SCSI: Write error

[Fernando Jose P de Souz]

in my Syslog has a following messages:

Oct 21 14:20:33 slavemm vmunix: blkno: 379384, sectno: 1782802, offset: 388489216, bcount: 8192.
Oct 21 14:20:33 slavemm vmunix: SCSI: Write error -- dev: b 31 0x021002, errno: 126, resid: 4096,
Oct 21 14:20:33 slavemm vmunix: blkno: 379312, sectno: 1782658, offset: 388415488, bcount: 4096.
Oct 21 14:20:33 slavemm vmunix: SCSI: Write error -- dev: b 31 0x021002, errno: 126, resid: 16384,
Oct 21 14:20:33 slavemm vmunix: blkno: 379296, sectno: 1782626, offset: 388399104, bcount: 16384.
Oct 21 14:20:33 slavemm vmunix: SCSI: Async write error -- dev: b 31 0x021002, errno: 126, resid: 8192,
Oct 21 14:20:33 slavemm vmunix: blkno: 11001896, sectno: 23027826, offset: 11265941504, bcount: 8192.
Oct 21 14:20:33 slavemm vmunix: blkno: 11006776, sectno: 23037586, offset: 11270938624, bcount: 8192.
Oct 21 14:20:33 slavemm vmunix: blkno: 11006768, sectno: 23037570, offset: 11270930432, bcount: 8192.
Oct 21 14:20:33 slavemm vmunix: blkno: 8525552, sectno: 18075138, offset: 8730165248, bcount: 8192.
Oct 21 14:20:33 slavemm vmunix: blkno: 4669744, sectno: 10363522, offset: 4781817856, bcount: 8192.
Oct 21 14:20:33 slavemm vmunix: blkno: 4583584, sectno: 10191202, offset: 4693590016, bcount: 8192.
Oct 21 14:20:33 slavemm vmunix: SCSI: Write error -- dev: b 31 0x020002, errno: 126, resid: 8192,
Oct 21 14:20:33 slavemm vmunix: SCSI: Async write error -- dev: b 31 0x021002, errno: 126, resid: 8192,
Oct 21 14:20:34 slavemm above message repeats 5 times
Oct 21 14:20:33 slavemm vmunix: blkno: 17972568, sectno: 36969170, offset: 18403909632, bcount: 8192.
Oct 21 14:20:33 slavemm vmunix: SCSI: Write error -- dev: b 31 0x020002, errno: 126, resid: 16384,
Oct 21 14:20:33 slavemm vmunix: blkno: 379296, sectno: 1782626, offset: 388399104, bcount: 16384.
Oct 21 14:20:33 slavemm vmunix: blkno: 379472, sectno: 1782978, offset: 388579328, bcount: 16384.
Oct 21 14:20:33 slavemm vmunix: blkno: 379456, sectno: 1782946, offset: 388562944, bcount: 16384.
Oct 21 14:20:33 slavemm vmunix: blkno: 379440, sectno: 1782914, offset: 388546560, bcount: 16384.
Oct 21 14:20:33 slavemm vmunix: blkno: 379424, sectno: 1782882, offset: 388530176, bcount: 16384.
Oct 21 14:20:33 slavemm vmunix: blkno: 379408, sectno: 1782850, offset: 388513792, bcount: 16384.
Oct 21 14:20:33 slavemm vmunix: blkno: 379392, sectno: 1782818, offset: 388497408, bcount: 16384.
Oct 21 14:20:33 slavemm vmunix: SCSI: Write error -- dev: b 31 0x020002, errno: 126, resid: 8192,
Oct 21 14:20:33 slavemm vmunix: SCSI: Write error -- dev: b 31 0x020002, errno: 126, resid: 16384,
Oct 21 14:20:35 slavemm above message repeats 6 times

[Steven E. Protter]

You have a bad block on the disk.

Your disk will eventually, probably fail.

ITs a good time to do an Ignite backup if its vg00 or other backups otherwise and plan for replacement prior to failure.

SEP

[RAC_1]

Looks like a error on the disk. To get what disk it is do ll /dev/dsk/* | grep 0x020002

If this is a SCSI disk, is that properly terminated? Are cables connecting it OK?? check for loose connections, reseating it.

Seems to be /dev/dsk/c2t0d0. You may run stm, dd and diskinfo on this disk to know if it has any problems.

dd if=/dev/dsk/cxtxdx of=/dev/null bs=4k
Check if it returns any error.

Anil

[Ranjith_5]

Hi Fernando,

The errors pointing to a hardware problem.

The reasons can be the following.

1.)one of your hard disk is having physical errors on it.

2.) SCSI termination loose.

3.) SCSI data cable problem.

4). If this disk is a part of an external storage device , worst case it cud be a problem with the slot of the stoarage device too.

Find out the disk which is giving error with the minor number 0x020002. Use the command
# ls -al /dev/dsk|grep 0x020002

Check out the above things.

Hope this works..

regards,
Syam

[Sanjay_6]

Hi Fernando,

If this is 11.11 OS read below.

Check out this link from itrc,

http://www2.itrc.hp.com/service/cki/docDisplay.do?docLocale=en_US&docId=200000075968215

The doc id is CMTKBRC00012383

you may be needing a patch "PHKL_30622". This is the lates patch in the category suggested in the above link,

http://www2.itrc.hp.com/service/cki/patchDocDisplay.do?patchId=PHKL_30622

Hope this helps.

Regds

[Prashant Zanwar_4]

You have a bad blcok associated to disk as it is showing according to me.

31 is LVM major..

so you can try,

ls -altrR /dev/vg* | grep -i 0x020002

then drill down to disk level if anything is wrong there.

Thanks
Prashant

[Prashant Zanwar_4]

Just sorry for above, it is a disk major..

ls -altrR /dev/rdsk | grep -i 20002, which device it is? have a check and then do

diskinfo /dev/rdsk/cxtxdx. It is surely turning bad I believe.

Thanks
Prashant

[Sanjay_6]

Hi,

the disks reporting the errors in this case are,

/Quote/

Oct 21 14:20:33 slavemm vmunix: SCSI: Write error -- dev: b 31 0x021002, errno: 126, resid: 4096,

/endQuote/

Disk 0x021002 -> c2t1d0

/Quote/

Oct 21 14:20:33 slavemm vmunix: SCSI: Write error -- dev: b 31 0x020002, errno: 126, resid: 8192,

/EndQuote/

Disk 0x020002 -> c2t0d0

Hope this helps.

regds