On Mon, Aug 3, 2015 at 2:55 PM, Anton Ivanov <Anton.Ivanov@xxxxxxx> wrote:
> Hi all,
>
>
>
> I am running Kubernetes on CoreOS and use rbd binary exctracted from
> ceph/ceph-docker/config image to map images to CoreOS host. Everything works
> just fine except when trying to unmap the volume. It doesn’t unmap and gives
> following error:
>
>
>
> $ rbd showmapped
>
> id pool image snap device
>
> 0 rbd dcops-db - /dev/rbd0
>
> $ sudo rbd unmap /dev/rbd0
>
> rbd: '/dev/rbd0' is not an rbd device
>
> rbd: unmap failed: (22) Invalid argument
>
>
>
> When I was running strace against rbd unmap command I found out that it
> fails while trying to get stat of this file:
>
>
>
> lstat("/sys/devices/rbd/0/0", 0x7ffdfdc13f10) = -1 ENOENT (No such file or
> directory)
>
>
>
> On other machine running ArchLinux instead of looking for either
> /sys/devices/rbd/0/0 or /sys/devices/rbd/0/254 (which are values of minor
> and major), rbd unmap looks for /sys/devices/rbd/0/minor and reads value
> from there. And rbd unmap works as supposed.
>
>
>
> CoreOS runs kernel 4.0.5 #2 SMP Fri Jul 10 06:25:01 UTC 2015 x86_64
> Intel(R) Xeon(R) CPU E5620 @ 2.40GHz GenuineIntel GNU/Linux
>
> ArchLinux where rbd works runs kernel 4.1.3-1-ARCH #1 SMP PREEMPT Wed Jul 22
> 20:37:12 CEST 2015 x86_64 GNU/Linux
>
>
>
> rbd version on both machines is the same
>
> $ rbd --version
>
> ceph version 0.94.2 (5fb85614ca8f354284c713a2f9c610860720bbf3)
>
>
>
> What can I try to fix the unmap issue?
Can you attach full strace and ltrace outputs for both cases? Also,
what's the output of "cat /sys/module/rbd/parameters/single_major?
Thanks,
Ilya
_______________________________________________
ceph-users mailing list
ceph-users@xxxxxxxxxxxxxx
http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
