• Status: Solved
  • Priority: Medium
  • Security: Public
  • Views: 2524
  • Last Modified:

RAID 5 problem on IBM x225

Hallo experts

I have big problem with one old server - IBM x225

We had Raid 5 with four idetical SCSI HDD, unfortunatelly, two of them is broken down (one few days ago, the second one today - I see it only by means of orange lights on the HDDs and of course, server stoped working and not possible to boot OS).

When I start the server I got this message:

Warning: 1 logical drive offline
Warning: 3 online drive defunct

Raid controller is: ServerRaid-6i Adaptec

Its there any way I could get at least some data from these disk?

thank a lot for help
0
AndrejArdan
Asked:
AndrejArdan
  • 3
  • 2
  • 2
  • +1
3 Solutions
 
Andrej PirmanCommented:
Hi,
I suggest you be very carefull or get expert call, because you may broke RAID5 very easily!

If your message is correct, with 1/4 disks offline, and 3/4 online, RAID5 should still be functional, but degraded. If 2/4 disks are offline, then it should be defunct.

Now, if you have backup, you may try putting manually ONLINE the disk which failed last. Sometimes drive fails offline and is normally functional after manually puting it online back again. Try it (with backup, of course!)
If your data is important and you do not have backup, your last resort (before playing with RAID!!!) is some good Data Restore company. They may image your physical drive into image and retreive data back. In my country, such a procedure for 4 HDD array would cost between 1000 and 2000 EUR.
0
 
AndrejArdanAuthor Commented:
Thanks, can you pls help me how to get it online again? because in raid controller utility its not possible to do anything. I only see, that 1 logical drive is offline. I am considering, if some kind of OS booted directly from CD could at least get me to the data, just like WinPE or so on.
0
 
DavidCommented:
Well, first, if the data is worth $2000+ then pay for professional recovery now.  Every moment a any of the disks is powered up, you risk further damage (assuming media failure like some crud is flying around smashing the platters at thousands of RPM making things worse ...


Having said that, here is a cheap way to attempt recovery
 1. Since you ran in degraded mode for at least several days, then the first disk that died is useless for you, so don't even *think* about trying to use that.
 2. Run diagnostics on the disk that failed.  Good diagnostics. You need to know exact nature of problem.  if disk doesn't spin up, or can't be seen by BIOS, then you need a data recovery firm, but only to get data back on that disk, you will want another disk with exact same data at same block number.  This will cost somewhat less than what you were quoted for whole thing.
3. If that disk can be recovered, then you need to run a reconstructor to extrapolate the XOR parity and live data into a recovered volume using scratch disk.   runtime.org's reconstuctor will work.

So that is it.  It comes down to a simple thing.  Get the last disk that died to a point where you have block-level image.  If that is not possible, or you won't pay the money, then forget about it, and hopefully some cheapskate that wouldn't pop for proper backup hardware and hot spare will be to blame, not you.
0
Get quick recovery of individual SharePoint items

Free tool – Veeam Explorer for Microsoft SharePoint, enables fast, easy restores of SharePoint sites, documents, libraries and lists — all with no agents to manage and no additional licenses to buy.

 
MS_help_guyCommented:
This would be the best time go to back up, may be...
0
 
Andrej PirmanCommented:
You can boot into ServeRaid Support CD and there you will find utility to manage your RAID array. And physical disks.

First, examine what's wrong with physical drives.

If you are willing to play roulette, inside this utility you have an option to mark off-line disk back to on-line. It will prompt you that data lose is possible, but...as I say, risk is yours, you may try. Sometimes it is possible (and NOT rare!) that disks simply detach from RAID array and they will function normally after you manually mark them on-line again.
Of course, automatic REBUILD procedure will occure after you make disks online again, and it might take several hours to complete. Be patient.
0
 
AndrejArdanAuthor Commented:
Thanks for help, I hope it will help us, tomorrow I will try do it. Fortunatelly we have almost full backup (95% of data) of that server.
0
 
Andrej PirmanCommented:
Good luck then :)

So if you sucessfully mark THE LAST FAILED (4th) off-line disk back to on-line, and if automatic REBUILD procedure suceeds, I would imidiatelly!!! order 2 new disks, and:
- while 4th disk is still offline, replace it with new disk, wait 15 minutes and if autorebuild does not start, mark it on-line manually and wait for auto rebuild
- the next day, when rebuild finishes, I'd replave 3rd disk too. Since it is now online, first go to controller utility and either mark it "prepare for removal" or "offline". Then pull it out, insert new disk, and wait 15 minutes if auto rebuild will start. If not, mark it manually as online, and auto rebuild should start.
0

Featured Post

Industry Leaders: We Want Your Opinion!

We value your feedback.

Take our survey and automatically be enter to win anyone of the following:
Yeti Cooler, Amazon eGift Card, and Movie eGift Card!

  • 3
  • 2
  • 2
  • +1
Tackle projects and never again get stuck behind a technical roadblock.
Join Now