[CentOS] Raid Questions

Mace Eliason

meliason at shaw.ca
Mon May 1 17:19:32 UTC 2006


Hi,

I have a centos 4.2 server setup and running as a zimbra email server.

It has 2 18.4Gig SCSI hard drives setup in software raid 1 for mirroring.

I had tested the unit before production and found that if I disconnected 
one of the drives the other wouldn't boot

I search and found that I had to modify the boot (I don't remember 
exactly at the moment) 

After testing I am sure it worked.

On the weekend the power went out and turns out one drive had failed and 
the server wouldn't boot from the other drive.

I am not sure at what point the other drive failed could be days could 
be weeks ago.

What is the proper way to get the system running again?

I can fix the failed drive.  (damaged pin).  When I get it running again 
will the info on the drive that won't boot be mirror back to the other 
drive?

I am sure the drive that won't boot due to grub and raid is the most 
current.

Thanks for any help

Mace



More information about the CentOS mailing list