on 2/12/2008 2:00 PM Jimmy Bradley spake the following:
OK, I've figured out which drive is bad. Before I go any furtherlet me give a brief HD configuration explanation. The primary IDE has a 80gig(Master) HD, and an internal zip drive(Slave) connected to it. The two 200gig Hard Drives are connect to a PCI ATA controller card(Not a promise card), one is master, the other one is slave. It is the slave 200gig drive that is going bad. Now, I have a feeling this would be too easy. When I set the machine up, Cent OS set the drives up using LVM. I only have about 20 to 30 gigs of data saved on the machine right now. Maybe not quite that much. Like I said, I have a feeling that hoping this will work would be too easy, but what do you think my chances would be, that I could disconnect the drive that is going bad, boot the machine up, and it fully boot up, and all my data be intact, and LVM automatically resize the home partition down to around 200gigs instead of 405gigs?
Jim
Slim to none! There is no automatic fix for this one. You need to remove the failed drive from LVM the proper way, but you should be able to do it live if your drive will hold on long enough.
The easiest way would be to add another drive of equal or larger size, add it to the volume group, move the volumes off the bad drive and then remove the bad drive.