On Wed, Sep 15, 2021 at 09:24:08PM +0200, Paul Gevers wrote:
> I looked at the results of the autopkgtest of you package on amd64
> because with a recent upload of glibc the autopkgtest of ocfs2-tools
> fails in testing. It seems to me that the failures are related to the
> worker that the test runs on. ci-worker13 fails, while the other workers
> are OK. We recently changed the setup of ci-worker13, to have /tmp/ of
> the host on tmpfs as that speeds up testing considerably is a lot of
> cases. I copied some of the output at the bottom of this report, but I'm
> not 100% sure that the /tmp there (the one inside the lxc testbed) *is*
> on tmpfs.
>
> Don't hesitate to contact us at debian...@lists.debian.org if you need
> help debugging this issue.
>
> Paul
>
> https://ci.debian.net/data/autopkgtest/testing/amd64/o/ocfs2-tools/15277216/log.gz
>
>
> autopkgtest [19:14:22]: test basic: [-----------------------
>
> === disk ===
> 200+0 records in
> 200+0 records out
> 209715200 bytes (210 MB, 200 MiB) copied, 0.109005 s, 1.9 GB/s
>
> === mkfs ===
> mkfs.ocfs2 1.8.6
> mkfs.ocfs2: Could not open device
> /tmp/autopkgtest-lxc.8neywhcx/downtmp/autopkgtest_tmp/disk: Invalid argument
> autopkgtest [19:14:23]: test basic: -----------------------]
Yes, tmpfs seems to be the problem since it doesn't support O_DIRECT that
is being requested here:
static void
open_device(State *s)
{
s->fd = open64(s->device_name, O_RDWR | O_DIRECT);
if (s->fd == -1) {
com_err(s->progname, 0,
"Could not open device %s: %s",
s->device_name, strerror (errno));
exit(1);
}
}
--
Valentin
