| *** igordc has quit IRC | 00:46 | |
| kata-irc-bot | <eric.ernst> @fupan @xu ptal? https://github.com/kata-containers/packaging/pull/304 | 03:02 |
|---|---|---|
| kata-irc-bot | <xu> will take a look :slightly_smiling_face: | 03:02 |
| kata-irc-bot | <eric.ernst> I don’t bash good. Feedback appreciated. :) | 03:03 |
| *** igordc has joined #kata-dev | 03:14 | |
| *** eernst has joined #kata-dev | 03:32 | |
| *** eernst has quit IRC | 03:50 | |
| *** eernst has joined #kata-dev | 04:15 | |
| *** eernst has quit IRC | 04:55 | |
| *** igordc has quit IRC | 07:21 | |
| *** stackedsax has joined #kata-dev | 07:23 | |
| *** sameo has joined #kata-dev | 08:47 | |
| *** davidgiluk has joined #kata-dev | 09:03 | |
| *** lpetrut has joined #kata-dev | 09:17 | |
| *** gwhaley has joined #kata-dev | 09:31 | |
| *** jugs has quit IRC | 10:06 | |
| *** jugs has joined #kata-dev | 10:06 | |
| *** jodh_ has joined #kata-dev | 10:30 | |
| kata-irc-bot | <graham.whaley> OK folks, as per https://github.com/kata-containers/ci/issues/98#issuecomment-453144585 and the soon-to-be dis-use of pullapprove, I've started to enable github '2 acks' and 'codeowners approval' to the repos. It is already on for proxy, and I've just enabled for tests and runtime. I guess I may as well go 'whole hog', and will do proxy/shim/agent whilst I am there. **if** we find this causes disruption (it should not, the tests on | 11:15 |
| kata-irc-bot | the proxy repo worked fine), then you can undo it (if you have the power) in the settings/branches/branch_protection_rules/master dialog on the repos on github... just so you know ;) @thierry | 11:15 |
| *** jodh has joined #kata-dev | 11:21 | |
| *** jodh_ has quit IRC | 11:23 | |
| *** stackedsax has quit IRC | 11:38 | |
| *** gwhaley has quit IRC | 11:58 | |
| *** sameo has quit IRC | 13:06 | |
| *** sameo has joined #kata-dev | 13:24 | |
| *** gwhaley has joined #kata-dev | 13:30 | |
| *** eernst has joined #kata-dev | 14:53 | |
| kata-irc-bot | <eric.ernst> @samuel.ortiz @julio.montes @jose.carlos.venegas.m ping? | 15:53 |
| eernst | hey sameo :) | 15:54 |
| kata-irc-bot | <samuel.ortiz> @jose.carlos.venegas.m I'm expecting pod resources requirements to be honoured at the host level. | 15:54 |
| eernst | agreed - that's the most important part. After that, getting it right in guest is the next bit. | 15:55 |
| eernst | cpuset in particular is phase 1 | 15:55 |
| kata-irc-bot | <jose.carlos.venegas.m> @samuel.ortiz so @julio.montes patch should bring a good starting point | 15:57 |
| kata-irc-bot | <jose.carlos.venegas.m> It will restrict the VM at cpu and cpuset cgroup level (by merging all the constrains of its containers) | 15:58 |
| kata-irc-bot | <jose.carlos.venegas.m> memory is missing there | 15:58 |
| kata-irc-bot | <jose.carlos.venegas.m> but seems that is more complex topic, if the VM reach the limit of memory it is killed, so all the containers of the pod | 15:59 |
| kata-irc-bot | <eric.ernst> I think we should be checking pod level constraints as well. | 15:59 |
| kata-irc-bot | <eric.ernst> This isn't utilized much today, but does exist. | 15:59 |
| kata-irc-bot | <jose.carlos.venegas.m> @eric.ernst pod level? | 16:00 |
| kata-irc-bot | <eric.ernst> ie, someone could, theoretically, put a pod constraint in, and there could be 3 containers running unconstrained in it. | 16:00 |
| kata-irc-bot | <samuel.ortiz> @eric.ernst Where is that defined in CRI? | 16:00 |
| *** devimc has joined #kata-dev | 16:00 | |
| kata-irc-bot | <eric.ernst> will google after this call. | 16:01 |
| kata-irc-bot | <eric.ernst> https://docs.google.com/document/d/1CyVuEYOqhBdCdTvnhCbIh1zxw3M6wtedKVpY67fisP4/edit | 16:01 |
| kata-irc-bot | <eric.ernst> I will write about it at the above document. | 16:01 |
| kata-irc-bot | <eric.ernst> I think we should have a wholistic doc about our resource management here. | 16:01 |
| kata-irc-bot | <eric.ernst> and I want to add parts about the pod overhead for runtiemClass here too. | 16:01 |
| kata-irc-bot | <eric.ernst> and make sure we add bits about jailer here too. | 16:01 |
| *** devimc_ has joined #kata-dev | 16:06 | |
| *** eernst has quit IRC | 16:08 | |
| *** devimc has quit IRC | 16:09 | |
| *** devimc_ is now known as devimc | 16:09 | |
| *** stackedsax has joined #kata-dev | 16:12 | |
| *** lpetrut has quit IRC | 16:57 | |
| *** jodh has quit IRC | 17:34 | |
| *** igordc has joined #kata-dev | 17:45 | |
| kata-irc-bot | <eric.ernst> @jose.carlos.venegas.m can you PTAL @ https://github.com/kata-containers/runtime/issues/1184 ? | 17:55 |
| kata-irc-bot | Action: jose.carlos.venegas.m looks | 17:56 |
| *** gwhaley has quit IRC | 18:01 | |
| *** davidgiluk has quit IRC | 20:24 | |
| *** djinni has quit IRC | 21:25 | |
| *** djinni has joined #kata-dev | 21:27 | |
| *** eernst has joined #kata-dev | 21:27 | |
| *** m3thos has joined #kata-dev | 21:46 | |
| *** eernst has quit IRC | 21:56 | |
| *** sameo has quit IRC | 22:27 | |
| *** devimc has quit IRC | 22:39 | |
| *** m3thos has quit IRC | 23:56 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!