On Sat, 2018-03-10 at 22:56 +0200, Jaco Kroon wrote:
> On 22/02/2018 18:46, Bart Van Assche wrote:
> > On 02/22/18 02:58, Jaco Kroon wrote:
> > > We've been seeing sporadic IO lockups on recent kernels.
> >
> > Are you using the legacy I/O stack or blk-mq? If you are not yet using
> > blk-mq, can you switch to blk-mq + scsi-mq + dm-mq? If the lockup is
> > reproducible with blk-mq, can you share the output of the following
> > command:
> >
> > (cd /sys/kernel/debug/block && find . -type f -exec grep -aH . {} \;)
>
> Looks like the lockups are far more frequent with everything on mq.
> Just to confirm:
>
> CONFIG_SCSI_MQ_DEFAULT=y
> CONFIG_DM_MQ_DEFAULT=y
>
>
> Please find attached the output from the requested.
>
> http://downloads.uls.co.za/lockup/lockup-20180310-223036/ contains
> additional stuff, surrounding that.
Thanks, that helps. In block_debug.txt I see that only for /dev/sdm a
request got stuck:
$ grep 'busy=[^0]' block_debug.txt
./sdm/hctx0/tags:busy=9
But I can't see in the output that has been shared which I/O scheduler has
been configured nor which SCSI LLD is involved. Can you please also share
that information, e.g. by providing the output of the following commands:
cat /sys/block/sdm/queue/scheduler
find /sys -name sdm # provides the PCI ID
lspci
Thanks,
Bart.
