| jeblair | jhesketh: i think that's mostly already there -- if a job definition appears in a branch of a project repo (this is not applicable to a config repo), it gets an implied branch matcher for that branch | 00:00 |
|---|---|---|
| jeblair | (this probably would have been a good thing to include in the email) | 00:00 |
| jeblair | * caveat: some of the branch-dependent logic isn't fully implemented yet (there are some todos in the configloader file) | 00:01 |
| jhesketh | ah, okay.. | 00:02 |
| jeblair | so, iow, if nova has a special gate job defined in their repo, the definition that appears in the 'master' branch of nova will run on master changes, and mitaka on stable/mitaka changes, etc. | 00:03 |
| jhesketh | jeblair: will that create a varient though? so where you have python27 defined in master with a branch matcher for a stable branch, and it defined again in the branch itself, will they be considered two jobs? | 00:04 |
| jhesketh | what I'm leaning towards, is that other than config-repos, we should probably lean away from using branch matches and instead define the jobs in their branches (in the openstack use case) | 00:04 |
| jeblair | jhesketh: yeah, it would be as if you had two definitions of python27 both with stable branch matchers. | 00:04 |
| jhesketh | okay | 00:05 |
| jeblair | jhesketh: and yes, i agree, i think that's the more natural way to use the feature | 00:05 |
| jhesketh | :-) | 00:05 |
| jhesketh | jeblair: so if you have python27 defined in master with no branch conditions, will that also be ran against the stable branch | 00:05 |
| jeblair | jhesketh: if it's in a project repo, i think it won't match; so only the stable branch version will run | 00:06 |
| jeblair | jhesketh: (that's probably not what we want to do with the actual python27 job though -- that probably wants to be defined in project-config, with explicit branch matchers) | 00:07 |
| jhesketh | okay, so the master branch matcher is also implied the same as a stable branch | 00:07 |
| jeblair | jhesketh: yep | 00:07 |
| jeblair | there is probably a hole here with, for example, devstack-gate, which i suggested we might want to make a project repo, but we also may want to leave it branchless, yet, we want repos with stable branches to be able to run jobs it defines... | 00:08 |
| jeblair | i bet we can work it out though. :) | 00:08 |
| clarkb | tempest is branchless too | 00:09 |
| jeblair | (maybe a fallback on master (that seems very zuulish); or maybe explicit branch:[]) | 00:09 |
| jhesketh | yeah, I suspect the flexibilty in config/project repos allows us to do some funky things | 00:09 |
| jeblair | yep. we are drawing more coastline on our map, but there still be dragons. :) | 00:10 |
| jhesketh | from a project's point of view, it's still tricky to see what jobs will run for which projects against which branches in which pipelines etc. because of the matchers | 00:11 |
| jeblair | yeah, and "well, the version that's right there in the tree you're looking at" is pretty intuitive | 00:12 |
| jhesketh | I wonder if defining the branches as part of the project-pipeline definition gains us simplicity there... however I'm not sure it works against cross-branch jobs or other complicated integration tests | 00:12 |
| jeblair | it's only weird for complex things like devstack | 00:12 |
| jhesketh | right | 00:13 |
| jhesketh | it gets pretty complex for any deployment projects too | 00:13 |
| jhesketh | but they need to test pretty complex scenarios, so that's kinda expected | 00:13 |
| jeblair | jhesketh: i think the project-pipeline also needs to be branch-specific (i believe that's one of the TODOs). that certainly will have an interaction here. | 00:13 |
| jeblair | (it is valid to include a project-pipeline snippet in a project-repo .zuul.yaml -- it needs to be because that would be the only way to run a job you define in a project repo) | 00:14 |
| jeblair | (it just doesn't do any implied branch matching yet) | 00:14 |
| jhesketh | jeblair: if we do project-pipeline-branch mapping, do we need branch matching? | 00:15 |
| jeblair | jhesketh: i think so, because i think we want to still be able to say "this job should run on trusty nodes when run on stable/foo" and not have to do that in the project-pipeline | 00:17 |
| openstackgerrit | James E. Blair proposed openstack-infra/zuul feature/zuulv3: Run zuul v3 launcher https://review.openstack.org/433347 | 00:18 |
| jeblair | it can run as a daemon! ^ | 00:18 |
| jhesketh | jeblair: right, so we need it to define things like nodes that applies to all stable branches | 00:21 |
| jhesketh | hmm okay | 00:21 |
| *** jamielennox is now known as jamielennox|away | 01:09 | |
| *** jamielennox|away is now known as jamielennox | 01:17 | |
| *** saneax is now known as saneax-_-|AFK | 01:35 | |
| *** tristanC_ is now known as tristanC | 04:56 | |
| *** saneax-_-|AFK is now known as saneax | 05:26 | |
| *** abregman has joined #zuul | 06:25 | |
| *** isaacb has joined #zuul | 07:28 | |
| *** isaacb has quit IRC | 07:50 | |
| *** isaacb has joined #zuul | 07:53 | |
| *** isaacb has quit IRC | 09:14 | |
| *** isaacb has joined #zuul | 09:24 | |
| *** hashar has joined #zuul | 09:54 | |
| *** isaacb has quit IRC | 10:14 | |
| *** isaacb has joined #zuul | 10:49 | |
| *** Cibo has quit IRC | 11:18 | |
| *** Cibo has joined #zuul | 11:30 | |
| *** hashar is now known as hasharLunch | 12:22 | |
| *** hasharLunch is now known as hashar | 13:06 | |
| *** jamielennox is now known as jamielennox|away | 13:21 | |
| *** Cibo_ has quit IRC | 13:36 | |
| *** isaacb has quit IRC | 14:01 | |
| *** isaacb has joined #zuul | 14:09 | |
| *** saneax is now known as saneax-_-|AFK | 14:28 | |
| *** TheJulia has quit IRC | 14:41 | |
| *** patrickeast has quit IRC | 14:42 | |
| *** patrickeast has joined #zuul | 14:44 | |
| *** adam_g has quit IRC | 14:44 | |
| *** TheJulia has joined #zuul | 14:46 | |
| *** adam_g has joined #zuul | 14:48 | |
| *** isaacb has quit IRC | 15:05 | |
| *** isaacb has joined #zuul | 15:06 | |
| *** abregman is now known as abregman|brb | 15:08 | |
| *** isaacb has quit IRC | 15:11 | |
| *** isaacb has joined #zuul | 15:12 | |
| *** herlo has quit IRC | 15:16 | |
| *** isaacb has quit IRC | 15:22 | |
| *** herlo has joined #zuul | 15:23 | |
| *** herlo has joined #zuul | 15:23 | |
| *** isaacb has joined #zuul | 15:28 | |
| openstackgerrit | Merged openstack-infra/nodepool master: Make web status text/plain https://review.openstack.org/428501 | 15:36 |
| *** abregman|brb is now known as abregman | 15:50 | |
| *** isaacb has quit IRC | 15:53 | |
| openstackgerrit | David Shrewsbury proposed openstack-infra/nodepool feature/zuulv3: WIP: Implement node cleanup https://review.openstack.org/433736 | 15:53 |
| *** abregman is now known as abregman|mtg | 15:54 | |
| *** abregman|mtg is now known as abregman | 16:06 | |
| openstackgerrit | Jan Kundrát proposed openstack-infra/zuul master: Fix change number extraction on new enough Gerrit master https://review.openstack.org/433748 | 16:13 |
| mordred | jeblair: ^^ | 16:15 |
| jeblair | mordred: thx, approved with comment. | 16:37 |
| openstackgerrit | Merged openstack-infra/zuul master: Fix change number extraction on new enough Gerrit master https://review.openstack.org/433748 | 16:40 |
| openstackgerrit | James E. Blair proposed openstack-infra/zuul feature/zuulv3: Re-enable test_abandoned_not_timer https://review.openstack.org/427985 | 17:10 |
| openstackgerrit | James E. Blair proposed openstack-infra/zuul feature/zuulv3: Reset merger repo when performing an update https://review.openstack.org/433798 | 17:10 |
| jeblair | SpamapS: i know you're traveling today, but for when you get a sec: i think you found a zuul bug in your test, so i fixed the bug, rebased your test on it and updated the test ^ | 17:20 |
| *** abregman has quit IRC | 17:43 | |
| *** hashar has quit IRC | 17:59 | |
| SpamapS | jeblair: oh fun! :) | 17:59 |
| SpamapS | jeblair: oh my. Glad you're paying attention. :) | 18:01 |
| SpamapS | jeblair: I'll go ahead and update 430568 in kind | 18:03 |
| jeblair | SpamapS: cool, thanks | 18:03 |
| openstackgerrit | Clint 'SpamapS' Byrum proposed openstack-infra/zuul feature/zuulv3: Re-enable test_timer https://review.openstack.org/430968 | 18:12 |
| openstackgerrit | Clint 'SpamapS' Byrum proposed openstack-infra/zuul feature/zuulv3: Re-enable test_timer_smtp https://review.openstack.org/430983 | 18:12 |
| openstackgerrit | Clint 'SpamapS' Byrum proposed openstack-infra/zuul feature/zuulv3: Re-enable test_idle https://review.openstack.org/430568 | 18:12 |
| jeblair | SpamapS, mordred, jhesketh: it occurs to me that the requirement for having zuul's config in a zuul-managed repo makes the third-party ci case (as well as the "i want to run zuul locally and point it at review.o.o" case) difficult. so i'm thinking of writing a quick 'git' driver so you can point zuul at a plain old (ie, not gerrit- or github- or whatever- managed) git repo so that it can fetch its config from there. | 18:25 |
| jeblair | i don't think it will need to be more more than a few lines of code | 18:25 |
| mordred | jeblair: that makes sense - but is there any need to have that file in git if it's not in a zuul managed repo? if we're going to let people do that, might as well let them just point at a directory and manage the file with config management? | 18:28 |
| jeblair | mordred: yes! zuul has no idea how to load a config file if it's not in git :) | 18:29 |
| jeblair | mordred: zuul.conf and the tenant config (which points to the git repos with the rest of the config) are expected to be on disk, but everything else has to come from a git repo | 18:30 |
| jeblair | mordred: i mean, of course, we could add a second path through config loading to do that, but it would be much more complicated than a git driver | 18:30 |
| jeblair | mordred: also.. and perhaps this is more germaine -- the launchers need to be able to clone/checkout the config repos because that's where the job playbooks are | 18:32 |
| jeblair | mordred: so i don't think we'd get very far down that path... | 18:32 |
| jeblair | it's worth noting that puppet-openstackci users likely have their own project-config repos, so they are (like us) using config-management to update their zuul configs (which are kept in git). this would allow pretty much the same kind of setup in the future. | 18:34 |
| mordred | jeblair: cool | 18:35 |
| SpamapS | It's also pretty easy to get your config management to just commit changed files. | 18:37 |
| SpamapS | if that's how you want to do it. | 18:37 |
| SpamapS | And that lets zuul be efficient and just wait for changes in git. | 18:37 |
| SpamapS | so +1 from me for a light weight git driver | 18:37 |
| openstackgerrit | David Shrewsbury proposed openstack-infra/nodepool feature/zuulv3: Add generator API method for node iteration https://review.openstack.org/433252 | 18:44 |
| *** saneax-_-|AFK is now known as saneax | 19:06 | |
| *** saneax is now known as saneax-_-|AFK | 20:14 | |
| *** jamielennox|away is now known as jamielennox | 20:17 | |
| *** openstackgerrit has quit IRC | 20:18 | |
| *** openstackgerrit has joined #zuul | 20:22 | |
| openstackgerrit | David Shrewsbury proposed openstack-infra/nodepool feature/zuulv3: Disconnect from ZooKeeper at shutdown https://review.openstack.org/433919 | 20:22 |
| Shrews | Easy review, with the added fun of #TuesdayPoemDay ^^^^ | 20:23 |
| pabelanger | Shrews: lulz | 20:30 |
| pabelanger | nice | 20:30 |
| *** hashar has joined #zuul | 20:31 | |
| *** abregman has joined #zuul | 20:39 | |
| mordred | Shrews: I'm not approving that - just because I think everyone should read your commit message | 21:06 |
| mordred | SpamapS, jeblair: at least the two of you :) | 21:07 |
| Shrews | SpamapS was my inspiration, with his ode to Rust | 21:07 |
| Shrews | and sometimes I get tired of writing code | 21:08 |
| jeblair | Shrews: you've found the recipe for collecting all the code review +2s but no workflow +1s. :) | 21:11 |
| openstackgerrit | David Shrewsbury proposed openstack-infra/nodepool feature/zuulv3: WIP: Implement node cleanup https://review.openstack.org/433736 | 21:18 |
| Shrews | ~/c 2 | 21:18 |
| Shrews | oops | 21:18 |
| *** abregman has quit IRC | 21:22 | |
| openstackgerrit | David Shrewsbury proposed openstack-infra/nodepool feature/zuulv3: WIP: Implement node cleanup https://review.openstack.org/433736 | 21:25 |
| openstackgerrit | David Shrewsbury proposed openstack-infra/nodepool feature/zuulv3: WIP: Implement node cleanup https://review.openstack.org/433736 | 21:43 |
| openstackgerrit | James E. Blair proposed openstack-infra/zuul feature/zuulv3: Run zuul v3 launcher https://review.openstack.org/433347 | 21:52 |
| openstackgerrit | James E. Blair proposed openstack-infra/zuul feature/zuulv3: Add Git driver https://review.openstack.org/433942 | 21:52 |
| openstackgerrit | James E. Blair proposed openstack-infra/zuul feature/zuulv3: Remove gearman settled check https://review.openstack.org/433943 | 21:52 |
| jeblair | SpamapS, mordred: there's the git driver | 21:53 |
| SpamapS | Woot | 21:53 |
| mordred | jeblair: you're a madman | 21:57 |
| mordred | jeblair: also, I know what it is - but the commit subject "Add Git driver" implies that before that zuul didn't know how to talk to git, which amuses me | 21:58 |
| mordred | jeblair: you'd think __ne__ on the base object class would be the exact impl you have there | 21:59 |
| mordred | (sorry, in the first patch of that stack) | 21:59 |
| mordred | jeblair: while I'm bothering you - weren't you or SpamapS working on test_dependent_behind_dequeue recently? it started failing in 'Run zuul v3 launcher' | 22:02 |
| openstackgerrit | Merged openstack-infra/zuul feature/zuulv3: Implement node equality https://review.openstack.org/430491 | 22:04 |
| jeblair | mordred: i don't think we're actively working on it, we've just identified it's a random problem. we'll have to recheck for now. | 22:06 |
| jeblair | (until we make that pot of coffee) | 22:06 |
| mordred | jeblair: I have a pot of coffee - but I just sneezed some of it out onto the floor and counter and table and other counter and chair | 22:06 |
| mordred | so it's probably not the right thing | 22:07 |
| jeblair | whatever it takes | 22:08 |
| Shrews | jeblair: for our FakeProviderManager, should we expect cleanupServer() to throw the NotFound exception in testing? | 22:10 |
| jeblair | Shrews: yes | 22:11 |
| Shrews | jeblair: ok. so i guess i should catch that and go ahead and delete the znode if NotFound is returned? | 22:12 |
| Shrews | oh, yeah. looks like that's what the old code does | 22:12 |
| openstackgerrit | Antoine Musso proposed openstack-infra/zuul master: Gerrit trailing delay is now configurable https://review.openstack.org/343562 | 22:13 |
| openstackgerrit | Antoine Musso proposed openstack-infra/zuul master: Status: apply default style by default https://review.openstack.org/276480 | 22:15 |
| openstackgerrit | Antoine Musso proposed openstack-infra/zuul master: Do not import paramiko before daemonization https://review.openstack.org/343530 | 22:21 |
| *** hashar has quit IRC | 22:24 | |
| openstackgerrit | Merged openstack-infra/zuul master: Status: apply default style by default https://review.openstack.org/276480 | 22:25 |
| openstackgerrit | James E. Blair proposed openstack-infra/zuul feature/zuulv3: Squelch ref-replication gerrit warnings https://review.openstack.org/433959 | 22:29 |
| jeblair | mordred or pabelanger: can you +3 427985 ? | 22:29 |
| SpamapS | mordred: I believe there's a deadlock or something looping forever due to races in that test | 22:31 |
| SpamapS | I will be sitting in the Reno airport looking at that in about an hour :) | 22:31 |
| openstackgerrit | David Shrewsbury proposed openstack-infra/nodepool feature/zuulv3: Disconnect from ZooKeeper at shutdown https://review.openstack.org/433919 | 22:32 |
| openstackgerrit | David Shrewsbury proposed openstack-infra/nodepool feature/zuulv3: Add generator API method for node iteration https://review.openstack.org/433252 | 22:32 |
| openstackgerrit | David Shrewsbury proposed openstack-infra/nodepool feature/zuulv3: Set Node image_id and launcher attributes https://review.openstack.org/433242 | 22:32 |
| openstackgerrit | David Shrewsbury proposed openstack-infra/nodepool feature/zuulv3: Implement node cleanup https://review.openstack.org/433736 | 22:32 |
| mordred | jeblair: on it | 22:33 |
| openstackgerrit | Merged openstack-infra/zuul feature/zuulv3: Reset merger repo when performing an update https://review.openstack.org/433798 | 22:34 |
| openstackgerrit | Monty Taylor proposed openstack-infra/zuul feature/zuulv3: Remove TODO comment that seems to be done https://review.openstack.org/433961 | 22:36 |
| mordred | jeblair: ^^ I +A'd 427985 - but it seemed there was amissed comment removal in it - rather than go through a review cycle, I just pushed up a patch ^^ | 22:36 |
| mordred | jeblair: it's also possible I read things wrong and there is still a todo there | 22:36 |
| jeblair | mordred: i think you're right, thanks | 22:37 |
| * mordred is useful | 22:37 | |
| openstackgerrit | Merged openstack-infra/zuul feature/zuulv3: Re-enable test_abandoned_not_timer https://review.openstack.org/427985 | 22:40 |
| *** _ari__ has joined #zuul | 22:41 | |
| *** NTolerance has joined #zuul | 22:41 | |
| *** jasondotstar has joined #zuul | 22:42 | |
| *** nt has quit IRC | 22:42 | |
| *** Shrews has quit IRC | 22:42 | |
| *** _ari_ has quit IRC | 22:42 | |
| *** jasondotstar_ has quit IRC | 22:42 | |
| *** greghaynes has quit IRC | 22:42 | |
| *** saneax-_-|AFK has quit IRC | 22:42 | |
| *** jamielennox has quit IRC | 22:42 | |
| *** fungi has quit IRC | 22:42 | |
| *** NTolerance is now known as nt | 22:42 | |
| openstackgerrit | James E. Blair proposed openstack-infra/zuul feature/zuulv3: Squelch ref-replication gerrit warnings https://review.openstack.org/433959 | 22:44 |
| openstackgerrit | James E. Blair proposed openstack-infra/zuul feature/zuulv3: Add Git driver https://review.openstack.org/433942 | 22:44 |
| openstackgerrit | James E. Blair proposed openstack-infra/zuul feature/zuulv3: Remove gearman settled check https://review.openstack.org/433943 | 22:44 |
| openstackgerrit | James E. Blair proposed openstack-infra/zuul feature/zuulv3: Correct getGitwebUrl https://review.openstack.org/433964 | 22:44 |
| mordred | jeblair: darn. I didn't catch you before you fixed the pep8 - there's a funny typo in the commit message in https://review.openstack.org/#/c/433959/ | 22:45 |
| jeblair | mordred: heh, i actually haven't fixed the pep8 yet, i just noticed that | 22:46 |
| pabelanger | jeblair: mordred: mind reviewing https://review.openstack.org/#/q/topic:enable-tests too? re-enables tests for nodepool feature/zuulv3 | 22:46 |
| *** _ari__ has quit IRC | 22:46 | |
| *** TheJulia has quit IRC | 22:46 | |
| *** morgan has quit IRC | 22:46 | |
| *** mattclay has quit IRC | 22:46 | |
| *** auggy has quit IRC | 22:46 | |
| *** bstinson has quit IRC | 22:46 | |
| openstackgerrit | James E. Blair proposed openstack-infra/zuul feature/zuulv3: Squelch ref-replication gerrit warnings https://review.openstack.org/433959 | 22:46 |
| openstackgerrit | James E. Blair proposed openstack-infra/zuul feature/zuulv3: Correct getGitwebUrl https://review.openstack.org/433964 | 22:46 |
| *** bstinson has joined #zuul | 22:46 | |
| jeblair | mordred: fixed both | 22:47 |
| *** _ari_ has joined #zuul | 22:47 | |
| jeblair | pabelanger: i will do so in a few mins | 22:48 |
| *** Shrews has joined #zuul | 22:49 | |
| *** greghaynes has joined #zuul | 22:49 | |
| *** fungi has joined #zuul | 22:49 | |
| openstackgerrit | James E. Blair proposed openstack-infra/zuul feature/zuulv3: Add (minimal) support for topic-changed event https://review.openstack.org/433966 | 22:50 |
| *** mattclay has joined #zuul | 22:55 | |
| openstackgerrit | James E. Blair proposed openstack-infra/zuul feature/zuulv3: Squelch ref-replication gerrit warnings https://review.openstack.org/433959 | 22:56 |
| openstackgerrit | James E. Blair proposed openstack-infra/zuul feature/zuulv3: Add Git driver https://review.openstack.org/433942 | 22:56 |
| openstackgerrit | James E. Blair proposed openstack-infra/zuul feature/zuulv3: Remove gearman settled check https://review.openstack.org/433943 | 22:56 |
| openstackgerrit | James E. Blair proposed openstack-infra/zuul feature/zuulv3: Run zuul v3 launcher https://review.openstack.org/433347 | 22:56 |
| openstackgerrit | James E. Blair proposed openstack-infra/zuul feature/zuulv3: Add (minimal) support for topic-changed event https://review.openstack.org/433966 | 22:56 |
| openstackgerrit | James E. Blair proposed openstack-infra/zuul feature/zuulv3: Correct getGitwebUrl https://review.openstack.org/433964 | 22:56 |
| openstackgerrit | James E. Blair proposed openstack-infra/zuul feature/zuulv3: Remove misleading log statements https://review.openstack.org/433970 | 22:56 |
| jeblair | that was mostly a rebase | 22:56 |
| jeblair | hopefully votes will carry | 22:57 |
| *** TheJulia has joined #zuul | 22:57 | |
| *** auggy has joined #zuul | 22:57 | |
| *** morgan_ has joined #zuul | 23:16 | |
| openstackgerrit | Merged openstack-infra/nodepool feature/zuulv3: Re-enable test_node_net_name test https://review.openstack.org/433227 | 23:17 |
| jeblair | pabelanger, Shrews: i left a comment on 433235 | 23:18 |
| openstackgerrit | Merged openstack-infra/nodepool feature/zuulv3: Re-enable test_node_vhd_image test https://review.openstack.org/433233 | 23:18 |
| openstackgerrit | James E. Blair proposed openstack-infra/zuul feature/zuulv3: Remove misleading log statements https://review.openstack.org/433970 | 23:24 |
| openstackgerrit | James E. Blair proposed openstack-infra/zuul feature/zuulv3: Add __repr__ to jobtree https://review.openstack.org/430493 | 23:24 |
| jeblair | i have a zuulv3 scheduler and launcher running locally! (they are not running jobs or anything, but they are running and watching gerrit) | 23:33 |
| openstackgerrit | James E. Blair proposed openstack-infra/zuul feature/zuulv3: Rename zuul-server zuul-scheduler https://review.openstack.org/433980 | 23:36 |
| SpamapS | coffee ordered | 23:50 |
| SpamapS | time to debug tests in the corner of the baggage claim area. :) | 23:50 |
| openstackgerrit | David Shrewsbury proposed openstack-infra/nodepool feature/zuulv3: Disconnect from ZooKeeper at shutdown https://review.openstack.org/433919 | 23:55 |
| openstackgerrit | David Shrewsbury proposed openstack-infra/nodepool feature/zuulv3: Add generator API method for node iteration https://review.openstack.org/433252 | 23:55 |
| openstackgerrit | David Shrewsbury proposed openstack-infra/nodepool feature/zuulv3: Implement node cleanup https://review.openstack.org/433736 | 23:55 |
| Shrews | jeblair: I had to adjust the test in 433252 there since I can't guarantee order from the new nodeIterator() method | 23:56 |
| Shrews | it could (will/did) randomly fail | 23:56 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!