How to Detect Failed Hard Drive in LVM Cluster

richip · 05-25-2012, 09:04 AM

It seems one (or more) of my harddisks / hard drives in an LVM Volume Group has failed and I need to take it out. The problem is I don't know which drive failed. How do I determine which of my hard drives has gone bad? Is there a way to recover the contents on the volume and logical groups in the drives which hasn't gone bad? Any suggestions on how to take out the harddisk? Will normal lvm apps just allow me to remove the failed physical drive from the volume group?

lithos · 05-26-2012, 03:43 AM

Hi,

I don't know LVM's but if your LVM is created on RAID1 or RAID5 then you will be able to replace the drive and contents will be restored from a healthy drive. Otherwise I think it's lost.

If you're running a production server, then it will probably not be possible to shutdown and give a drive test (with bootable CD).
- HD diags
- UBCD

syg00 · 05-26-2012, 05:37 AM

Quote:

Originally Posted by richip

It seems one (or more) of my harddisks / hard drives in an LVM Volume Group has failed and I need to take it out.

How did you come to that conclusion ?.
Quick google was enlightening - this popped back up. Seemed sensible when I first saw it a few years back.

skipper123 · 06-01-2012, 08:30 AM

Now that depends on which Hardware you are using. If you have a dell server, you might want to use the omreport tool to find out which hardrive has failed. If you are using HP Server, you might want to use hpacucli. These tools needs to be installed before you can run them. You can find more info if you google for these tools.