| *** jangutter has quit IRC | 00:42 | |
| *** igordc has quit IRC | 01:01 | |
| *** armstrongs has joined #zuul | 01:32 | |
| *** armstrongs has quit IRC | 01:46 | |
| *** jeliu_ has joined #zuul | 02:20 | |
| *** jeliu_ has quit IRC | 02:27 | |
| *** bhavikdbavishi has joined #zuul | 02:34 | |
| *** bhavikdbavishi1 has joined #zuul | 02:37 | |
| *** bhavikdbavishi has quit IRC | 02:38 | |
| *** bhavikdbavishi1 is now known as bhavikdbavishi | 02:38 | |
| *** logan- has quit IRC | 03:11 | |
| *** logan- has joined #zuul | 03:15 | |
| *** michael-beaver has quit IRC | 03:21 | |
| *** saneax has joined #zuul | 03:48 | |
| *** jamesmcarthur has quit IRC | 04:50 | |
| *** bolg has joined #zuul | 05:00 | |
| *** igordc has joined #zuul | 05:07 | |
| *** pcaruana has joined #zuul | 05:08 | |
| *** jamesmcarthur has joined #zuul | 05:29 | |
| *** jamesmcarthur has quit IRC | 05:37 | |
| *** raukadah is now known as chandankumar | 06:03 | |
| *** bolg has quit IRC | 06:59 | |
| *** igordc has quit IRC | 07:16 | |
| *** tosky has joined #zuul | 07:21 | |
| *** jangutter has joined #zuul | 07:32 | |
| *** jamesmcarthur has joined #zuul | 07:33 | |
| *** jangutter has quit IRC | 07:33 | |
| *** jangutter has joined #zuul | 07:33 | |
| *** jamesmcarthur has quit IRC | 07:38 | |
| *** igordc has joined #zuul | 07:47 | |
| *** hashar has joined #zuul | 07:54 | |
| *** igordc has quit IRC | 08:06 | |
| *** jangutter has quit IRC | 08:15 | |
| *** jangutter_ has joined #zuul | 08:15 | |
| *** jangutter_ has quit IRC | 08:20 | |
| *** jangutter has joined #zuul | 08:24 | |
| *** panda has quit IRC | 08:54 | |
| *** jamesmcarthur has joined #zuul | 08:56 | |
| *** panda has joined #zuul | 08:57 | |
| *** jamesmcarthur has quit IRC | 09:00 | |
| *** bolg has joined #zuul | 09:04 | |
| *** jamesmcarthur has joined #zuul | 09:28 | |
| *** bjackman has joined #zuul | 09:29 | |
| *** jamesmcarthur has quit IRC | 09:32 | |
| sshnaidm|rover | If I have job defined with specific branches, not including master: https://review.opendev.org/#/c/670168/2/zuul.d/standalone-jobs.yaml - can I still run it on master when specifying "branches: master" for job vars in pipeline config? | 09:35 |
|---|---|---|
| sshnaidm|rover | because it doesn't seem to work: https://review.opendev.org/#/c/670176/1/zuul.d/layout.yaml | 09:36 |
| *** tosky__ has joined #zuul | 10:17 | |
| *** tosky has quit IRC | 10:17 | |
| *** tosky__ is now known as tosky | 10:18 | |
| openstackgerrit | Tobias Henkel proposed zuul/zuul master: Add support for smart reconfigurations https://review.opendev.org/652114 | 10:19 |
| *** hashar has quit IRC | 10:37 | |
| openstackgerrit | Tobias Henkel proposed zuul/zuul master: Add --check-config option to zuul scheduler https://review.opendev.org/542160 | 10:43 |
| *** electrofelix has joined #zuul | 10:45 | |
| zbr | i wonder if there is a way to make the success/failure url more dynamic. if one is unsing them for pointing to unittest results, the link would be invalid if the job failed to produce them. | 10:51 |
| *** bhavikdbavishi has quit IRC | 10:51 | |
| zbr | ideally we should be able to decide which file to point to at the end of the build, not at the start. based on what happens the "outcome" could be a different file. | 10:53 |
| zbr | this could lower the time needed to find the source of the error. | 10:53 |
| mordred | zbr: the plan there is to improve the build page in the web dashboard to be able to display "artifact" urls returned by the job via zuul_return. (you can already return them and they're already in the json in the rest api - they're just not rendered in the build page) - then we'll start linking to build pages - and on the build page there can be any number of arbitrary named urls pointing to specific results | 11:05 |
| mordred | zbr: the main next step is improving the build page to display links to the artifact urls | 11:05 |
| mordred | mhu: I approved the openapi patch, in case you want to add api docs for the new auth related endpoints | 11:06 |
| openstackgerrit | Simon Westphahl proposed zuul/zuul master: Spec for allowing circular dependencies https://review.opendev.org/643309 | 11:08 |
| mordred | tobiash: I don't suppose any of the finger clients support ssl connections do they? :) | 11:13 |
| zbr | mordred: cool. glad to hear that. | 11:14 |
| *** jamesmcarthur has joined #zuul | 11:29 | |
| openstackgerrit | Merged zuul/zuul master: Run cleanup playbooks in job thread https://review.opendev.org/670888 | 11:32 |
| *** jamesmcarthur has quit IRC | 11:33 | |
| *** hashar has joined #zuul | 11:42 | |
| openstackgerrit | Tobias Henkel proposed zuul/zuul master: Add support for smart reconfigurations https://review.opendev.org/652114 | 11:49 |
| openstackgerrit | Tobias Henkel proposed zuul/zuul master: Add --check-config option to zuul scheduler https://review.opendev.org/542160 | 11:50 |
| tobiash | mordred: the finger gateway is a finger gateway that will support it ;) | 11:50 |
| mordred | tobiash: :) | 11:50 |
| tobiash | doh | 11:51 |
| tobiash | you get it | 11:51 |
| mordred | tobiash: I meant cli clients for the folks using it to stream the logs directly | 11:51 |
| mordred | yeah. | 11:51 |
| tobiash | I don't suppose it | 11:51 |
| mordred | nothing would stop us from writing a client that does :) | 11:51 |
| mordred | finger for the 21st century! | 11:51 |
| tobiash | but you could create your own finger client with one line of bash and openssl s_client | 11:51 |
| mordred | hahaha. yes | 11:51 |
| openstackgerrit | Merged zuul/zuul master: web: add OpenAPI documentation https://review.opendev.org/535541 | 11:52 |
| *** bjackman has quit IRC | 11:53 | |
| mordred | tristanC: ^^ woot. time to document the rest of the API now I guess | 11:54 |
| *** hashar_ has joined #zuul | 11:56 | |
| *** hashar has quit IRC | 11:56 | |
| tobiash | mordred: I've responded to your comment on the codeowners patch | 11:59 |
| mordred | cool | 11:59 |
| mordred | tobiash: also - that's a nice pile of code in there - and neat that it's starting to use graphql! | 11:59 |
| tobiash | mordred: so the target for this is essentially to evaluate the green checkmark for required reviews as close as possible because github lacks the required info in their api | 12:00 |
| tobiash | mordred: we switched to graphql because of bugs in subteam handling in the rest api... | 12:00 |
| tobiash | but actually mocking graphql using graphene was much easier than I initially thought | 12:01 |
| mordred | tobiash: from your experience with graphql so far, do you think it will improve our existing gh api interactions? or not worth it at the moment? | 12:01 |
| mordred | yeah - I thought the graphql mocking was pretty neat | 12:01 |
| tobiash | mordred: I think we should explore that where we make more requests in succession, but there is no need to completely convert to graphql | 12:02 |
| mordred | ++ | 12:03 |
| *** jamesmcarthur has joined #zuul | 12:03 | |
| *** hashar_ has quit IRC | 12:05 | |
| *** hashar_ has joined #zuul | 12:06 | |
| *** jamesmcarthur has quit IRC | 12:08 | |
| *** bhavikdbavishi has joined #zuul | 12:10 | |
| *** jamesmcarthur has joined #zuul | 12:15 | |
| *** snapiri has quit IRC | 12:18 | |
| *** snapiri has joined #zuul | 12:18 | |
| fbo | Hi, about nodes labels, let's say I have two nodes. First with labels "fedora30, paris" and the other one with "fedora30, tokyo". Is there a way to tell zuul to pick up in a nodeset the node with labels fedora30 && tokyo ? | 12:22 |
| fbo | https://zuul-ci.org/docs/zuul/user/config.html#attr-nodeset seems to allow only one label | 12:22 |
| *** wznoinsk has joined #zuul | 12:22 | |
| tobiash | fbo: no, that's currently not possible | 12:23 |
| wznoinsk | tobiash, do you think such a function would be useful (upstream openstack or otherwise for others) ? | 12:23 |
| fbo | tobiash: ok, what about the difficulty to implement that ? is there any blocker ? | 12:24 |
| tobiash | useful probably yes, but the current algorithms are not really prepared for this use case so it could get hard to get it right without bigger modifications | 12:25 |
| openstackgerrit | Merged zuul/zuul master: web: add tenant and project scoped, JWT-protected actions https://review.opendev.org/576907 | 12:26 |
| fbo | tobiash: thanks for the informations. We will look if we can find a workaround or provide that feature. | 12:27 |
| openstackgerrit | Merged zuul/zuul master: Allow operator to generate auth tokens through the CLI https://review.opendev.org/636197 | 12:45 |
| AJaeger | mordred: have you seen https://review.opendev.org/670895 ? Feel free to -2 if you don't like it... | 12:46 |
| mordred | AJaeger: lgtm - thanks! | 12:46 |
| AJaeger | great | 12:46 |
| *** badboy has quit IRC | 12:48 | |
| *** hwangbo has quit IRC | 12:49 | |
| *** michael-beaver has joined #zuul | 13:06 | |
| openstackgerrit | Sorin Sbarnea proposed zuul/zuul-jobs master: Install system dependencies for tox-molecule https://review.opendev.org/671029 | 13:13 |
| *** jamesmcarthur has quit IRC | 13:19 | |
| fungi | mordred: telnet-ssl is probably a passable "finger client" too if you're not above entering finger protocol commands (which are exceedingly simple) | 13:20 |
| zbr | fungi: mordred: please let me know if https://review.opendev.org/#/c/671029/ looks ok. | 13:33 |
| Shrews | fbo: tobiash: my initial instinct on matching multiple labels (AND logic) would be exceedingly difficult with the current codebase. | 13:34 |
| Shrews | but not opposed to someone exploring it | 13:35 |
| Shrews | there have been enough feature requests on nodepool in the recent past, i wonder if we should begin planning version 4.0 | 13:37 |
| *** bolg has quit IRC | 13:39 | |
| mordred | yeah - because as I think about the example from above, it makes me think of region-affinity - and the ability to say "I want a fedora30 node and I don't care what cloud-region it's in" or "I want a fedora30 node in ovh-gra" | 13:39 |
| mordred | but that's a whole big mess and not something to enter in to lightly without a ton of thought and design | 13:40 |
| Shrews | mordred: is that a real request from someone or hypothetical? | 13:40 |
| mordred | Shrews: just me thinking based off of the tags in the example being "paris" and "tokyo" | 13:41 |
| Shrews | we should at least begin tracking feature requests somewhere. i'll start doing that for us | 13:42 |
| mordred | Shrews: ++ | 13:42 |
| Shrews | ah | 13:42 |
| clarkb | it would be good to understand the use casebeyond region affinity too. Is the motivation cost or maybe locality? I think different scenarios change what that looks like | 13:48 |
| fbo | Shrews: that a real user request, we help them to move from Jenkins, the POC is almost done but we just discovered that issue. With Jenkins they can match on multiple labels. | 13:49 |
| clarkb | with cost for example nodepool could fogure that out without region aaffinity if clouds advertise their costs | 13:49 |
| clarkb | and data locality might tie into zuuls artifact system and so on | 13:49 |
| fbo | and that not just about the region it can also be about the disk size or whatever hardware specification | 13:49 |
| fbo | so they expect a basic && matching on the labels | 13:50 |
| Shrews | so region affinity and node characteristics. probably separate solutions for those | 13:51 |
| fbo | I was thinking that th logic should only be in Zuul where a nodeset would allow to provide a list of labels where a simple && will apply to select the right node from nodepool | 13:52 |
| Shrews | fbo: no, that logic would lie within nodepool | 13:52 |
| Shrews | it's a request-response communication channel between zuul and nodepool. zuul doesnt do any node selection | 13:54 |
| fbo | ok well I don't know well that part | 13:54 |
| fbo | Shrews: where it is done in the Zuul code, I mean the call to nodepool ? | 13:55 |
| Shrews | it creates a request via zookeeper | 13:58 |
| Shrews | https://opendev.org/zuul/zuul/src/branch/master/zuul/nodepool.py#L107 | 13:58 |
| fbo | thanks | 13:59 |
| Shrews | fbo: to be clear, what you want to do would require changes to both zuul and nodepool, but the actual node selection logic is all nodepool | 14:01 |
| clarkb | it is also a somewhat different approach | 14:02 |
| clarkb | with jenkins you are asking jenkins to give the job any resource which matches some criteria (ignoring the other criteria) | 14:02 |
| clarkb | with zuul+nodepool the job says I need exactly this give it to me | 14:03 |
| clarkb | So we may have to do quite a bit of tweaking to represent the less concrete node requests | 14:04 |
| *** bolg has joined #zuul | 14:04 | |
| tobiash | yes, this use case seems more targeted to static nodes | 14:04 |
| tobiash | for cloud providers I don't really see the need for such a mechanism | 14:05 |
| fbo | tobiash: clarkb yes exact it is related to static nodes | 14:09 |
| *** bolg has quit IRC | 14:13 | |
| corvus | fbo: as a workaround you may be able to add a "fedora30-tokyo" label to the node? | 14:16 |
| corvus | obviously not a scalable solution, but since static nodes can have multiple lables, you could add the cartesian product of all of them | 14:17 |
| tobiash | multiple labels on static nodes however currently don't work well yet as well (https://review.opendev.org/667371 tries to address this) | 14:18 |
| corvus | tobiash: yes, though fbo already suggested that the node had 2 labels | 14:19 |
| fbo | corvus: tobiash yes and it advised in the doc to have one pool per label so it does not ease the situation | 14:19 |
| fbo | https://zuul-ci.org/docs/nodepool/configuration.html#attr-providers.[static].pools | 14:19 |
| tobiash | fbo: that's exactly what 667371 tries to fix | 14:20 |
| tobiash | fbo: so maybe you want to try this | 14:20 |
| corvus | right, but that may be a shorter path to get something working than a redesign (which, i agree with Shrews, would be good). i'm just trying to help come up with short *and* long-term options. | 14:21 |
| Shrews | oh static driver. that multi label logic is OR. the request is for AND | 14:21 |
| Shrews | adding support for both to static will be... interesting | 14:22 |
| corvus | Shrews: yeah, i agree, i don't think we get multi-label *requests* without a redesign. but there's a workaround for simple cases with concatenation: labels: ['fedora30', 'tokyo', 'fedora30-tokyo']. you want fedora30 && tokyo, you ask for fedora30-tokyo. that would work today, it just won't work with perfect efficiency because of 667371. but there is hope we can improve that without a redesign. | 14:24 |
| tobiash | yes, so corvus' suggestion combined with 667371 would probably the easiest short term option | 14:24 |
| Shrews | yeah | 14:25 |
| fbo | ok then I'll have a deeper look at 667371 and propose that concat solution. thanks everybody ! | 14:26 |
| *** bolg has joined #zuul | 14:34 | |
| *** bolg has quit IRC | 14:35 | |
| *** bolg has joined #zuul | 14:36 | |
| *** bolg has quit IRC | 14:42 | |
| *** jeliu_ has joined #zuul | 14:44 | |
| *** jeliu_ has quit IRC | 14:44 | |
| *** jeliu_ has joined #zuul | 14:45 | |
| *** bhavikdbavishi has quit IRC | 14:52 | |
| *** jamesmcarthur has joined #zuul | 15:03 | |
| *** bolg has joined #zuul | 15:08 | |
| *** bolg has quit IRC | 15:12 | |
| *** saneax has quit IRC | 15:20 | |
| *** chandankumar is now known as raukadah | 15:26 | |
| *** hashar_ has quit IRC | 15:31 | |
| *** bolg has joined #zuul | 15:37 | |
| openstackgerrit | Jeff Liu proposed zuul/zuul-operator master: [WIP] Verify Operator Pod Running https://review.opendev.org/670395 | 15:39 |
| openstackgerrit | Stephen Finucane proposed zuul/zuul master: web: Add warning about incompleteness of OpenAPI spec https://review.opendev.org/671086 | 15:41 |
| *** bolg has quit IRC | 15:42 | |
| *** igordc has joined #zuul | 15:47 | |
| *** mattw4 has joined #zuul | 16:17 | |
| *** mattw4 has quit IRC | 16:19 | |
| *** mattw4 has joined #zuul | 16:19 | |
| *** hwangbo has joined #zuul | 16:26 | |
| *** jamesmcarthur has quit IRC | 16:29 | |
| *** hashar has joined #zuul | 16:30 | |
| *** jamesmcarthur has joined #zuul | 16:31 | |
| *** electrofelix has quit IRC | 16:35 | |
| openstackgerrit | Merged zuul/zuul master: web: Add warning about incompleteness of OpenAPI spec https://review.opendev.org/671086 | 16:52 |
| *** aluria has quit IRC | 17:06 | |
| *** hashar has quit IRC | 17:12 | |
| *** igordc has quit IRC | 17:17 | |
| *** jamesmcarthur has quit IRC | 17:43 | |
| *** igordc has joined #zuul | 17:44 | |
| *** electrofelix has joined #zuul | 18:00 | |
| *** tosky has quit IRC | 18:01 | |
| *** electrofelix has quit IRC | 18:03 | |
| *** jamesmcarthur has joined #zuul | 18:07 | |
| openstackgerrit | Matthieu Huin proposed zuul/zuul master: Zuul CLI: allow access via REST https://review.opendev.org/636315 | 18:13 |
| openstackgerrit | Matthieu Huin proposed zuul/zuul master: Add Authorization Rules configuration https://review.opendev.org/639855 | 18:13 |
| openstackgerrit | Matthieu Huin proposed zuul/zuul master: Web: plug the authorization engine https://review.opendev.org/640884 | 18:14 |
| *** hashar has joined #zuul | 18:15 | |
| *** jamesmcarthur has quit IRC | 18:25 | |
| *** hashar has quit IRC | 18:36 | |
| *** jamesmcarthur has joined #zuul | 18:42 | |
| *** jamesmcarthur has quit IRC | 18:51 | |
| *** jamesmcarthur has joined #zuul | 18:52 | |
| *** jamesmcarthur_ has joined #zuul | 18:53 | |
| *** jamesmcarthur_ has quit IRC | 18:54 | |
| *** mattw4 has quit IRC | 18:55 | |
| *** jamesmcarthur has quit IRC | 18:57 | |
| *** jamesmcarthur has joined #zuul | 18:58 | |
| *** jamesmcarthur_ has joined #zuul | 18:59 | |
| *** hashar has joined #zuul | 19:00 | |
| *** jamesmcarthur has quit IRC | 19:03 | |
| *** jamesmcarthur_ has quit IRC | 19:04 | |
| *** jamesmcarthur has joined #zuul | 19:05 | |
| *** jamesmcarthur_ has joined #zuul | 19:06 | |
| *** jamesmcarthur has quit IRC | 19:10 | |
| *** jamesmcarthur_ has quit IRC | 19:25 | |
| *** jamesmcarthur has joined #zuul | 19:28 | |
| *** jamesmcarthur_ has joined #zuul | 19:29 | |
| *** jamesmcarthur_ has quit IRC | 19:32 | |
| *** jamesmcarthur_ has joined #zuul | 19:33 | |
| *** jamesmcarthur has quit IRC | 19:33 | |
| *** jamesmcarthur has joined #zuul | 19:34 | |
| *** jamesmcarthur_ has quit IRC | 19:37 | |
| *** jamesmcarthur has quit IRC | 19:53 | |
| *** jamesmcarthur has joined #zuul | 19:53 | |
| *** jamesmcarthur_ has joined #zuul | 19:54 | |
| *** jamesmcarthur has quit IRC | 19:58 | |
| *** hashar has quit IRC | 20:02 | |
| *** mattw4 has joined #zuul | 20:04 | |
| *** rfolco is now known as rfolco_l8r | 20:39 | |
| *** jamesmcarthur_ has quit IRC | 20:45 | |
| *** sgw has quit IRC | 20:45 | |
| *** pcaruana has quit IRC | 21:01 | |
| daniel2 | Still having issues with the zuul docker stuff. I restored most of the original config. This is the errors I'm getting https://shafer.cc/paste/view/raw/e527fb92 Thats the entire docker compose debug output basically. | 21:35 |
| *** sgw has joined #zuul | 21:38 | |
| corvus | daniel2: something is wrong with mysql | 21:40 |
| corvus | (the scheduler is waiting for mysql to start, and everything else is waiting for the scheduler to start gearman) | 21:41 |
| corvus | daniel2: mysql_1 | Version: '10.4.6-MariaDB-1:10.4.6+maria~bionic' socket: '/var/run/mysqld/mysqld.sock' port: 0 mariadb.org binary distribution | 21:41 |
| corvus | daniel2: that should say "port: 3306" | 21:41 |
| daniel2 | oh weird.. | 21:41 |
| corvus | daniel2: since it doesn't, i think that means it's not listening on tcp | 21:41 |
| corvus | daniel2: any chance something is listening on that port? (container or otherwise?) | 21:43 |
| daniel2 | corvus: no, at this point nothing is running on this server except docker and ssh | 22:40 |
| daniel2 | corvus: stopping and starting the containers fixed it | 22:47 |
| daniel2 | i think it timed out too soon | 22:47 |
| corvus | daniel2: yay? :) | 22:52 |
| SpamapS | first boot of mysql takes a bit | 23:16 |
| SpamapS | especially if you have slow IO | 23:16 |
| SpamapS | it makes binlogs and stuff | 23:16 |
| *** tobiash has quit IRC | 23:51 | |
| *** tobiash has joined #zuul | 23:52 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!