Re: Go Team CI down again?
Hello there,
Just did a little volume cleanup on the host (docker volume prune)
this has freed 8GB and should buy us more time.
On 09/01/2023 09:44, Thomas Goirand wrote:
On 1/6/23 16:24, Nilesh Patra wrote:
Hi Thomas/Fausin,
Looks like the CI choked again, all pipelines end with:
| 2023/01/04 14:54:54 build only (Debian packaging disables tests):
v2ray.com/core
| 2023/01/04 14:54:54 write /dev/stdout: no space left on device,
*fs.PathError, &fs.PathError{Op:"write", Path:"/dev/stdout", Err:0x1c}
| 2023/01/04 14:54:54 exit status 1, *exec.ExitError,
&exec.ExitError{ProcessState:(*os.ProcessState)(0xc00000e120),
Stderr:[]uint8(nil)}
| Cleaning up project directory and file based variables
For instance see [1]. Looks like the disk is full? Could you please
consider checking?
[1]:
https://salsa.debian.org/go-team/packages/golang-gvisor-gvisor/-/jobs/3740886
Hi,
Indeed, I see that there's no space available on the server.
The system disk is 20GB, so that's quite small, but there's a much
bigger data /dev/sdb disk that isn't in use. How should I do? Should I
mount /var/lib/docker on it? Using ext4 or xfs? What should I do with
the current running docker runner?
The runner mount host directory /srv/{gopath,cache} into running
container. They contains a cache and a Go workspace
build from the whole Debian unstable archive. Those directories are
inside /dev/sdb, that's why extra space was allocated to /dev/sdb.
(Maybe too much indeed)
I guess it could be great to add extra volume space for /var/lib/docker
but also we should clean the docker volumes from time to time.
Cheers,
Thomas Goirand (zigo)
Cheers,
--
⢀⣴⠾⠻⢶⣦⠀ Aloïs Micard
⣾⠁⢠⠒⠀⣿⡁ Debian Developer
⢿⡄⠘⠷⠚⠋⠀ https://creekorful.org
⠈⠳⣄⠀⠀⠀⠀ DA4A A436 9BFA E299 67CD E85B F733 E871 0859 FCD2
Reply to: