SCSI Reset Errors

[TAMILVANAN]

Hi Folks,
Suddenly one of my N-CLass Server which is connected to a HASS box and a SC10 array got hanged.The solution for my problem was then was recycle the power of the system. After I restarted I saw the below SCSI Errors in OLDsyslog.log file of my system.
Can u just help me to interpret this message.
Thanx & Regards,
V.Tamilvanan

Jun 22 14:58:04 gemini vmunix: SCSI: Resetting SCSI -- lbolt: 129243527, bus:
8
Jun 22 14:58:04 gemini vmunix: SCSI: Reset detected -- lbolt: 129243527, bus:
8
Jun 22 14:58:04 gemini vmunix: LVM: vg[10]: pvnum=0 (dev_t=0x1f08b000) is POWE
RFAILED
Jun 22 14:58:43 gemini vmunix: SCSI: Resetting SCSI -- lbolt: 129247427, bus:
8
Jun 22 14:58:43 gemini vmunix: SCSI: Reset detected -- lbolt: 129247427, bus:
8

P.S. This thread ahs been moved from General to Disk Array.- Hp Forum moderator

[Michael Tully]

Hi,

In most cases you will find that this particular error relates to actual timeouts between your server and the disk array that is servicing. To fix this problem have a look at increasing the SCSI timeout for that LUN or even all of your LUN's in your disk array. The use of the 'pvchange -t' will look after this. There is a man page for this as well.
As an example I would look at increasing this value to 180.
pvchange -t 180 /dev/dsk/c2t0d2

HTH
~Michael~

[TAMILVANAN]

Hi Michael,
Thanx for your help. I identified the disk and ran dd to find for errors. It got terminated with I/O error. Now I am planning to replace the disk.
Thanx & Regards,
V.Tamilvanan.

[Vincent Farrugia]

Hello,

IO error does not necessarily mean a faulty harddisk. It could be a faulty SCSI cable or terminator. It could be the SCSI card.

HTH,
Vince