SCSI Read error in dmesg

[CCIL]

I have HP - N class server with two virtual partitions. in one of the Virtual partitions, the output of dmesg shows "SCSI: Read error -- dev: c 188 0x031000, errno: 22, resid: 7,
blkno: 32, sectno: 2292, offset: 32768, bcount: 7."

syslog also shows crtical event code 199

[Alexander M. Ermes]

Hi there.
Check your syslog files.
/var/adm/syslog/syslog.log
If you find that error in this file during the last 24 hours, you should check your scsi connections / cables and devices.
Sorry, but i cannot read your attachment,
so i have to guess.
Rgds
Alexander M. Ermes

[Jochen Heuer]

Hello!

You have a read error trying to access the raw disk with minor number 0x031000. Run

$ ll /dev/rdsk | grep "0x0310000"

to find out which disk. Then you can try to read from the whole disk:

$ dd if=/dev/rdsk/cXtYdZ of=/dev/null bs=1024k

If this aborts with an i/o error your disk is probably broken.

Jochen

[Armin Feller]

Hi,

if the 'dd' run fine and you can not find any read-error.

# dd if=/dev/rdsk/cXtYdZ of=/dev/null bs=1k

Then please check if there are any timeouts set.

# lvdisplay -v /dev/vgXY/lvolX | grep -i time
# pvdisplay -v /dev/dsk/cXtYdZ | grep -i time

Remember on each disk which contains filesystems you should NEVER set the timeout other then default (30sec).

Then check if SCSI, LVM, FILESYSTEM (vxfs) patches are installed.

Hope that helps.

Regards,
Armin

[T G Manikandan]

Just do a ls -l in /dev/dsk

It should give you the correct disk device

Just do a dd on that disk

ALso do a diskinfo on that disk and check whether the disk gives out the proper information.

REvert