| *** jamesmcarthur has joined #zuul | 01:20 | |
| *** jamesmcarthur has quit IRC | 01:23 | |
| *** openstackgerrit has quit IRC | 01:30 | |
| *** openstackgerrit has joined #zuul | 01:45 | |
| openstackgerrit | Tristan Cacqueray proposed openstack-infra/nodepool master: Dockerfile: add support for arbritary uid https://review.openstack.org/651424 | 01:45 |
|---|---|---|
| *** bhavikdbavishi has quit IRC | 02:11 | |
| tristanC | mordred: with ^ i was able to add nodepool services to the zuul-operator. it's almost feature complete for integration in software factory now :) | 02:48 |
| tristanC | mordred: would be nice to get the project created to have ci and image publication in place | 02:53 |
| *** bhavikdbavishi has joined #zuul | 02:57 | |
| clarkb | tristanC: the &> is a bashism iirc. May want to use 2>&1 so that the shebang is valid on images where sh is not bash | 03:14 |
| clarkb | also if you set user at build time which that change does why do you need to dynamically update /etc/passwd? | 03:16 |
| clarkb | couldnt you write /etc/passwd at build time and avoid the entrypoint entirely? | 03:16 |
| clarkb | or is the arbitrary uid important and maybe we drop the user dorective at build time? | 03:18 |
| tristanC | clarkb: k8s may not respect image user config when using securityContext | 03:22 |
| tristanC | clarkb: build time user is for local runtime usage, not sure why user isn't added to passwd at build time though | 03:23 |
| openstackgerrit | Tristan Cacqueray proposed openstack-infra/nodepool master: Dockerfile: add support for arbritary uid https://review.openstack.org/651424 | 03:24 |
| openstackgerrit | Tristan Cacqueray proposed openstack-infra/zuul master: Dockerfile: add support for arbritary uid https://review.openstack.org/650246 | 03:24 |
| clarkb | I see so the uid may be arbitray in some.cases but we also want to set it to not root for the other cases | 03:27 |
| tristanC | clarkb: it seems like a reasonable thing to not run zuul/nodepool as root for the other cases too, but that's optional | 03:28 |
| clarkb | ya I have no opposition to that. Just making sure I understand why we need the USER directive and the entrypoint | 03:28 |
| clarkb | at first glance they seem mutually exclusive | 03:29 |
| clarkb | but it is because the USER set uid may be overridde | 03:29 |
| clarkb | I cant vote now because I am on my phonr but I'll leave a review in the morning | 03:29 |
| tristanC | clarkb: thanks! | 03:30 |
| *** openstackstatus has quit IRC | 04:35 | |
| *** openstackstatus has joined #zuul | 04:37 | |
| *** ChanServ sets mode: +v openstackstatus | 04:37 | |
| *** quiquell|off is now known as quiquell|rover | 04:48 | |
| *** pcaruana has joined #zuul | 05:06 | |
| *** bjackman_ has joined #zuul | 05:26 | |
| *** bhavikdbavishi has quit IRC | 05:54 | |
| *** bhavikdbavishi has joined #zuul | 05:55 | |
| *** quiquell|rover is now known as quique|rover|brb | 06:28 | |
| openstackgerrit | Tristan Cacqueray proposed openstack-infra/nodepool master: Add python-path option to node https://review.openstack.org/637338 | 06:38 |
| *** gtema has joined #zuul | 06:47 | |
| *** sshnaidm|pto is now known as sshnaidm | 07:06 | |
| *** quique|rover|brb is now known as quiquell|rover | 07:11 | |
| *** pcaruana has quit IRC | 07:34 | |
| *** pcaruana has joined #zuul | 07:35 | |
| openstackgerrit | Merged openstack-infra/zuul master: tox: Fix indentation https://review.openstack.org/649040 | 07:48 |
| *** jpena|off is now known as jpena | 07:52 | |
| *** quiquell|rover is now known as quique|rover|bbl | 07:54 | |
| openstackgerrit | Luigi Toscano proposed openstack-infra/zuul-jobs master: stage-output: fix dotfiles handling https://review.openstack.org/648658 | 07:58 |
| openstackgerrit | Luigi Toscano proposed openstack-infra/zuul-jobs master: stage-output: fix dotfiles handling https://review.openstack.org/648658 | 08:05 |
| openstackgerrit | Merged openstack-infra/zuul master: Assert ara database is loaded from ansible venv https://review.openstack.org/650917 | 08:09 |
| *** bhavikdbavishi has quit IRC | 08:24 | |
| *** themroc has joined #zuul | 08:29 | |
| *** hashar has joined #zuul | 08:38 | |
| *** hashar has quit IRC | 08:39 | |
| *** bhavikdbavishi has joined #zuul | 09:04 | |
| *** electrofelix has joined #zuul | 09:28 | |
| *** quique|rover|bbl is now known as quiquell|rover | 09:50 | |
| *** bhavikdbavishi has quit IRC | 10:53 | |
| *** jpena is now known as jpena|lunch | 11:34 | |
| *** bhavikdbavishi has joined #zuul | 11:52 | |
| *** rlandy has joined #zuul | 11:55 | |
| *** rlandy is now known as rlandy|ruck | 11:56 | |
| *** hashar has joined #zuul | 12:01 | |
| openstackgerrit | Tristan Cacqueray proposed openstack-infra/zuul master: Dockerfile: add support for arbritary uid https://review.openstack.org/650246 | 12:19 |
| openstackgerrit | Tristan Cacqueray proposed openstack-infra/nodepool master: Dockerfile: add support for arbritary uid https://review.openstack.org/651424 | 12:20 |
| *** jpena|lunch is now known as jpena | 12:31 | |
| openstackgerrit | Tobias Henkel proposed openstack-infra/zuul master: WIP: Support emitting warnings via zuul_return https://review.openstack.org/651526 | 12:54 |
| *** bjackman_ has quit IRC | 12:57 | |
| *** yolanda_ has joined #zuul | 12:58 | |
| pabelanger | Hmm, I think might have a bug with noop jobs and github returning the wrong ref_url: https://dashboard.zuul.ansible.com/t/ansible/build/b144ba131f364e5887f5509aa1dd5c51 | 12:59 |
| pabelanger | note is says https://api.github.com/repos/ansible-network/resource_module_builder/pulls/24 but should maybe be https://github.com/ansible-network/resource_module_builder/pulls/24 | 12:59 |
| pabelanger | sorry: https://github.com//ansible-network/resource_module_builder/pull/24 | 13:00 |
| pabelanger | tobiash: you seen that by chance?^ | 13:00 |
| *** irclogbot_0 has joined #zuul | 13:04 | |
| *** altlogbot_2 has joined #zuul | 13:08 | |
| tobiash | pabelanger: I never looked at the build results of noop ;) | 13:13 |
| *** pwhalen has joined #zuul | 13:30 | |
| *** bhavikdbavishi has quit IRC | 13:33 | |
| *** quiquell|rover is now known as quiquell|lunch | 13:38 | |
| *** yolanda_ has quit IRC | 13:49 | |
| *** quiquell|lunch is now known as quiquell|rover | 14:02 | |
| Shrews | clarkb: the squashed change lgtm. just a few unnecessary whitespace changes in a couple of files | 14:12 |
| Shrews | hopefully we can try that again today, though i'm going to be in and out with a couple of appointments today | 14:13 |
| *** openstackgerrit has quit IRC | 14:14 | |
| clarkb | I'll be around shortly ina spot to watch it if we give it another go | 14:14 |
| AJaeger | clarkb: please self-approve once you're ready - I'll +2 now... | 14:15 |
| *** hashar has quit IRC | 14:22 | |
| Shrews | https://review.openstack.org/651339 helps stabilize nodepool tests if anyone has a moment today. easy review | 14:38 |
| clarkb | that eill make a good test of the image builds | 14:43 |
| clarkb | I'll review it once image change is in | 14:43 |
| *** ianychoi has quit IRC | 14:56 | |
| *** quiquell|rover is now known as quiquell|off | 15:01 | |
| clarkb | mordred: tobiash fungi if you can review https://review.openstack.org/#/c/651339/1 and if it looks good +2 but don't +3 I can add the approval vote once the image roles update | 15:01 |
| tobiash | clarkb: lgtm | 15:03 |
| mordred | clarkb: ++ | 15:06 |
| *** openstackgerrit has joined #zuul | 15:08 | |
| openstackgerrit | Merged openstack-infra/zuul-jobs master: Only alias registries on build nodes https://review.openstack.org/651393 | 15:08 |
| mordred | clarkb: ^^ woot | 15:10 |
| clarkb | cool approving the other change now | 15:11 |
| *** gtema has quit IRC | 15:11 | |
| *** jangutter_ is now known as jangutter | 15:15 | |
| *** themroc has quit IRC | 15:15 | |
| clarkb | I think skopeo has the same issue | 15:16 |
| clarkb | localhost | time="2019-04-10T15:15:48Z" level=fatal msg="Invalid source name docker://[2607:ff68:100:54:f816:3eff:fef2:fc69]:5000/zuul/nodepool:latest: invalid reference format" | 15:16 |
| clarkb | mordred: ^ | 15:16 |
| clarkb | I bet everyone is importing that docker distribution/reference lib regex and using it | 15:17 |
| clarkb | because it is the "standard" | 15:17 |
| clarkb | it just completely fails at ipv6ing | 15:17 |
| openstackgerrit | Clark Boylan proposed openstack-infra/zuul-jobs master: Revert "Only alias registries on build nodes" https://review.openstack.org/651579 | 15:22 |
| clarkb | Shrews: AJaeger mordred ^ fyi | 15:22 |
| AJaeger | clarkb: I'm reaching out to "my" skopeo expert now... | 15:23 |
| clarkb | AJaeger: https://github.com/moby/moby/issues/39033 may be relevant | 15:28 |
| *** jamesmcarthur has joined #zuul | 15:32 | |
| AJaeger | clarkb: which skope command do you use? Can you leave an example and I link to this discussion... | 15:33 |
| clarkb | AJaeger: http://logs.openstack.org/39/651339/1/gate/nodepool-upload-image/143d1ce/ara-report/file/2bdc78fb-bf65-4b40-a93a-3decdda15759/#line-1 that is the command that fails | 15:34 |
| mordred | clarkb: AWESOME | 15:34 |
| clarkb | fwiw the docker push commands also failed | 15:35 |
| fungi | consistency! | 15:35 |
| * fungi sighs | 15:35 | |
| clarkb | but I think we half expected skopeo to work so that is the more noteworthy item :) | 15:35 |
| clarkb | I think we may want to go back to the drawing board on this one | 15:46 |
| openstackgerrit | Merged openstack-infra/zuul-jobs master: Revert "Only alias registries on build nodes" https://review.openstack.org/651579 | 15:46 |
| clarkb | /etc/hosts isn't a good hack for anything on localhost and skopeo not working means we need something for localhost | 15:46 |
| clarkb | and we want to avoid localhost edits of /etc/hosts because we don't want executors in untrusted contexts to be able to ssh back into the executor somehow and I like /etc/hosts being read only as a layer of protection there | 15:47 |
| clarkb | Real dns is likely massive overkill for these ephemeral setups | 15:48 |
| clarkb | we could create an ipv4 label :/ | 15:49 |
| clarkb | or put effort into fixing https://github.com/moby/moby/issues/39033 | 15:50 |
| clarkb | AJaeger: my hunch is that because the urls are docker:// skopeo parses them using the docker rules which don't allow the [] escapes. But I bet docker:// is how you tell skopeo to talk to a docker registry | 15:54 |
| clarkb | maybe ^ hints at another option here. Run a different registry with different push/pull tool | 15:54 |
| mordred | the docker demons on nodes need to be able to talk to the registries | 15:55 |
| mordred | otherwise the jobs aren't good at testing that docker images work as they would in the wild | 15:55 |
| clarkb | mordred: hrm thats a fair point | 15:55 |
| clarkb | and we also want ot push to dockerhub | 15:55 |
| mordred | yeah | 15:56 |
| mordred | so - I think effort in to fixing the moby issue potentially - or yeah, an ipv4 label which would be sadmaking | 15:56 |
| clarkb | we could potentially use the intermediate registry for everything | 15:58 |
| clarkb | that may have scaling problems but could have permanent real dns record that made ipv6 happier | 15:58 |
| AJaeger | clarkb: timing is bad, I'll relay once I hear something from my colleague | 15:59 |
| clarkb | no worries I don't think this is so much a rush as me just throwing ideas out in a fit of brainstorming | 15:59 |
| clarkb | I guess I can make a PR to fix the regex issue in docker | 16:01 |
| clarkb | I'll start on that now | 16:01 |
| clarkb | at the very least maybe that will get the wheels spinning upstream | 16:01 |
| mordred | clarkb: the problem with using the intermediate registry for everything is that on the consume side jobs want to just be able to say "I want zuul/zuul:latest" and have that result in the correct speculative state for that buildset | 16:01 |
| mordred | clarkb: that said - we COULD spin up a dns server on the buildset registry host and configure the remote hosts to talk to it via dns. of course, that still doesn't help with localhost | 16:02 |
| clarkb | mordred: ya, but we do know who are parent change(s) are and could be smarter about pulling just their images I think | 16:03 |
| * mordred is mostly thinking out loud | 16:03 | |
| clarkb | s/are/our/ | 16:03 |
| *** bhavikdbavishi has joined #zuul | 16:03 | |
| mordred | clarkb: it's important that job content be able to pull images themselves | 16:03 |
| mordred | clarkb: such as the docker-compose one | 16:03 |
| mordred | clarkb: I think the upstream moby patch will be super helpful | 16:04 |
| clarkb | mordred: ya I'm forking the docker distribution repo now | 16:04 |
| clarkb | the regex is actually quite simple. I just need to figure out how to write it in go | 16:05 |
| clarkb | should have it up shortly | 16:05 |
| *** jamesmcarthur has quit IRC | 16:06 | |
| *** jamesmcarthur has joined #zuul | 16:08 | |
| pabelanger | so, I've noticed a few times that zuul misses github events, I actually haven't looked yet to see where they could be dropping. EG: github side, network or zuul side. However, I am wondering if there might be a way to audit open PRs for a project, and if there was no zuul comment after X mins, to leave a recheck comment, or some other process to try and enqueue said PR. The idea would be to not have users | 16:28 |
| pabelanger | keep track if zuul didn't run a PR, then do the recheck themself | 16:28 |
| pabelanger | Oh, so zuul did get the event from github | 16:31 |
| pabelanger | but raised an exception | 16:31 |
| pabelanger | http://paste.openstack.org/show/749130/ | 16:31 |
| tobiash | pabelanger: permission problem? | 16:32 |
| pabelanger | tobiash: nope, zuul has been gating that repo for a while | 16:33 |
| tobiash | was that pr created? | 16:33 |
| pabelanger | yes | 16:33 |
| mordred | maybe race condition on the GH side | 16:33 |
| pabelanger | I can get the payload from github app | 16:33 |
| mordred | like - it sent the payload but when we went to query things about it it wasn't showing up yet | 16:34 |
| pabelanger | Oh, maybe | 16:34 |
| mordred | pabelanger: also, per your earlier thing, I think obviosly fixing exceptions is important :) - then we've got some stuff sketched out related to HA scheduler that should allow us to start running more than one event receiver to allow not missing stuff | 16:34 |
| pabelanger | mordred: Yah, nice, cool. I'll fix exception first, since that is the issue here | 16:35 |
| mordred | which is to say - I think first priority should be to make not missing thigns in the first place rock solid before we start having to poll things regularly to see what might have been missed ... but I agree, it should not be on the users to try to keep track of this in their heads | 16:35 |
| mordred | like - that's a terrible experience for the users for sure :) | 16:35 |
| mordred | pabelanger: maybe it just needs a retry/backoff something something? | 16:35 |
| pabelanger | agree, and yah I think a retry should be okay | 16:36 |
| mordred | ++ | 16:36 |
| pabelanger | let me see if I can write up a patch | 16:36 |
| mordred | \o/ | 16:36 |
| tobiash | but we need to be careful there as this loop is time critical | 16:36 |
| mordred | yeah | 16:36 |
| tobiash | so retrying is only a good idea for a few seconds | 16:36 |
| mordred | tobiash: yup. although we should only ever need to retry if gh-side is backed up somehow - getting a NotFound at that point in the logic should be very uncommon, one would hope | 16:37 |
| mordred | and I'm guessing more common on public github than on a private ghe | 16:37 |
| tobiash | yes, except you really have a permission issue | 16:37 |
| tobiash | which can be easily the case if not using app auth | 16:38 |
| mordred | ah - this is a good point | 16:38 |
| pabelanger | we actually aready retry, upto 5 times it looks like in that function, we just don't see to catch the exception | 16:40 |
| pabelanger | so, maybe we just need to catch it and hope we get it on next loop | 16:40 |
| pabelanger | http://git.zuul-ci.org/cgit/zuul/tree/zuul/driver/github/githubconnection.py#n1141 | 16:41 |
| pabelanger | tobiash: what do you think? | 16:41 |
| tobiash | ++ | 16:42 |
| tobiash | I think the target of that loop probably was exactly catching this case | 16:42 |
| tobiash | (and possibley infrastructure related issues) | 16:42 |
| mordred | yeah | 16:43 |
| tobiash | so when you catch that exception, we should not only catch NotFoundError | 16:43 |
| *** altlogbot_2 has quit IRC | 16:45 | |
| tobiash | mordred: as you did the initial implementation of the command module override in zuul this might be interesting for you: https://review.openstack.org/650276 | 16:48 |
| mordred | tobiash: yeah- I keep looking at that | 16:48 |
| tobiash | thanks | 16:49 |
| mordred | tobiash: do you know what impact, if any, this has on ordering between stdout and stderr streams? | 16:49 |
| mordred | tobiash: one of the reason we were originally combining them is to get the closest approximation of "foo.sh 2>&1" - so that log lines would be interleaved appropriately. I'm guessing that the select([]) call is gonna do mostly the same? | 16:50 |
| tobiash | mordred: I think that should be mostly the same | 16:51 |
| mordred | nod | 16:53 |
| openstackgerrit | Paul Belanger proposed openstack-infra/zuul master: Handle project not found exception better https://review.openstack.org/651609 | 16:54 |
| pabelanger | tobiash: mordred: how about ^ | 16:54 |
| pabelanger | oh, some backscoll, reading now | 16:54 |
| pabelanger | tobiash: ah, so catch all exceptions... | 16:55 |
| tobiash | Yepp ;) | 16:55 |
| mordred | pabelanger: otherwise I think that looks solid | 16:56 |
| pabelanger | k, fixing | 16:57 |
| pabelanger | is, except github3.exceptions good enough there? | 16:57 |
| tobiash | I think yes | 16:57 |
| pabelanger | github3.exceptions.GitHubException seems to be base | 16:59 |
| tobiash | k | 16:59 |
| pabelanger | all others parent to it | 16:59 |
| openstackgerrit | Paul Belanger proposed openstack-infra/zuul master: Handle github exceptions better when trying to get a PR https://review.openstack.org/651609 | 17:06 |
| clarkb | https://github.com/docker/distribution/pull/2899 ok that took entirely too much time | 17:06 |
| clarkb | mordred: AJaeger Shrews ^ fyi | 17:07 |
| pabelanger | mordred: tobiash: ^updated | 17:07 |
| *** jpena is now known as jpena|off | 17:07 | |
| pabelanger | clarkb: might have thoughts too^ | 17:07 |
| clarkb | pabelanger: reviewed. Just a fix for the log string (and commit message update) | 17:28 |
| pabelanger | fixing | 17:33 |
| openstackgerrit | Merged openstack-infra/zuul master: Fix zuul-stream functional test https://review.openstack.org/650297 | 17:34 |
| openstackgerrit | Paul Belanger proposed openstack-infra/zuul master: Handle github exceptions better when trying to get a PR https://review.openstack.org/651609 | 17:36 |
| pabelanger | updated | 17:36 |
| clarkb | +2 thanks | 17:37 |
| pabelanger | np! | 17:37 |
| pabelanger | ty | 17:37 |
| pabelanger | mordred: tobiash: if you'd like to review again^ | 17:38 |
| mordred | ++ | 17:38 |
| mordred | pabelanger: thanks! | 17:38 |
| pabelanger | also, for spacex fans, another falcon heavy launch today! about 4 hours out | 17:39 |
| clarkb | 6:30 Eastern is whne launch windows starts roughly iirc | 17:40 |
| clarkb | pabelanger: this one has customer payload too | 17:40 |
| pabelanger | yah, that was for webcast starting | 17:40 |
| pabelanger | clarkb: yah, should be awesome | 17:41 |
| *** jamesmcarthur has quit IRC | 17:42 | |
| *** jamesmcarthur has joined #zuul | 18:12 | |
| *** brennen has left #zuul | 18:13 | |
| openstackgerrit | Tobias Henkel proposed openstack-infra/zuul master: Support emitting warnings via zuul_return https://review.openstack.org/651526 | 18:21 |
| *** electrofelix has quit IRC | 18:24 | |
| *** tobiash has quit IRC | 18:30 | |
| *** tobiash has joined #zuul | 18:30 | |
| *** tobiash has quit IRC | 18:35 | |
| *** tobiash has joined #zuul | 18:40 | |
| openstackgerrit | Merged openstack-infra/nodepool master: Fix race with test_provider_removal unit test https://review.openstack.org/651339 | 18:42 |
| *** tobiash has quit IRC | 18:46 | |
| *** hashar has joined #zuul | 18:49 | |
| *** tobiash has joined #zuul | 18:51 | |
| *** themroc has joined #zuul | 18:56 | |
| -openstackstatus- NOTICE: Restarting Gerrit on review.openstack.org to pick up new configuration for the replication plugin | 19:06 | |
| *** rlandy|ruck is now known as rlandy|ruck|brb | 19:41 | |
| pabelanger | the other thing I have noticed, is sometime as soon as I create a PR, I apply by gate label, because it is a trival fix. Most of the time, once check jobs report back, it will get enqueued into gate. But every so often, zuul won't move it to gate. Again, I haven't looked in the logs... mostly just sharing my thoughts :) | 19:43 |
| *** bhavikdbavishi has quit IRC | 19:44 | |
| fungi | shout out to zuul for its ability to test the insane velocity of openstack development through yet another major coordinated release! http://lists.openstack.org/pipermail/openstack-discuss/2019-April/004906.html | 19:48 |
| pabelanger | ++ | 19:49 |
| fungi | merging an *average* of 155 changes a day | 19:50 |
| fungi | that's more than a change every 10 minutes for ~6 months | 19:51 |
| fungi | (not considering how many patchsets on average go into each change, how many jobs are run on average for each change, et cetera) | 19:51 |
| pabelanger | yah, when I looked this morning openstack peaked at 1.2 kjps | 19:52 |
| fungi | there are only a handful of software efforts in the world, floss or proprietary, approaching that scale | 19:52 |
| fungi | earlier today the openstack release team basically pushed a button and then watched as zuul.openstack.org built, signed, published and announced release artifacts for dozens of complex services and many more supporting modules. the only failures were in generating release notes for two who hadn't upgraded off a deprecated sphinx theme | 19:57 |
| fungi | start to finish was roughly 30 minutes | 19:57 |
| fungi | this is an activity which only a few years ago when openstack was a fraction its current size took days of hand coordination | 19:58 |
| *** jamesmcarthur has quit IRC | 19:59 | |
| pabelanger | Yah, the cool part for me, is starting to make those bits more generic for any zuul operator to consume (release team). I looked into it a few months back, just haven't dug more into it | 20:00 |
| fungi | i'd love to see more projects start providing their own release signing keys | 20:02 |
| fungi | for openstack it's already just a zuul secret | 20:02 |
| fungi | master key is kept on independent hardware, rotated twice annually, dedicated signing subkey is generated and encrypted as a zuul secret kept in the tenant's config project | 20:03 |
| fungi | openstack uses the same signing subkey to sign its generated git tags and source tarballs | 20:05 |
| fungi | all from a handful of lines of configuration | 20:05 |
| *** rlandy|ruck|brb is now known as rlandy|ruck | 20:09 | |
| fdegir | mordred: i'm looking for your zuul presentation to present Zuul to my team | 20:21 |
| fdegir | mordred: they aren't available on your personal website - do you have it somewhere where I can get? | 20:21 |
| fdegir | and this is perhaps a more general comment - would it be possible to put zuul architecture diagram to zuul website so people can see things on a high level? | 20:23 |
| fdegir | something like this: https://storage.googleapis.com/cdn.thenewstack.io/media/2018/05/bc0977f2-zuul-screen-02-2.jpg | 20:23 |
| clarkb | fdegir: a similar diagram should be on the website | 20:23 |
| clarkb | let me find a link | 20:23 |
| * fdegir needs to improve his search skills... | 20:24 | |
| fdegir | clarkb: do you mean the one on the page https://zuul-ci.org/docs/zuul/admin/components.html ? | 20:25 |
| clarkb | https://zuul-ci.org/docs/zuul/admin/components.html | 20:26 |
| clarkb | ya that | 20:26 |
| fdegir | i like the ascii one more but this works too ;) | 20:26 |
| fdegir | thanks | 20:27 |
| mordred | fdegir: https://opendev.org/inaugust/inaugust.com/src/branch/master/src/zuulv3/test-the-future.rst | 21:04 |
| mordred | fdegir: is the latest intro version - but there are several versions of the talk in that directory | 21:04 |
| fdegir | mordred: thanks! | 21:07 |
| mordred | fdegir: (I need to do a better job organizing them for purpose) | 21:08 |
| pabelanger | mordred: clarkb: https://review.openstack.org/649371/ actually updates it to add optional db | 21:09 |
| *** hashar has quit IRC | 21:32 | |
| *** themroc has quit IRC | 21:32 | |
| SpamapS | components is missing the SQL db | 21:56 |
| SpamapS | pabelanger:derp, your answer was better | 21:57 |
| openstackgerrit | Merged openstack-infra/zuul master: Add database info for components graph https://review.openstack.org/649371 | 22:30 |
| clarkb | pabelanger: SpamapS ^ fyi | 22:39 |
| pabelanger | woot | 22:43 |
| mnaser | is there a generic ansible-lint job available? | 22:44 |
| clarkb | I want to say now beacuse we have a bunch of different rule exclusions in ours and pin to < 4? | 22:46 |
| clarkb | s/now/no/ | 22:46 |
| mnaser | clarkb: gotcha | 23:01 |
| SpamapS | yo dawg, I heard you like ansible so I wrote an ansible role to lint your ansible roles. | 23:13 |
| pabelanger | mnaser: I usually just wrap it in tox.ini as linters entry point | 23:14 |
| mnaser | yeah I figured as much | 23:15 |
| openstackgerrit | Tristan Cacqueray proposed openstack-infra/zuul-jobs master: Add ansible-lint job https://review.openstack.org/532083 | 23:21 |
| mnaser | oh good timing | 23:22 |
| mnaser | :P | 23:22 |
| tristanC | mnaser: you might be interested by this one too https://review.openstack.org/#/c/530682/, we are using it a lot to provide generic linters without bothering with tox | 23:25 |
| mnaser | tristanC: cool, I'll check it out | 23:26 |
| openstackgerrit | Clint 'SpamapS' Byrum proposed openstack-infra/nodepool master: Implement max-servers for AWS driver https://review.openstack.org/649474 | 23:36 |
| SpamapS | ^^ going to try this out before it merges anyway, but, that is kind of killing us at the moment and I"m not even sure it works.. the test is a little bit thin | 23:37 |
| SpamapS | I actually think it doesn't work, because it doesn't read the total number of servers already running. :-P | 23:37 |
| * SpamapS is sure the test is deeply flawed | 23:37 | |
| *** ianychoi has joined #zuul | 23:46 | |
| *** rlandy|ruck has quit IRC | 23:55 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!
