topic Re: 3 hard disk failed within 2 days. in General

3 hard disk failed within 2 days.

JI HUI — Mon, 01 Jul 2002 04:19:15 GMT

K380, HP-UX 10.20 two external hard disk storage
Within 2 days, 3 hard disk failed, what could be the cause?

thank you.

Re: 3 hard disk failed within 2 days.

Michael Tully — Mon, 01 Jul 2002 04:31:41 GMT

Please check to see if there are any messages in your /var/adm/syslog/syslog.log, 'dmesg' output or root's mail. It could be a coincidence or something else. How do you know that the disks have failed? If there are any messages, please post them so that they can interpreted.

Re: 3 hard disk failed within 2 days.

Sukant Naik — Mon, 01 Jul 2002 04:33:23 GMT

Hi Ji,

The problem is not with the harddisks.

The causes may be:
1. The SCCI cable connected to the Jamaica box might have gone bad. So the disks may be alright.

2. Check the power supply to the External Storage. Chances are the power supply unit might have got damaged.

3. The most important thing is take backup of rest of the data immediately.

-Sukant

Re: 3 hard disk failed within 2 days.

Paul Thomson_2 — Mon, 01 Jul 2002 04:34:39 GMT

Hi

Just wondering what errors were reported in the syslog at the time of these failures. Have you perhaps let your local HP representative look at these ?

Also, in the times I have seen disks fail, it was usually due to the age of the disks.

What was the age of the hardware.

Re: 3 hard disk failed within 2 days.

Shahul — Mon, 01 Jul 2002 04:41:56 GMT

Hi

U may be wrong. HDD may be OK. How U confirmed that HDD is gone? Hv U got some errors in syslog? Anyway.. Please follow this

First U check the power cable and SCSI cable of storage box. Whether they are fixed properly or not. Though SCSI cable fixed OK, It may not be a good one. Please get another one and check it.

Second check for errors in /var/adm/syslog/syslog.log. If U can see some errors related to HDD, Use dd command to verify HDD status. If the HDD is really gone, It must be because of some major power failure or power fluctuations. Please check ur power conditions before replaceing the HDD.

Best of luck
Shahul

Re: 3 hard disk failed within 2 days.

Paul Thomson_2 — Mon, 01 Jul 2002 04:47:28 GMT

Shahul

Good point
dd is a very useful way to check disk

dd if=/dev/dsk/c0t5d0 of=/dev/null bs=1024k

Generally the length of time this runs for shows how good, bad or ugly the disk may be.

Re: 3 hard disk failed within 2 days.

JI HUI — Mon, 01 Jul 2002 05:28:23 GMT

Thank you for your prompt reply. Sorry for any missing information
1. The error message shown in the console asked me to run "/opt/resmon/bin/resdata -R 100139438 -r /storage/events/disks/default/10_0.11.0 -n 100139012 -a "
The output of above command show that "This drive is not responding. Check cables, power supplies....."

2. Nothing in the dmesg or /var/adm/syslog/syslog.log indicated any failure related to hard disk.

Question:
If I run dd command, what i should observe?

Thank you

Re: 3 hard disk failed within 2 days.

Michael Tully — Mon, 01 Jul 2002 05:42:23 GMT

The longer the better runnind the 'dd' command.

You could also have a look at using the 'mstm' tool

# mstm
F2
Pick the hardware address for both the disks and SCSI interface and test it.
Highlight the address of the hardware you wish to test, and using the pull-down menus you can test the hardware. The output should indicate the problem. It is unusual for a SCSI card to fail, it is more likely disk.

Re: 3 hard disk failed within 2 days.

JI HUI — Mon, 01 Jul 2002 06:18:10 GMT

Ran xstm;
after remaping the system, the hard disks which failed, showed as "unknown" devices under scsi interface with the hardware path.

Question:
Does it mean it is a hard disk problem?
Why so many failure happened in such short period? Anything other than disk itself should be checked and how?

Thank you.

Re: 3 hard disk failed within 2 days.

Michael Tully — Mon, 01 Jul 2002 08:13:49 GMT

Yes it sounds like a disk problem I understand your concern. If you have a support contract, I would make them change the disk(s) and find out what is actually causing the problem(s). What about power? I've seen problems before with external disk arrays and power problems. Make sure you follow the advice in regards to creating good backups.

Re: 3 hard disk failed within 2 days.

JI HUI — Tue, 02 Jul 2002 00:40:47 GMT

I have powered off both server and external hard disk storage and powered on them again. Everything became OK!

So, is it a hard disk problem? power supply? or both? what tests should I do now?

Thank you!

Re: 3 hard disk failed within 2 days.

Michael Tully — Tue, 02 Jul 2002 02:05:32 GMT

Do have a support contract for this equipment?

If so call and get a CE on site to do the tests. Unless you are more familiar with hardware this is where to go next. Given that a power off/on has fixed the problem (most likely temporarily) get HP (or whoever does your support) to start changing components that are perhaps faulty.

SCSI card(s)
power supply in your external disk array
hard disk(s)
cables