SCSI reset lbolt error in syslog

[sysadm_1]

I am getting following error message in syslog intermittently

Nov 4 12:36:42 prod vmunix: SCSI: Reset detected -- lbolt: 311620234, bus: 0
Nov 4 12:36:42 prod vmunix: lbp->state: 6060
Nov 4 12:36:42 prod vmunix: lbp->offset: ffffffff
Nov 4 12:36:42 prod vmunix: lbp->uPhysScript: 81fbe000
Nov 4 12:36:42 prod vmunix: From most recent interrupt:
Nov 4 12:36:42 prod vmunix: ISTAT: 02, SIST0: 02, SIST1: 00, DSTAT: 80, DSPS:
0000000f
Nov 4 12:36:42 prod vmunix: lsp: 0000000000000000
Nov 4 12:36:42 prod vmunix: lbp->owner: 0000000000000000
Nov 4 12:36:42 prod vmunix: scratch_lsp: 0000000000000000
Nov 4 12:36:42 prod vmunix: Pre-DSP script dump [ffffffff81fbe008]:
Nov 4 12:36:42 prod vmunix: 81fbe5c0 80880000 000000e4 80000000
Nov 4 12:36:42 prod vmunix: 00000000 c0000004 00000000 00000000
Nov 4 12:36:42 prod vmunix: Script dump [ffffffff81fbe028]:
Nov 4 12:36:42 prod vmunix: 41000000 81fbe290 980dff00 0000000a
Nov 4 12:36:42 prod vmunix: 78350000 00000000 0e000000 81fbe540


Anybody faced this problem?

[KCS_1 · Accepted solution]

Hi,

Have a look at this attached file that explains how to solve your problem.

[Bernhard Mueller]

Hi,

most likely you have a hardware problem:

this could be (in decreasing likelyhood) a bad disk, a bad tape drive, a bent SCSI connector pin, a bad SCSI cable, improper termination of the SCSI bus, or even a duplicate target ID.

Above is assuming you have not changed anything and the system was running smoothly before, then you should check when it began and look for an indication which HW path it is.

Do you have EMS configured (default) and get warning mails to root?

Regards,
Bernhard

[Elmar P. Kolkman]

Yes, I have seen this before. In most cases it means there is a problem in the SCSI chain, especially if you get lots of these messages.

Are there other problems you have on this SCSI chain (c0) ?

[sysadm_1]

Hi

Server working fine and there is no other probs at all.

[sysadm_1]

Hi

Server working fine and there is no other probs at all.But this errors coming intermittently

[Rupesh_1]

Hi,
I guess this is either a Bus issues(i.e Could be the SCSI Cable, Terminator or most likely it could even be the disk )

I suggest you to change the SCSI cable and the Termintor for the particular bus first.

If the problem still exists then You could identify the disk , take a down time and Do a Pvmove and change the disk. This could avoide you from doing the restore of your files etc.. which you would have to do in case the disk fails.

[Rupesh_1]

Hi,
I guess this is either a Bus issues(i.e Could be the SCSI Cable, Terminator or most likely it could even be the disk )

I suggest you to change the SCSI cable and the Termintor for the particular bus first.

If the problem still exists then You could identify the disk , take a down time and Do a Pvmove and change the disk. This could avoide you from doing the restore of your files etc.. which you would have to do in case the disk fails.
In case you need to know the steps for the disk change pl feel free to contact me

cheers
Rups

[Bernhard Mueller]

Hi,

I would suggest you do not change anything before you can narrow down the cause a bit.

Check if there are any lines in syslog where you have a string like "dev:" indicating a specific device. If that is the case proceed to test the device, if not it could be a general bus issue like termination, broken pin, bad cable. Can you visually inspect the machine and check that all SCSI connectors and terminators are firmly in place?

Regards,
Bernhard

[Bernhard Mueller]

Also you can run dmesg

see if the output is any different from what you see in syslog.

Further, check the timestamps in syslog if they indicate that it happens e.g. every time you try to backup something to tape or run any other job.

Regards,
Bernhard

[sysadm_1]

Hi All,
Thanx a lot for quick response

I found the problem is with the Ultrium drive connected on BUS 0.
Replaced drive.Now no probs.

sysadm