| *** crobinso has quit IRC | 00:27 | |
| *** dklyle has quit IRC | 02:01 | |
| *** david-lyle has joined #kata-dev | 02:01 | |
| *** david-lyle has quit IRC | 02:18 | |
| *** fuentess has quit IRC | 02:53 | |
| kata-irc-bot | <dgibson> So.. who would I need to talk to to try to convince people that the tests repo should be integrated with the kata-containers repo for 2.x | 06:14 |
|---|---|---|
| *** fgiudici has joined #kata-dev | 07:57 | |
| kata-irc-bot | <fidencio> A good start would be trying to describe the problem you're facing, how it's related to the tests repo not being to integrated with the kata-containers one. | 08:00 |
| *** sgarzare has joined #kata-dev | 08:06 | |
| *** jodh has joined #kata-dev | 08:24 | |
| kata-irc-bot | <ssheribe> Hi, I'm looking at the delete process in openshift/k8s+crio, it seems the whole pod/sandbox/vm (& its containers) are deleted when the container representing the sandbox is stopped (StopContainer), hence, when it follows by the request to delete/remove (DeleteContainer) of the inner container it was already deleted. it seems to me that post-stop, maybe also other stages are not preformed this way, any idea how the agent itself 1.x/2.0 | 08:33 |
| kata-irc-bot | affect on this behavior? would it be possible to avoid waiting on the container and just delete upon request? | 08:33 |
| kata-irc-bot | <ssheribe> @fupan tagging you as it's seems to be related to this | 08:35 |
| kata-irc-bot | <fupan> If the container had been deleted, it did had the chance to do the post-stop actions, please see https://github.com/kata-containers/kata-containers/blob/74ebd96f99447a440b46acb019044cc94fe770b1/src/runtime/containerd-shim-v2/delete.go#L25 | 08:50 |
| kata-irc-bot | <ssheribe> Isn't it a problem? does post-stop should be executed also when it's deleted from here | 08:54 |
| *** davidgiluk has joined #kata-dev | 09:01 | |
| *** sameo has quit IRC | 09:02 | |
| kata-irc-bot | <ssheribe> Right, but for example the UnmountAll is also always failing, should it be skipped? | 09:06 |
| *** sameo has joined #kata-dev | 09:21 | |
| kata-irc-bot | <fupan> The fail would only produce an warning info and there’s no any side effect, and further, it would make sure there’s no mount leak. | 09:22 |
| *** fgiudici has quit IRC | 12:24 | |
| *** sameo has quit IRC | 13:03 | |
| *** sgarzare has quit IRC | 13:13 | |
| *** devimc has joined #kata-dev | 13:30 | |
| *** sameo has joined #kata-dev | 13:31 | |
| *** vgoyal has joined #kata-dev | 13:43 | |
| *** fgiudici has joined #kata-dev | 13:49 | |
| *** fuentess has joined #kata-dev | 13:58 | |
| *** sgarzare has joined #kata-dev | 14:16 | |
| *** crobinso has joined #kata-dev | 14:31 | |
| *** devimc has quit IRC | 15:26 | |
| *** devimc has joined #kata-dev | 15:26 | |
| *** dklyle has joined #kata-dev | 15:53 | |
| kata-irc-bot | <eric.ernst> Hey ya'll, I'm trying to finish 2.0 stable releases. It looks like http://jenkins.katacontainers.io/job/kata-containers-2.0-fedora-PR-crio-k8s-e2e-minimal/ has had some instability as of late. | 16:07 |
| kata-irc-bot | <eric.ernst> I'm wondering if we're missing any test backports to stable-2.0.0? I'd like to get this bug fix backported so I can complete the release. I see it isn't a blocking test -- WDYT about skipping? | 16:08 |
| kata-irc-bot | <eric.ernst> @fidencio @wmoschet either ya'll familiar? | 16:08 |
| *** ailan has joined #kata-dev | 16:22 | |
| kata-irc-bot | <fidencio> @eric.ernst ... I'm against skipping it. I'd rather have it failing and not blocking till some of us take a look at that. | 16:22 |
| kata-irc-bot | <eric.ernst> To clarify, sorry, still coffee'ing: OK to merge w/ it failing, or is someone looking? | 16:23 |
| kata-irc-bot | <eric.ernst> My guess is stable-2.0.0 has been neglected on tests side. | 16:23 |
| kata-irc-bot | <fidencio> ah, please, go ahead and merge with it failing | 16:23 |
| kata-irc-bot | <eric.ernst> OK, sounds good. | 16:23 |
| kata-irc-bot | <fidencio> yeah, I only focused on that on 2.0-dev branch | 16:24 |
| kata-irc-bot | <eric.ernst> Look for more simple ACKs soon as I bump 2.0 branches | 16:24 |
| kata-irc-bot | <fidencio> @eric.ernst tag me :slightly_smiling_face: | 16:24 |
| kata-irc-bot | <eric.ernst> will do! Thanks! | 16:26 |
| kata-irc-bot | <wmoschet> ```10:36:55 Get CRI-O sources 10:37:03 package github.com/cri-o/cri-o: no Go files in /tmp/jenkins/workspace/kata-containers-2.0-fedora-PR-crio-k8s-e2e-minimal/go/src/github.com/cri-o/cri-o 10:37:03 .ci/install_crio.sh: line 67: KUBERNETES: unbound variable 10:37:03 CRI-O not installed 10:37:03 containerd not installed``` | 16:57 |
| kata-irc-bot | <wmoschet> @fidencio looking/working on this issue? | 16:57 |
| kata-irc-bot | <wmoschet> in the build #150 of the job @eric.ernst pointed above | 16:57 |
| kata-irc-bot | <eric.ernst> To clarify @wmoschet - this is on stable-2.0.0 branch, so I'm guessing there are test changes on 2.0-dev that need to be backported. | 16:59 |
| kata-irc-bot | <fidencio> @wmoschet, I'm not, busy with other stuff right now :S | 17:00 |
| kata-irc-bot | <fidencio> I guess we never ever tested crio on stable-2.x, only on 2.0-dev, so whatever we have related to CRI-O should be backported there | 17:01 |
| kata-irc-bot | <fidencio> @wmoschet, I can take a look at that, if you prefer, but only Tomorrow | 17:01 |
| kata-irc-bot | <wmoschet> @eric.ernst thanks for the clarification. @fidencio it seems not urgent but I can have a look at it today | 17:02 |
| kata-irc-bot | <fidencio> @wmoschet: f2de04635fd1d8109dd4f0a824795f847b5e5788 may be the one that needs to be backported | 17:03 |
| kata-irc-bot | <fidencio> I've opened https://github.com/kata-containers/tests/pull/3161 | 17:05 |
| kata-irc-bot | <fidencio> Let me run the tests there, but it may be that simple | 17:05 |
| kata-irc-bot | <wmoschet> @fidencio that may solve the problem which is about KUBERNETES not being set | 17:08 |
| *** sgarzare has quit IRC | 17:08 | |
| kata-irc-bot | <fidencio> Yep! Let's see. I'll keep trying to run it successfully after a meeting that I have in 5 | 17:09 |
| kata-irc-bot | <fidencio> @eric.ernst, I'll be afk for an hour or so | 17:09 |
| *** fgiudici has quit IRC | 17:48 | |
| *** pcaruana has quit IRC | 17:51 | |
| *** jodh has quit IRC | 18:01 | |
| *** pcaruana has joined #kata-dev | 18:03 | |
| kata-irc-bot | Action: fidencio is back | 18:47 |
| kata-irc-bot | <fidencio> @wmoschet, @gabriela.cervantes.te, @eric.ernst: The problem was ... when I setup that job I set it up to run on `2.0-dev` and `stable-2.*` for `kata-containers/kata-containers`, while I only set it to `2.0-dev` on `kata-containers/tests` side. | 18:56 |
| *** fuentess has quit IRC | 18:56 | |
| kata-irc-bot | <fidencio> https://github.com/kata-containers/tests/pull/3161 solves the issue | 18:56 |
| kata-irc-bot | <fidencio> and now I have to run to yet another short meeting | 18:57 |
| *** fuentess has joined #kata-dev | 18:59 | |
| *** fuentess has quit IRC | 19:04 | |
| *** fuentess has joined #kata-dev | 19:05 | |
| kata-irc-bot | <eric.ernst> review help for 2.x releases :thred: | 19:10 |
| kata-irc-bot | <eric.ernst> https://github.com/pulls?q=is%3Aopen+is%3Apr+org%3Akata-containers+author%3Aegernst++2.0.1 | 19:10 |
| kata-irc-bot | <eric.ernst> https://github.com/pulls?q=is%3Aopen+is%3Apr+org%3Akata-containers+author%3Aegernst+2.1-alpha0 | 19:10 |
| kata-irc-bot | <eric.ernst> @fidencio @jose.carlos.venegas.m ^^ | 19:11 |
| kata-irc-bot | <fidencio> @eric.ernst, done! | 19:14 |
| kata-irc-bot | <fidencio> I wonder when we should rename the stable-2.0.0 branch to stable-2.0 | 19:15 |
| kata-irc-bot | <wmoschet> I was thinking the same @fidencio, all others are stable-x.y | 19:26 |
| kata-irc-bot | <eric.ernst> oh yes definitely it is an error, but it is what exists now. | 19:51 |
| kata-irc-bot | <eric.ernst> i think we'd need to fix CI, etc. | 19:51 |
| kata-irc-bot | <eric.ernst> @fidencio i was afraid of what else needs fixing to address that. | 19:51 |
| kata-irc-bot | <jose.carlos.venegas.m> @eric.ernst done | 19:52 |
| kata-irc-bot | <jose.carlos.venegas.m> @eric.ernst I hope not a lot, one thing I remember is one of the release scripts is that afte the first alpha release it will create branches based in the name "stable-x.y" | 19:53 |
| kata-irc-bot | <jose.carlos.venegas.m> @eric.ernst we can create a copy branch and see that all works then remove the old branch ? | 19:54 |
| kata-irc-bot | <eric.ernst> i'm happy to see that created, etc. If the CI won't be mad, I can do it. | 19:54 |
| kata-irc-bot | <eric.ernst> Can someone ack the CI? | 19:54 |
| kata-irc-bot | <jose.carlos.venegas.m> @eric.ernst the link you pasted is not working for me | 19:55 |
| kata-irc-bot | <jose.carlos.venegas.m> could you check to see what else is missing to ack? | 19:55 |
| kata-irc-bot | <eric.ernst> 2.0.1 and a 2.1-alpha0 PR | 19:56 |
| kata-irc-bot | <eric.ernst> @salvador.fuentes @gabriela.cervantes.te you there? | 19:56 |
| kata-irc-bot | <eric.ernst> Curious -- what needs to be done to add CI jobs to a stable brancH? | 19:56 |
| kata-irc-bot | <fidencio> @eric.ernst but which repos are you missing acks? | 19:59 |
| kata-irc-bot | <eric.ernst> Ah, I see. None at this point. | 20:06 |
| kata-irc-bot | <eric.ernst> (looks at watch, considers merging even though 2.0.0 should be 2.0) | 20:06 |
| kata-irc-bot | <fidencio> @eric.ernst, sorry, I gave you the wrong impression. | 20:09 |
| kata-irc-bot | <fidencio> Let's have 2.0.1 released and as soon as we have that out we fix the branches | 20:09 |
| *** pcaruana has quit IRC | 20:15 | |
| *** fuentess has quit IRC | 20:22 | |
| *** davidgiluk has quit IRC | 20:27 | |
| *** crobinso has quit IRC | 20:31 | |
| *** fuentess has joined #kata-dev | 21:02 | |
| *** fuentess has quit IRC | 21:47 | |
| *** fuentess has joined #kata-dev | 21:59 | |
| *** fuentess has quit IRC | 21:59 | |
| *** fuentess has joined #kata-dev | 21:59 | |
| *** devimc has quit IRC | 22:22 | |
| *** sameo has quit IRC | 23:20 | |
| kata-irc-bot | <eric.ernst> heads up -- all the release actions have failed | 23:39 |
Generated by irclog2html.py 2.17.2 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!