mordred | pabelanger: yah - that syntax got deprecated in 2.0 in favor of something else | 00:04 |
---|---|---|
pabelanger | mordred: Ya, found it in the release notes. Surprisingly, I didn't see any deprecated warnings about it | 00:06 |
pabelanger | either way, all good. Updating playbooks now | 00:06 |
mordred | yay | 00:06 |
clarkb | is it now becomes: root? | 00:07 |
clarkb | or sudo: true? seems like there are 10 diffferent ways to do this thing depending on version | 00:07 |
mordred | yah. it's all going to one way | 00:28 |
mordred | clarkb: it's become: true - and then if you want to become a user that is not root, you use become_user: foo | 00:30 |
mordred | http://docs.ansible.com/ansible/become.html#become-privilege-escalation | 00:30 |
*** bcoca has quit IRC | 04:58 | |
*** saneax-_-|AFK is now known as saneax | 05:16 | |
*** abregman has joined #zuul | 07:19 | |
*** abregman has quit IRC | 08:26 | |
*** abregman has joined #zuul | 08:28 | |
*** abregman is now known as abregman|mtg | 08:38 | |
*** bhavik1 has joined #zuul | 08:42 | |
*** openstackgerrit has quit IRC | 08:48 | |
*** openstackgerrit has joined #zuul | 08:49 | |
*** hashar has joined #zuul | 08:52 | |
*** abregman|mtg has quit IRC | 09:07 | |
*** abregman has joined #zuul | 09:10 | |
*** abregman is now known as abregman|mtg | 09:14 | |
*** abregman|mtg is now known as abregman | 09:31 | |
openstackgerrit | Merged openstack-infra/zuul: Re-enable the duplicate_pipeline test https://review.openstack.org/384192 | 09:42 |
*** saneax is now known as saneax-_-|AFK | 11:28 | |
*** bhavik1 has quit IRC | 11:45 | |
*** bcoca has joined #zuul | 12:52 | |
*** hashar has quit IRC | 13:25 | |
*** hashar has joined #zuul | 13:25 | |
SpamapS | Well I'm going to need a whole lot of coding to distract me today.... time to start knocking down the zuul todos. | 13:32 |
Shrews | SpamapS: geez you're up early | 13:47 |
SpamapS | Shrews: early flight. :-P | 13:48 |
openstackgerrit | Clint 'SpamapS' Byrum proposed openstack-infra/zuul: Fix test_failed_change_in_middle to use tenants https://review.openstack.org/395607 | 13:57 |
SpamapS | Shrews: ^^ a fixed test! | 13:58 |
Shrews | SpamapS: so complex | 13:59 |
openstackgerrit | Paul Belanger proposed openstack-infra/zuul: Enable test_single_nonexistent_post_job test https://review.openstack.org/391692 | 14:01 |
openstackgerrit | Paul Belanger proposed openstack-infra/zuul: Enable 4 CRD unit tests https://review.openstack.org/393494 | 14:01 |
openstackgerrit | Paul Belanger proposed openstack-infra/zuul: Re-enable 3 disabled tests https://review.openstack.org/393452 | 14:01 |
openstackgerrit | Paul Belanger proposed openstack-infra/zuul: Re-enable test_file_head test https://review.openstack.org/393956 | 14:01 |
openstackgerrit | Paul Belanger proposed openstack-infra/zuul: Enable test_failed_change_in_middle test https://review.openstack.org/391687 | 14:01 |
openstackgerrit | Paul Belanger proposed openstack-infra/zuul: Move job running into own class https://review.openstack.org/385964 | 14:01 |
openstackgerrit | Paul Belanger proposed openstack-infra/zuul: Re-enable test_nonvoting_pipeline https://review.openstack.org/386769 | 14:01 |
openstackgerrit | Paul Belanger proposed openstack-infra/zuul: Re-enable test_merger_repack test https://review.openstack.org/393953 | 14:01 |
openstackgerrit | Paul Belanger proposed openstack-infra/zuul: Re-enable test_layered_templates test https://review.openstack.org/393445 | 14:01 |
openstackgerrit | Paul Belanger proposed openstack-infra/zuul: Re-enable test_job_from_templates_launched job https://review.openstack.org/393412 | 14:01 |
openstackgerrit | Paul Belanger proposed openstack-infra/zuul: Enable test_two_failed_changes_at_head test https://review.openstack.org/393950 | 14:01 |
openstackgerrit | Paul Belanger proposed openstack-infra/zuul: Re-enable test_can_merge unit test https://review.openstack.org/392953 | 14:01 |
openstackgerrit | Paul Belanger proposed openstack-infra/zuul: Re-enable test_queue_precedence test https://review.openstack.org/394191 | 14:01 |
openstackgerrit | Paul Belanger proposed openstack-infra/zuul: Enable test_statsd test https://review.openstack.org/393479 | 14:01 |
openstackgerrit | Paul Belanger proposed openstack-infra/zuul: Enable test_new_patchset_check / test_abandoned_check https://review.openstack.org/393458 | 14:01 |
openstackgerrit | Paul Belanger proposed openstack-infra/zuul: Re-enable test_source_cache unit test https://review.openstack.org/393278 | 14:01 |
openstackgerrit | Paul Belanger proposed openstack-infra/zuul: Re-enable test_one_job_project unit test https://review.openstack.org/393304 | 14:01 |
openstackgerrit | Paul Belanger proposed openstack-infra/zuul: Re-enable 4 disable tests https://review.openstack.org/393461 | 14:01 |
openstackgerrit | Paul Belanger proposed openstack-infra/zuul: Re-enable test_needed_changes_enqueue testing https://review.openstack.org/393001 | 14:01 |
openstackgerrit | Paul Belanger proposed openstack-infra/zuul: Lower the log level in tests https://review.openstack.org/395611 | 14:01 |
openstackgerrit | Paul Belanger proposed openstack-infra/zuul: Remove discover from test-requirements https://review.openstack.org/395612 | 14:01 |
openstackgerrit | Paul Belanger proposed openstack-infra/zuul: Ansible launcher: fix afs publisher https://review.openstack.org/395613 | 14:01 |
openstackgerrit | Paul Belanger proposed openstack-infra/zuul: Rename zuul_runner to command https://review.openstack.org/395614 | 14:01 |
openstackgerrit | Paul Belanger proposed openstack-infra/zuul: Use command module instead of zuul_runner https://review.openstack.org/395615 | 14:01 |
openstackgerrit | Paul Belanger proposed openstack-infra/zuul: Put script string in directly instead of in files https://review.openstack.org/395616 | 14:01 |
openstackgerrit | Paul Belanger proposed openstack-infra/zuul: Stop running commands with async https://review.openstack.org/395617 | 14:01 |
openstackgerrit | Paul Belanger proposed openstack-infra/zuul: Split playbook into vars, pre-playbook and playbook https://review.openstack.org/395618 | 14:01 |
openstackgerrit | Paul Belanger proposed openstack-infra/zuul: Enable pipelining for ansible-playbook https://review.openstack.org/395619 | 14:01 |
openstackgerrit | Paul Belanger proposed openstack-infra/zuul: Strip leading / from afs targets https://review.openstack.org/395620 | 14:01 |
openstackgerrit | Paul Belanger proposed openstack-infra/zuul: Ansible launcher: Fix afs publisher root detection https://review.openstack.org/395621 | 14:01 |
openstackgerrit | Paul Belanger proposed openstack-infra/zuul: Add a note to README.rst about contributing to Zuul v3 https://review.openstack.org/395622 | 14:01 |
openstackgerrit | Paul Belanger proposed openstack-infra/zuul: Default test log level to DEBUG except for testr https://review.openstack.org/395623 | 14:01 |
openstackgerrit | Paul Belanger proposed openstack-infra/zuul: Revert "Put script string in directly instead of in files" https://review.openstack.org/395624 | 14:01 |
openstackgerrit | Paul Belanger proposed openstack-infra/zuul: Generate shell scripts as a sequence https://review.openstack.org/395625 | 14:01 |
openstackgerrit | Paul Belanger proposed openstack-infra/zuul: Ansible launcher: use sequence-uuid in shell scripts https://review.openstack.org/395626 | 14:01 |
pabelanger | Um | 14:01 |
openstackgerrit | Paul Belanger proposed openstack-infra/zuul: Ansible launcher: import get_exception in ansible command https://review.openstack.org/395627 | 14:01 |
openstackgerrit | Paul Belanger proposed openstack-infra/zuul: Ansible launcher: don't close stdout in command module https://review.openstack.org/395628 | 14:01 |
openstackgerrit | Paul Belanger proposed openstack-infra/zuul: Add back timeout_var logic https://review.openstack.org/395629 | 14:01 |
Shrews | spammer! | 14:01 |
openstackgerrit | Paul Belanger proposed openstack-infra/zuul: Ansible launcher: remove keep_remote_files https://review.openstack.org/395630 | 14:01 |
pabelanger | that did go well | 14:01 |
mordred | pabelanger: did ya submit to master instead of feature/v3 ? | 14:02 |
pabelanger | mordred: that should be feature/v3 | 14:02 |
mordred | fascinating | 14:02 |
pabelanger | let me see what I did wrong | 14:02 |
mordred | well, I'm just going to assume you just wrote a LOT of code | 14:02 |
pabelanger | is it possible the merge from master to feature/v2 got rewritten | 14:03 |
pabelanger | https://review.openstack.org/#/c/389470/2 | 14:03 |
pabelanger | some how I think my patches created new reviews | 14:04 |
pabelanger | ya, looks like it | 14:05 |
pabelanger | mordred: do you mind +3 on https://review.openstack.org/#/c/389470/2 merge from master into feature/v3 | 14:06 |
pabelanger | then I should be able to rebase properly again | 14:07 |
mordred | pabelanger: done | 14:18 |
pabelanger | mordred: danke | 14:18 |
pabelanger | mordred: Hmm, 389470 still 'Needs Verified' even that jenkins has +1. Any ideas? | 14:26 |
mordred | pabelanger: I don't see it running in zuul at all | 14:29 |
mordred | pabelanger: or _any_ changes for zuul running in zuul | 14:29 |
mordred | pabelanger: Needs Verified is becuase it needs the final +2 from the gate job - but best I can tell, the gate job has not run | 14:30 |
pabelanger | Ah, yes. Was confusing it with 'Needs Workflow' | 14:30 |
pabelanger | let me check zuul debug logs and see | 14:30 |
pabelanger | http://paste.openstack.org/show/588559/ | 14:32 |
pabelanger | Ah | 14:33 |
pabelanger | looks like we are missing a few commits | 14:34 |
pabelanger | I mistakenly abandoned 385964 | 14:34 |
openstackgerrit | Paul Belanger proposed openstack-infra/zuul: Move job running into own class https://review.openstack.org/385964 | 14:40 |
*** saneax-_-|AFK is now known as saneax | 15:03 | |
*** saneax is now known as saneax-_-|AFK | 15:14 | |
*** saneax-_-|AFK is now known as saneax | 15:14 | |
openstackgerrit | Paul Belanger proposed openstack-infra/zuul: Merge remote-tracking branch 'origin/master' into feature/zuulv3 https://review.openstack.org/389470 | 15:16 |
SpamapS | pabelanger: oops, wrong branch? | 15:32 |
*** saneax is now known as saneax-_-|AFK | 15:32 | |
pabelanger | SpamapS: no, right branch, newer commits are breaking the feature/zuulv3 branch | 15:33 |
SpamapS | oh n/m, my backscroll just caught up | 15:33 |
pabelanger | at this point, I'm going to revert to the original merge attempt that jhesketh did | 15:33 |
*** saneax-_-|AFK is now known as saneax | 15:33 | |
pabelanger | okay, for now, I'm going to rebase enable-tests a top of TIP of feature/zuulv3. Until jhesketh is back online to help with 389470 | 15:46 |
pabelanger | apologies for the incoming spam, 18 patches | 15:47 |
openstackgerrit | Paul Belanger proposed openstack-infra/zuul: Enable test_single_nonexistent_post_job test https://review.openstack.org/391692 | 15:47 |
openstackgerrit | Paul Belanger proposed openstack-infra/zuul: Enable 4 CRD unit tests https://review.openstack.org/393494 | 15:47 |
openstackgerrit | Paul Belanger proposed openstack-infra/zuul: Re-enable 3 disabled tests https://review.openstack.org/393452 | 15:47 |
openstackgerrit | Paul Belanger proposed openstack-infra/zuul: Re-enable test_file_head test https://review.openstack.org/393956 | 15:47 |
openstackgerrit | Paul Belanger proposed openstack-infra/zuul: Enable test_failed_change_in_middle test https://review.openstack.org/391687 | 15:47 |
openstackgerrit | Paul Belanger proposed openstack-infra/zuul: Re-enable test_nonvoting_pipeline https://review.openstack.org/386769 | 15:47 |
openstackgerrit | Paul Belanger proposed openstack-infra/zuul: Re-enable test_merger_repack test https://review.openstack.org/393953 | 15:47 |
openstackgerrit | Paul Belanger proposed openstack-infra/zuul: Re-enable test_layered_templates test https://review.openstack.org/393445 | 15:47 |
openstackgerrit | Paul Belanger proposed openstack-infra/zuul: Re-enable test_job_from_templates_launched job https://review.openstack.org/393412 | 15:47 |
openstackgerrit | Paul Belanger proposed openstack-infra/zuul: Enable test_two_failed_changes_at_head test https://review.openstack.org/393950 | 15:47 |
openstackgerrit | Paul Belanger proposed openstack-infra/zuul: Re-enable test_can_merge unit test https://review.openstack.org/392953 | 15:47 |
openstackgerrit | Paul Belanger proposed openstack-infra/zuul: Re-enable test_queue_precedence test https://review.openstack.org/394191 | 15:47 |
openstackgerrit | Paul Belanger proposed openstack-infra/zuul: Enable test_statsd test https://review.openstack.org/393479 | 15:47 |
openstackgerrit | Paul Belanger proposed openstack-infra/zuul: Enable test_new_patchset_check / test_abandoned_check https://review.openstack.org/393458 | 15:47 |
openstackgerrit | Paul Belanger proposed openstack-infra/zuul: Re-enable test_source_cache unit test https://review.openstack.org/393278 | 15:47 |
openstackgerrit | Paul Belanger proposed openstack-infra/zuul: Re-enable test_one_job_project unit test https://review.openstack.org/393304 | 15:47 |
openstackgerrit | Paul Belanger proposed openstack-infra/zuul: Re-enable 4 disable tests https://review.openstack.org/393461 | 15:47 |
openstackgerrit | Paul Belanger proposed openstack-infra/zuul: Re-enable test_needed_changes_enqueue testing https://review.openstack.org/393001 | 15:47 |
pabelanger | okay, that should make things green again | 15:47 |
pabelanger | would love some reviews on ^ to avoid the noise of rebasing | 15:48 |
*** abregman has quit IRC | 15:53 | |
*** abregman has joined #zuul | 15:55 | |
*** saneax is now known as saneax-_-|AFK | 16:01 | |
*** saneax-_-|AFK is now known as saneax | 16:01 | |
*** saneax is now known as saneax-_-|AFK | 16:03 | |
*** hashar has quit IRC | 16:05 | |
pabelanger | abregman: I think we have a race condition in the test_duplicate_pipelines job | 16:06 |
pabelanger | http://logs.openstack.org/79/393479/3/check/gate-zuul-python27-ubuntu-xenial/01dceb9/console.html#_2016-11-09_15_51_59_184800 | 16:07 |
pabelanger | going to dig into it now | 16:07 |
*** saneax-_-|AFK is now known as saneax | 16:13 | |
openstackgerrit | Paul Belanger proposed openstack-infra/zuul: Fix race in TestDuplicatePipeline test https://review.openstack.org/395694 | 16:14 |
pabelanger | abregman: I believe ^ is our fix | 16:14 |
*** saneax is now known as saneax-_-|AFK | 16:22 | |
*** saneax-_-|AFK is now known as saneax | 16:22 | |
*** saneax is now known as saneax-_-|AFK | 16:23 | |
*** saneax-_-|AFK is now known as saneax | 16:23 | |
*** saneax is now known as saneax-_-|AFK | 16:33 | |
*** saneax-_-|AFK is now known as saneax | 16:34 | |
*** saneax is now known as saneax-_-|AFK | 16:37 | |
*** saneax-_-|AFK is now known as saneax | 16:37 | |
*** abregman_ has joined #zuul | 16:38 | |
abregman_ | pabelanger: sorry had a mtg. looking now. | 16:38 |
abregman_ | pabelanger: what self.orderedRelease() is used for? | 16:40 |
pabelanger | # Run one build at a time to ensure non-race order: | 16:40 |
*** saneax is now known as saneax-_-|AFK | 16:40 | |
pabelanger | which should be fine for this test, since we are making sure both pipelines right | 16:41 |
*** saneax-_-|AFK is now known as saneax | 16:41 | |
pabelanger | not which order they run in | 16:41 |
pabelanger | s/right/run | 16:41 |
pabelanger | pipelines are right* | 16:41 |
*** abregman has quit IRC | 16:41 | |
abregman_ | pabelanger: oh, cool. thanks | 16:43 |
*** abregman_ is now known as abregman | 16:43 | |
*** saneax is now known as saneax-_-|AFK | 16:46 | |
*** saneax-_-|AFK is now known as saneax | 16:46 | |
*** saneax is now known as saneax-_-|AFK | 16:50 | |
*** saneax-_-|AFK is now known as saneax | 16:50 | |
*** saneax is now known as saneax-_-|AFK | 16:52 | |
*** saneax-_-|AFK is now known as saneax | 16:53 | |
*** saneax is now known as saneax-_-|AFK | 16:54 | |
openstackgerrit | Merged openstack-infra/nodepool: Move kazooclient in to a fixture https://review.openstack.org/387577 | 17:02 |
openstackgerrit | Merged openstack-infra/zuul: Enable test_failed_change_in_middle test https://review.openstack.org/391687 | 17:03 |
openstackgerrit | Merged openstack-infra/zuul: Re-enable test_nonvoting_pipeline https://review.openstack.org/386769 | 17:05 |
jeblair | pabelanger: see possible surprise -1 on 391692 | 17:06 |
pabelanger | jeblair: looking | 17:09 |
pabelanger | ack | 17:09 |
jeblair | pabelanger: i'm looking at 393304 and wondering if we should delete that test too. i think it may be fairly redundant at this point. | 17:12 |
pabelanger | jeblair: I think so, we don't have anything that does a single job today. But, so far all of our things do a merge first | 17:15 |
jeblair | pabelanger: hrm. i guess we could keep it for now. :) | 17:18 |
openstackgerrit | James E. Blair proposed openstack-infra/zuul: Remove v3 project template test https://review.openstack.org/395722 | 17:20 |
pabelanger | okay, 391692 updated | 17:23 |
pabelanger | pushing up changes | 17:23 |
openstackgerrit | Paul Belanger proposed openstack-infra/zuul: Delete test_single_nonexistent_post_job test https://review.openstack.org/391692 | 17:23 |
openstackgerrit | Paul Belanger proposed openstack-infra/zuul: Enable 4 CRD unit tests https://review.openstack.org/393494 | 17:23 |
openstackgerrit | Paul Belanger proposed openstack-infra/zuul: Re-enable 3 disabled tests https://review.openstack.org/393452 | 17:23 |
openstackgerrit | Paul Belanger proposed openstack-infra/zuul: Re-enable test_file_head test https://review.openstack.org/393956 | 17:23 |
openstackgerrit | Paul Belanger proposed openstack-infra/zuul: Re-enable test_merger_repack test https://review.openstack.org/393953 | 17:23 |
openstackgerrit | Paul Belanger proposed openstack-infra/zuul: Re-enable test_layered_templates test https://review.openstack.org/393445 | 17:24 |
openstackgerrit | Paul Belanger proposed openstack-infra/zuul: Re-enable test_job_from_templates_launched job https://review.openstack.org/393412 | 17:24 |
openstackgerrit | Paul Belanger proposed openstack-infra/zuul: Enable test_two_failed_changes_at_head test https://review.openstack.org/393950 | 17:24 |
openstackgerrit | Paul Belanger proposed openstack-infra/zuul: Re-enable test_can_merge unit test https://review.openstack.org/392953 | 17:24 |
openstackgerrit | Paul Belanger proposed openstack-infra/zuul: Re-enable test_queue_precedence test https://review.openstack.org/394191 | 17:24 |
openstackgerrit | Paul Belanger proposed openstack-infra/zuul: Enable test_statsd test https://review.openstack.org/393479 | 17:24 |
openstackgerrit | Paul Belanger proposed openstack-infra/zuul: Enable test_new_patchset_check / test_abandoned_check https://review.openstack.org/393458 | 17:24 |
openstackgerrit | Paul Belanger proposed openstack-infra/zuul: Re-enable test_source_cache unit test https://review.openstack.org/393278 | 17:24 |
openstackgerrit | Paul Belanger proposed openstack-infra/zuul: Re-enable test_one_job_project unit test https://review.openstack.org/393304 | 17:24 |
openstackgerrit | Paul Belanger proposed openstack-infra/zuul: Re-enable 4 disable tests https://review.openstack.org/393461 | 17:24 |
openstackgerrit | Paul Belanger proposed openstack-infra/zuul: Re-enable test_needed_changes_enqueue testing https://review.openstack.org/393001 | 17:24 |
SpamapS | pabelanger: hey, can you please claim tests in https://storyboard.openstack.org/#!/story/2000773 | 17:28 |
SpamapS | pabelanger: we have a number of people who want to fix tests, but it's a waste for us all to work on the same ones. | 17:28 |
jeblair | ah, if we're ready to start using that story, i will write a note in the etherpad | 17:30 |
pabelanger | jeblair: I think we should land https://review.openstack.org/#/c/395694/ if possible too, fixes what seems to be a race condition on a recently enabled test | 17:30 |
pabelanger | SpamapS: sure, now to see how to claim a storey | 17:31 |
pabelanger | okay, I think I see how it works | 17:34 |
pabelanger | okay, added a bunch of tests | 17:39 |
pabelanger | Not sure how to avoid the very copy / paste of commit messages | 17:39 |
*** abregman has quit IRC | 17:42 | |
pabelanger | okay, I created 24 tasks | 17:47 |
jeblair | pabelanger: yeah, let's not ammend the commit messages just to add task headers. but maybe start using them going forward, and add them if we need another update for other reasons. | 18:01 |
pabelanger | jeblair: Yup, I was planning on manually closing those 25 tasks over another spam attack | 18:01 |
pabelanger | also, this is called a 'kanban board' right? | 18:04 |
pabelanger | trying to read up on some of these processes | 18:04 |
pabelanger | mordred: do you have a few moments to help land https://review.openstack.org/#/q/topic:enable-tests+status:open ? Think we are in good shape, for it to merge now | 18:19 |
mordred | pabelanger: sure! gimme just a sec | 18:21 |
mordred | pabelanger: (my gertty is syncing) | 18:23 |
pabelanger | mordred: danke | 18:24 |
mordred | pabelanger: that merge commit sure does look unhappy | 18:26 |
pabelanger | mordred: Ya, I opted to rebase away from it until jhesketh is online. Hopefully with his guidance we can get it going again | 18:27 |
openstackgerrit | Merged openstack-infra/zuul: Delete test_single_nonexistent_post_job test https://review.openstack.org/391692 | 18:36 |
openstackgerrit | Merged openstack-infra/zuul: Re-enable test_can_merge unit test https://review.openstack.org/392953 | 18:36 |
openstackgerrit | Merged openstack-infra/zuul: Re-enable test_needed_changes_enqueue testing https://review.openstack.org/393001 | 18:38 |
openstackgerrit | Merged openstack-infra/zuul: Re-enable test_source_cache unit test https://review.openstack.org/393278 | 18:38 |
mordred | pabelanger: ok - I merged all the ones that made sense - there are still 4 that jeblair still needs to review | 18:38 |
pabelanger | great! | 18:38 |
pabelanger | thank you | 18:38 |
openstackgerrit | Merged openstack-infra/zuul: Re-enable test_one_job_project unit test https://review.openstack.org/393304 | 18:39 |
openstackgerrit | Merged openstack-infra/zuul: Re-enable test_job_from_templates_launched job https://review.openstack.org/393412 | 18:39 |
jeblair | pabelanger, mordred: fyi i left a comment on (approved change) https://review.openstack.org/393461 | 18:39 |
mordred | jeblair: nod | 18:40 |
mordred | jeblair: I was about to +A https://review.openstack.org/#/c/382077 and its ancestor - but wanted to make sure you'd seen jhesketh's final comment | 18:45 |
pabelanger | jeblair: ack, I'll do that now | 18:45 |
mordred | pabelanger: test unhappy fails | 18:46 |
pabelanger | mordred: jeblair: could also use a review on 395694 to fix a boucing test | 18:46 |
pabelanger | mordred: ya, that is 395694 | 18:46 |
jeblair | mordred: roger, thanks. | 18:47 |
mordred | pabelanger: cool | 18:47 |
mordred | pabelanger: I can see how that patch would be related to the other failure | 18:47 |
pabelanger | yay for fixes | 18:48 |
clarkb | pabelanger: the waitforsettles before it is redundant, and I think you may need to hold tests? or did we stop needing to make that explicit? | 18:49 |
clarkb | (in practice a context switch between line 4675 and 4678 is likely uncommon but I think there is potentially a race there) | 18:50 |
pabelanger | clarkb: looking | 18:51 |
clarkb | oh the wait there may not be redundant because the wait in orderedRelease happens at the end of the loop not the beginning | 18:51 |
clarkb | but I think still possible race there if context switches between those lines | 18:52 |
jeblair | pabelanger, clarkb: ah, yeah, i think this test probably used to just have a count of the jobs run | 18:52 |
jeblair | but maybe switched to the new assertHistory method which is ordered by default | 18:52 |
pabelanger | assertHistory is new for this test | 18:52 |
jeblair | however, we can add 'ordered=False' to assertHistory, then the test shouldn't need to change | 18:52 |
jeblair | so basically two choices: we can do what clarkb suggests and add the hold/sequencing stuff to assert that things happen in order. or we could ignore ordering. | 18:53 |
pabelanger | okay, either works for me | 18:53 |
jeblair | since this testing multiple pipelines with no relationship, i don't think we actually care about ordering | 18:54 |
jeblair | so i lean toward ordered=False | 18:54 |
pabelanger | k | 18:54 |
jeblair | (if we validated ordering, it would essentially become a test of whether zuul ran pipelines in the same order all the time, which it will, for now, but that's not really a defined behavior) | 18:55 |
openstackgerrit | Paul Belanger proposed openstack-infra/zuul: Fix race in TestDuplicatePipeline test https://review.openstack.org/395694 | 18:56 |
clarkb | jeblair: any idea why assertHistory doesn't just check if listA == listB? if thats doable then we can also use collections.Counter to do unordered comparisons easily (then that method becomes like 3 lines | 18:56 |
clarkb | I guess you probably want assertEqual(list(inputA), list(inputB)) else assertEqual(Counter(inputA), Counter(inputB)) ? | 18:57 |
jeblair | clarkb: it's comparing a dictionary with an object. basically each key in the dictionary has to match a corresponding attribute on the object | 18:58 |
pabelanger | clarkb: 395694 should be good now | 18:58 |
clarkb | oh I bet its because the elements are distinct objects so that direct comparison always fails? that would explain it | 18:58 |
jeblair | clarkb: we could probably implement __eq__ methods on the object to make that work though? | 18:58 |
clarkb | jeblair: ya, not sure if that trouble is worth it though (now that I understand what it is doing there) | 18:58 |
openstackgerrit | Merged openstack-infra/zuul: Add getProjectBranches to Source https://review.openstack.org/382076 | 18:59 |
*** harlowja has quit IRC | 18:59 | |
openstackgerrit | Merged openstack-infra/zuul: Implement branch-specific in-repo job config https://review.openstack.org/382077 | 18:59 |
jeblair | clarkb: can you aprv 395694 when you are done? | 19:00 |
clarkb | yup just did | 19:00 |
openstackgerrit | Merged openstack-infra/zuul: Fix race in TestDuplicatePipeline test https://review.openstack.org/395694 | 19:02 |
openstackgerrit | Merged openstack-infra/zuul: Re-enable test_layered_templates test https://review.openstack.org/393445 | 19:07 |
openstackgerrit | Merged openstack-infra/zuul: Re-enable 3 disabled tests https://review.openstack.org/393452 | 19:08 |
pabelanger | Oh, hah | 19:09 |
pabelanger | guess we did need to force history | 19:09 |
pabelanger | http://logs.openstack.org/58/393458/6/gate/gate-zuul-python27-ubuntu-xenial/e9a217f/console.html#_2016-11-09_19_07_11_850332 | 19:09 |
openstackgerrit | Paul Belanger proposed openstack-infra/zuul: First hold jobs for test_duplicate_pipelines https://review.openstack.org/395772 | 19:24 |
pabelanger | clarkb: I _think_ that is what you first suggested right?^ | 19:24 |
clarkb | pabelanger: sort of, I think you still want to use orderedRelease() but ya the hold jobs thing at the beginning is what you need to make orderedRelease do what you want | 19:26 |
pabelanger | ah, okay | 19:26 |
pabelanger | let me add that back | 19:26 |
clarkb | pabelanger: then you can remove the hold_jobs_in_build = False and the release() | 19:27 |
clarkb | basically swap ^ with orderedRelease() | 19:27 |
pabelanger | Ah, I understand now | 19:28 |
openstackgerrit | Paul Belanger proposed openstack-infra/zuul: First hold jobs / orderedRelease for test_duplicate_pipelines https://review.openstack.org/395772 | 19:29 |
jeblair | pabelanger, clarkb: i'd rather make the message check unordered... | 19:47 |
jeblair | since i still think that testing that dup1 fires before dup2 isn't really something we want to do | 19:47 |
clarkb | also fine by me | 19:48 |
pabelanger | sure, let me see how we did it before | 19:48 |
jeblair | i don't have a suggestion for a concise way of doing that, other than just a loop and some counters or something | 19:50 |
openstackgerrit | Paul Belanger proposed openstack-infra/zuul: test_duplicate_pipelines messages are also unordered https://review.openstack.org/395772 | 19:52 |
pabelanger | jeblair: clarkb: ^ is how we did it in zuulv2 | 19:52 |
*** hashar has joined #zuul | 19:52 | |
jeblair | ah yeah that. | 19:52 |
jeblair | at some point, we should probably make a helper function for messages like assertHistory | 19:53 |
pabelanger | makes sense | 19:53 |
hashar | hello there :] | 19:57 |
hashar | I am not much active nowadays, but was wondering what is the status of Zuul v2 / master branch. I got a few patches that might be worth including | 19:58 |
openstackgerrit | Merged openstack-infra/zuul: test_duplicate_pipelines messages are also unordered https://review.openstack.org/395772 | 19:58 |
jeblair | hashar: every now and then we look at important bugfixes we should merge, but generally are trying to keep it fairly stable while we focus on v3. | 20:00 |
hashar | so v2/master is in maintenance mode really | 20:01 |
hashar | I should find some time and actually look at v3 | 20:01 |
jeblair | hashar: that would be great, but note that it doesn't do much useful yet. : | 20:02 |
jeblair | er :) | 20:02 |
hashar | :D | 20:02 |
hashar | maybe I will rebase the few patches I have and hope eventually one will sneak approve them :D | 20:02 |
hashar | on another topic, I welcome the meeting/sessions dedicated solely to Zuul | 20:03 |
hashar | and will try to follow them | 20:03 |
jeblair | hashar: oh cool. how's the time for you? or do you plan on reading logs? | 20:04 |
hashar | I will probably catch the email announcement ahead of the meeting | 20:05 |
hashar | then depending on the topic attempt to join | 20:05 |
hashar | but iirc that is 22:00 UTC which is 11pm local | 20:05 |
hashar | or a bit late when you get kids that wake you up at 7am no matter what! | 20:05 |
hashar | then your meetings have agenda, minutes and logs so it is easy to stay up to date. I just have to do it | 20:06 |
openstackgerrit | Merged openstack-infra/zuul: Enable test_new_patchset_check / test_abandoned_check https://review.openstack.org/393458 | 20:14 |
openstackgerrit | Merged openstack-infra/zuul: Re-enable 4 disable tests https://review.openstack.org/393461 | 20:22 |
openstackgerrit | Merged openstack-infra/zuul: Enable test_statsd test https://review.openstack.org/393479 | 20:23 |
openstackgerrit | Merged openstack-infra/zuul: Enable 4 CRD unit tests https://review.openstack.org/393494 | 20:23 |
openstackgerrit | Merged openstack-infra/zuul: Enable test_two_failed_changes_at_head test https://review.openstack.org/393950 | 20:23 |
openstackgerrit | Merged openstack-infra/zuul: Re-enable test_merger_repack test https://review.openstack.org/393953 | 20:24 |
openstackgerrit | Merged openstack-infra/zuul: Re-enable test_file_head test https://review.openstack.org/393956 | 20:24 |
openstackgerrit | Merged openstack-infra/zuul: Re-enable test_queue_precedence test https://review.openstack.org/394191 | 20:24 |
pabelanger | great | 20:24 |
pabelanger | that should be the stack now | 20:24 |
pabelanger | let me see what is left over | 20:25 |
openstackgerrit | David Shrewsbury proposed openstack-infra/nodepool: WIP: Transition ZK API from dict to object model https://review.openstack.org/394592 | 20:27 |
openstackgerrit | Paul Belanger proposed openstack-infra/zuul: Add tenant support to zuul client https://review.openstack.org/393850 | 20:27 |
Shrews | jeblair: i think 394592 ^^^^ is mostly there now. I just want to run the builder by hand to test various things now before I un-WIP it. | 20:27 |
Shrews | turned out to be quite a large change, but simplifies the code and API considerably, I think | 20:28 |
pabelanger | jeblair: mordred: 391681 enables webapp status again, if you wanted to check if the json format is what you expected | 20:28 |
openstackgerrit | Paul Belanger proposed openstack-infra/zuul: Remove includes and parameter functions https://review.openstack.org/394644 | 20:33 |
openstackgerrit | David Shrewsbury proposed openstack-infra/nodepool: WIP: Transition ZK API from dict to object model https://review.openstack.org/394592 | 20:37 |
openstackgerrit | David Shrewsbury proposed openstack-infra/nodepool: WIP: Transition ZK API from dict to object model https://review.openstack.org/394592 | 20:38 |
openstackgerrit | David Shrewsbury proposed openstack-infra/nodepool: WIP: Transition ZK API from dict to object model https://review.openstack.org/394592 | 20:39 |
Shrews | grr | 20:39 |
mordred | Shrews: grr indeed | 20:45 |
Shrews | i'm just trying to get my daily patch set count up to match pabelanger. i'm falling way behind | 20:47 |
mordred | tell me about it | 20:48 |
pabelanger | happy to help with zuulv3! | 20:50 |
pabelanger | jeblair: could use some guidance on race condition with test_live_reconfiguration. I _think_ we actually have the race condition in zuulv2 too, but with zuulv3 looks like we are hitting it more often. It looks like, the project-merge code is running as we reconfigure the scheduler, which ends up getting cancelled: | 20:55 |
pabelanger | http://logs.openstack.org/88/393488/2/check/dev-zuul-coverage/14e24dd/console.html.gz#_2016-11-03_21_03_56_368758 | 20:55 |
pabelanger | my question is, should we self.gearman_server.hold_jobs_in_queue = True, reconfigure, then release | 20:56 |
pabelanger | or account for the possible abort in the unit test | 20:57 |
clarkb | pabelanger: maybe rework it so that the reconfiguration is additive and not subtractive (though we may want to test both thigns there hrm) | 20:58 |
jeblair | pabelanger: i'm not sure that the old way of doing reconifuration makes sense | 20:58 |
jeblair | (and yes, that's the process of replanting forests with conifurs) | 20:59 |
clarkb | jeblair: you still have the case of any central job config updating right? I agree it makes less sense for project defined tests | 20:59 |
jeblair | clarkb: yeah. that is similar. though i think instead of being triggered by an event, it maybe should just be triggered by a merge to that repo. | 21:00 |
jeblair | clarkb: so maybe the change is really in the triggering, but still reloading the pipelines, etc, stays similar.... | 21:01 |
jeblair | so maybe we should proceed with this for now with that assumption, and see if it holds. | 21:01 |
jeblair | clarkb, pabelanger: we shouldn't be running any ojebs during the reconfiguration in that test. | 21:02 |
jeblair | oh, wait, it's 'hold build' | 21:03 |
jeblair | so yeah, we should be running jobs | 21:03 |
jeblair | they should not be canceled though | 21:03 |
pabelanger | Ya, project-merge is getting run | 21:03 |
pabelanger | okay, I'll see why it is cancelled | 21:03 |
jeblair | Shrews: looks wonderful! i think that will set us up very nicely to reincorporate the builder now. | 21:12 |
jeblair | Shrews: we should rebase my patch series 3839620 through 383967 on that | 21:13 |
jeblair | then go from there | 21:13 |
Shrews | sweeeeet | 21:14 |
pabelanger | jeblair: correct me if I am wrong, but it looks like when we reconfigure the scheduler, we take the old tenant and reenqueue changes into the new tenant, which in doing so, cancels the original job? | 21:25 |
jeblair | pabelanger: that's generally how it should work, aside from the canceling the build part. it's supposed to say "oh, we should have a build of this job, and we do, so move the build over and attach it to the buildset for the item" | 21:27 |
jhesketh | Morning | 21:33 |
*** harlowja has joined #zuul | 21:49 | |
pabelanger | jeblair: Oh, I think it exposed a bug | 21:54 |
*** hashar has quit IRC | 22:16 | |
openstackgerrit | Paul Belanger proposed openstack-infra/zuul: Re-enable test_live_reconfiguration test https://review.openstack.org/393488 | 22:16 |
pabelanger | jeblair: okay, I believe ^ addresses the cancel job issue | 22:17 |
mordred | pabelanger: it's neat when tests expose bugs! | 22:50 |
mordred | jhesketh: morning! | 22:50 |
jhesketh | o/ | 22:50 |
*** harlowja_ has joined #zuul | 23:08 | |
*** harlowja has quit IRC | 23:12 | |
*** saneax-_-|AFK is now known as saneax | 23:15 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!