Degraded RAID 1 Array:PowerEdge 2800

Hello to everyone. I hope someone can shed some light on this fairly unusual situation.

I have a PowerEdge 2800 which is due to be replaced before May this year, it's a great server and done a great job over the last 7 years.

Ok to the problem; Server Administrator was predicting a failure on 0:0

In the server we have:

0:0 Maxtor 36GB 15k 0:1 Fujitsu 73GB 15k 1:2 Something .. 1:3 Something ..

So, I popped 0:0 and the server was working fine with a degraded array, and after 15 seconds I inserted it back in to the PE2800. After approximately 45 minutes the RAID 1 array was rebuilt, but it still said predicted failure. So I thought I best order some new drives - I received them today; 2 x 146GB 15k Seagate Cheetahs.

Ok, so I popped 0:0 drive and inserted one of the 146GB cheetah's and the server crashed - would not boot at all, and was giving me an error saying the logical drive had failed.

So, in the PERC4i BIOS, I forced drive 0:1 online, and the server is now back up and running albeit with a degraded array.

So this is where I'm at, what choices do I have?

Insert the old drive with a predicted failure, leave it alone, or try the 146GB cheetah again. I don't really fancy doing any of them, but I will order another 73GB Fujitsu in the meantime.

What I want to know is why did the server crash after inserting the cheetah drive? I have swapped hot pluggable hard drives many-a-time in the past in many dell severs and never experienced anything like this.

Currently running degraded, so any replies would be massively appreciated.

Thanks a lot.