On Sat, Oct 29, 2016 at 03:53:25PM -0700, Nicholas A. Bellinger wrote:
> Hi Johannes & Zhu,
>
> On Tue, 2016-10-18 at 23:29 -0700, Nicholas A. Bellinger wrote:
> > On Tue, 2016-10-18 at 19:35 +0200, Johannes Thumshirn wrote:
> > > On Tue, Oct 18, 2016 at 09:01:34AM +0200, Johannes Thumshirn wrote:
> > >
> > > [...]
> > >
> > > > >
> > > > > This is likely the missing SCF_ACK_KREF assignment in >= v4.1.y:
> > > > >
> > > > > http://www.spinics.net/lists/target-devel/msg13530.html
> > >
> > > Sorry to disappoint you but it didn't fix my issue. Is there any debug
> > > data I
> > > can give you, or do you have an advice where I could start looking?
> > >
> > > I have to admit I only tested the patch on our downstream kernel and not
> > > the
> > > upstream kernel, I'll repeat the tests on 4.8 final tomorrow.
> > >
> >
> > Is it possible to generate vmcore to have a look in crash + gdb..?
> >
> >
>
> Just curious if you've been able to verify the above patch for
> v4.8.y iscsi-target ports on your specific config(s)..?
>
> To confirm, using v4.1.y + patch I've not run into any further
> se_cmd->cmd_kref leaks and/or hung tasks with iscsi-target ports due to
> high backend device I/O latency, resulting in host timeouts + ABORT_TASK
> + session reinstatement to occur while waiting for outstanding se_cmd
> backend I/O to complete.
Unfortunately v4.8.5 which has above patch included (I double checked to be
sure) shows the same effects.
Here's my "reproducer":
VM1 (target) VM2 (Initiator)
============ ===============
dd if=/dev/sda of=/dev/null bs=1 (given sda
is the iscsi LUN)
Wait a little bit (~10 secs)
dmsetup suspend dm-0
watch -n1 'ps aux |\
grep -E "\[iscsi" | grep -v grep'
I'll build a kernel with all target stuff builtin and full debug symbols.
Let's see if I can find out something.
Thanks,
Johannes
--
Johannes Thumshirn Storage
jthumsh...@suse.de +49 911 74053 689
SUSE LINUX GmbH, Maxfeldstr. 5, 90409 Nürnberg
GF: Felix Imendörffer, Jane Smithard, Graham Norton
HRB 21284 (AG Nürnberg)
Key fingerprint = EC38 9CAB C2C4 F25D 8600 D0D0 0393 969D 2D76 0850
--
To unsubscribe from this list: send the line "unsubscribe linux-scsi" in
the body of a message to majord...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
