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

I just set up a RAID 1 array, 2 2tb drives mirrored.

I am wondering what will happen if one were to fail.

Would unplugging one of them be a valid test?

I am using the Onboard controller of an asus M4A89TD pro series motherboard.

Thanks, Blake

share|improve this question
up vote 4 down vote accepted

Unless the controller and the drives support hot-swap unplugging one while the system is running is very risky. They'll probably survive but you could potentially permanently upset the controller on the drive or, worse, the machine's I/O controller.

So unless everything supports hot-swap, unplug the drive while the machine is not running then start it up again, and to test the resync process again power off the machine before changing what is plugged in. If all is working as it should the controller will see the drive reappear and start a full resync.

Remember that the resync will take quite some time - unless the controller uses a write-intent map (which it almost certainly won't as it is both an advanced technique and something that can have significant performance issues so will be turned off by default if the controller supports it) or some similar technique it will read then entire 2Tb from the "good" drive and write it back to the drive that has been reinserted into the array. During the time it takes for this to happen you array is operating in a degraded state and if something goes wrong with the other drive in that period you may end up with a badly corrupt filesystem on the surviving drive.

Long story short: Yes. But be careful.
Refreshing and testing your backups first would not be a bad idea.

share|improve this answer

Sort of. It is a valid test of failure (the ability of your system to keep running), but not a valid test of your controller's repair mechanism (it's ability to assimilate a replacement drive) unless you also format or otherwise wipe the disk before re-inserting it.

I would test this for a scratch volume before placing it into production, and document the procedure to correctly attach the "new" drive to volume without losing the data on volume. Most recent controllers can handle this automatically for the same drive inserted to the same bay, but sometimes you have to jump through some hoops; it's nice to have that set of instructions ready ahead of time.

share|improve this answer

Yes, but that risky.

drives will be not in sync, then when you connect them both again, hard to predict. most likely you will need too format disconnected drive and join it to raid1

test it on disks with data you do not afraid to loose

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.