| josefwells | Hmm, not what I expected, but I can see that | 18:47 |
| corvus | a build is one execution of a job; a buildset is a collection of builds for the same change | 18:47 |
| josefwells | I rewrote the prepare-workspace-git for my world of NFS and LSF, so I have a unique disk area per-build.. but it should really be buildset | 18:48 |
| josefwells | The interesting bit being lots of this: {{ zuul_workspace_root }}/{{ zj_project.value.src_dir }}/{{ zuul.build }} | 18:51 |
| josefwells | but I think to collect multiple repo dependencies and runs, I need to change it to: | 18:51 |
| josefwells | {{ zuul_workspace_root }}/{{ zuul.buildset }}/{{ zj_project.value.src_dir }} | 18:51 |
| josefwells | but I can see that evolve as I understand more about how this works | 18:51 |
| corvus | josefwells: i gotta say at first glance that doesn't seem like it's going to match zuul's expectations. zuul prepares a workspace for a job, not a buildset. | 18:51 |
| corvus | josefwells: you may be using multiple jobs where you should be using just one. | 18:52 |
| josefwells | seems like zuul prepares a value.checkout, and the jobs do the rest | 18:52 |
| *** smyers has quit IRC | 18:53 | |
| *** patrick34 has joined #zuul | 19:07 | |
| *** patrick34 has quit IRC | 19:19 | |
| *** patrick34 has joined #zuul | 19:19 | |
| *** saneax has quit IRC | 19:30 | |
| openstackgerrit | Matthieu Huin proposed zuul/zuul master: Web UI: add Autoholds, Autohold page https://review.opendev.org/c/zuul/zuul/+/768199 | 19:32 |
| mhu | Hey zuul-maint, a few reasonanbly sized changes for zuul-client that'd be nice to merge before adding some subcommands: https://review.opendev.org/c/zuul/zuul-client/+/789007 https://review.opendev.org/c/zuul/zuul-client/+/788340 https://review.opendev.org/c/zuul/zuul-client/+/788832 | 19:36 |
| *** patrick34 has quit IRC | 19:42 | |
| *** smyers has joined #zuul | 20:04 | |
| avass | this could be a good reason to add type annotations in the future if it ever gets upstreamed: https://github.com/facebookincubator/cinder#static-python | 20:05 |
| fungi | could also be an interesting source of confusion if zuul started using openstack cinder (block storage) and facebook cinder (performance-tuned python) | 20:17 |
| tosky | running cinder on cinder | 20:18 |
| *** hashar has joined #zuul | 20:24 | |
| *** nils has joined #zuul | 20:28 | |
| *** hashar has quit IRC | 20:50 | |
| *** hashar has joined #zuul | 20:52 | |
| openstackgerrit | James E. Blair proposed zuul/zuul-operator master: Support imagePrefix and versions https://review.opendev.org/c/zuul/zuul-operator/+/785279 | 21:09 |
| openstackgerrit | James E. Blair proposed zuul/zuul-operator master: Support fingergw https://review.opendev.org/c/zuul/zuul-operator/+/785300 | 21:09 |
| openstackgerrit | James E. Blair proposed zuul/zuul-operator master: Add docs https://review.opendev.org/c/zuul/zuul-operator/+/785083 | 21:09 |
| openstackgerrit | James E. Blair proposed zuul/zuul-operator master: Support zuul-preview https://review.opendev.org/c/zuul/zuul-operator/+/785760 | 21:09 |
| openstackgerrit | James E. Blair proposed zuul/zuul-operator master: Add support for zuul-registry https://review.opendev.org/c/zuul/zuul-operator/+/785761 | 21:09 |
| openstackgerrit | James E. Blair proposed zuul/zuul-operator master: Remove extra 2 minute wait from tests https://review.opendev.org/c/zuul/zuul-operator/+/785762 | 21:09 |
| openstackgerrit | James E. Blair proposed zuul/zuul-operator master: Add allowUnsafeConfig database setting https://review.opendev.org/c/zuul/zuul-operator/+/785764 | 21:09 |
| openstackgerrit | James E. Blair proposed zuul/zuul-operator master: Pass through environment to scheduler, web and launcher https://review.opendev.org/c/zuul/zuul-operator/+/785988 | 21:09 |
| openstackgerrit | James E. Blair proposed zuul/zuul-operator master: Allow terminationGracePeriodSeconds to be configurable https://review.opendev.org/c/zuul/zuul-operator/+/785989 | 21:09 |
| openstackgerrit | James E. Blair proposed zuul/zuul-operator master: Flake8 cleanups https://review.opendev.org/c/zuul/zuul-operator/+/786349 | 21:09 |
| openstackgerrit | James E. Blair proposed zuul/zuul-operator master: Add keystore password support https://review.opendev.org/c/zuul/zuul-operator/+/790182 | 21:09 |
| corvus | the last change is new; the rest are rebased on it | 21:10 |
| corvus | (that's inserted at the point in the stack where it switches from being pinned to 4.1.0 to latest zuul) | 21:11 |
| *** vishalmanchanda has quit IRC | 21:12 | |
| openstackgerrit | James E. Blair proposed zuul/zuul master: Decrypt secrets on the executors https://review.opendev.org/c/zuul/zuul/+/789986 | 21:20 |
| avass | corvus: append empty dict? https://review.opendev.org/c/zuul/zuul/+/789986/2/tests/unit/test_connection.py | 21:29 |
| corvus | avass: maintains test compatibility with the old version; plus it does have a small bit of menaning: a playbook with no secrets, verses no playbooks. | 21:33 |
| corvus | (basically, it's a list of the secrets for each playbook, so, for example [{}, {}, {}] means 3 playbooks with 0 secrets; and [] means zero playbooks. | 21:34 |
| avass | ah, yeah it just looks weird | 21:35 |
| corvus | we used to get the {} directly from the gearman job serialization, but with that now encrypted, we have to get it from the executor's internal data structure, and that uses None if there are no secrets for a playbook (rightly so, i think :) | 21:35 |
| corvus | yep, no argument there :) | 21:36 |
| *** hashar has quit IRC | 22:22 | |
| *** rlandy is now known as rlandy|bbl | 22:24 | |
| *** nils has quit IRC | 22:41 | |
| *** tosky has quit IRC | 23:17 | |
| *** smyers_ has joined #zuul | 23:18 | |
| *** smyers- has joined #zuul | 23:20 | |
| *** smyers has quit IRC | 23:21 | |
| *** smyers- is now known as smyers | 23:21 | |
| *** smyers_ has quit IRC | 23:22 | |
| *** Shrews has quit IRC | 23:28 | |
| *** dustinc has quit IRC | 23:34 | |
| *** hamalq has quit IRC | 23:42 | |
| openstackgerrit | James E. Blair proposed zuul/zuul master: Support multiple semaphores https://review.opendev.org/c/zuul/zuul/+/790189 | 23:46 |
Generated by irclog2html.py 2.17.2 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!