Monday, 2021-04-26

*** auk_ has quit IRC02:32
*** amorenoz has joined #kata-dev05:36
*** hbrueckner has joined #kata-dev06:49
*** sgarzare has joined #kata-dev06:51
*** amorenoz has quit IRC07:09
*** amorenoz has joined #kata-dev07:10
*** amorenoz has quit IRC07:11
*** amorenoz has joined #kata-dev07:14
*** amorenoz has quit IRC07:14
*** amorenoz has joined #kata-dev07:17
*** sameo has joined #kata-dev07:22
*** fgiudici has joined #kata-dev07:25
*** jodh has joined #kata-dev07:46
*** openstack has joined #kata-dev07:50
*** ChanServ sets mode: +o openstack07:50
*** jodh has quit IRC07:54
*** jodh has joined #kata-dev07:57
*** davidgiluk has joined #kata-dev08:01
*** pcaruana has quit IRC10:45
*** pcaruana has joined #kata-dev10:55
*** brtknr_ has joined #kata-dev10:59
brtknr_hi all11:00
brtknr_does kata play well with selinux?11:00
kata-irc-bot<fidencio> heya!11:14
kata-irc-bot<fidencio> brtknr_, it should, at least when using CRI-O / podman.11:14
brtknr_what about containerd?11:15
kata-irc-bot<fidencio> brtknr_, I guess they'd need to implement passing the correct labels down to the kata-containers processes.11:15
kata-irc-bot<fidencio> We, as part of Red Hat, did the work for kata-containers, CRI-O, and podman, and we're regularly testing on CRI-O environment, so we know it does work on this stack.11:16
fidenciobrtknr_: I can point you to the code on CRI-O side and you can either submit a PR or an issue to containerd, would this help?11:19
brtknr_fidencio: yes that could be handy11:19
fidenciobrtknr_: the CRI-O code was introduced in https://github.com/cri-o/cri-o/pull/3861 and then fixed on https://github.com/cri-o/cri-o/pull/388911:22
fidenciobrtknr_: the containerd patch should be fairy simple11:22
kata-irc-bot<fidencio> @aadam, @christophe, @cmeadors, two questions: 1. Shall we have a kata-backlog channel where or are you fine being poked here for suggestions on the process? 2. Can we add issues from kata-containers/tests repo whi contains "need-review" as part of the search used for the backlog scrubbing meeting? For 2.x, those are the main repos we care about, and that we really should be paying attention there: `kata-containers`  and `tests`.11:30
kata-irc-bot<aadam> For me at least this channel is good for asking questions. Regarding kata-containers/tests, I thought our filter already looks at all the repositories:11:35
kata-irc-bot<aadam> is:issue is:open org:kata-containers created:>2020-01-01 -linked:pr label:needs-review sort:created-asc11:35
kata-irc-bot<fidencio> Aha, the date! I got surprised by facing way too many issues opened against the test repos with the "needs-review" label, but it's due to the date11:41
kata-irc-bot<fidencio> https://github.com/kata-containers/tests/issues?q=is%3Aissue+is%3Aopen+label%3Aneeds-review11:41
*** devimc has joined #kata-dev12:39
*** fuentess has joined #kata-dev13:30
*** khyr0n has joined #kata-dev13:41
*** snir has quit IRC14:02
*** snir has joined #kata-dev14:03
*** fgiudici_ has joined #kata-dev14:03
*** fgiudici has quit IRC14:06
kata-irc-bot<christophe> We decided to cleanup the old stuff, and @james.o.hunt I believe took care of the actual cleanup. So the "official" approach would be to close the older stuff ;)14:46
*** pcaruana has quit IRC14:50
*** pcaruana has joined #kata-dev14:50
*** dklyle has joined #kata-dev14:54
*** sgarzare has quit IRC15:28
*** hbrueckner has quit IRC15:55
brtknr_anyone know what this means: Failed to negoatiate QMP capabilities16:07
brtknr_trying to use virtiofsd16:07
brtknr_and thanks fr the link fidencio16:07
*** egernst has quit IRC16:11
*** egernst has joined #kata-dev16:12
kata-irc-bot<eric.ernst> Hey brtknr_! What version of QEMU you using?16:14
*** crobinso has joined #kata-dev16:33
kata-irc-bot<chelsea.e.mafrica> I'm getting some failed compiler checks because of a couple files generated by ttrpc-compiler. Specifically the files are src/agent/protocols/src/agent_ttrpc.rs and src/agent/protocols/src/health_ttrpc.rs, and the failure is redundant field names. I'm not super familiar with rust yet and can't figure out how these files are generated. Can anyone suggest how to fix this?16:36
kata-irc-bot<chelsea.e.mafrica> ```error: redundant field names in struct initialization   --> protocols/src/agent_ttrpc.rs:32:13    | 32 |             client: client,    |             ^^^^^^^^^^^^^^ help: replace it with: `client`    |    = note: `-D clippy::redundant-field-names` implied by `-D warnings`    = help: for further information visit https://rust-lang.github.io/rust-clippy/master/index.html#redundant_field_names```16:36
kata-irc-bot<chelsea.e.mafrica> https://github.com/kata-containers/kata-containers/pull/173016:37
*** fgiudici_ has quit IRC16:52
*** jodh has quit IRC17:02
kata-irc-bot<eric.ernst> womp womp: https://github.com/kata-containers/kata-containers/issues/175517:33
kata-irc-bot<eric.ernst> I want to git blame (s/blame/high-five) who did the CLH.go handling -- it handles viritofs correctly and is nice and clean17:34
kata-irc-bot<fidencio> :slightly_smiling_face:17:34
kata-irc-bot<fidencio> Please, /cc me in the PR and I'll make sure we get this in before the release17:34
kata-irc-bot<eric.ernst> Def will do.17:37
kata-irc-bot<eric.ernst> thinking through how to handle it now....17:37
kata-irc-bot<eric.ernst> Noticing now that there's a nice virtiofsd interface used in CLH, but not used in QEMU.  Seems that we should require two implementations.18:00
kata-irc-bot<eric.ernst> @jose.carlos.venegas.m did you consider having QEMU utilize this interface as well?18:01
kata-irc-bot<eric.ernst> seems we could deduplicate18:01
kata-irc-bot<jose.carlos.venegas.m> @eric.ernst yes, I wanted to have it in both implementation, but seems I forgot to added as well there when we enabling clh. Using the interface still not fixes the issue you see, right ?18:20
kata-irc-bot<eric.ernst> Migrating to that is a bigger change that I didn't make just yet.18:21
kata-irc-bot<eric.ernst> I'm just reimpl what you have in Virtiofsd Stop18:21
kata-irc-bot<salvador.fuentes> any idea @james.o.hunt @bergwolf18:38
kata-irc-bot<jose.carlos.venegas.m> ok, if you are reimplementing for now, let me give  a try in the meanwhile you we can get your fix for the release18:38
kata-irc-bot<eric.ernst> sounds good.  thanks.18:51
kata-irc-bot<eric.ernst> @jose.carlos.venegas.m recall this? https://github.com/kata-containers/kata-containers/blame/main/src/runtime/virtcontainers/virtiofsd.go#L270-L27319:02
kata-irc-bot<eric.ernst> why only if in error?19:02
kata-irc-bot<eric.ernst> Do we want to set it to zero after we do the kill, so we don't fail 'next time' ?19:04
*** davidgiluk has quit IRC19:04
kata-irc-bot<jose.carlos.venegas.m> @eric.ernst I dont remember :S  but using yeah looks like a error19:34
kata-irc-bot<eric.ernst> ok.19:38
kata-irc-bot<eric.ernst> agreed.19:38
kata-irc-bot<eric.ernst> i have initial PR up if you want to TAL: https://github.com/kata-containers/kata-containers/pull/1756  Works okay in my initial testing.19:38
kata-irc-bot<eric.ernst> Need to eat, but will add details on how I reproduce/test.19:39
kata-irc-bot<jose.carlos.venegas.m> sure19:39
kata-irc-bot<jose.carlos.venegas.m> provecho!19:39
kata-irc-bot<jose.carlos.venegas.m> Looks good at the moment19:42
*** Yarboa has quit IRC20:02
*** Yarboa has joined #kata-dev20:05
*** devimc has quit IRC21:38
kata-irc-bot<eric.ernst> @jose.carlos.venegas.m -- PTAL: https://github.com/kata-containers/kata-containers/issues/175721:42
kata-irc-bot<eric.ernst> can assign to you if you want, otherwise help-wanted for now.21:42
*** amorenoz has quit IRC23:10
*** amorenoz has joined #kata-dev23:11
*** sameo has quit IRC23:16
*** fuentess has quit IRC23:31
*** fuentess has joined #kata-dev23:37

Generated by irclog2html.py 2.17.2 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!