L V resync failed

[shridhar_1]

i got these errors
why this had happened
now lv is resync properly


OS: HPUX - B.11.11, HW: 9000/800/rp3410 RAM: 4094 MB
CPU: 2 x 800.004 MHz, PA-RISC2.0, v.



# more /var/adm/syslog/syslog.log | grep error
Nov 25 13:28:45 KBXSBUP vmunix: LVM: Failed to automatically resync PV 1f055000 error: 5
Nov 25 13:35:41 KBXSBUP vmunix: LVM: Failed to automatically resync PV 1f055000 error: 5
Nov 25 13:36:38 KBXSBUP vmunix: LVM: Failed to automatically resync PV 1f055000 error: 5
Nov 25 13:45:28 KBXSBUP vmunix: LVM: Failed to automatically resync PV 1f055000 error: 5
Nov 25 13:46:26 KBXSBUP vmunix: LVM: Failed to automatically resync PV 1f055000 error: 5
Nov 25 13:51:18 KBXSBUP vmunix: LVM: Failed to automatically resync PV 1f055000 error: 5
Nov 25 13:52:19 KBXSBUP vmunix: LVM: Failed to automatically resync PV 1f055000 error: 5
Nov 25 14:04:56 KBXSBUP vmunix: LVM: Failed to automatically resync PV 1f055000 error: 5
Nov 25 14:05:53 KBXSBUP vmunix: LVM: Failed to automatically resync PV 1f055000 error: 5
Nov 25 14:07:23 KBXSBUP vmunix: LVM: Failed to automatically resync PV 1f055000 error: 5
Nov 25 14:08:24 KBXSBUP vmunix: LVM: Failed to automatically resync PV 1f055000 error: 5
Nov 25 14:18:39 KBXSBUP vmunix: LVM: Failed to automatically resync PV 1f055000 error: 5

[Muthukumar_5]

It is a SCSI error. May be one of the disk having problem. Same type of error(s).

http://forums1.itrc.hp.com/service/forums/questionanswer.do?threadId=367067

hth.

[Steven E. Protter]

Shalom shridhar,

You have/had two physical paths to the same logical volume. For some time one of those pathes failed. You report that the resync is complete and this is no longer a problem.

Therefore the messages should stop.

if they continue, you still have a problem and need to look into why one of the paths is gone. Perhaps a scsi or fiber card or cable has failed.

Perhaps you need to manually resynch the PV.

SEP

[RAC_1]

Check the disk 1f055000.
The error 5 is i/o error. run lvdisplay/pvdisplay and dd tests to check disk.
Also check disk in stm.

The disk in question is c5t5d0

[shridhar_1]

thanx

now the output of
lvdisplay & pvdisplay are ok
there are no stale pe
disks are ok
can u guide me for any
precautions that i should take
so that above problem does not repeat


shridhar

[Sยภเl Kย๓คг · Accepted solution]

Hi,
Use this small command line scrip to find any more stale PE are there

#for i in ` vgdisplay -v|grep "LV Name"|awk '{print $3}'`
> do
> lvdisplay -v $i |grep stale
> done

You need to consult your hardware vendor to come and have a look at the system. Possible look at cable connectivity, scsi terminations, storage system health check etc..

Regards,
Sunil

[shridhar_1]

i hav already checked there r no stale PE

disks are in HP DS 2300

ioscan output is ok disks are claimed

is ther any necesity to replace the hard disk


thanx

shri

[Tonya Underwood]

You should really dd the drive....

dd if=/dev/rdsk/c5t5d0 of=/dev/null bs=64k

Let this run until completion... if you see an I/O error, the disk needs to be replaced. If it completes with no I/O error, disk media should be OK...

As stated earlier, it's also a good idea to take a look at stm logtool.

[shridhar_1]

thanx