*** jamielennox is now known as jamielennox|away | 00:38 | |
*** jamielennox|away is now known as jamielennox | 03:22 | |
*** bhavik1 has joined #zuul | 05:21 | |
*** Cibo_ has quit IRC | 07:51 | |
*** Cibo_ has joined #zuul | 07:53 | |
*** Cibo_ has quit IRC | 08:05 | |
*** hashar has joined #zuul | 08:31 | |
*** isaacb has joined #zuul | 08:52 | |
*** openstackgerrit has quit IRC | 09:02 | |
*** isaacb has quit IRC | 09:09 | |
mordred | jeblair: the first time we start v3 with all of openstack projects imported, it's sure gonna be fun to watch that generation chug :) | 10:01 |
---|---|---|
mordred | morning jhesketh | 10:31 |
*** adamw has quit IRC | 10:31 | |
*** adamw has joined #zuul | 10:33 | |
jhesketh | o/ mordred | 10:42 |
jhesketh | a little past morning, but all the same :-) | 10:43 |
*** hashar is now known as hasharLunch | 10:44 | |
mordred | jhesketh: bah. it's morning in both of the timezones I'm in right now ;) | 10:46 |
jhesketh | Heh that's fair :-) | 10:48 |
*** Cibo_ has joined #zuul | 11:00 | |
*** bhavik1 has quit IRC | 11:08 | |
*** Cibo_ has quit IRC | 11:39 | |
*** _ari_ is now known as _ari_|gone | 11:59 | |
dmsimard | pabelanger: would love your input on https://review.openstack.org/#/c/442370/ -- it's something we're hitting in RDO land | 13:27 |
pabelanger | dmsimard: likely want jeblair to comment on it | 13:30 |
dmsimard | ack | 13:30 |
dmsimard | jeblair: if you have a chance ^ ? tl;dr have Zuul properly fallback to HEAD instead of 'master' because 'master' might not be the HEAD or might not even exist at all (ex: github.com/ansible/ansible has no master branch) | 13:31 |
Shrews | pabelanger: jeblair: issues on nodepool-launcher. investigating | 13:32 |
pabelanger | Ah, I haven't checked this morning | 13:33 |
*** openstackgerrit has joined #zuul | 13:37 | |
openstackgerrit | David Shrewsbury proposed openstack-infra/nodepool feature/zuulv3: Unlock request if it disappears https://review.openstack.org/447501 | 13:37 |
Shrews | pabelanger: jeblair: ^^^ that should fix the bug | 13:38 |
Shrews | going to restart it to get it going again | 13:38 |
pabelanger | ++ | 13:41 |
mordred | dmsimard: definitely one jeblair should look at - and likely want to forward port to v3 - not sure myself what the full ramifications might be | 13:47 |
dmsimard | mordred: interesting, it looked straightforward enough to naive me | 13:48 |
pabelanger | Shrews: I'm not seeing nl01.o.o logging anything | 13:49 |
Shrews | pabelanger: me neither | 13:49 |
Shrews | pabelanger: though it's running. | 13:49 |
pabelanger | except our exceptions | 13:50 |
Shrews | pabelanger: got to head out... at the car shop. back online in a bit | 13:50 |
pabelanger | ack | 13:50 |
openstackgerrit | Merged openstack-infra/zuul master: Update hacking dependency https://review.openstack.org/442533 | 13:52 |
mordred | dmsimard: it is entirely possible that is is ... but any time cloning behavior gets involved, I definitely want jeblair to look | 13:59 |
mordred | Shrews: didn't we have something late last week where something went to the bad place and stopped logging and it was because of a thread deadlock? | 14:00 |
mordred | Shrews: is it possible we didn't roll the fix for that out? | 14:00 |
*** hasharLunch is now known as hashar | 14:05 | |
*** herlo has joined #zuul | 14:23 | |
Shrews | mordred: mmm, not that i immediately recall? | 14:34 |
Shrews | i mean, nodepool-launcher is clearly doing things (it drained the request pool), but just not logging | 14:37 |
Shrews | lsof doesn't show any log files opened for the process | 14:42 |
Shrews | ???? | 14:42 |
mordred | Shrews: :( | 14:43 |
pabelanger | we did merge a change recently for to refactor base apps | 14:43 |
pabelanger | logs in job look correct | 14:44 |
* Shrews so confused | 14:45 | |
pabelanger | mordred: Shrews: 445674 seems to be the issue | 14:52 |
pabelanger | I don't think we are call setup_logging when we run as a daemon | 14:52 |
Shrews | oh, hrm | 14:53 |
Shrews | for some reason, that I didn't spot before, that also removes SIGUSR2 handling | 14:55 |
Shrews | pabelanger: i think we should revert | 14:55 |
pabelanger | Shrews: was going to suggest the same | 14:56 |
Shrews | oh, i see sigusr2 now, in the NodepoolDaemonApp | 14:56 |
openstackgerrit | David Shrewsbury proposed openstack-infra/nodepool feature/zuulv3: Revert "Refactor nodepool apps into base app" https://review.openstack.org/447534 | 15:00 |
Shrews | jamielennox: fyi ^^^^ | 15:03 |
Shrews | I'm going to stop nodepool-launcher. If something goes wrong, we'll have no logs. | 15:12 |
mordred | Shrews: +2 | 15:13 |
*** jamielennox has quit IRC | 15:17 | |
openstackgerrit | Tobias Henkel proposed openstack-infra/zuul feature/zuulv3: Change mutex to counting semaphore https://review.openstack.org/442563 | 15:20 |
clarkb | fwiw there is one major reason to not change HEAD when using git. HEAD is not something that gets updated on push/pull | 15:36 |
clarkb | support non master head in zuul is probably a good idea ebcause some people will do it anyways, but may we should have a note that says "you really don't want to do this because half the world breaks when you do" | 15:37 |
*** jamielennox has joined #zuul | 16:19 | |
*** jamielennox is now known as jamielennox|away | 16:21 | |
jeblair | Shrews, jamielennox|away: left a comment on 447534 | 16:28 |
*** Cibo_ has joined #zuul | 16:50 | |
*** hashar is now known as hasharAway | 16:59 | |
openstackgerrit | Paul Belanger proposed openstack-infra/nodepool feature/zuulv3: Remove SSH support from nodepool https://review.openstack.org/446683 | 16:59 |
jeblair | Shrews: what's the fix for the failure in 447534 you're referring to? | 17:13 |
Shrews | jeblair: c4bcfd8 | 17:14 |
jeblair | Shrews: zuul tests the proposed change merged to the head of the branch | 17:15 |
jlk | Shrews: how did you guess my password? | 17:15 |
jeblair | Shrews: so whether the original was built on that or not doesn't matter; that fix is in the tree being tested | 17:15 |
Shrews | jeblair: then maybe something new. wasn't sure how gerrit did the revert | 17:16 |
jeblair | Shrews: there is a further problem. i have just been able to modify the test to expose it, and i understand it now | 17:16 |
Shrews | sweet | 17:16 |
jeblair | Shrews: i will push up a modified (failing) test in a little while with an explanation | 17:16 |
SpamapS | FYI: I have discovered a great way to audit C programs line by line.... port them to Rust! :) | 17:48 |
SpamapS | https://github.com/SpamapS/rubble <-- (my rust port of bubblewrap ;) | 17:48 |
clarkb | SpamapS: there is even a compiler for that | 17:48 |
SpamapS | not even close to done. :) | 17:48 |
SpamapS | clarkb: no, the point is to _read_ each line, understand it, and then do the equivalent in Rust. :) | 17:48 |
SpamapS | clarkb: BUT, I did not know that, so.. COOL | 17:48 |
clarkb | I mean technically thats what a compiler does :P | 17:48 |
SpamapS | clarkb: except compilers are really bad at doing security audits and providing me with human readable action items. | 17:49 |
SpamapS | But yes, technically I am compilering | 17:49 |
clarkb | https://news.ycombinator.com/item?id=12056230 | 17:49 |
openstackgerrit | Merged openstack-infra/nodepool feature/zuulv3: Revert "Refactor nodepool apps into base app" https://review.openstack.org/447534 | 17:52 |
*** Cibo_ has quit IRC | 17:57 | |
clarkb | https://github.com/jameysharp/corrode/blob/master/src/Language/Rust/Corrode/C.md really makes me miss haskell now | 18:00 |
SpamapS | clarkb: neat | 18:01 |
clarkb | (it should be noted most people don't write haskell that way in my experience, but when they do its so nice) | 18:02 |
SpamapS | It should probably be noted that most people don't write Haskell. :-/ | 18:07 |
SpamapS | (which is sad because it's a lovely language) | 18:07 |
pabelanger | jeblair: should we considering removing our auth settings (username, password, auth-url, etc) from a provider in favor of using os-client-config in nodepool.yaml? | 18:08 |
clarkb | pabelanger: I think that if we intend on not supporting nodepool.yaml's settings as first class (pretty sure we don't anymore and occ is required at this point) then we should do that | 18:08 |
clarkb | pabelanger: so my vote is yes remove them and explciitly require clouds.yaml | 18:08 |
SpamapS | +1 for using the standard config file. | 18:09 |
pabelanger | clarkb: right, I guess I should be asking that. Are they first class or should be just expect clouds.yaml | 18:09 |
clarkb | right now they are mostly just there to confuse people I think :) | 18:10 |
clarkb | (I myself have been confused by settings left over in nodepool.yaml that are now mostly clouds.yaml's responsibility) | 18:10 |
pabelanger | agree | 18:10 |
Shrews | pabelanger++ fyi, mordred was just mentioning doing just that when last i chatted with him | 18:13 |
openstackgerrit | James E. Blair proposed openstack-infra/nodepool feature/zuulv3: Break test_node_assignment_at_quota https://review.openstack.org/447630 | 18:13 |
jeblair | Shrews: commit message explains problem, i hope | 18:14 |
Shrews | jeblair: yeah, just reading. i was narrowing in on that bit of code | 18:14 |
Shrews | fun bug | 18:15 |
jeblair | Shrews: yeah, i got as far as considering maybe we can move addition to the nodeset into the request hander... i haven't gotten far enough to figure out if there are downsides to that | 18:15 |
jeblair | pabelanger: yeah that sounds good | 18:16 |
*** Cibo_ has joined #zuul | 18:17 | |
*** jamielennox|away is now known as jamielennox | 18:24 | |
*** dkranz has joined #zuul | 18:24 | |
Shrews | jeblair: before we restart n-l, can we get this fix merged too? https://review.openstack.org/447501 | 18:26 |
Shrews | assuming it looks ok | 18:26 |
*** Cibo_ has quit IRC | 18:26 | |
jeblair | Shrews: will do. do you want to take over the work on test_node_assignment_at_quota or would you like me to work on a solution? | 18:28 |
Shrews | jeblair: i can take it on if you want | 18:29 |
jeblair | Shrews: it's yours :) | 18:29 |
jeblair | Shrews: 501 +3 | 18:29 |
Shrews | jeblair: ugh, that new bug is biting us pretty often now :( | 18:36 |
jlk | Apologies for the incoming spew. | 18:58 |
openstackgerrit | Jesse Keating proposed openstack-infra/zuul feature/zuulv3: support github pull reqeust labels https://review.openstack.org/444511 | 18:58 |
openstackgerrit | Jesse Keating proposed openstack-infra/zuul feature/zuulv3: Allow github trigger to match on branches/refs https://review.openstack.org/445625 | 18:58 |
openstackgerrit | Jesse Keating proposed openstack-infra/zuul feature/zuulv3: Set filter according to PR/Change in URL https://review.openstack.org/446782 | 18:58 |
openstackgerrit | Jesse Keating proposed openstack-infra/zuul feature/zuulv3: Support for dependent pipelines with github https://review.openstack.org/445292 | 18:58 |
openstackgerrit | Jesse Keating proposed openstack-infra/zuul feature/zuulv3: Add basic Github Zuul Reporter. https://review.openstack.org/443323 | 18:58 |
openstackgerrit | Jesse Keating proposed openstack-infra/zuul feature/zuulv3: Configurable SSH access to GitHub https://review.openstack.org/444034 | 18:58 |
openstackgerrit | Jesse Keating proposed openstack-infra/zuul feature/zuulv3: Add 'push' and 'tag' github webhook events. https://review.openstack.org/443947 | 18:58 |
openstackgerrit | Jesse Keating proposed openstack-infra/zuul feature/zuulv3: Add 'pr-comment' github webhook event https://review.openstack.org/443959 | 18:58 |
openstackgerrit | Jesse Keating proposed openstack-infra/zuul feature/zuulv3: Support for github commit status https://review.openstack.org/444060 | 18:58 |
openstackgerrit | Jesse Keating proposed openstack-infra/zuul feature/zuulv3: Better merge message for GitHub pull reqeusts https://review.openstack.org/445644 | 18:58 |
openstackgerrit | Jesse Keating proposed openstack-infra/zuul feature/zuulv3: Encapsulate determining the event purpose https://review.openstack.org/445242 | 18:58 |
openstackgerrit | Jesse Keating proposed openstack-infra/zuul feature/zuulv3: Allow using webapp from connections https://review.openstack.org/439831 | 18:58 |
openstackgerrit | Jesse Keating proposed openstack-infra/zuul feature/zuulv3: Merge pull requests from github reporter https://review.openstack.org/444463 | 18:58 |
openstackgerrit | Jesse Keating proposed openstack-infra/zuul feature/zuulv3: Support GitHub PR webhooks https://review.openstack.org/439834 | 18:58 |
openstackgerrit | Jesse Keating proposed openstack-infra/zuul feature/zuulv3: GitHub file matching support https://review.openstack.org/446113 | 18:58 |
openstackgerrit | Jesse Keating proposed openstack-infra/zuul feature/zuulv3: Log GitHub API rate limit https://review.openstack.org/446150 | 18:58 |
Shrews | jeblair: i believe we can just append to nodeset at build time, rather than when it becomes ready. | 19:02 |
openstackgerrit | Paul Belanger proposed openstack-infra/nodepool feature/zuulv3: Remove legacy openstack settings from nodepool.yaml https://review.openstack.org/447647 | 19:02 |
openstackgerrit | David Shrewsbury proposed openstack-infra/nodepool feature/zuulv3: Fix test_node_assignment_at_quota https://review.openstack.org/447630 | 19:10 |
openstackgerrit | David Shrewsbury proposed openstack-infra/nodepool feature/zuulv3: Fix test_node_assignment_at_quota https://review.openstack.org/447630 | 19:11 |
*** hasharAway is now known as hashar | 19:22 | |
*** _ari_|gone is now known as _ari_ | 19:31 | |
openstackgerrit | Paul Belanger proposed openstack-infra/nodepool feature/zuulv3: Remove legacy openstack settings from nodepool.yaml https://review.openstack.org/447647 | 19:47 |
pabelanger | okay that removes username, password, auth-url, project-name, project-id, service-type and service-name | 19:58 |
pabelanger | while keeping region-name, api-timeout and image-type for the moment | 19:58 |
openstackgerrit | Merged openstack-infra/zuul feature/zuulv3: Fix testtools.run syntax for README.rst https://review.openstack.org/447219 | 20:04 |
openstackgerrit | Paul Belanger proposed openstack-infra/nodepool feature/zuulv3: Remove keypair from provider section https://review.openstack.org/447673 | 20:09 |
clarkb | pabelanger: I don't think we can remove region name fwiw | 20:09 |
clarkb | pabelanger: since that is necessary info to know where in a cloud to boot insances | 20:09 |
jeblair | pabelanger: left a couple comments on that | 20:10 |
*** pleia2_ has joined #zuul | 20:18 | |
*** toabctl_ has joined #zuul | 20:22 | |
openstackgerrit | Paul Belanger proposed openstack-infra/nodepool feature/zuulv3: Remove keypair from provider section https://review.openstack.org/447673 | 20:27 |
openstackgerrit | Paul Belanger proposed openstack-infra/nodepool feature/zuulv3: Remove legacy openstack settings from nodepool.yaml https://review.openstack.org/447647 | 20:27 |
openstackgerrit | Paul Belanger proposed openstack-infra/nodepool feature/zuulv3: Remove SSH support from nodepool https://review.openstack.org/446683 | 20:27 |
openstackgerrit | Paul Belanger proposed openstack-infra/nodepool feature/zuulv3: Remove keypair from provider section https://review.openstack.org/447673 | 20:28 |
openstackgerrit | Paul Belanger proposed openstack-infra/nodepool feature/zuulv3: Remove legacy openstack settings from nodepool.yaml https://review.openstack.org/447647 | 20:28 |
*** toabctl has quit IRC | 20:29 | |
*** pleia2 has quit IRC | 20:29 | |
*** toabctl_ is now known as toabctl | 20:29 | |
openstackgerrit | Paul Belanger proposed openstack-infra/nodepool feature/zuulv3: Remove keypair from provider section https://review.openstack.org/447673 | 20:38 |
openstackgerrit | Paul Belanger proposed openstack-infra/nodepool feature/zuulv3: Remove legacy openstack settings from nodepool.yaml https://review.openstack.org/447647 | 20:38 |
openstackgerrit | Merged openstack-infra/nodepool feature/zuulv3: Unlock request if it disappears https://review.openstack.org/447501 | 20:39 |
pabelanger | looks like our experimental job is broken, but I think I updated then nodepool side correctly | 20:39 |
pabelanger | I'll look at zuul shortly | 20:39 |
Shrews | pabelanger: i stopped nodepool-launcher this morning | 20:42 |
Shrews | unless you restarted it | 20:43 |
pabelanger | Shrews: I have not | 20:43 |
Shrews | perhaps you weren't referring to the v3 jobs | 20:44 |
pabelanger | jeblair: Shrews: looking at zuul, I don't see where ZooKeeperConnectionConfig is used any more, aside from our integration job. Can we just remove that now? | 20:45 |
Shrews | dunno about zuul. it's used in nodepool though | 20:46 |
pabelanger | Shrews: I think we opted for just letting kazoo client parse it for zuul | 20:47 |
pabelanger | 127.0.0.1:2181:/chroot | 20:47 |
clarkb | ya since the format we were going to use was one already accepted directly by kazoo | 20:48 |
openstackgerrit | Paul Belanger proposed openstack-infra/zuul feature/zuulv3: [WIP] Fix zuul-nodepool-integration job https://review.openstack.org/447683 | 20:48 |
pabelanger | ya, we should consider making the same change for nodepool | 20:50 |
pabelanger | to keep them in sync | 20:50 |
Shrews | well, for nodepool, that would make the code *inconsistent* since we read everything from the config into objects. i think it was, originally, w/o the class. but the class was added for consistency, iirc | 21:15 |
Shrews | but that was many reviews ago | 21:15 |
Shrews | i can go either way | 21:16 |
*** herlo has quit IRC | 21:44 | |
jhesketh | Morning | 21:54 |
jeblair | it's zuul meeting time in #openstack-meeting-alt | 22:00 |
*** hashar has quit IRC | 22:01 | |
rbergeron | oh hey! eetin' time | 22:03 |
rbergeron | err. meetin' time | 22:03 |
Shrews | rbergeron: right on both accounts | 22:04 |
rbergeron | aye. i just got past "failing to pick up my kid from school and then rushing over late like an idiot" time | 22:04 |
Shrews | jamielennox: now that you're awake (-ish?), just re-pointing out we had to revert one of your changes due to logging issues | 22:26 |
jamielennox | Shrews: yea, i did see the message, have had a few meetings this morning so haven't had a chance to look closely yet | 22:27 |
jamielennox | Shrews: but i swapped the order of the logging setup? | 22:28 |
Shrews | jamielennox: i think that's what jeblair speculated. not sure | 22:28 |
jamielennox | annoying that that would make a difference but i can look into it, i hadn't noticed it in my env but there were lots of things running there and i easily could have missed it | 22:28 |
openstackgerrit | Paul Belanger proposed openstack-infra/zuul feature/zuulv3: Add ansible-lint to tox (pep8) https://review.openstack.org/438276 | 22:40 |
SpamapS | jlk: If you can add https://review.openstack.org/441441 to your review queue... it would help to get your perspective on role variable naming, given your position in the ansible community. Thanks. :) | 23:35 |
jlk | reading | 23:36 |
*** jamielennox is now known as jamielennox|away | 23:36 | |
*** jamielennox|away is now known as jamielennox | 23:40 | |
jamielennox | jeblair, pabelanger: so SpamapS and i were talking last week about deployment variables, and i think neither of us knew enough about the longer term plans to come to a conclusion | 23:47 |
jamielennox | so i'm thinking about the concept of a stdlib of roles provided by zuul, and the tree nature that jobs currently have | 23:47 |
jamielennox | looking at https://github.com/openstack-infra/zuul/blob/feature/zuulv3/.zuul.yaml and a bit of the discussion i've had with people we are expecting that jobs will inherit from some standard base | 23:49 |
jamielennox | this base job is very specific to a deployment, eg: success-url: http://zuulv3-dev.openstack.org/logs/{build.uuid}/ | 23:50 |
jamielennox | how do you see a stdlib tox job working across deployments when they must be parented to this base job to be provisioned and logs collected? | 23:51 |
jeblair | jamielennox: i think that we should aim for *roles* to certainly be a unit of reusability. i'm not sure that we will succeed in making jobs as reusable. | 23:52 |
jeblair | jamielennox: so i think we will end up with a tox role in the stdlib (along with the git repo setup, and log copying, and 'make', etc, roles) | 23:53 |
clarkb | jeblair: I'm reviewing the nodepool change around allocation mentioned in the meeting earlier and is the stuff changed in https://review.openstack.org/#/c/447630/3/nodepool/tests/test_nodepool.py purely for readability? | 23:54 |
jamielennox | jeblair: ok, that's a sufficient answer, it also means we get control over all the jobs available in our deploy and can test them out before rolling htem publicly | 23:54 |
jeblair | clarkb: nope, it's critical for exposing the bug. | 23:55 |
clarkb | jeblair: it appears to be equivalent though? | 23:55 |
jeblair | clarkb: the first paragraph of the commit message should roughly correspond with that | 23:55 |
jamielennox | jeblair: so then a more specific version of the same question, i want us to define out own project-config repo (obviously), however we have multiple developers with their own dev environments, we (potentially) have a staging environment, vagrant envs etc | 23:55 |
jeblair | clarkb: it's the one-at-a-time nature that's important | 23:55 |
jamielennox | is there a way currently that i can template out something from project-config such that we could reuse it in all those environments without having to redefine all the jobs per env | 23:56 |
clarkb | jeblair: gotcha, shouldn't there be an explicit guard between deleting 1 and deleting 2 then? | 23:56 |
clarkb | something like did we start booting 3? | 23:56 |
jamielennox | with the caveat i don't want to have to maintain a centralized list of everyones development environment with their URLs hardcoded | 23:56 |
jeblair | clarkb: yeah, that's a good idea i think | 23:57 |
jamielennox | jeblair: my current solution is: https://github.com/jamielennox/zuul/commit/eb9a63ca342f64c6a973b15b06a9e36789f19a55 | 23:57 |
jeblair | jamielennox: you didn't submit that upstream :( | 23:58 |
jamielennox | which would allow me when deploying zuul to have my deployment tooling drop a yml vars file in /etc/zuul with things like logs_url | 23:58 |
jamielennox | jeblair: no, it was caught up in the middle of that conversation with SpamapS | 23:58 |
jamielennox | and i then heard you might already have a solution | 23:58 |
clarkb | jamielennox: I left a comment based on ^ | 23:59 |
clarkb | er jeblair ^ | 23:59 |
jamielennox | i wanted to check i wasn't crazy first - though i don't mind being told i'm crazy upstream either | 23:59 |
jeblair | jamielennox: :) | 23:59 |
jeblair | jamielennox: have you read the conversation i had with SpamapS? | 23:59 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!