error in syslog of rx5670 HP-UX 11i v2 (11.11) server

[Md. Khairul Alam Prodha_4]

Hi,

We have rx5670 Two Node Cluster Server with VA7110 external storage HP-UX 11i v2 (11.11)running Oracle 9i Online Telecome Billing Application.

We are getting following error message in syslog:
Nov 14 10:15:19 dtcellm vmunix: SCSI: Async write error -- dev: b 31 0x050100, errno: 126, resid: 8192,
Nov 14 10:15:19 dtcellm vmunix: SCSI: Read error -- dev: b 31 0x050100, errno: 126, resid: 8192,
Nov 14 10:15:19 dtcellm above message repeats 24 times
Nov 14 10:15:19 dtcellm vmunix: blkno: 47212040, sectno: 94424080, offset: 48345128960, bcount: 8192.
Nov 14 10:15:19 dtcellm vmunix: blkno: 27540040, sectno: 55080080, offset: 28201000960, bcount: 8192.
Nov 14 10:15:19 dtcellm vmunix: blkno: 26237952, sectno: 52475904, offset: 26867662848, bcount: 8192.
Nov 14 10:15:19 dtcellm vmunix: SCSI: Write error -- dev: b 31 0x050100, errno: 126, resid: 1024,

Could any one help how to solve it?

Thanks

[Arunvijai_4]

Check the connectivity of external storage. Also, run #ioscan -fnC disk and make sure its displayed on your server.

-Arun

[Md. Khairul Alam Prodha_4]

Hi Arun,

ioscan and all hardware and FC connection ok only some time we geting this error message in syslog.

[LiPEnS]

Hi
This is problem with SCSI device. This is disk (dev: b 31) with device file (0x050100):
#ll -R /dev/ |grep 31 |grep 0x050100

Regards

[Adisuria Wangsadinata_1]

Hi,

Check the device (c5t0d1) on your system :

# ioscan -nfCdisk
# ioscan -nf |grep NO_HW
# ioscan -nf |grep SCAN

Check the event.log at /var/opt/resmon/log, check the last available log about the disk.

If the disk still with CLAIMED status, try to backup the system first.

If the disk a part of VA7110, check the LED status on your VA 7110 (is there any amber light appear on your VA 7110 ?) and try to get the info log with command below (CommandView's command) :

# armdsp -i [hostname]
# armdsp [your_VA_alias]
# armdsp [your_VA_alias]

Check man page of armdsp for more details.

If you have a support with HP, log a hardware call.

Hope this information can help you.

Cheers,
AW

[saju_2]

Hi

The EMS notification for u is coming like

Oct 26 11:36:05 dtcellm EMS [1906]: ------ EMS Event Notification ------ Value: "SERIOUS (4)" for Resource: "/adapters/events/TL_adapter/0_2_1_0" (Threshold: >= " 3") Execute the following command to obtain event details: /opt/resmon/bin/resdata -R 124911618 -r /adapters/events/TL_adapter/0_2_1_0 -n 124911621 -a


The H/W path 0_2_1_0 corrpsponds to the FC card. So it seems that the FC lost connectivity with the storage , that is why all the parimary paths are getting switched to the alternate paths.

Pls check the FC at the port 0/2/1/0

(Sep 5 04:30:35 dtcellm vmunix: 0/2/1/0 td)

Regards
CS

[Devender Khatana]

Hi,

This is a disk and the device file for the same can be found in /dev/dsk directory after greping for this minor no. (0x050100)

Once located the file take backup and then you can do various diagnostics on the disk.

The device file for this shall be /dev/dsk/c5t0d1 but not sure. Find it out and after backup you can do various tests on it. One of that can be a dd read test.

Also take some time to assign points to responses to your earlier questions as per the relevance of the answeres. It will help others and will also get you prompt responses.

HTH,
Devender