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.

We have a RAID5 array and limited resources.

It has been suggested, for data destruction of a SAS drive with a bad sector, to remove ALL drives from the RAID5 array, and then place this 1 drive into the RAID5 to access and wipe the data. The server we are speaking of is an MFSYS25 Intel Modular Server.

My question: How could the RAID be preserved? My thoughts are, the RAID would need to be deleted and the disk would need to be added to its own RAID0 array or as a hot swap, and then use DBAN to wipe the drive.

The last thing I want to do is cause all the data in the RAID5 to be lost due to negligence or incorrect analysis.

share|improve this question

1 Answer 1

If you have a bad drive in a RAID5 (or just about any RAID), pull the bad drive, put in a new one, let it rebuild. Then either use the tool of your choice to securely wipe the bad drive, or just smash it with a hammer.

So, the person is telling you to remove all the drives, just so you can access the one that is bad, and wipe it? Is that because you don't have another way to plug a SAS drive into another system? That's adding a huge amount of unnecessary risk to physically dismantle your RAID array, and hope it comes back online when you plug all the drives back in. (you labeled each one with the port that it was in, right?)

If you are concerned with data that may be on that disk, either send it out to be destroyed, physically destroy it yourself, or plug it into another system and run a tool like DBAN. Do not take apart an array for a simple task like that.

EDIT: Since you need to return the defective disc, purchase a SAS to SATA adapter, have it overnighted and wipe the drive using one of your workstations. Still a cheaper and safer alternative to the original proposed plan.

share|improve this answer
    
Thanks for the information. The RAID array was recently rebuilt successfully with a warranted hard disk drive and is in a healthy state. We are trying to wipe the old drive now, and it is a SAS drive. We are having difficulty because all of our alternative servers are SATA, which does not recognize SAS. –  Radioactive 1 Apr 28 at 17:09
    
This idea was proposed since it was a low-cost option. However, the risk seems to outweigh money. Restoring all VMs on this RAID array would at best take 2 days. The issue is that this server is the only one with a SAS controller, all other servers have SATA controllers. If we were to complete this, we would of course label the drives properly. I'm just thinking there would be risk namely because the RAID5 would need deleted and re-added. I am cringing already at this; I just need to justify. –  Radioactive 1 Apr 28 at 17:20
    
You are absolutely correct. Physically destroy the disk yourself, its free. –  DanBig Apr 28 at 17:21
    
The downside to this is: the disk is a warranty claim. The vendor is expecting the returned, faulty disk back within a week. We are concerned sending sensitive data to the vendor although they claim to destroy all data on returned drives. –  Radioactive 1 Apr 28 at 17:32
    
Take a magnet to it. :P –  Nathan C Apr 28 at 17:53

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.