Home / server / Questions / 167307
Accepted
Nathan Friend
Asked: 2010-08-06 02:13:36 +0800 CST

1950 Replacing drives in a predictive fail state

  • 772

I have a Dell 1950 server with two 70Gb SAS drives. Both drives are showing an amber flashing light, and OpenManage displays both drives in a predictive fail state.

Dell have sent me to two new drives under warranty, the new drives are 300Gb.

1) What is the procedure for replacing the dives, I would like to keep the server running without restarting if possible.

2) Can I RAID controller cope with different capacity drives?

Cheers,

Nathan.

raid dell-poweredge

2 Answers

  1. Best Answer

    Replace one drive, rebuild the array and replace the other. You will end up with 70GB virtual volume though, which will let you create a second one using the rest of the space available.

    The PERC5i/6i controller can easily handle the different sizes of course. Just make sure you do the rebuild out of hours, so the server is not overloaded, and the rebuild overhead doesn't affect the users.

    And of course, before anything else - make sure you have a valid, tested backup.

    • 3

  2. I'd be tempted to remove the drives, put in the new drives, format them and create a 300 gig array, then restore from backup so you don't have a 70 gig RAID and leftover 230 gig volume. Plus you have a chance that if the drives are predicting failure, you may have a total failure during the rebuild, leaving you in position of more downtime than you originally planned for.

    • 1