| *** annabelleB has quit IRC | 01:19 | |
| *** eernst has joined #kata-dev | 02:56 | |
| *** kata-irc-bot2 has joined #kata-dev | 03:54 | |
| *** kata-irc-bot has quit IRC | 03:55 | |
| *** eernst has quit IRC | 04:15 | |
| *** eernst has joined #kata-dev | 04:16 | |
| *** eernst has quit IRC | 04:45 | |
| *** sjas_ has joined #kata-dev | 04:57 | |
| *** sjas has quit IRC | 04:58 | |
| *** marcov__ is now known as marcov | 06:35 | |
| *** jodh has joined #kata-dev | 07:34 | |
| *** sjas_ is now known as sjas | 07:42 | |
| *** lpetrut has joined #kata-dev | 07:54 | |
| *** gwhaley has joined #kata-dev | 07:58 | |
| *** marcov has quit IRC | 08:02 | |
| *** marcov has joined #kata-dev | 08:03 | |
| *** davidgiluk has joined #kata-dev | 08:05 | |
| *** gwhaley has quit IRC | 11:10 | |
| * davidgiluk manages to get podman to run kata, just had to tweak /etc/containers/libpod.conf to set cgroup_manager = "cgroupfs" | 11:15 | |
| marcov | davidgiluk: nice! just tried and I confirm this | 11:28 |
|---|---|---|
| marcov | This can be documented then :) | 11:30 |
| davidgiluk | marcov: I wont claim enough to know what the effects of changing that cgroup option are, except it doesn't pass the unsupported --ssystemd-cgroup to kata-runtime | 11:30 |
| marcov | davidgiluk: way back cgroupfs was the default option, in fact the conf file on my system still have that | 11:31 |
| marcov | git blame on libpod.conf gave this: https://github.com/containers/libpod/pull/1254 | 11:34 |
| davidgiluk | ah, I only picked it up recently to play with it | 11:36 |
| davidgiluk | marcov: Again, me not knowing what to expect, I don't quite understand how 'standard' the standard is supposed to be, so passing new flags to the runtime by default feels odd | 11:36 |
| *** devimc has joined #kata-dev | 11:50 | |
| marcov | davidgiluk: the --systemd-cgroup flag is in runc since a couple of year. | 11:52 |
| davidgiluk | marcov: But I don't think is in something like the oci spec? | 11:54 |
| devimc | davidgiluk, marcov currently --systemd-cgroup option is meaningless | 11:57 |
| marcov | devimc: why? | 12:01 |
| marcov | devimc: ah, you mean, that option in kata-runtime? | 12:04 |
| devimc | marcov, yes | 12:05 |
| davidgiluk | kata rejects it, it took me a while to hunt where it had come from | 12:07 |
| marcov | davidgiluk: what version of kata? | 12:07 |
| marcov | davidgiluk: you can try `kata-runtime --help` and the option should be there | 12:07 |
| marcov | in fact you dont even need to change the cgroup mgr option in libpod.conf | 12:08 |
| davidgiluk | marcov: 1.3.0-rc1 | 12:09 |
| devimc | davidgiluk, please update to 1.3.0 | 12:09 |
| davidgiluk | ah ok, optinos creeping in after rc! | 12:10 |
| marcov | hehe | 12:10 |
| davidgiluk | I'd looked at podman a few weeks back and come back to it today | 12:10 |
| *** gwhaley has joined #kata-dev | 12:32 | |
| davidgiluk | devimc: Thanks! Yes updating to the release version fixed that | 12:33 |
| kata-irc-bot2 | <graham.whaley> @anne heh, I see we still have the old s/k a t a/j e s s f r a z/ issue going on here (that threw me for a minute ;) ) | 12:36 |
| devimc | davidgiluk, np, just remember that systemd cgroups won't be applied using kata | 12:36 |
| davidgiluk | devimc: Does that mean I'm still better telling podman to use cgroupfs or is the effect the same? | 12:38 |
| devimc | davidgiluk, is the same | 12:39 |
| devimc | cgroups are applied inside the VM | 12:39 |
| davidgiluk | ok, great | 12:40 |
| kata-irc-bot2 | <graham.whaley> Hi @salvador.fuentes et. al. I'm going to hook the metrics CI into the tests and agent repos now (it is already enabled on proxy and shim). It looks stable enough to get some more coverage. I'm going to leave the runtime repo just a little longer. | 12:55 |
| kata-irc-bot2 | <graham.whaley> oh, is already on agent - will add tests repo. | 12:56 |
| *** dims_ has quit IRC | 12:58 | |
| *** annabelleB has joined #kata-dev | 13:26 | |
| *** eernst has joined #kata-dev | 13:27 | |
| *** eernst has quit IRC | 13:45 | |
| kata-irc-bot2 | <salvador.fuentes> Hi @graham.whaley, sure thanks. | 13:49 |
| annabelleB | gwhaley: oof, why will that “feature” not leave! davidgiluk: FYI, when you type k-a-t-a in IRC, it pings Jessie Fraz on Slack. Fantastic Tier 1 support, but a lot of pings to poor Jess ;) | 14:22 |
| davidgiluk | haha | 14:23 |
| davidgiluk | annabelleB: That's a shame for the kat* channel :-) | 14:23 |
| gwhaley | I've never figured out the real connection though @annabelleB - so, *something* must do that translation ... but, it makes no sense, does it? I know, it has been dug into - I seem to remember some oddity to do with 'last name being a noun' or something.... but??? :-/ | 14:23 |
| annabelleB | jbryce: IRC bridge bug has returned | 14:24 |
| gwhaley | bug, feature, ai, sentientness?? | 14:24 |
| jbryce | it's actually slack api lookup laziness | 14:25 |
| annabelleB | gwhaley: I played with making other user names that would sit alpahbetically between jess and k-a-t-a, but nothing changed. I did however accidentally ping a woman with the github user name kat when I filed an issue and brought her into this whole mess :) | 14:25 |
| marcov | oh wait, I always thought that as an easter egg, not a bug :D | 14:29 |
| kata-irc-bot2 | <jose.carlos.venegas.m> lol | 14:34 |
| jodh | I think whoever fixes this one is gonna have to do a write up for medium.com! :) | 14:34 |
| gwhaley | medium, as in clairvoyant jodh? :-) | 14:35 |
| jodh | gwhaley: hahaaa! | 14:35 |
| *** lpetrut has quit IRC | 15:59 | |
| *** davidgiluk has quit IRC | 16:08 | |
| *** annabelleB has quit IRC | 16:14 | |
| *** annabelleB has joined #kata-dev | 16:16 | |
| *** annabelleB has quit IRC | 16:22 | |
| *** davidgiluk has joined #kata-dev | 16:24 | |
| *** annabelleB has joined #kata-dev | 16:28 | |
| *** gwhaley has quit IRC | 16:46 | |
| *** dims has joined #kata-dev | 16:49 | |
| *** jodh has quit IRC | 17:03 | |
| *** marcov has quit IRC | 18:05 | |
| *** annabelleB has quit IRC | 18:18 | |
| *** annabelleB has joined #kata-dev | 18:25 | |
| *** annabelleB has quit IRC | 19:07 | |
| kata-irc-bot2 | <salvador.fuentes> hi @mnaser, today we have seen that jenkins is launching VMs that do not have nested virtualization enabled, could you help us check? | 19:26 |
| kata-irc-bot2 | <mnaser> @salvador.fuentes for all or for some? | 19:27 |
| kata-irc-bot2 | <salvador.fuentes> @mnaser, just some of them, maybe around 50% of the machines | 19:27 |
| kata-irc-bot2 | <mnaser> is this jenkins or zuul or both | 19:27 |
| kata-irc-bot2 | <salvador.fuentes> currently just jenkins | 19:27 |
| kata-irc-bot2 | <mnaser> do you know if demand has increased or anyone has made any changes to the flavor type used? | 19:29 |
| *** davidgiluk has quit IRC | 19:46 | |
| kata-irc-bot2 | <salvador.fuentes> I think we have same demand and no, we havent made changes to the flavor used. at least thats what I saw on the jenkins config | 19:48 |
| kata-irc-bot2 | <mnaser> looking into this | 20:01 |
| kata-irc-bot2 | <salvador.fuentes> thaks | 20:07 |
| *** devimc has quit IRC | 21:04 | |
| *** eernst has joined #kata-dev | 22:35 | |
| *** eernst has quit IRC | 22:42 | |
| *** eernst has joined #kata-dev | 22:43 | |
| *** eernst has quit IRC | 22:48 | |
| *** eernst has joined #kata-dev | 22:48 | |
| *** eernst has quit IRC | 22:54 | |
| *** eernst has joined #kata-dev | 23:00 | |
| *** eernst has quit IRC | 23:05 | |
| *** eernst has joined #kata-dev | 23:05 | |
| *** eernst has quit IRC | 23:12 | |
| *** annabelleB has joined #kata-dev | 23:24 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!