Server Fault is a question and answer site for system and network administrators. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

We have 9 machines at the moment. One of them had an electrical fault which fried the board but the drives seem ok. I'm not entirely sure how I would even attempt to recover data from it. The drives were in RAID 5 and I'm not sure if hotswapping a drive into another server will keep its data, fry it, or just not work at all.

Any advice would be helpful.

share|improve this question
up vote 10 down vote accepted

Before you try anything, I would plug the drives into a non-RAID-controller and dumping them byte by byte (e.g. via dd) onto some other storage. Then you can do whatever you want - you have a backup.

share|improve this answer
+1 A few years ago I screwed up a (consumer level hardware) raid array trying to do a plug into equivalent controller recovery. – Dan Neely Nov 29 '11 at 16:48
of course depending on the controller you might end up with a "backup" of random bytes - YMMV – Jim B Nov 29 '11 at 18:22
Better yet after your dd, attempt recovery from the backup. That way, if it turns out your dd didn't work, you can do it again. – derobert Dec 5 '11 at 21:48

I have tried that before on IBM servers, and it didn't turn out pretty well.

I would recommend that you contact the vendor of the server/RAID controller and ask for further assistant there.

share|improve this answer
I'm going to do that first, thanks for the reply – samdunne Nov 29 '11 at 13:26

If your drives are OK, you can just insert them in another server with similar RAID configuration and it should work. It is recommended to do this on a testing (non-production) server.

Even if you have one faulty drive, you should be able to get your data.

share|improve this answer
Thanks, gonna try that now :D – samdunne Nov 29 '11 at 12:24
+1: I did this with HP kit, it worked out well. – Stu Thompson Nov 29 '11 at 13:55
@StuThompson: HP has made a point through the years of keeping the on-disk RAID format the same. Basically with HP if the drives can be hooked up to another of their RAID controllers it will work. With most other vendors this is not true. – freiheit Dec 2 '11 at 18:29

If you place the drives in a server of the same model and with the same RAID controller, you might be able to get the data out of the array.

But this strongly depends on the RAID controller's make and model (which you really should add to your question).

share|improve this answer
The issue is that the RAID card holds the configuration of the array. There should be a vendor specific way of copying the array config from one card to the other. Whether this is possible or not if the original card is dead would be the question. – AngerClown Nov 29 '11 at 13:52
Some cards also let you recover the RAID configuration from the drives. – Massimo Nov 29 '11 at 14:17

As others have mentioned it might be possible to recover the drives by using the same model RAID card. I've done that with HP drives and raid cards without issue. It's vital that you label the drives so that you keep each drive in the same slot/order. With IBMs some raid cards this works on, some it doesn't.

share|improve this answer

You should restore your backups to known good hardware if you need to be up and running any time soon.

After a fault like that, you should run the drives in a test server before redeploying them to make sure that they really are OK. I would be very hesitant to rush any component from a server that died like that back into production.

share|improve this answer

Your Answer


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.