Disk write and read errors

[Sheriff Andy]

Hello,

I have a J5600 HP-UX 11.0 workstation and I am getting both read errors on all 4 drives and one of the drives I am getting both read and write errors. Very similar to the following thread;

http://forums1.itrc.hp.com/service/forums/questionanswer.do?threadId=226575

I have run dd on all of the disks multiple times and there are no errors. However I get the following message in my syslog.log;

Feb 16 06:00
SCSI: Resetting SCSI -- lbolt: 1877967359, bus: 3
SCSI: Reset detected -- lbolt: 1877967359, bus: 3
Feb 15 16:19:52 Moe EMS [2126]: ------ EMS Event Notification ------ Value: "MAJORWARNING (3)" for Resource: "/storage/events/disks/default/10_0_15_1.3.0" (Threshold: >= " 3") Execute the following command to obtain event details: /opt/resmon/bin/resdata -R 139329538 -r /storage/events/disks/default/10_0_15_1.3.0 -n 139329543 -a
Feb 15 16:19:52 Moe EMS [2126]: ------ EMS Event Notification ------ Value: "MAJORWARNING (3)" for Resource: "/storage/events/disks/default/10_0_15_1.6.0" (Threshold: >= " 3") Execute the following command to obtain event details: /opt/resmon/bin/resdata -R 139329547 -r /storage/events/disks/default/10_0_15_1.6.0 -n 139329544 -a
Feb 15 16:19:52 Moe EMS [2126]: ------ EMS Event Notification ------ Value: "MAJORWARNING (3)" for Resource: "/storage/events/disks/default/10_0_15_1.5.0" (Threshold: >= " 3") Execute the following command to obtain event details: /opt/resmon/bin/resdata -R 139329544 -r /storage/events/disks/default/10_0_15_1.5.0 -n 139329545 -a

My question is, is there any way to tell if this is a hardware issue or a driver issue.

Appreciate all of your insight,
Sheriff

[LoC_1]

Its possible its a timeout or a hardware issue. The first thing to do would be to run the command
/opt/resmon/bin/resdata -R 139329547 -r /storage/events/disks/default/10_0_15_1.6.0 -n 139329544 -a
as noted in syslog and you ill get more detail on the error.
Have a look at the error or post and we can decode it .

[Sheriff Andy]

This is basically what I get when I run the command.

http://docs.hp.com/en/diag/ems/scsi.htm#100091

This is for the other error;

http://docs.hp.com/en/diag/ems/scsi.htm#100091

Thx

[Sheriff Andy]

Sorry,

This is the 2nd link;
http://docs.hp.com/en/diag/ems/disk_em.htm#17

[A. Clay Stephenson · Accepted solution]

Because this is occurring on multiple devices, I'm more inclined to think the problem is not the disks themselves. It could be the controller or it could be something as simple as a bad/missing terminator. Often, SCSI buses with only one terminator will work almost perfectly. I rather doubt you have software problems because I assume the box has worked well for some time in the past. I would check the cables and make sure that you have exactly 2 terminators -- at the physical ends of the bus. You might also find that termination is improperly enabled on a drive in the middle of the SCSI bus.

Make sure that the following requirements are met:
2 terminators -- 1 at each end. The controller is generally self-terminated but make sure that the termination is enabled.
At least one device (anywhere on the bus) must supply termination power.
Substitute a known good terminator.

If all this fails, I would suspectr the controller -- these should be cheap on the used market.

[Sheriff Andy]

Here is some more of the log details.

[LoC_1]

From that error it looks like a driver issue and not a hardware problem.

I would probably install the latest driver scsi patch (assuming the drives are scsi PHKL_32091) and the lvm patch (assuming your using lvm PHKL_34013) along with the required dependencies

[Sheriff Andy]

A Clay & Louis,

Thanks for your great insight on this. This gives me 2 things to look for. I will make sure to check the termination on the drives. I can also say that we don't have the latest scsi and lvm patches.

Thx so much,
Sheriff