Managing Hard Disk Crash

[Alice Daniel]

What is the best way to recover from a hard disk drive crash? Should I have a separate disk for multiple systems on multiple models of the alpha servers (ds20e, es40s and es45)? Would cloning work and if so, do I still need a separate disk per system?

[Ivan Ferreira]

>>> What is the best way to recover from a hard disk drive crash?

Backup/Restore of course, but, is your question oriented to recover the operating system? If so, again "backup/restore". You can use vdump/vrestore or btcreate.

You can use LSM Mirroring.

[Rob Leadbeater]

Hi Alice,

Personally, I would ensure that all of the disks in the servers are connected to a RAID controller of some form, preferably a KZPDC (SmartArray) controller.

Then if you have a disk failure, you won't get any problems at the OS level.

Of course, if you are backing things up properly with vdump, then restoring is straightforward...

Cheers,

Rob

[Kapil Jha]

to recover.
1.LSM Mirroring.
2.LSM evacuation mechanism.(you will have to have some spare disks)
3.periodic tape backup is the best way.
4.periodically see the counters of the error for a disk(hwmgr -get attrib -id HWID),and change the disk beofore it crashed.
Kapil