SmartArray: Is this procedure valid/supported?

[chuckk281]

Javier had a SmartArray question:

**********************

Hello Team,

During years I’ve used and heard a lot of times the following procedure, in Proliants/blades SmartArray with system disk in a RAID1 configuration, as a fast way to go back in case of problems:

             - Shutdown server and extract one disk

             - Boot up the server and perform tasks: install service pack, update a driver, install something, etc……..

             - If all works fine plug back the extracted disk while the server is started to re-synchronize the mirror.

Is this procedure valid/supported and if so, is it documented/written/where (I’m unable to find it out)?

I suppose that the perfect way is plug back the extracted disk after shutdown the server and press F1 to rebuild disk during next startup.

Thanks in advance.

*************

Reply from Eirik:

***************

I have customers who do that all the time.

Remove one disk while server is off.

If config is good insert disk while system is running, then it rebuilds automatically.

If config is bad shutdown, remove active disk and insert old disk, boot and insert the disk you removed, auto rebuild.

*************

More info from  Hiro:

************

This wiki is really great to find Smart Array related information.

https://hpedia.hp.com/wiki/HP_Smart_Array

You can find the link to the whitepaper in the Smar Array wiki.

I think it’s you are looking for.

RAID 1(+0): breaking mirrors and rebuilding drives

http://h10032.www1.hp.com/ctg/Manual/c00378986.pdf

Reverting to a previous copy of a mirrored set using the manual method

************

Your comments?

[Johan Guldmyr]

I take it hpedia.hp.com is HP internal only?