AutoRaid 12H

[Paul Dray]

Hi to All, and HELP.
I have a disk in slot a6 which is giving errors ( See Attached ) and is still being used. I wish to swap this for a new one.

I believe I have to down the disk first, but, there may be recovery maps on it.

Q 1. How do I find out if this disk has recovery maps on it ?
Q 2. If so, how do I move them to another disk ?
Q 3. How long does this downing process take ?

Once downing has completed i beleive i am free to swap the disk and then require to add the disk back into the array using ARRAYCFG -Da6 -a.

Q 4. How long does this process take ?

Q 5. How can I show that the disk is back in and being used ?

Q 6. How can I show that I have redudancy back ?

Q 7. If recovery maps are an issue ( as Above ) how do i move them back ?

Any help would be greatfully received.

Thanks.

Paul Dray.

[Brian M. Fisher]

If your AutoRaid is configured with Hot-spare enabled, all you need to do is to remove the disk, wait for the rebuild to ocurr then insert your new disk. This can be done on the fly although performance will go down until the rebuild has completed. I have successfully done this with no problems.
But as always make sure you have a GOOD backup first.

Brian
<*(((>< er

[Insu Kim]

In my opinion, It's not that serious.

If auto include option is enabled, you can see this in front panel or sam, all that you have to do is simply to swap a fishy module to a new one.
auto include is a default on a new disk array.

If auto include is not, you should include a new disk manually.

[Antoanetta Naghiu]

The previous posts are right. For more details, check:

http://www.hp.com/essd/model12H_writews.html.http://my1.itrc.hp.com/cm/QuestionAnswer/1,1150,0x856a7e990647d4118fee0090279cd0f9,00.html

[Nelson Broskey_1 · Accepted solution]

Paul:
I recommend downing any suspect disk before
removing it. For mech at A6 the command is:
arraycfg -D A6 -d -R

Firmware HP54 and higher includes Data Resiliency (recovery maps). Recovery information is on all disks and not an issue
when removing any disk.. this feature is for
near-complete recovery (4 secs old) of configuration information in the event BOTH controllers and all 4 batteries fail and a
normal shutdown has not occurred.

Downing a disk takes as long as it takes (nice answer huh?) and depends on how much
data is on the bound luns...and how much
unassigned space the array has (note: deleted
files will not increase "free" space on luns,
you must DELETE luns to return space to the
array).

Arraydsp -a will show you the status of mech A6 after it is replaced. It
should show included and rebuild or balancing
information will also be reported.

Redundancy will be lost during "downing" of
disk and will be returned if sufficient unassigned space exists, or if active spare
is enabled.

Maps are updated to all disks every 4 secs
by default and will be written to the new
mech once rebuild/balancing is complete.

I think that covers all your questions.??