Take the 2-minute tour ×
Server Fault is a question and answer site for professional system and network administrators. It's 100% free, no registration required.

I am running a RAID 1+0 on a HP ProLiant DL360 G5 with 6 disks (each 146 GB).

The mirror groups looks as follows:

Group 1

  • 146 GB 1-Port SAS Drive at Port 1l: Box1:Bay1
  • 146 GB 1-Port SAS Drive at Port 1l: Box1:Bay2
  • 0 GB 1-Port SAS Drive at Port 1l: Box1:Bay3

Group 2

  • 146 GB 1-Port SAS Drive at Port 1l: Box1:Bay4
  • 146 GB 1-Port SAS Drive at Port 2l: Box1:Bay5
  • 0 GB 1-Port SAS Drive at Port 2l: Box1:Bay6

Unfortunately the disks in bay 3 and bay 6 are not working anymore (almost since the same time). Is there a chance to get the system back online again by replacing disk 3 + 6 by new ones?

* UPDATE *

It's a MS Server 2003 R2. I've got the information that the disks are broken from the HP smart start CD diagnostics tool. The following versions are in place: * ACU Version 8.60.7.0 * Diagnostic Module Version 5.2.52.0 * INFOMGR Version 5.9-29.0

Server was already off (pressing power button) when people involved me. ROM/RAM firmware revision is 2.10

share|improve this question
    
Is this a Linux system or Windows system? Where is this output coming from? hpacucli? –  ewwhite Feb 25 '13 at 15:48
1  
Also, when did the drives fail? –  ewwhite Feb 25 '13 at 15:58
    
Thanks for the answers so far. Will check system / firmware details tomorrow. Is then the mirroring between Group 1 and Group 2 and the striping within the group itself? –  user161767 Feb 25 '13 at 20:47
    
@YvesSturzenegger the mirrored pairs are striped... –  ewwhite Feb 25 '13 at 20:52
    
@YvesSturzenegger We need the controller and server firmware revisions. Did you try the powering-down as I suggested? Are there any filesystem errors on the server? –  ewwhite Feb 26 '13 at 8:41
show 2 more comments

2 Answers

If disk 3 and 6 have failed and they won't come back online, then no, you can not get your data back.

share|improve this answer
    
Not necessarily... It depends upon what's being shown at the OS level. This is an old system, and people tend not to keep firmware updated. There were early firmware revisions of that controller that falsely marked drives as failed... –  ewwhite Feb 25 '13 at 15:57
add comment

If you can shut the system down gracefully, you can be able to jog the disks through a simple power-down operation.

You're correct in that the mirror groups mate RAID1 pairs from each group, as in this example.

If you can get the system off and the drives spun down (also remove power), there's some hope that they'll spin up and give the controller a chance to perform an array recovery.

Of course, at the same time, you'll want to get spares or replacements ready or on the way.

Can you share the specific server model, firmware, OS and other details? There are cases of early firmware revisions of that Smart Array P400 giving false-positive failure indications on drives, so a comprehensive evaluation of your firmware and system health is in order.


Edit:

You say the Smart Array P400 RAID controller's firmware version is 2.10 from May 2007. The current version of firmware for that system is 7.18!!

You are severely behind. Also, I suggested pulling the power cables to the system for a few minutes and trying a cold start. Take close note of the RAID controller messages as the system boots. Or does the system even boot properly? Do the physical disks have error lights illuminated?

Can you try a firmware update of all of the components using the bootable HP Firmware DVD just to rule-out bad firmware interactions or a false-positive disk failure?

share|improve this answer
    
I just updated the firmware (7.22). Unfortunately, it's still not working. Looks like the disks are broken. RAID controller error message is: "disk drives are failed and should be replaced". Error lights are red for the two disks. System boots up to the point where it says "Non-system disk or disk error". –  Yves Feb 27 '13 at 9:37
    
Oh well... I doubt that the drives failed at the same time, so going forward, you will need to add the HP monitoring agents so that you can obtain email or SNMP alerts to drive failures... or at the very least, give the system a quick visual check to look for red lights. They RAID controller may have given you an F1 option to attempt data recovery during post. Reboot and try that. If it doesn't, your data is gone. –  ewwhite Feb 27 '13 at 9:48
add comment

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.