Hard Drive Errors on Raid-5

[allanc]

One of my clients has a Dell PowerEdge 1800 (Windows Server 2003) with 3 hard drives on RAID-5.

One of the drives is failing the Dell 'confidence' diagnostics within a couple of minutes.
However the RAID-5 still displays 'good' as its status and has not degraded its status or taken this drive offline.

I know that when a drive's status has been downgraded, you swap the drive and then choose 'rebuild' from the RAID menu.
My question is what are the specific steps if I want to replace the drive proactively (before it fails)?

Thank you in advance.

 

[Ccomp5950]

Just swap the drive out with a new one and tell it to rebuilt itself.

it will take a couple of hours depending on size.

 

[allanc]

Just swap the drive out with a new one and tell it to rebuilt itself.

it will take a couple of hours depending on size.

I will be swapping the drive tomorrow.
Thank you for the info.

 

[jeremyc]

better swap it soon

we lose exchange db's when that happens...
you dont want to spend all weekend running eseutil!

 

[allanc]

we lose exchange db's when that happens...
you dont want to spend all weekend running eseutil!

Are you referring to the swap drive process itself or the 'confidence' test of the diags failing?

 

[MobileGeeks]

Swap it out dude, I would do all three drives since SATA drives are cheap.

This of course will be done over three visits onsite.


Thanks.


Mobilegeeks

Computer Repairs

 

[NETWizz]

Make sure to failover/spin down if there is an option for that in the Dell Open Manage Server administrator.

If there is, the array should then show up as degraded. As soon as the disk is replaced, it should just rebuild.

 

[Blues]

Swap the drive like the rest suggest I don't know much about the PowerEdge 1800 what kind of RAID controller card it runs or drives it uses but my experience with RAID% drive swaps is they are hot swappable and the rebuild is automatic.

 

[Tony_Scarpelli]

Swap it out dude, I would do all three drives since SATA drives are cheap.

This of course will be done over three visits onsite.


Thanks.


Mobilegeeks

Computer Repairs

Mobilegeeks,

Have you ever heard of the term infant mortality as it relates to engineering or new equipment?

Google it.

http://www.cvarc.org/tech/burnin.html
http://www.lifetime-reliability.com...e-strategy-101/equipment-reliability-101.html

If there is no reason to suspect the other drives, keep them. Or maybe put a hot spare in, but I wouldn't put new drives in an old server without burning them in some how.