Error message in syslog on D390

[Robert T. Lott]

Getting the following Error messages it seems every 24 hrs but volume group appear to recover. It the same PV & Vg every time...

Thx, pls see attachment:

[RAC_1 · Accepted solution]

You have problems with disk c0t1d0. Check it with diskinfo/dd/pvdisplay.
dd if=/dev/rdsk/c0t1d0 of=/dev/null bs=1024k
Also what is this disk??

[AwadheshPandey]

visit this thread, http://forums1.itrc.hp.com/service/forums/questionanswer.do?threadId=962008

[Alex Lavrov.]

Can you please post the info not in M$WORD format?

Alex.

[Andrew Rutter]

hi robert,

this appears to be a disk dying by the looks of it, if its just started happening.

check the disk which belongs to vg01? is there just one?
check ioscan output/diskinfo and test with dd/cstm
if so backup this vg and get ready to replace it.

Andy

[Alex Lavrov.]

Call HP and get the disk replaced.

Even if dd will be ok and the problem will seem go away, I strongly do not recommend to continue and use this disk, since it can "die" any time.

Alex.

[Stephen Keane]

20051018.002242 CRFORCEP (1592)

-- Oct 18 00:13:09 ars05048 vmunix: blkno: 420072, sectno: 840144, offset: 430153728, bcount: 4096.
-- Oct 18 00:13:09 ars05048 vmunix: SCSI: Write error -- dev: b 31 0x001200, errno: 126, resid: 4096,
-- Oct 18 00:13:09 ars05048 vmunix: SCSI: Async write error -- dev: b 31 0x001200, errno: 126, resid: 4096,
-- Oct 18 00:13:13 ars05048 above message repeats 2 times
-- Oct 18 00:13:09 ars05048 vmunix: blkno: 421744, sectno: 843488, offset: 431865856, bcount: 4096.
-- Oct 18 00:13:09 ars05048 vmunix: blkno: 421740, sectno: 843480, offset: 431861760, bcount: 4096.
-- Oct 18 00:13:09 ars05048 vmunix: blkno: 421732, sectno: 843464, offset: 431853568, bcount: 4096.
-- Oct 18 00:13:09 ars05048 vmunix: SCSI: Read error -- dev: b 31 0x001200, errno: 126, resid: 1024,
-- Oct 18 00:13:09 ars05048 vmunix: SCSI: Write error -- dev: b 31 0x001200, errno: 126, resid: 4096,
-- Oct 18 00:13:14 ars05048 above message repeats 2 times
-- Oct 18 00:13:09 ars05048 vmunix: blkno: 8, sectno: 16, offset: 8192, bcount: 1024.
PM Oct 18 00:13:09 ars05048 vmunix: LVM: VG 64 0x010000: PVLink 31 0x001200 Failed! The PV is not accessible.
-- Oct 18 00:13:09 ars05048 vmunix:
-- Oct 18 00:13:14 ars05048 above message repeats 108 times
-- Oct 18 00:13:09 ars05048 vmunix: LVM: VG 64 0x010000: PVLink 31 0x001200 Recovered.
-- Tue Oct 18 00:15:01 EDT 2005
-- ars05048 Tue Oct 18 00:15:04 EDT 2005
-- Oct 18 00:21:30 ars05048 vmunix: SCSI: Request Timeout; Abort Tag -- lbolt: 139121121, dev: 1f001100, io_id: e6590a
-- Oct 18 00:21:44 ars05048 vmunix: SCSI: Request Timeout; Abort Tag -- lbolt: 139121321, dev: 1f001000, io_id: e6591f
-- Oct 18 00:21:44 ars05048 vmunix: SCSI: Request Timeout -- lbolt: 139121421, dev: 1f001000

[Kent Ostby]

The "dd" command that RAC mentions above is a decent although not full proof test.

If it comes back with "I/O error" (even though the in records and out records match), then you definitely have a disk problem.

In either case, you need to get HP to take a look at your log files and more than likely replace the disk.

Best regards,

Kent M. ostby