| openstackgerrit | Tristan Cacqueray proposed openstack-infra/zuul master: config: add playbooks to job.toDict() https://review.openstack.org/621343 | 00:19 |
|---|---|---|
| mordred | mrhillsman: I just pushed up patches to gophercloud and cloud-provider-openstack and as a result got test results from your zuul - thanks! | 00:19 |
| mordred | mrhillsman: fwiw, it was MUCH easier to figure out what was wrong with my gophercloud patch from openlab than it was from travis | 00:20 |
| mrhillsman | ;) | 00:22 |
| mrhillsman | we hope to get that latest zuul up and running to make it even easier | 00:23 |
| mrhillsman | i really like the extra tabs | 00:23 |
| mrhillsman | hopefully our xcommunity work will yield some more zuul patches | 00:24 |
| mrhillsman | a k8s community member messaged me on how to get involved and i pointed them here | 00:25 |
| openstackgerrit | Tristan Cacqueray proposed openstack-infra/zuul master: config: add playbooks to job.toDict() https://review.openstack.org/621343 | 01:03 |
| openstackgerrit | Tristan Cacqueray proposed openstack-infra/zuul master: config: add tenant.toDict() method and REST endpoint https://review.openstack.org/621344 | 01:03 |
| openstackgerrit | Tristan Cacqueray proposed openstack-infra/zuul master: config: add tenant.toDict() method and REST endpoint https://review.openstack.org/621344 | 02:35 |
| openstackgerrit | Tristan Cacqueray proposed openstack-infra/nodepool master: Implement a Runc driver https://review.openstack.org/535556 | 02:54 |
| *** bhavikdbavishi has joined #zuul | 03:07 | |
| openstackgerrit | Tristan Cacqueray proposed openstack-infra/nodepool master: Implement zookeeper-auth https://review.openstack.org/619155 | 03:37 |
| openstackgerrit | Tristan Cacqueray proposed openstack-infra/nodepool master: print-zk: convert zookeeper_servers to a list https://review.openstack.org/629754 | 03:37 |
| openstackgerrit | Tristan Cacqueray proposed openstack-infra/zuul master: Implement zookeeper-auth https://review.openstack.org/619156 | 03:39 |
| openstackgerrit | Tristan Cacqueray proposed openstack-infra/nodepool master: Implement zookeeper-auth https://review.openstack.org/619155 | 04:50 |
| openstackgerrit | Tristan Cacqueray proposed openstack-infra/zuul master: Implement zookeeper-auth https://review.openstack.org/619156 | 04:50 |
| openstackgerrit | Tristan Cacqueray proposed openstack-infra/zuul master: config: add tenant.toDict() method and REST endpoint https://review.openstack.org/621344 | 04:53 |
| openstackgerrit | Tristan Cacqueray proposed openstack-infra/nodepool master: Implement an OpenShift resource provider https://review.openstack.org/570667 | 05:05 |
| openstackgerrit | Tristan Cacqueray proposed openstack-infra/nodepool master: Implement an OpenShift Pod provider https://review.openstack.org/590335 | 05:47 |
| openstackgerrit | Tristan Cacqueray proposed openstack-infra/zuul master: web: add jobs graph rendering https://review.openstack.org/537869 | 06:25 |
| *** quiquell|off is now known as quiquell | 07:19 | |
| *** bhavikdbavishi has quit IRC | 07:35 | |
| *** hashar has joined #zuul | 07:44 | |
| *** pcaruana has joined #zuul | 07:51 | |
| *** avass has joined #zuul | 07:52 | |
| *** bhavikdbavishi has joined #zuul | 07:57 | |
| *** hashar has quit IRC | 08:06 | |
| *** jpena|off is now known as jpena | 08:08 | |
| *** gtema has joined #zuul | 08:38 | |
| avass | tobiash: Is pywinrm supposed to be pre-installed on the executor docker image? because I'm getting an error from ansible that there's no winrm module | 08:54 |
| tobiash | no, you need to install it separately afaik | 08:55 |
| *** logan- has quit IRC | 09:21 | |
| *** logan- has joined #zuul | 09:28 | |
| *** logan- has quit IRC | 09:28 | |
| *** themroc has joined #zuul | 10:06 | |
| *** hashar has joined #zuul | 10:09 | |
| jkt | pabelanger: ah, I haven't realized that you're from RH, I didn't mean my question about timeline like that :) | 10:24 |
| *** sshnaidm|afk is now known as sshnaidm | 10:31 | |
| *** openstackgerrit has quit IRC | 11:05 | |
| *** electrofelix has joined #zuul | 11:28 | |
| *** dkehn has quit IRC | 11:35 | |
| *** hashar has quit IRC | 11:38 | |
| *** bhavikdbavishi has quit IRC | 12:00 | |
| *** hashar has joined #zuul | 12:41 | |
| *** jpena is now known as jpena|lunch | 12:43 | |
| *** bhavikdbavishi has joined #zuul | 13:04 | |
| *** rcarrillocruz has joined #zuul | 13:24 | |
| *** bhavikdbavishi has quit IRC | 13:31 | |
| ssbarnea|rover | is there are way to avoid this http://git.openstack.org/cgit/openstack-infra/zuul/tree/zuul/ansible/callback/zuul_stream.py#n304 -- is really annoying to see ERROR: All items complete -- on logs as this is highligted as an error my my regex-es. | 13:41 |
| ssbarnea|rover | not sure why "all items complete" counts as an error. | 13:41 |
| *** rlandy has joined #zuul | 13:42 | |
| *** jpena|lunch is now known as jpena | 13:45 | |
| *** dkehn has joined #zuul | 13:52 | |
| *** dkehn has quit IRC | 14:13 | |
| *** electrofelix has quit IRC | 14:31 | |
| Shrews | corvus: tobiash: care to put another set of eyes on the openshift driver changes? https://review.openstack.org/570667 | 14:46 |
| tobiash | Shrews: I commented already today with some nits and a missing field in the docs | 14:47 |
| tobiash | I wasn't sure if that's -1 worthy | 14:47 |
| tobiash | I'd say yes with option to fix this in a followup | 14:47 |
| *** dkehn has joined #zuul | 14:48 | |
| Shrews | tobiash: ah, missed that. thx | 14:51 |
| *** _ari_ has joined #zuul | 14:59 | |
| *** openstackgerrit has joined #zuul | 15:00 | |
| openstackgerrit | Merged openstack-infra/nodepool master: print-zk: convert zookeeper_servers to a list https://review.openstack.org/629754 | 15:00 |
| *** rcarrillocruz has quit IRC | 15:25 | |
| avass | tobiash: is it possible to make a job only run if a job it depends on is not skipped? | 15:27 |
| corvus | Shrews, tobiash, tristanC: looks great. let's make sure there's at least a patch for the missing image-pull doc before we approve so we don't forget about that. | 15:37 |
| corvus | Shrews, tobiash: the depend-on could user another +2 | 15:37 |
| corvus | avass: if a job depends on a job that is skipped, i don't think it should run; do you see that it does? | 15:38 |
| avass | yeah it still does, the documentation is a bit ambiguous but it seems like it runs as long as it does not fail | 15:38 |
| avass | unless i missed something | 15:39 |
| corvus | avass: just to make sure i'm clear: C depends on B which depends on A; A fails, b is skipped, c runs (but should be skipped). right? | 15:42 |
| avass | A is dependent on file changes B should not run if A is skipped | 15:43 |
| corvus | avass: A has a "file:" attribute in the job definition? | 15:44 |
| avass | yeah | 15:44 |
| corvus | avass: so in the report, A doesn't run at all, right? it doesn't show up as "SKIPPED" it just isn't there? | 15:44 |
| avass | exactly | 15:45 |
| corvus | avass: one more question -- you're not using zuul_return.child_jobs are you? (where, in job A, you return a 'child_jobs' attribute to zuul to tell it not to run B) | 15:45 |
| avass | nope | 15:46 |
| corvus | avass: what version are you running? | 15:53 |
| avass | we pulled today | 15:53 |
| *** quiquell is now known as quiquell|off | 15:54 | |
| corvus | avass: i just wrote a unit test for this, and i see that zuul should report back this error: Unable to freeze job graph: Job child-job depends on parent-job which was not run. | 15:54 |
| corvus | (i'm not sure that error is what we want either, since it's not running *any* jobs, but it makes me think i haven't quite reproduced your situation) | 15:55 |
| openstackgerrit | James E. Blair proposed openstack-infra/zuul master: WIP: Test a child job depending on a parent that did not run https://review.openstack.org/629885 | 15:56 |
| avass | B is not inherited from A, only set as dependent. if that help | 15:57 |
| corvus | avass: ^ does the 'job-dependency.yaml' file in that change look like the right configuration? | 15:57 |
| corvus | (yeah, my use of parent/child here is confusing, sorry. i did not use inheritance) | 15:57 |
| avass | yes, only that dependecies is set in the child job definition | 15:58 |
| openstackgerrit | James E. Blair proposed openstack-infra/zuul master: WIP: Test a child job depending on a parent that did not run https://review.openstack.org/629885 | 15:59 |
| corvus | avass: like that? | 15:59 |
| avass | yes | 16:00 |
| corvus | the child-job did run this time, so i think we've got a reproducer | 16:00 |
| openstackgerrit | Merged openstack-infra/zuul-jobs master: Add install and deploy openshift roles. https://review.openstack.org/608610 | 16:00 |
| corvus | avass: so i think we have one bug where we are ignoring the dependencies attribute when it's set there, and if we fix that, we'd probably end up with the error message from earlier. i'm assuming it would be preferable to simply not run the child job in that case, so we'll have to change that too. | 16:02 |
| avass | Maybe it could be configurable later. But it is nice if it behaves the same way at least. | 16:04 |
| *** sshnaidm has quit IRC | 16:10 | |
| *** sshnaidm has joined #zuul | 16:11 | |
| sshnaidm | do you have any example how to configure user data for nodepool? https://zuul-ci.org/docs/nodepool/configuration.html#attr-providers.[openstack].pools.labels.instance-properties | 16:26 |
| clarkb | sshnaidm: the openstack nodepool config does it to set a flag for ovh instances and glean. Let me get a link | 16:28 |
| clarkb | sshnaidm: https://git.openstack.org/cgit/openstack-infra/project-config/tree/nodepool/nl04.openstack.org.yaml#n67 | 16:28 |
| sshnaidm | clarkb, but it's not user-data, right? | 16:31 |
| clarkb | user-data is just a type of instance property I thought. But maybe it is distinct in nova? | 16:32 |
| clarkb | hrm with osc it is its own flag so maybe it is distinct | 16:32 |
| clarkb | ya confirmed with nova api docs. metadata and user_data are separate fields | 16:34 |
| sshnaidm | clarkb, so for passing some cloud config, what should be done? I tried like: instance-properties: | 16:35 |
| sshnaidm | user-data: userdata.txt | 16:35 |
| clarkb | I don't think that will work. That is what I assumed will work but rereading api docs I'm pretty sure it won't | 16:36 |
| clarkb | nodepool likely needs explicit support for this which I dn't think it has currently | 16:36 |
| sshnaidm | clarkb, oh, I see.. | 16:36 |
| sshnaidm | clarkb, maybe you know - if I have 2 users in image, how can I assign public keys to both of them with keypair? | 16:37 |
| sshnaidm | because currently it's assigned to first user only | 16:37 |
| clarkb | sshnaidm: that would be a cloud init configuration item aiui | 16:37 |
| clarkb | ubuntu sets it for the ubuntu and root users so should be possible | 16:38 |
| clarkb | if you are using glean, glean only supports root user | 16:38 |
| sshnaidm | clarkb, yeah, it's what I tried - to inject cloud init config using instance-properties in nodepool.. :) | 16:38 |
| sshnaidm | I need the nodepool to assign pub keys to all users.. | 16:39 |
| clarkb | nodepool doesn't assign any keys to anything. That is up to the image (and nova to pass the data) | 16:40 |
| sshnaidm | clarkb, yeah, I mean when it spawns a vm on openstack with key_name: ... | 16:41 |
| clarkb | yes that is up to your cloud init config which you can set in the image | 16:42 |
| clarkb | nova's api doesn't distinguish keys per user in metadata. But you can instruct your init tooling to apply the key to multiple users | 16:42 |
| sshnaidm | clarkb, but how do I set it on the image? | 16:43 |
| clarkb | sshnaidm: write the config file as part of the nodepool image builds with dib would be one way | 16:43 |
| sshnaidm | clarkb, and then I need to build it.. and if I have already built image? | 16:44 |
| *** avass has quit IRC | 16:45 | |
| clarkb | if you have an already built image that you cannot modify it will depend on however it is already configured. | 16:46 |
| *** themroc has quit IRC | 16:48 | |
| *** gtema has quit IRC | 16:54 | |
| *** hashar has quit IRC | 17:13 | |
| *** avass has joined #zuul | 17:21 | |
| avass | thanks for the help earlier | 17:21 |
| clarkb | sshnaidm: reading further on cloud-init docs it appears that if you want to side load user configs then you need to set user-data. SO probably the thing to do is add support for user data to nodepool and in the meantime modify the image? | 17:26 |
| *** jpena is now known as jpena|off | 17:32 | |
| *** panda is now known as panda|off | 17:53 | |
| *** electrofelix has joined #zuul | 18:15 | |
| *** electrofelix has quit IRC | 18:20 | |
| *** nhicher has joined #zuul | 18:27 | |
| *** rcarrillocruz has joined #zuul | 18:27 | |
| corvus | avass: thanks for the report -- i'll try to work on fixing those issues | 18:43 |
| *** electrofelix has joined #zuul | 19:02 | |
| *** electrofelix has quit IRC | 19:04 | |
| *** rcarrillocruz has quit IRC | 19:15 | |
| *** rcarrillocruz has joined #zuul | 19:19 | |
| *** rcarrillocruz has quit IRC | 19:39 | |
| openstackgerrit | David Shrewsbury proposed openstack-infra/nodepool master: Add a timeout for the image build https://review.openstack.org/629923 | 19:47 |
| Shrews | clarkb: i *think* that will prevent the situation you noticed yesterday ^^^. It changes build log buffer writing a bit to not flush per-line but rather by chunk, but I think that's ok? | 19:49 |
| Shrews | readline() blocking seems to be a common thing folks have issues with. it was easier to remove it entirely | 19:51 |
| clarkb | Shrews: thanks. | 19:52 |
| *** openstackgerrit has quit IRC | 20:50 | |
| *** openstackgerrit has joined #zuul | 20:52 | |
| openstackgerrit | Merged openstack-infra/zuul-base-jobs master: Add ensure-output-dirs to base jobs https://review.openstack.org/628674 | 20:52 |
| openstackgerrit | Merged openstack-infra/zuul-base-jobs master: Add fetch-output to base jobs https://review.openstack.org/628975 | 20:52 |
| openstackgerrit | Merged openstack-infra/zuul-base-jobs master: Ignore errors on ssh key removal https://review.openstack.org/628976 | 20:52 |
| *** rlandy is now known as rlandy|biab | 21:25 | |
| *** pcaruana has quit IRC | 21:46 | |
| *** avass has quit IRC | 22:07 | |
| *** rlandy|biab is now known as rlandy | 22:36 | |
| openstackgerrit | Monty Taylor proposed openstack-infra/zuul-jobs master: Add role to move docs and artifacts to log root https://review.openstack.org/629571 | 22:48 |
| *** etp has quit IRC | 22:57 | |
| *** etp has joined #zuul | 22:59 | |
| *** etp has quit IRC | 23:14 | |
| *** etp has joined #zuul | 23:16 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!