Solved Why use ZFS Hot Spares?

Phishfry

Phishfry

I have been reading alot about ZFS as I am about ready to setup the drives in my 24-bay Chenbro Chassis.
What I don't understand is why you would use Hot Spares. It just seems like you are putting hours on the drives for nothing.
Hot Spares are not automatically used in a failure so what is the advantage?
I understand having cold spare drives mounted in trays but what do you save by having them live?
Is this usage best for off-site/remote data storage where swapping in a drive would not be possible?
 
usdmatt said:
zfsd() should be able to handle a replace, so the benefit is starting a resilver immediately and minimising the degraded time.
Click to expand...
Look at the formulas for reliability of a RAID array, one example is in the original Garth Gibson paper. The reliability of an array improves directly with lowering the MTTR or Mean Time To Repair. So starting the resilvering as soon as possible is very important. Even if ZFS doesn't start the resilvering by itself, having the spare already physically in place makes it possible for an admin to start that process faster than having to wait until they are on site and have a spare. The typical time to get a spare installed is often several days, and that's for systems that have very good maintenance contracts with well paid and trained service personnel.
 
You must log in or register to reply here.
Back
Top