| *** ssbarnea has quit IRC | 00:25 | |
| ianw | have there been webui changes recently? I've noticed that just leaving it with once change "pulled down", eventually it refreshes at some point and decides to expand out another change all together | 01:19 |
|---|---|---|
| ianw | it's almost like as the screen changes distance, it unrolls whatever change is at the point in the screen that my originally expanded one was, if that makes sense | 01:20 |
| clarkb | ianw I think that ue a bug in the react rewrite | 01:21 |
| clarkb | ya I wonder if it uses an index on thr list that it thinks will be stable | 01:21 |
| ianw | yeah, that would explain it quite well | 01:23 |
| *** swest has quit IRC | 02:31 | |
| tristanC | that's the issue indeed, the 'key' property seems to be in charge of keeping track of component state, and it's currently set with list indexes | 02:44 |
| *** caphrim007_ has joined #zuul | 02:55 | |
| caphrim007_ | any zuul folks around? | 02:55 |
| openstackgerrit | Tristan Cacqueray proposed openstack-infra/zuul master: web: uses queues uid to preserve state on change https://review.openstack.org/614933 | 02:59 |
| *** bhavikdbavishi has joined #zuul | 02:59 | |
| tristanC | ianw: while i'm at it, are you aware of other issues/regressions with the new webui? | 03:05 |
| tristanC | caphrim007_: may I help you? | 03:05 |
| caphrim007_ | tristanC: I'm having a helluva time trying to determine why this error is popping up http://paste.openstack.org/show/733926/ | 03:06 |
| ianw | tristanC: not that i've noticed, yet :) | 03:06 |
| caphrim007_ | https://github.com/F5Networks/f5-ansible/tree/devel/.zuul.d | 03:06 |
| caphrim007_ | is my zuul config that it's mentioning | 03:07 |
| caphrim007_ | but, i clearly have names indented under my job and project so... | 03:07 |
| tristanC | caphrim007_: i don't think you are supposed to keep playbooks in zuul.d directory, maybe try moving those out of the way? | 03:08 |
| caphrim007_ | k, trying | 03:08 |
| tristanC | caphrim007_: it seems like the issue is that part: https://github.com/F5Networks/f5-ansible/blob/devel/.zuul.d/playbooks/job-playbook.yaml#L3-L5 | 03:10 |
| tristanC | that dictionary object doesn't have a single key as expected | 03:10 |
| caphrim007_ | then so i understand | 03:10 |
| caphrim007_ | zuul is recursively looking for all yaml files? | 03:11 |
| tristanC | caphrim007_: iirc, zuul only loads files ending with '.yaml', recursively | 03:11 |
| caphrim007_ | ok, checking | 03:11 |
| tristanC | caphrim007_: perhaps you could work around this by using '.yml' extension? | 03:11 |
| tristanC | though it's probably better to just move the playbook out of the zuul.d for now | 03:11 |
| caphrim007_ | tristanC: you hit the nail on the head. that was the problem | 03:13 |
| caphrim007_ | geez, i sat here for an hour smacking my head about that. ugh. thanks so much for the help tristanC | 03:14 |
| tristanC | caphrim007_: you're welcome, glad i could help :-) | 03:17 |
| tristanC | ianw: 614933 seems to fix the expand shifting position issue, to quickly reproduce you can filter, expand a change, and remove the filter | 03:18 |
| tristanC | ianw: without 614933, after removing the filter, the first change of the pipelines is expanded, and with the fix, the filtered change stays expanded after removing the filter | 03:19 |
| *** bhavikdbavishi1 has joined #zuul | 04:07 | |
| *** bhavikdbavishi has quit IRC | 04:09 | |
| *** bhavikdbavishi1 is now known as bhavikdbavishi | 04:09 | |
| *** pbrobinson has quit IRC | 04:09 | |
| *** toabctl has quit IRC | 04:13 | |
| *** toabctl has joined #zuul | 04:14 | |
| *** caphrim007_ has quit IRC | 04:54 | |
| *** threestrands has quit IRC | 05:40 | |
| *** pcaruana has joined #zuul | 07:20 | |
| *** goern has joined #zuul | 07:43 | |
| *** bhavikdbavishi has quit IRC | 08:25 | |
| *** ssbarnea has joined #zuul | 09:15 | |
| *** panda|off is now known as panda | 10:03 | |
| *** sshnaidm|off has quit IRC | 10:10 | |
| *** sshnaidm has joined #zuul | 10:14 | |
| *** sshnaidm is now known as sshnaidm|off | 10:24 | |
| *** EmilienM is now known as EvilienM | 10:25 | |
| *** electrofelix has joined #zuul | 11:00 | |
| frickler | is there a way to do negative filtering on http://zuul.openstack.org/builds , like all builds with "Result != SUCCESS" or "Job != *tripleo*"? | 11:37 |
| goern | hey all, has anyone worked on using a zuul job running on openshift to build, waiting for the new image and push the image to a docker registry?! | 11:41 |
| openstackgerrit | Vieri proposed openstack-infra/zuul-base-jobs master: Update min tox version to 2.0 https://review.openstack.org/615148 | 11:56 |
| *** zul has quit IRC | 12:06 | |
| *** strigazi has joined #zuul | 12:07 | |
| strigazi | Hello, I'm trying to encrypt a secret with tools/encrypt_secret.py from openstack-infra/zuul but I can't find the zuul server url. It is not zuul.openstack.org | 12:07 |
| openstackgerrit | Vieri proposed openstack-infra/zuul-sphinx master: Update min tox version to 2.0 https://review.openstack.org/615162 | 12:11 |
| openstackgerrit | Vieri proposed openstack-infra/zuul master: Update min tox version to 2.0 https://review.openstack.org/615164 | 12:13 |
| openstackgerrit | Vieri proposed openstack-infra/zuul-jobs master: Update min tox version to 2.0 https://review.openstack.org/615172 | 12:22 |
| strigazi | I tried also to fetch kolla's public key that should and it is not there: curl https://zuul.openstack.org/api/tenant/openstack/key/kolla.pub | 12:23 |
| strigazi | per documentation it should be found in: They can be fetched at the path /api/tenant/<tenant>/key/<project>.pub where <project> is the canonical name of a project and <tenant> is the name of a tenant with that project. | 12:23 |
| strigazi | https://zuul-ci.org/docs/zuul/user/encryption.html | 12:24 |
| *** zul has joined #zuul | 12:37 | |
| Shrews | strigazi: tools/encrypt_secret.py --infile file_with_secret --tenant openstack https://zuul.openstack.org openstack/kolla | 13:02 |
| *** panda is now known as panda|lunch | 13:10 | |
| *** panda|lunch is now known as panda | 13:48 | |
| *** rfolco is now known as rfolco|off | 14:11 | |
| openstackgerrit | Monty Taylor proposed openstack-infra/nodepool master: Consume rate limiting task manager from openstacksdk https://review.openstack.org/612169 | 14:27 |
| mordred | Shrews: ^^ and there's an update to the nodepool patch to show consuming that | 14:28 |
| *** jimi|ansible has quit IRC | 14:31 | |
| Shrews | mordred: yeah, but those test failures concern me | 14:41 |
| Shrews | no idea what's going on | 14:41 |
| mordred | Shrews: me either. | 14:48 |
| *** j^2 has joined #zuul | 14:57 | |
| *** ssbarnea has quit IRC | 15:08 | |
| *** ssbarnea has joined #zuul | 15:10 | |
| *** chandankumar is now known as chkumar|off | 15:14 | |
| openstackgerrit | James E. Blair proposed openstack-infra/zuul master: Merger: automatically add new hosts to the known_hosts file https://review.openstack.org/608453 | 15:19 |
| openstackgerrit | James E. Blair proposed openstack-infra/zuul master: WIP: support foreign required-projects https://review.openstack.org/613143 | 15:22 |
| mrhillsman | can someone point me to the docs where i can run a job that requires multiple nodes; is that a thing? | 15:35 |
| clarkb | mrhillsman: yes, you create a nodeset with multiple nodes in it | 15:35 |
| clarkb | mrhillsman: https://zuul-ci.org/docs/zuul/user/config.html#nodeset the docs for nodesets | 15:36 |
| mrhillsman | thx | 15:36 |
| mrhillsman | was just doing a good ol docs search :) | 15:36 |
| logan- | https://github.com/openstack-infra/openstack-zuul-jobs/blob/master/zuul.d/nodesets.yaml#L98-L111 too | 15:37 |
| mrhillsman | thx | 15:38 |
| *** AJaeger has quit IRC | 15:46 | |
| *** AJaeger has joined #zuul | 15:48 | |
| *** caphrim007 has joined #zuul | 15:52 | |
| caphrim007 | morning folks. i had a question | 15:57 |
| caphrim007 | if i have single project, then i can make many jobs for that single project | 15:57 |
| caphrim007 | and can each of those jobs have its own nodeset? | 15:57 |
| caphrim007 | and, do those jobs run in parallel? or sequentially? | 15:57 |
| clarkb | caphrim007: each job can have its own nodeset and they will run in parallel as long as nodepool has available resources that allow it to spin up those nodesets in parallel | 15:58 |
| caphrim007 | clarkb: cool thanks for the info | 15:58 |
| caphrim007 | would i need to change zuul-executor to run X jobs? does it have a default limit on the number of jobs it runs at once? | 15:59 |
| clarkb | caphrim007: the executor has a few governors in place where it will stop accepting new jobs once memory, cpu, and disk thresholds are met ( I forget the specific details, but the idea is you shouldn't have to tune that yourself, isntead th executor will do as much work as it can without running out of system resources) | 16:00 |
| caphrim007 | clarkb: gotcha. thanks! | 16:00 |
| clarkb | caphrim007: the openstack zuul instance has 11 executors to keep up with our demand. I think that roughly comes out to ~100 test nodes per executor and ~70 jobs per executor at any time given our average use | 16:02 |
| clarkb | http://grafana.openstack.org/d/T6vSHcSik/zuul-status?orgId=1 has specific numbers. In general though that is a function of the size of the instances too | 16:02 |
| caphrim007 | clarkb: ok. i don't have anywhere near that much...but as I get more resources in my cloud accounts I would want to ramp it up to run my full test suite as fast as possible | 16:03 |
| clarkb | yup and you should be able to scale the executors horizontally as necessary to achieve that | 16:03 |
| corvus | clarkb: are you happy with the nodepool sha from yesterday? | 16:07 |
| clarkb | corvus: yes, I haven't noticed any problems and we appear to be using our full capactiy. Let me double check that images uploaded properly though | 16:07 |
| clarkb | | 0000000012 | 0000000001 | rax-dfw | ubuntu-bionic | ubuntu-bionic-1541137033 | 7144fc56-6ba4-4a41-9f73-b11ecd3c35bc | ready | 00:09:58:52 | | 16:08 |
| clarkb | that lgtm. I am happy with nodepool as is running now | 16:08 |
| corvus | zuul 3.3.0 and nodepool 3.3.1 tagged | 16:10 |
| *** AJaeger_ has joined #zuul | 16:13 | |
| *** AJaeger has quit IRC | 16:15 | |
| *** pcaruana has quit IRC | 16:17 | |
| *** AJaeger_ is now known as AJaeger | 16:20 | |
| *** zul has quit IRC | 16:53 | |
| *** panda is now known as panda|off | 17:36 | |
| -openstackstatus- NOTICE: OpenStack infra's mirror nodes stopped accepting connections on ports 8080, 8081, and 8082. We will notify when this is fixed and jobs can be rechecked if they failed to communicate with a mirror on these ports. | 18:11 | |
| *** electrofelix has quit IRC | 18:24 | |
| *** pcaruana has joined #zuul | 18:27 | |
| *** jimi|ansible has joined #zuul | 18:46 | |
| openstackgerrit | Andreas Jaeger proposed openstack/pbrx master: Use publish-to-pypi https://review.openstack.org/615296 | 18:49 |
| *** EvilienM is now known as EmilienM | 18:53 | |
| -openstackstatus- NOTICE: The firewall situation with ports 8080, 8081, and 8082 on mirror nodes has been resolved. You can recheck jobs that have failed to communicate to the mirrors on those ports now. | 18:55 | |
| *** pcaruana has quit IRC | 19:08 | |
| *** caphrim007 has quit IRC | 20:12 | |
| goern | heyall, its there a plan to extend the github driver to also understand checks in addition to statuses? | 20:57 |
| mordred | goern: yes - as well as to support in-line comments from zuul on the PR | 21:03 |
| corvus | goern: yes -- jlk was first implementing support for that in github3.py (the library zuul uses to interact with github) | 21:04 |
| mordred | goern: my understanding is that each require updates to the pygithub3 library we use - and jlk, who works at github now, has been working on getting that done | 21:04 |
| mordred | corvus: kinx | 21:04 |
| mordred | jinz | 21:04 |
| mordred | WOW - I cannot type jinx apparently | 21:04 |
| corvus | mordred: your keyboard is geenksed. | 21:05 |
| mordred | corvus: clearly | 21:05 |
| goern | cool, thx ya. any timeframe for that? | 21:05 |
| goern | or something where help is needed? other than alpha testing :) | 21:05 |
| corvus | lemme look up the pr | 21:05 |
| corvus | goern: https://github.com/sigmavirus24/github3.py/pull/888 | 21:06 |
| mordred | goern: once that support has landed in the library, we'll need someone to update the gh driver to use it - that hasn't started yet | 21:06 |
| mordred | corvus: although, I suppose someone could start on that with a depends-on that PR :) | 21:07 |
| corvus | mordred: true :) | 21:07 |
| goern | #888 looks pretty complete and close to merge?! | 21:08 |
| corvus | yeah, hopefully jlk can weigh in on what else is necessary :) | 21:08 |
| goern | and the gh driver... any pointer and some spare time to mentor someone? :) | 21:09 |
| corvus | goern: i can help a bit with general driver questions (and it closely follows the gerrit driver, so that's quite a bit actually), jlk and tobiash are more expert in github details | 21:13 |
| goern | nice, tobias is in my TZ, so I will give him a ping | 21:14 |
| openstackgerrit | James E. Blair proposed openstack-infra/zuul master: DNM: extra debugging https://review.openstack.org/615330 | 21:30 |
| *** rlandy has quit IRC | 21:41 | |
| openstackgerrit | Merged openstack/pbrx master: Use publish-to-pypi https://review.openstack.org/615296 | 21:47 |
| *** j^2 has quit IRC | 22:05 | |
| openstackgerrit | James E. Blair proposed openstack-infra/zuul master: DNM: extra debugging https://review.openstack.org/615330 | 22:19 |
| *** ianychoi has quit IRC | 23:02 | |
| openstackgerrit | James E. Blair proposed openstack-infra/zuul master: WIP: Set relative priority of node requests https://review.openstack.org/615356 | 23:24 |
| clarkb | corvus: what is with the ssh host key debugging? are we having problems with that again? | 23:30 |
| openstackgerrit | James E. Blair proposed openstack-infra/zuul master: Merger: automatically add new hosts to the known_hosts file https://review.openstack.org/608453 | 23:30 |
| corvus | clarkb: nah, it's me trying to make the merger automatically add ssh keys to known hosts. that change ^ | 23:31 |
| corvus | clarkb: the quickstart test was failing on that and i couldn't figure out why without crazy debugging. i figured it out. it was because the actual username comes from the url that we give the merger. | 23:32 |
| corvus | clarkb: regarding the change itself -- in the scheduler, paramiko is set up to ignore unknown host keys. but the *git* subprocess in the merger can't do that, because it relies on the openssh client. there's a new feature in openssh to auto-add unknown keys, but it's too bleeding edge for us. so instead, i'm trying to have the merger perform a paramiko connection solely to set up a known_hosts file before | 23:34 |
| corvus | we perform our first git operation on a repo in the merger. | 23:34 |
| corvus | clarkb: long story short -- a new user setting up zuul shouldn't have to do anything with ssh host keys on any component of zuul | 23:34 |
| corvus | (it's one of the worst parts of bootstrapping a zuul right now) | 23:35 |
| clarkb | interesting, and I guess if you wanted to have verification and control you could preconfigure a key? | 23:35 |
| corvus | also -- makes it easier to add new connections (as long as they aren't that version of gerrit that does weird things with host keys) | 23:35 |
| clarkb | woudl zuul honor a host keys file if it was already there? | 23:35 |
| corvus | clarkb: exactly. yes. | 23:35 |
| corvus | clarkb: paramiko and openssh will still throw a fit if the ssh key is *wrong* | 23:35 |
| corvus | so if it changes after the first connection, you're still safe. | 23:36 |
| corvus | if you're paranoid about the first connection, then prepopulate known_hosts. i'd also be happy to make an option to disable this (if you know you always want to be paranoid and don't want to accidentally accept a host key on first connect) | 23:37 |
| corvus | though i think for a default, the behavior i'm proposing (accept on first connect) is good for most cases. not everyone runs their zuul across the public internet. :) | 23:38 |
| clarkb | For myself I'm usually less paranoid about making a first connection out of the bluethat would be hard for someone to know to itnercept (of course maybe I should be more paranoid) | 23:38 |
| clarkb | checking that things don't change later is my biggest concern, but I'm also sure there are individuals that would like to supply the keys themselves as well | 23:39 |
| corvus | yep. we can accomodate all cases | 23:44 |
| *** ianychoi has joined #zuul | 23:55 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!