Server Fault is a question and answer site for system and network administrators. It's 100% free, no registration required.

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 a 4 x 100gig EBS volumes used as dynamic disks in a raid 0 (striped) drive attached to a Windows2008 R2 Datacentre 64bit Amazon EC2 instance.

This all works fine, except we have had 3 occurrences within 5 weeks where one of the disk suddenly becomes unreadable/unavailable and it is offline in disk manager, and you can't reactivate it back into the striped drive using the disk manager.

The only way we can seem to get it working again is to restart the EC2 windows instance. All disks/volumes of the striped drive then come back online once it boots up again.

Amazon haven't yet been able to answer the question for us and I was wondering if anyone else had experienced the same problem or had a solution ?

share|improve this question

If it's the same disk every time, I can say I've had this problem before as well. Best I can tell, the server hosting that volume could be either slow or otherwise wonky. You might want to try making a second raid0, copying all the data over, and smashing the first. It's brute force, but effective.

Another option is to recreate with raid5. It will be slower, but at least you won't be offlined by a single disk missing, and it's much easier to recover with a new volume with the system still active (albeit with a degraded raid5).

Our solution is to have primary and secondary servers, where if the primary messes up we can fail over. Usually the problem system will clear itself up after a while, and the original system becomes usable again.

share|improve this answer
This has happend on 2 different windows 2008 instances and different raid0 volumes in the last 5 weeks. Amazon are now saying we need to update the Redhat PVR driver which includes fixes for Raid1 and Raid5. Thats all very well, but telling us to go Raid1 is essentially like telling us they dont know why Raid0 continually reports the drive as unavailable and therefore you need to put a plaster on it by using raid1 or raid5. – user93900 Sep 13 '11 at 8:56

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.