| *** tosky has quit IRC | 00:36 | |
| *** jamesmcarthur has joined #zuul | 01:17 | |
| *** cloudnull has quit IRC | 01:29 | |
| *** jamesmcarthur has quit IRC | 01:42 | |
| *** jamesmcarthur has joined #zuul | 01:42 | |
| *** jamesmcarthur has quit IRC | 01:47 | |
| *** zenkuro has quit IRC | 02:18 | |
| *** jamesmcarthur has joined #zuul | 02:36 | |
| *** jamesmcarthur has quit IRC | 02:37 | |
| *** jamesmcarthur has joined #zuul | 03:44 | |
| *** jamesmcarthur has quit IRC | 04:15 | |
| *** jamesmcarthur has joined #zuul | 04:17 | |
| *** saneax has joined #zuul | 04:32 | |
| *** amotoki has quit IRC | 04:44 | |
| *** amotoki has joined #zuul | 04:44 | |
| *** evrardjp has quit IRC | 05:33 | |
| *** evrardjp has joined #zuul | 05:35 | |
| *** ykarel has joined #zuul | 05:41 | |
| *** ykarel_ has joined #zuul | 05:54 | |
| *** ykarel has quit IRC | 05:57 | |
| *** ykarel__ has joined #zuul | 06:05 | |
| *** ykarel__ is now known as ykarel | 06:07 | |
| *** ykarel_ has quit IRC | 06:08 | |
| *** jamesmcarthur has quit IRC | 06:11 | |
| *** jamesmcarthur has joined #zuul | 06:12 | |
| *** jamesmcarthur has quit IRC | 06:17 | |
| *** jamesmcarthur has joined #zuul | 06:42 | |
| *** reiterative has quit IRC | 06:55 | |
| *** reiterative has joined #zuul | 06:56 | |
| *** jfoufas1 has joined #zuul | 07:46 | |
| *** rpittau|afk is now known as rpittau | 07:47 | |
| openstackgerrit | Merged zuul/zuul master: tutorial: Rework quick-start and prepare for other tutorials https://review.opendev.org/c/zuul/zuul/+/732066 | 07:48 |
|---|---|---|
| *** jcapitao has joined #zuul | 07:51 | |
| *** jpena|off is now known as jpena | 07:52 | |
| *** hashar has joined #zuul | 08:22 | |
| *** piotrowskim has joined #zuul | 08:24 | |
| *** jamesmcarthur has quit IRC | 08:39 | |
| *** tosky has joined #zuul | 08:49 | |
| openstackgerrit | Merged zuul/zuul-jobs master: upload-artifactory: no_log upload task https://review.opendev.org/c/zuul/zuul-jobs/+/768111 | 08:51 |
| *** nils has joined #zuul | 09:05 | |
| *** rishabhhpe has joined #zuul | 09:11 | |
| rishabhhpe | Hi Team , Need help i had created the VM using nodepool in openstack third party CI -- but i am able to login to VM passwordless using private IP but with public IP it is asking for password. can anyone help me with this | 09:13 |
| *** ykarel_ has joined #zuul | 09:13 | |
| *** ykarel has quit IRC | 09:16 | |
| guillaumec | corvus, tobiash: thanks ! | 09:22 |
| *** zenkuro has joined #zuul | 09:22 | |
| openstackgerrit | daniel.pawlik proposed zuul/zuul master: Added timestamp field, changed duration field type; improve doc https://review.opendev.org/c/zuul/zuul/+/756284 | 09:39 |
| danpawlik | tobiash, ianw: hey, I have done a rebase on previous accepted PS: https://review.opendev.org/c/zuul/zuul/+/756284 . Please check it in free time. Thanks | 09:45 |
| *** rishabhhpe has quit IRC | 09:50 | |
| *** rishabhhpe has joined #zuul | 09:51 | |
| *** rishabhhpe has quit IRC | 10:00 | |
| *** rishabhhpe has joined #zuul | 10:01 | |
| *** ykarel_ is now known as ykarel | 10:06 | |
| openstackgerrit | Andy Ladjadj proposed zuul/zuul-jobs master: [ensure-python] install python version only if not present https://review.opendev.org/c/zuul/zuul-jobs/+/770656 | 10:14 |
| *** jamesmcarthur has joined #zuul | 10:36 | |
| *** sshnaidm|afk is now known as sshnaidm|ruck | 10:40 | |
| *** jamesmcarthur has quit IRC | 10:41 | |
| mhu | hello zuul-maint, here's a fix so that pyjwt 2.0.0 can be used in zuul https://review.opendev.org/c/zuul/zuul/+/768312 | 10:45 |
| mhu | also the fix for the fringe case when someone wants to have a autohold request on a gerrit change if the change number is below 10: https://review.opendev.org/c/zuul/zuul/+/768229 | 10:47 |
| openstackgerrit | Guillaume Chauvel proposed zuul/zuul master: [DNM] zuul_stream zuul_console debug https://review.opendev.org/c/zuul/zuul/+/770780 | 10:49 |
| openstackgerrit | Andy Ladjadj proposed zuul/zuul-jobs master: [ensure-python] install python version only if not present https://review.opendev.org/c/zuul/zuul-jobs/+/770656 | 10:56 |
| tobiash | mhu: commented on 768312, what do you think? | 11:05 |
| openstackgerrit | Andy Ladjadj proposed zuul/zuul-jobs master: [ensure-python] install python version only if not present https://review.opendev.org/c/zuul/zuul-jobs/+/770656 | 11:06 |
| mhu | tobiash, yeah it's fair. I just hope they keep the breaking changes to major versions :) | 11:06 |
| tobiash | mhu: at least they claim it | 11:07 |
| openstackgerrit | Andy Ladjadj proposed zuul/zuul-jobs master: [ensure-python] install python version only if not present https://review.opendev.org/c/zuul/zuul-jobs/+/770656 | 11:07 |
| openstackgerrit | Matthieu Huin proposed zuul/zuul master: Pin pyjwt to 2.0.0 and fix issues due to version bump https://review.opendev.org/c/zuul/zuul/+/768312 | 11:18 |
| *** jcapitao is now known as jcapitao_lunch | 11:25 | |
| *** jamesmcarthur has joined #zuul | 11:37 | |
| *** jamesmcarthur has quit IRC | 11:42 | |
| *** jamesmcarthur has joined #zuul | 11:53 | |
| *** hashar is now known as hasharLunch | 11:54 | |
| *** jamesmcarthur has quit IRC | 11:59 | |
| *** jamesmcarthur has joined #zuul | 12:09 | |
| *** jamesmcarthur has quit IRC | 12:14 | |
| openstackgerrit | Andy Ladjadj proposed zuul/zuul-jobs master: [ensure-python] install python version only if not present https://review.opendev.org/c/zuul/zuul-jobs/+/770656 | 12:21 |
| *** rishabhhpe has quit IRC | 12:22 | |
| *** hasharLunch is now known as hashar | 12:22 | |
| *** dry has quit IRC | 12:27 | |
| *** dry has joined #zuul | 12:29 | |
| *** jpena is now known as jpena|lunch | 12:33 | |
| *** rfolco has left #zuul | 12:35 | |
| openstackgerrit | Matthieu Huin proposed zuul/zuul master: Bump pyjwt to 2.0.0 https://review.opendev.org/c/zuul/zuul/+/768312 | 12:35 |
| *** rlandy has joined #zuul | 12:58 | |
| *** jcapitao_lunch is now known as jcapitao | 13:00 | |
| *** jpena|lunch is now known as jpena | 13:29 | |
| *** jamesmcarthur has joined #zuul | 13:51 | |
| openstackgerrit | Tristan Cacqueray proposed zuul/zuul master: Get executor job params https://review.opendev.org/c/zuul/zuul/+/607078 | 13:52 |
| openstackgerrit | Tristan Cacqueray proposed zuul/zuul master: Separate out executor server from runner https://review.opendev.org/c/zuul/zuul/+/607079 | 13:53 |
| *** jamesmcarthur has quit IRC | 13:55 | |
| openstackgerrit | Tristan Cacqueray proposed zuul/zuul master: Move repository preparation into common class https://review.opendev.org/c/zuul/zuul/+/648642 | 14:00 |
| openstackgerrit | Tristan Cacqueray proposed zuul/zuul master: Separate out executor concerns from AnsibleJob https://review.opendev.org/c/zuul/zuul/+/648643 | 14:01 |
| *** ykarel has quit IRC | 14:02 | |
| mnaser | tobiash, corvus: i can have alex polish up https://review.opendev.org/c/zuul/zuul/+/708233 -- is there an approach you think we can go about this? maybe we can make it an extra installable package so to be more useful (or a seperate repo so it can be published with its own images?) | 14:13 |
| tobiash | mnaser: if you want to go as far as a real package I think a dedicated repo would be best suited | 14:14 |
| openstackgerrit | Lida Liu proposed zuul/zuul master: Fix errors regarding parsing timestamps https://review.opendev.org/c/zuul/zuul/+/765990 | 14:14 |
| tobiash | (aka adding tests etc) | 14:15 |
| tobiash | since we use that in production I'd support adding an extra repo for this so we can work together on tests etc | 14:16 |
| openstackgerrit | Alfredo Moralejo proposed zuul/zuul-jobs master: Rename config repos file config for CentOS Stream https://review.opendev.org/c/zuul/zuul-jobs/+/770815 | 14:16 |
| openstackgerrit | Guillaume Chauvel proposed zuul/zuul master: Fix zuul_stream test setup https://review.opendev.org/c/zuul/zuul/+/770817 | 14:25 |
| openstackgerrit | Guillaume Chauvel proposed zuul/zuul master: zuul_stream: Improve chdir error management https://review.opendev.org/c/zuul/zuul/+/770818 | 14:25 |
| openstackgerrit | Benedikt Löffler proposed zuul/zuul-jobs master: Pass environment variables to 'tox envlist config' task https://review.opendev.org/c/zuul/zuul-jobs/+/770819 | 14:25 |
| openstackgerrit | Tobias Henkel proposed zuul/zuul master: Required SQL reporters https://review.opendev.org/c/zuul/zuul/+/630472 | 14:32 |
| *** nhicher has quit IRC | 14:45 | |
| *** fbo has quit IRC | 14:45 | |
| *** nhicher has joined #zuul | 14:46 | |
| *** fbo has joined #zuul | 14:46 | |
| *** cloudnull has joined #zuul | 14:55 | |
| *** ykarel has joined #zuul | 15:02 | |
| *** hashar has quit IRC | 15:33 | |
| openstackgerrit | Jonas Sticha proposed zuul/nodepool master: aws: add support for uploading diskimages https://review.opendev.org/c/zuul/nodepool/+/735217 | 15:59 |
| *** cloudnull has quit IRC | 15:59 | |
| *** cloudnull has joined #zuul | 16:05 | |
| *** nils has quit IRC | 16:13 | |
| *** nils has joined #zuul | 16:16 | |
| openstackgerrit | Tristan Cacqueray proposed zuul/zuul master: runner: implement prep-workspace https://review.opendev.org/c/zuul/zuul/+/607082 | 16:18 |
| *** ykarel is now known as ykarel|away | 16:24 | |
| *** jfoufas1 has quit IRC | 16:27 | |
| *** hashar has joined #zuul | 16:31 | |
| *** jamesmcarthur has joined #zuul | 16:33 | |
| *** saneax has quit IRC | 16:33 | |
| *** zbr3 has joined #zuul | 16:38 | |
| *** zbr3 has quit IRC | 16:39 | |
| *** zbr9 has joined #zuul | 16:40 | |
| *** zbr has quit IRC | 16:40 | |
| *** zbr9 is now known as zbr | 16:40 | |
| *** ykarel|away has quit IRC | 16:46 | |
| *** jpena is now known as jpena|off | 17:07 | |
| corvus | mnaser, tobiash: new project sounds good to me. if we were to add an aws proxy, would we want that to be a separate project, or would we want to use the same codebase? | 17:07 |
| *** jcapitao has quit IRC | 17:12 | |
| tobiash | I think the same codebase would make sense | 17:24 |
| tobiash | a generic zuul-logs-proxy might make sense that could be extended to support various backends if needed | 17:25 |
| corvus | ok. just wanted to get the name right if we're going to be make a new repo :) | 17:36 |
| tobiash | or another idea would be to even integrate that into zuul-web so we can combine that with the upcomi g auth support | 17:38 |
| *** piotrowskim has quit IRC | 17:38 | |
| corvus | tobiash: yeah, though we have tried to avoid adding cloud backends to zuul itself | 17:41 |
| tobiash | that's true | 17:42 |
| *** rpittau is now known as rpittau|afk | 17:42 | |
| corvus | if we do think that's a good idea, then i think we probably need to reconsider that everywhere, and potentially reconsider the idea of making storage a first-class concept (as avass was recently suggesting) | 17:43 |
| corvus | personally, i think the "keep zuul simple and support external systems with ansible" approach has worked well; but maybe we hit the limit here | 17:43 |
| tobiash | at least when having the auth support I think how role based access to tenants and logs can be done | 17:45 |
| tobiash | the auth stack is a good starting point to enable authn not just to the admin api but in general as well | 17:47 |
| tobiash | I'm not sure yet though how the log stuff fits into this picture then | 17:47 |
| corvus | tobiash: yeah, we always assumed that the log storage area would have sufficient authorization features of its own, but it seems like that's not true in practice | 17:48 |
| tobiash | at least we now build native authn support directly into zuul | 17:48 |
| tobiash | but I think a zuul-logs-proxy could also just re-use the zuul-web config and its token handling | 17:49 |
| corvus | and that at least with swift, you need to put something in front of it, and that something doesn't exist | 17:49 |
| tobiash | if we want to keep that separated | 17:49 |
| corvus | tobiash: oh, im looking at the swift proxy and i just noticed it handles writes... that's surprising to me | 17:49 |
| corvus | tobiash: i thought this was read only, to restrict who can access the logs | 17:50 |
| tobiash | yes, that's something we needed for adding post processing results after the job | 17:50 |
| corvus | that seems like it might be pretty site-specific | 17:51 |
| corvus | tobiash: why not give the post-processor cloud credentials so it can do it directly? | 17:52 |
| tobiash | we use this for decoupling our post processing pipeline from the object store but that could be handled differently as well | 17:52 |
| tobiash | when we implemented that the post processing pipeline was not part of our core project and we didn't want them to be able to do anything they want with the cloud storage | 17:53 |
| tobiash | so we added the functionality to add files but not overwrite or delete | 17:53 |
| tobiash | but in the upstream version that can be removed | 17:54 |
| corvus | k | 17:54 |
| tobiash | meanwhile our post processing pipeline is part of our project so we have no real need for this anymore | 17:54 |
| tobiash | (other than a simple way to not having to guess the object store location from the log url it gets from zuul) | 17:55 |
| corvus | you could return it as an artifact | 17:56 |
| tobiash | but at least what I think would be important is that we can mid term integrate this proxy into the authn of zuul-web by either integrating it into zuul-web or re-using the authn functionality of zuul-web | 17:57 |
| tobiash | I guess re-using would be possible so we can keep this separate | 17:58 |
| tobiash | which is also useful since this enables scale out independent of zuul-web | 17:58 |
| corvus | yeah, i think we could do either, and even if we start this out as a separate project, we can probably integrate it into zuul-web in the future. | 17:59 |
| tobiash | ++ | 17:59 |
| corvus | i'd like to see a thread on zuul-discuss about this though, since i think it would be worthwhile finding out if we need this for backends other than swift | 17:59 |
| corvus | mnaser: ^ if you're offering to push this forward, can you do that (or ask alex)? | 18:00 |
| corvus | basically, we have a few choices about how to handle this, so i think it's worth a little bit of discussion to figure out the road ahead | 18:01 |
| tobiash | makes sense | 18:02 |
| *** jamesmcarthur has quit IRC | 18:26 | |
| *** hashar is now known as hasharAway | 18:29 | |
| *** hasharAway has quit IRC | 18:34 | |
| corvus | tobiash: https://review.opendev.org/765990 looks ready to go if you want to re-review | 18:41 |
| *** akrpan-pure has joined #zuul | 18:48 | |
| akrpan-pure | I just added a bunch of source repositories from opendev.org to my zuul instance, and the scheduler is now crashing on load | 18:49 |
| akrpan-pure | Specifically, it's trying to load jobs from keystone and then eventually (after like 5-10 minutes) fails with "AttributeError: 'MergeJob' object has no attribute 'updated'" | 18:49 |
| akrpan-pure | So now zuul's web UI won't load :/ | 18:50 |
| akrpan-pure | My guess is that it went back to an older branch and loaded some zuul config that wasn't updated (given it was going through various stable/* branches) | 18:50 |
| tobiash | akrpan-pure: that error message is a bit misleading but usually it means that the merger or executor couldn't fetch that repo | 18:51 |
| akrpan-pure | Ohhhh that's actually really good to know, thank you, lemme poke around a bit more then | 18:51 |
| tobiash | akrpan-pure: there is also a default timeout of I think 10min which might not be enough with large repos | 18:52 |
| fungi | or large repos over slow connections | 18:52 |
| akrpan-pure | tobiash & fungi: I think that's more what I'm hitting, because I can confirm the repo exists and it's finding the branches | 18:52 |
| tobiash | But you will find a hint into this direction in the merger or executor logs | 18:53 |
| openstackgerrit | Tristan Cacqueray proposed zuul/zuul master: runner: implement prep-workspace https://review.opendev.org/c/zuul/zuul/+/607082 | 18:53 |
| akrpan-pure | Oooh: /var/lib/zuul/executor/review.opendev.org is using 1484MB (limit=1024) | 18:53 |
| akrpan-pure | That seems like a relevant log line | 18:53 |
| akrpan-pure | (in the executor log, it's zuul.ExecutorDiskAccountant) | 18:53 |
| openstackgerrit | Tristan Cacqueray proposed zuul/zuul master: runner: add configuration schema https://review.opendev.org/c/zuul/zuul/+/640672 | 18:54 |
| openstackgerrit | Tristan Cacqueray proposed zuul/zuul master: runner: add prepare-workspace command line interface https://review.opendev.org/c/zuul/zuul/+/644770 | 18:54 |
| openstackgerrit | Tristan Cacqueray proposed zuul/zuul master: runner: add execute sub-command https://review.opendev.org/c/zuul/zuul/+/630944 | 18:54 |
| corvus | akrpan-pure: that log line sounds suspicious. that suggests that there may be a misconfiguration with the different build/git repo/cache directories | 18:58 |
| akrpan-pure | corvus: possibly, yeah. I'm tweaking the zuul config (bumping that up to 4096) | 18:58 |
| corvus | akrpan-pure: that log line should only emit for build directories, which means that it thinks "/var/lib/zuul/executor/review.opendev.org" is a build, which it clearly isn't | 18:59 |
| akrpan-pure | FWIW I'm using a software factory deployment, not just raw zuul | 18:59 |
| akrpan-pure | Ohhhhh okay | 18:59 |
| akrpan-pure | That would be an issue then wouldn't it | 18:59 |
| corvus | a build dir should be something like "/var/lib/zuul/builds/$UUID" | 18:59 |
| akrpan-pure | I have "git_dir=/var/lib/zuul/executor/" in my zuul.conf | 19:00 |
| corvus | akrpan-pure: and the job disk limit should not affect startup, only when it actually gets around to running logs | 19:00 |
| corvus | er running jobs | 19:00 |
| akrpan-pure | Alright. Let me remove all the repos I added, then add them back one at a time | 19:01 |
| akrpan-pure | Hopefully that'll unstick things | 19:01 |
| corvus | akrpan-pure: as for what to put in zuul.conf: i recommend nothing :) the defaults should work | 19:02 |
| akrpan-pure | corvus: I'm just using whatever software factory put in there by default. I can chase them down in #softwarefactory if it seems like it's something wrong on their deployment side? | 19:02 |
| akrpan-pure | Cool, after commenting out all the repos it loads fine | 19:03 |
| corvus | akrpan-pure: can you paste a copy of your zuul.conf with the secret info blanked out to paste.openstack.org? | 19:04 |
| akrpan-pure | Sure, gimme a bit | 19:04 |
| tristanC | akrpan-pure: corvus: iirc the ExecutorDiskAccountant emits warning for git_dir, but that shouldn't prevent job from running | 19:05 |
| corvus | tristanC: it should not do that | 19:06 |
| akrpan-pure | corvus: http://paste.openstack.org/show/801639/ | 19:08 |
| corvus | akrpan-pure: thanks, gimme a minute to pull on some threads :) | 19:08 |
| akrpan-pure | Oh wow, even after just adding one project (opendev.org/openstack/nova) it fails with that | 19:09 |
| akrpan-pure | FWIW, full stack trace: http://paste.openstack.org/show/801640/ | 19:10 |
| akrpan-pure | From /var/log/zuul/scheduler.log | 19:10 |
| openstackgerrit | Guillaume Chauvel proposed zuul/zuul master: Fix zuul_stream test setup https://review.opendev.org/c/zuul/zuul/+/770817 | 19:10 |
| tobiash | akrpan-pure: can you paste the executor log? | 19:11 |
| akrpan-pure | Un momento | 19:11 |
| tobiash | corvus: I think we should catch that attribute error and emit a better error message :) | 19:12 |
| corvus | tobiash: no objection :) | 19:12 |
| akrpan-pure | Ah-ha, in there I see "stderr: 'Timeout: the command "git fetch -f --prune --tags origin" did not complete in 300 secs.'" | 19:12 |
| akrpan-pure | I think that the openstack repos are just straight up too big | 19:13 |
| tobiash | jepp that's the timeout | 19:13 |
| akrpan-pure | Is that configurable somewhere? | 19:13 |
| tobiash | I thought the default was 10min | 19:13 |
| tobiash | yes, it is | 19:13 |
| tobiash | https://zuul-ci.org/docs/zuul/discussion/components.html#attr-merger.git_timeout | 19:14 |
| corvus | akrpan-pure: can you provide the output of 'ls -l /var/lib/zuul/builds' ? and also /var/lib/zuul/executor ? | 19:14 |
| corvus | akrpan-pure, tobiash: i agree changing merger.git_timeout should help the immediate startup problem | 19:15 |
| tristanC | corvus: here the merge_root is passed to the diskaccountant: https://opendev.org/zuul/zuul/src/branch/master/zuul/executor/server.py#L2645 | 19:15 |
| akrpan-pure | var/lib/zuul/builds is empty, var/lib/zuul/executor returns "cinder-ci.dev.purestorage.com review.opendev.org" | 19:15 |
| tristanC | corvus: which emits a warning and calls stopJobDiskFull here: https://opendev.org/zuul/zuul/src/branch/master/zuul/executor/server.py#L153 | 19:16 |
| akrpan-pure | Adding a config line in merger to bump timeout to 3000 (10x as long) | 19:16 |
| tristanC | corvus: but stopJobDiskFull can't stop the job because the path isn't part of the job root and it has no job unique id | 19:16 |
| corvus | tristanC: https://opendev.org/zuul/zuul/src/branch/master/zuul/executor/server.py#L147-L150 is why i would expect it not to emit the warning | 19:18 |
| corvus | tristanC, akrpan-pure: i think it's the trailing / | 19:19 |
| corvus | akrpan-pure: can you try removing the trailing / on "git_dir=/var/lib/zuul/executor/" in your config file? | 19:19 |
| akrpan-pure | corvus: Sure, one sec | 19:19 |
| corvus | akrpan-pure: that should cause that warning mesage to no longer appear (and we'll know it worked if we don't see it once the size of the repos being cloned into /var/lib/zuul/executor is > 5GB) | 19:21 |
| corvus | akrpan-pure: and if that turns out to be the case, then the good news is i don't think it actually affects the functionality of the diskaccountant; it's just an annoying warning msg. | 19:22 |
| corvus | tristanC: and if so, then if softwarefactory generated that line, you might want to drop the trailing '/'. and in zuul, we should probably normalize those paths before checking them. | 19:23 |
| openstackgerrit | Tristan Cacqueray proposed zuul/zuul master: executor: prevent false positive diskaccountant warning https://review.opendev.org/c/zuul/zuul/+/770851 | 19:24 |
| tristanC | corvus: well i'd rather fix the root cause because it's rather common to terminate a directory name by a slash | 19:25 |
| akrpan-pure | I can tentatively say it worked? Now I'm just seeing "INFO zuul.ExecutorServer: Re-registering as job is within its limits 0.34 <= 20.0, 78.9% <= 5.0, running, 47.8% <= 5.0, 0 <= 20" which seems like a good sign | 19:27 |
| corvus | tristanC: why don't you use the defaults in softwarefactory? | 19:28 |
| tristanC | corvus: that should be changed, that zuul.conf template is quite old... | 19:32 |
| akrpan-pure | Also, apologies, a slightly more "RTFM" question: I'm trying to import jobs from various openstack repos (I'm trying to get access to the devstack jobs). Is the proper way to add them as source repositories to the existing project, or to add a project for each upstream openstack project I need jobs from? | 19:32 |
| corvus | tristanC: ah, probably from before we updated those with sane defaults | 19:32 |
| tristanC | corvus: i think that was required for the migration to zuul v2 to zuul v3 where we supported both services on a single node in different home | 19:33 |
| tristanC | from* zuul v2 | 19:33 |
| akrpan-pure | Seeing as it's complaining about "missing projects" maybe I've been doing it wrong. I'll see about adding them as projects instead | 19:34 |
| tristanC | akrpan-pure: that's quite a trick question, i think we discussed about supporting configuration import from foreign project (not registered in zuul tenant config), but that's not there yet. so you have to add all required project | 19:35 |
| akrpan-pure | tristanC: so add a "- project: name: openstack/nova" (so on and so forth) for each one? I'll give it a shot | 19:35 |
| corvus | akrpan-pure: yes and building on what tristanC said, for devstack it's going to be a long list, and i don't know how to build it other than trial and error and keep adding what it says is missing. now, if you'd like to be a hero, please send that list when you're done to one of the mailing lists and maybe you'll save some time for the next person. | 19:36 |
| corvus | so far, everyone i've helped get this far has disappeared before doing that. :( | 19:36 |
| akrpan-pure | corvus: will do, my main goal for this whole undertaking is to document how to make a third party CI system since there's like... no posts on it | 19:36 |
| akrpan-pure | Good to hear that there's a need for docs | 19:37 |
| akrpan-pure | Well not "good" but you get what I mean | 19:37 |
| openstackgerrit | Tristan Cacqueray proposed zuul/zuul master: executor: prevent false positive diskaccountant warning https://review.opendev.org/c/zuul/zuul/+/770851 | 19:40 |
| openstackgerrit | Tristan Cacqueray proposed zuul/zuul master: executor: prevent false positive diskaccountant warning https://review.opendev.org/c/zuul/zuul/+/770851 | 19:40 |
| openstackgerrit | Merged zuul/zuul master: Fix errors regarding parsing timestamps https://review.opendev.org/c/zuul/zuul/+/765990 | 20:00 |
| *** hashar has joined #zuul | 20:54 | |
| openstackgerrit | Merged zuul/zuul master: executor: prevent false positive diskaccountant warning https://review.opendev.org/c/zuul/zuul/+/770851 | 20:56 |
| akrpan-pure | Hopefully last question for the day: after a bunch of dependency resolution, I've finally got it down to one last set of errors. It seems like some of the .zuul.d files are looking for projects by the full name opendev.org/openstack/cinder instead of just by openstack/cinder | 21:13 |
| akrpan-pure | Should I be calling my projects the full name with opendev.org? I've seen both naming methods come up, so I feel like fixing one would break the other | 21:14 |
| clarkb | I want to say pabelanger (who isn't here right now) had some experience with ^ and had thought through some of the implications | 21:14 |
| akrpan-pure | Zuul is basically spitting out "Unknown project" errors in the web UI | 21:15 |
| akrpan-pure | Ahhh, it's specifically in the "roles" section that it uses the full name | 21:18 |
| clarkb | akrpan-pure: I think if you name your connection to review.opendev.org or git.opendev.org as opendev.org canonically in the zuul connection config then that substitution will be automatic. Unless you have another connection with that name and the lookup is ambiguous | 21:19 |
| clarkb | and I think that is what pabelanger ran into. Having a bunch of local forks and upstream repos with the same names meant that using specific names is often desireable | 21:19 |
| akrpan-pure | So I have a connection opendev-third-party-ci with baseurl https://review.opendev.org. Are you saying I should call the connection opendev.org and keep the baseurl where it is? | 21:20 |
| clarkb | akrpan-pure: https://zuul-ci.org/docs/zuul/reference/drivers/gerrit.html#attr-%3Cgerrit%20connection%3E.canonical_hostname is the setting to set | 21:22 |
| *** hamalq has joined #zuul | 21:22 | |
| akrpan-pure | clarkb: ahh, set that to opendev.org? Will do | 21:23 |
| clarkb | ya | 21:23 |
| clarkb | akrpan-pure: but that assumes other connections don't have projects called openstack/cinder | 21:23 |
| akrpan-pure | clarkb: good to know, luckily I'm literally just using this for third party CI so I shouldn't have any conflicts | 21:23 |
| *** hashar has quit IRC | 21:26 | |
| fungi | yeah, if your only connection is to opendev.org then it shouldn't hurt | 21:27 |
| *** nils has quit IRC | 21:49 | |
| openstackgerrit | Guillaume Chauvel proposed zuul/zuul master: zuul_stream: Improve chdir error management https://review.opendev.org/c/zuul/zuul/+/770818 | 21:54 |
| openstackgerrit | Guillaume Chauvel proposed zuul/zuul master: zuul_stream: close streamer when task is skipped https://review.opendev.org/c/zuul/zuul/+/735476 | 21:54 |
| *** ikhan has joined #zuul | 21:57 | |
| openstackgerrit | Guillaume Chauvel proposed zuul/zuul master: tutorial: Add "gate your first patch" https://review.opendev.org/c/zuul/zuul/+/732067 | 23:00 |
| *** rlandy is now known as rlandy|bbl | 23:28 | |
| akrpan-pure | corvus: If you're still here, just checking... is the list of openstack projects I have to set up supposed to be in the ~100's scale? Parsing through the errors, I've found 128 projects | 23:42 |
| akrpan-pure | I'm fine if it is, I just want to make sure I'm actually going down the right path | 23:42 |
| clarkb | akrpan-pure: if you are trying to run devstack jobs then yes I think its tendrils go far and wide | 23:42 |
| akrpan-pure | I am, yeah. Thanks for the confirmation! | 23:43 |
| *** sshnaidm|ruck is now known as sshnaidm|afk | 23:49 | |
Generated by irclog2html.py 2.17.2 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!