| *** sameo has joined #kata-dev | 04:25 | |
| *** sameo has quit IRC | 06:31 | |
| *** jodh has joined #kata-dev | 06:59 | |
| *** pcaruana has joined #kata-dev | 07:16 | |
| *** sgarzare has joined #kata-dev | 07:18 | |
| *** sameo has joined #kata-dev | 07:33 | |
| *** sameo has joined #kata-dev | 07:33 | |
| *** davidgiluk has joined #kata-dev | 07:57 | |
| *** gwhaley has joined #kata-dev | 08:10 | |
| *** gemini_117 has joined #kata-dev | 09:49 | |
| gemini_117 | hi, does kata-runtime support block device hotplug, such as /dev/ | 09:50 |
|---|---|---|
| gemini_117 | sda | 09:51 |
| gemini_117 | i found kata support add scsi device at create time , for example: | 09:54 |
| gemini_117 | INFO[0004] {"arguments":{"driver":"raw","file":{"driver":"file","filename":"/dev/sda"},"node-name":"drive-ce59ad8ad4e080a2"},"execute":"blockdev-add"} arch=amd64 command=run container=Foo name=kata-runtime pid=32264 source=virtcontainers subsystem=qmp | 09:54 |
| gemini_117 | but i'm not sure whether support hotplug block-dev at kata running. | 09:55 |
| *** gemini_117 has quit IRC | 10:01 | |
| *** gwhaley has quit IRC | 11:01 | |
| *** gwhaley has joined #kata-dev | 12:12 | |
| *** devimc has joined #kata-dev | 13:26 | |
| *** ananos has joined #kata-dev | 13:47 | |
| ananos | Hi guys, trying to run kata with firecracker on aarch64 | 13:50 |
| ananos | I've made a couple of quick "hacky" changes, to check whether this is feasible -- and I'm stuck at the point where the guest is spawned, but the agent can not communicate with the host | 13:50 |
| ananos | the exact same setup (if I replace configuration.toml with the -qemu.toml equivalent works as expected | 13:52 |
| ananos | the exact same setup (if I replace configuration.toml with the -qemu.toml equivalent works as expected) | 13:52 |
| *** fuentess has joined #kata-dev | 13:58 | |
| *** dklyle has joined #kata-dev | 14:50 | |
| *** sameo has quit IRC | 15:03 | |
| *** igordc has joined #kata-dev | 15:27 | |
| *** igordc has quit IRC | 15:34 | |
| kata-irc-bot3 | <eric.ernst> Probably to do with hybrid vsock. @julio.montes | 15:43 |
| fuentess | stefanha, davidgiluk: hi guys, we are getting a weird behavior when trying to run dnf update under a kata-container with virtiofs, have documented the issue here: https://github.com/kata-containers/tests/issues/2008, do you think you can take a look? | 15:48 |
| stefanha | fuentess: Looking... | 15:49 |
| stefanha | Oct 17 23:15:45 virtiofs-test virtiofsd[61195]: lo_lookup(parent=975, name=README;5da8f604) | 15:50 |
| stefanha | That filename looks suspicious | 15:50 |
| stefanha | Hmm...the logs show that it continues by creating that file. | 15:50 |
| fuentess | the issue is kind of weird, because it can be workarounded. Eg. if I first run `dnf update dnf`and then `dnf update` to update all other packages, everything goes well, but if I try to run `dnf update` after the container creation, the command gets hanged | 15:52 |
| stefanha | fuentess: I've replied on GitHub. It would be interesting to see the /proc/PID/stack of the hung process. | 15:53 |
| stefanha | fuentess: DAX is being used. The pread64(2) syscall is hanging. So I guess something happened in the DAX code. | 15:53 |
| fuentess | ok, thanks stefanha, I'll post the info | 15:54 |
| *** sgarzare has quit IRC | 16:05 | |
| *** sgarzare has joined #kata-dev | 16:10 | |
| *** canyounot has joined #kata-dev | 16:34 | |
| *** jodh has quit IRC | 16:58 | |
| *** gwhaley has quit IRC | 17:00 | |
| *** sgarzare has quit IRC | 17:10 | |
| *** davidgiluk has quit IRC | 19:02 | |
| *** pcaruana has quit IRC | 20:16 | |
| *** canyounot has quit IRC | 21:09 | |
| *** ananos has quit IRC | 22:37 | |
| *** dklyle has quit IRC | 22:42 | |
| *** david-lyle has joined #kata-dev | 22:42 | |
| *** devimc has quit IRC | 22:50 | |
| *** fuentess has quit IRC | 23:02 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!