URGENT: Predictive Disk failure

[Juancho]

I restarted my ML370 and a disk error was reported. What should I do?
I have 4 hotplug disk, with two raid 1 mirrors configured (Disk1 + Disk2, Disk3 + Disk4).
Disk4 is giving the error and now has a flashing red light. (I attached ADU report)

I have another server that also has the same configuration but it's a backup server. I want to extract one of this disks and put it in server1. What steps should I take? Obviously server2's disk has data. Should I format it first before inserting it in server1?

Thanks a lot!

[David Paris_1]

You can do that because the controller will detect a new disk and start the rebuilding, so the previous information on that disk will be lost.

[Juancho]

But server2's disk4 has array data in it!
How am I sure that smartarray will copy mirror server1.disk3 -> server2.disk4, and not server2.disk4 -> server1.disk3 ??

Thanks for your time.

[David Paris_1]

Because the controller have the information that the the disk 4 is on predictive failure.
And the agents will detect that the disk 4 have changed of status.

And i have do that on the past with no problems, but if you wont you can format the other disk but the result will be the same.

But if you don´t wont to take any risk´s backup Disk3.

[Juancho]

Thanks again.

[Viswanath Adapa]

Hi,

Checked ADU report attached.
Server is an ML370 G3
Has a Smart Array 641 in Slot 2

SCSI Port 1 Drive ID 0 has exceeded the following threshold
Spare Part # 404709-001

Replacing disk with a known good HDD would rebuild the array automatically.
However, it is always advisable to have a complete backup of data.


ADU report indicated that the controller firmware is out of date.

You may upgrade the firmware to the latest from the following URL :

http://h18007.www1.hp.com/support/files/storage/us/download/26775.html




Thanks and Regards
Vishu