URGENT - Errors in dmesg

[Tal Drigov_2]

Hello,

One of the serves performed PANIC reboot.
After the servers has booted I saw these messages in the dmesg:


SCSI: First party detected bus hang -- lbolt: 12213, bus: 3
lbp->state: 3060
lbp->offset: 40
lbp->uPhysScript: f4000000
From most recent interrupt:
ISTAT: 29, SIST0: 00, SIST1: 00, DSTAT: 84, DSPS: 0000000a
lsp: 0000000000000000
lbp->owner: 0000000042fff600
bp->b_dev: cb035002
scb->io_id: 300262f
scb->cdb: 12 00 00 00 80 00
lbolt_at_timeout: 12113, lbolt_at_start: 11613
lsp->state: 5
scratch_lsp: 0000000042fff600
Pre-DSP script dump [fffffffff4000020]:
00000000 00000000 41050000 f4000290
980dff00 0000000a 78351000 00000000
Script dump [fffffffff4000040]:
0e000005 f4000540 e0100004 f40007f0
870b0000 f40002d8 98080000 00000005
SCSI: Resetting SCSI -- lbolt: 12313, bus: 3
SCSI: Reset detected -- lbolt: 12313, bus: 3
SCSI: Unexpected Disconnect -- lbolt: 12818, dev: cb035002, io_id: 3002638
mvfs: Ok: INFO: MVFS: Setting mvfs_largeinit to 2 based on physical memory size of 2097152KB
SCSI: Unhandled interrupt -- lbolt: 160477, dev: bc026000
lbp->state: 4060
lbp->offset: ffffffff
lbp->uPhysScript: f4002000
From most recent interrupt:
ISTAT: 2a, SIST0: c8, SIST1: 00, DSTAT: 80, DSPS: 00058000
lsp: 0000000042261700
bp->b_dev: bc026000
scb->io_id: 202c5b4
scb->cdb: 28 00 00 61 5b 00 00 00 80 00
lbolt_at_timeout: 0, lbolt_at_start: 0
lsp->state: 5
lbp->owner: 0000000042ffff00
bp->b_dev: bc024000
scb->io_id: 202c5b5
scb->cdb: 28 00 00 54 fb 80 00 00 80 00
lbolt_at_timeout: 0, lbolt_at_start: 0
lsp->state: 5
scratch_lsp: 0000000042261700
Pre-DSP script dump [000000004400a000]:
09010000 00090000 e25c0004 f40027f0
80080000 f4002090 80080000 f4002090
Script dump [000000004400a020]:
09010000 00058000 e25c0004 f40027e8
80080000 f4002090 80080000 f4002090
SCSI: Resetting SCSI -- lbolt: 160777, bus: 2
SCSI: Reset detected -- lbolt: 160777, bus: 2


Which disk is it ?!?

[Noobianz]

Dev = cb035002

cb : 03 : 5 : 0 : 0 : 2

major number = 203 (cb HEX)
controller = 03, target = 5, lun=0,slice = 2

c3t5d0s2

http://forums1.itrc.hp.com/service/forums/questionanswer.do?threadId=943878

You could try a pvchange -t to increase the timeout value for the disk.

pvchange -t 180 /dev/dsk/c3t5d0s2'

[AwadheshPandey]

See this thread http://forums1.itrc.hp.com/service/forums/questionanswer.do?threadId=232452&admit=-682735245+1187784360335+28353475

[Amit Parui]

Hi Tal,

Seems you are using an Itanium Server because the disk which is giving lbolt messages is - c3t5d0s2. This seems to be the root volume as in case of Itanium servers. As suggested above try - pvchange -t and also check the disk whether its ok without any stale extents -
1) ioscan -fnC -> see for Claimed
2) pvdisplay -v -> see for Stale Extents
3) diskinfo -v -> see for i/o

Hope with this you will get the disk !!!