| *** eernst has joined #kata-dev | 00:16 | |
| *** david-lyle has joined #kata-dev | 00:18 | |
| *** eernst_ has joined #kata-dev | 00:18 | |
| *** gmmaharaj has joined #kata-dev | 00:19 | |
| *** gmmaha has quit IRC | 00:20 | |
| *** gmmaharaj is now known as gmmaha | 00:20 | |
| *** eernst has quit IRC | 00:21 | |
| *** dklyle has quit IRC | 00:21 | |
| *** eernst_ has quit IRC | 00:42 | |
| *** EricRen has joined #kata-dev | 02:52 | |
| kata-irc-bot1 | <xwlpt> Had someone evaluated the 9p vs nfs solution for filesystem type volume for kata? | 02:58 |
|---|---|---|
| stefanha | xwlpt: gwhaley may have run benchmarks | 03:05 |
| kata-irc-bot1 | <xwlpt> Is there any test result? @graham.whaley Thanks | 03:15 |
| *** eernst has joined #kata-dev | 03:52 | |
| *** eernst has quit IRC | 04:07 | |
| *** sgarzare has joined #kata-dev | 08:12 | |
| *** jodh has joined #kata-dev | 08:21 | |
| *** tmhoang has joined #kata-dev | 08:32 | |
| *** djinni has quit IRC | 08:54 | |
| *** djinni has joined #kata-dev | 08:57 | |
| *** davidgiluk has joined #kata-dev | 09:02 | |
| *** djinni has quit IRC | 09:06 | |
| *** gwhaley has joined #kata-dev | 09:09 | |
| *** djinni has joined #kata-dev | 09:16 | |
| *** sameo has joined #kata-dev | 09:40 | |
| kata-irc-bot1 | <graham.whaley> hi @xwlpt @stefanha - let me dig up that original PR/Issue on github where I did some eval... | 10:28 |
| kata-irc-bot1 | <graham.whaley> this is the Issue with details https://github.com/kata-containers/runtime/issues/279 and will point you to a gist on how to re-create the setup https://gist.github.com/grahamwhaley/03d818bc5d20dcbab59cbf897d9375bf | 10:29 |
| kata-irc-bot1 | <graham.whaley> that was nfs-vsock btw - if you mean just plain nfs, then I've not evaluated that... but you can either steal some method from the above, or maybe try the filesystem test from the https://github.com/kata-containers/tests/tree/master/metrics/report | 10:31 |
| kata-irc-bot1 | <xwlpt> I see. Thanks for that. Actually it need time to setup the env and do the test. So is there any doc that have the test result of 9p/ nfs over vsock vs host , then we can see how much performance degraded. | 10:34 |
| kata-irc-bot1 | <xwlpt> @graham.whaley | 10:34 |
| kata-irc-bot1 | <graham.whaley> @xwlpt - it was quite some time ago I ran the test. I don't have the old results, and I think it is worth re-doing with updated Kata components and kernels etc. Also you may wish to compare against virtio-fs as well, also some work @stefanha is involved in. https://github.com/kata-containers/runtime/pull/1016 | 10:40 |
| kata-irc-bot1 | <xwlpt> @graham.whaley Thanks. | 10:51 |
| EricRen | hello guys, may i ask a question about kata-runtime? | 11:50 |
| EricRen | I'm using containerd + kata, using "/containerd/root" and "/containerd/state" as root and state path for containerd respectively. The interesting point is, "/containered" is ext4 fs on *LVM* block device. When using kata as runtime to run a container, runtime errored out - "level=error msg="QMP command failed" arch=amd64 command=create". | 11:50 |
| EricRen | The preceeding kata-runtime log shows kata is trying to attach the *LVM* block device, but failed with "level=info msg="{\"error\": {\"class\": \"GenericError\", \"desc\": \"Duplicate node name\"}}"", I don't understand why kata have to deal with the *LVM* block device. More logs follows: | 11:50 |
| EricRen | Feb 13 19:30:04 eric-dm kata-runtime[8870]: time="2019-02-13T19:30:04.334124926+08:00" level=info msg="device details" arch=amd64 command=create container=69c93ac7dfcce9df549e0cd54801e6b54e8b8a202ac0b250328384d031b23b77 device-major=252 device-minor=0 mount-point=/containerd name=kata-runtime pid=8870 sandbox=a0b83e3ca070566a1d0142a9555de4caa30547d0c284321158ab826873d94eab source=virtcontainers subsystem=container | 11:50 |
| EricRen | Feb 13 19:30:04 eric-dm kata-runtime[8870]: time="2019-02-13T19:30:04.348360544+08:00" level=info msg="{\"arguments\":{\"cache\":{\"direct\":false},\"driver\":\"raw\",\"file\":{\"driver\":\"file\",\"filename\":\"/dev/dm-0\"},\"node-name\":\"drive-077673d035cf9b29\"},\"execute\":\"blockdev-add\"}" arch=amd64 command=create container=69c93ac7dfcce9df549e0cd54801e6b54e8b8a202ac0b250328384d031b23b77 name=kata-runtime pid=8870 | 11:50 |
| EricRen | sandbox=a0b83e3ca070566a1d0142a9555de4caa30547d0c284321158ab826873d94eab source=virtcontainers subsystem=qmp | 11:50 |
| EricRen | I'm using kata-runtime 1.4.0. Thanks in advance :-) | 11:52 |
| *** gwhaley has quit IRC | 11:59 | |
| *** EricRen has quit IRC | 13:00 | |
| *** fuentess has joined #kata-dev | 13:34 | |
| *** devimc has joined #kata-dev | 13:40 | |
| *** tmhoang has quit IRC | 13:44 | |
| *** EricRen has joined #kata-dev | 13:46 | |
| *** tmhoang has joined #kata-dev | 13:47 | |
| kata-irc-bot1 | <tuan.hoang1> Hi, do we have a planned timeline for upgrading to 4.19 kernel ? | 13:48 |
| *** gwhaley has joined #kata-dev | 13:50 | |
| kata-irc-bot1 | <graham.whaley> hi @tuan.hoang1 - see https://github.com/kata-containers/runtime/pull/1111 I pushed yesterday to get this updated and landed. sboeuf has rebased, but the CIs are not happy. I looked earlier and nudged, but it seems I will need to look again. Any help figuring out if that kernel version broke the CI most welcome ;) | 13:52 |
| kata-irc-bot1 | <tuan.hoang1> Nice thanks. I'm checkig it out | 13:55 |
| *** tmhoang has quit IRC | 14:10 | |
| *** tmhoang has joined #kata-dev | 14:13 | |
| *** eernst has joined #kata-dev | 14:32 | |
| *** devimc has quit IRC | 14:32 | |
| *** devimc has joined #kata-dev | 14:33 | |
| *** eernst has quit IRC | 14:36 | |
| *** lpetrut has joined #kata-dev | 14:38 | |
| *** lpetrut has quit IRC | 16:00 | |
| *** tmhoang has quit IRC | 16:46 | |
| *** eernst has joined #kata-dev | 17:00 | |
| *** eernst has quit IRC | 17:03 | |
| *** sgarzare has quit IRC | 17:46 | |
| *** jodh has quit IRC | 18:04 | |
| *** gwhaley has quit IRC | 18:19 | |
| *** eernst has joined #kata-dev | 19:27 | |
| *** eernst has quit IRC | 19:28 | |
| *** eernst has joined #kata-dev | 19:42 | |
| *** eernst has quit IRC | 19:42 | |
| *** eernst has joined #kata-dev | 19:42 | |
| *** davidgiluk has quit IRC | 20:09 | |
| *** sameo has quit IRC | 20:32 | |
| *** eernst has quit IRC | 20:35 | |
| *** eernst has joined #kata-dev | 20:41 | |
| *** stackedsax has joined #kata-dev | 20:51 | |
| *** sameo has joined #kata-dev | 20:53 | |
| *** EricRen has quit IRC | 22:27 | |
| *** fuentess has quit IRC | 22:37 | |
| *** eernst has quit IRC | 22:39 | |
| *** sameo has quit IRC | 22:41 | |
| *** eernst has joined #kata-dev | 22:44 | |
| *** eernst has quit IRC | 22:47 | |
| *** eernst has joined #kata-dev | 22:55 | |
| *** eernst has quit IRC | 22:57 | |
| *** eernst has joined #kata-dev | 22:57 | |
| *** eernst has quit IRC | 23:00 | |
| *** eernst_ has joined #kata-dev | 23:00 | |
| *** david-lyle has quit IRC | 23:02 | |
| *** dklyle has joined #kata-dev | 23:02 | |
| *** eernst_ has quit IRC | 23:05 | |
| *** devimc has quit IRC | 23:17 | |
| *** devimc has joined #kata-dev | 23:28 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!