SCSI Power failed on Disk Drives

[Yan Wong]

Hi folks,

I have an HP box running on 10.20. The dmesg command gives the following message:
.....
.....
SCSI: Resetting SCSI -- lbolt: 526588929, bus: 0

SCSI: Reset detected -- lbolt: 526588965, bus: 0
LVM: vg[4]: pvnum=0 (dev_t=0x1f000200) is POWERFAILED
LVM: vg[0]: pvnum=0 (dev_t=0x1f006000) is POWERFAILED
LVM: PV 0 has been returned to vg[4].
LVM: PV 0 has been returned to vg[0].

I check the logical volumns partitioned from the above disk drives is alright. These drives are internal disks.

how do I fix this message?

[Eugeny Brychkov]

Yan,
try increasing disk timeout interval with 'pvchange -t'. First device is c0t0d2 and second is c0t6d0. If server has huge disk activity on the bus then disk with lower SCSI priority will timeout. Highest 7..0,15..8 Lowest. So usually disk with id=6 does not timeout. If you have this powerfail messages for many disks I would suspect whole SCSI bus or SCSI controller
Eugeny

[Yan Wong]

Will a system reboot solve this powerfailed message? It only happen once so far. So, a reboot can reset the whole SCSI bus.

Right?

[KCS_1]

Hi,

In my opinion,

first,I recommend backup your all file system on Volume Group, cause will protected your data on your the VG in your system.


and then look at,

# vgdisplay vg04 and vg00

# strings /etc/lvmtab
what's your disk device name in vg04 and vg00?

# diskinfo -v c#T#D

may you see all descriptions of all disks on the system.

# ioscan -funCdisk
is it output all right?
do you got a message "NO_HW"?

if, above status are good,check your SCSI terminator and Cable on the system.

good luck!

[T G Manikandan]

dmesg shows the messages on the system from the last reboot.
the messages could be right now or from the time of last reboot.
it is your disk which is creating problems.

Just do a
#cd /dev/dsk
#ls -l 000200
#ls -l 006000
These are the hard disks that are causing problems.

If you notice scsi reset messages in the syslog,
When you tell it comes back,
usually there is a scsi hung,followed by a scsi reset.
check the scsi cables and the scsi termination.

If the error is followed by a message about pv returned to vg, then usually be to a low timeout value on the disk driver. By default, this timeout is 30 seconds.

change it by using pvchange.

Also check your I/O on the specific disk.
use sar -d to check it.

Also this might be due to a hard disk going bad.
Just do a dd on the disk and check whether the disk has gone bad.

#dd if=/dev/rdsk/c?t?d? of=/dev/null bs=4096k





Thanks