[stgt] Help pinpointing cause of tgtd instability

[ronnie sahlberg]

I have tried to write a small tool to generate a huge amount of READ10
tasks, then using both
ABORT_TASK as well as ABORT_TASK_SET
but I can not get it to crash here.


I have attached the ABORT_TASK/ABORT_TASK_SET torture program here in
case someone wants to
try to tweak it to reproduce the crashes.


It links with libiscsi.

regards
ronnie sahlberg


On Sat, Dec 10, 2011 at 8:34 PM, FUJITA Tomonori
<fujita.tomonori at lab.ntt.co.jp> wrote:
> On Thu, 8 Dec 2011 13:33:04 -0500
> John Pletka <jpletka at abraxis.com> wrote:
>
>> What is the side effect of disabling the task manager?  Any important
>> functionality lost?  Will it need to be periodically restarted to
>
> I think that if the task management fails, a Linux initiator closes
> the connection and try to reconnect then if it works, the initiator
> sends the same command again. So the file system doesn't detect
> failure. If reconnecting fails, the command fails, then the file
> system hits the I/O error.
>
>
>> reset whatever the task manager was doing? This is going into a
>> low-priority production server, so anything except data corruption can
>> be dealt with.
>
> --
> To unsubscribe from this list: send the line "unsubscribe stgt" in
> the body of a message to majordomo at vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
-------------- next part --------------
A non-text attachment was scrubbed...
Name: tmgmttorture.c
Type: text/x-csrc
Size: 3214 bytes
Desc: not available
URL: <http://lists.wpkg.org/pipermail/stgt/attachments/20111219/29244f4e/attachment-0002.c>