[stgt] tgtd segfault with software RAID, hard resetting link

[Tomasz Chmielewski]

Tomasz Chmielewski schrieb:

>> Apr  7 04:02:54 san3 kernel: [153756.406006] tgtd[20545]: segfault at 
>> 31 ip 40c32d sp 75df00c0 error 4 in tgtd[400000+25000]
>> Apr  7 04:02:55 san3 tgtd: conn_close(100) connection closed, 0x257df38 2
>> Apr  7 04:02:55 san3 tgtd: conn_close(106) sesson 0x25906c0 1
> 
> To reproduce:
> 
> 1. On the initiator, start to write (i.e. with dd)
> 2. On the target, use hdparm -Y to temporarily disable the drive:
> 
> i=1
> while [ $i -ne 100 ] ; do
>     hdparm -Y /dev/sdd
>     i=$((i+1))
> done
> 
> 
> tgtd will segfault after one or two minutes.

Also, note that I can only reproduce it with software RAID.

That means, the drive which is "stopped" with "hdparm -Y" (or, which 
throws SATA timeouts/hard resets) has to be a part of the software RAID 
array; /dev/mdX has to be made a target to the initiators.

Otherwise, tgtd doesn't crash for me (i.e. if I make /dev/sdX a target 
to the initiators and stop this drive with "hdparm -Y").


I wonder what makes that difference.



-- 
Tomasz Chmielewski
http://wpkg.org



--
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