Subscribe

Bad RaidLabels for FAS 2040

Hello,

I've been experiencing some difficulties with raid volumes recently and was hoping someone might have some insight. I've looked around on the discussions here and have found some similar topics but not quite what I'm looking for.

I recently had to aquire a new shelf from our warehouse (PSU problems) and stuck my disks into the new shelf. (I didn't properly prep disks for re-assignement) Unfortunately, the new shelf had Data OnTap 8.0* on it. We use automated FAS Load Tools that are scripted to zero and build our filers. The script is designed to run from Data OnTap 7.3.4. (7.3.4 uses raid labels of 9, 8+ uses raid label 10).

I am able to sucessfully zero the filer, without the script and I was also able to install DOT 7.3.4, but was unable to "revert_to". ... So I have a blank filer using DOT 7.3.4 and 12 disks that are labled with a raid 10. Since all 12 disks are the wrong label, I am unable to create a flexible root volume.

I can't seem to use any "aggr" command. I can't seem to find a way to "relable" the disks.

In a nutshell....

* My Filer is a FAS 2040 with 12 disks (600GB each).

* I recently swapped my disks into a new shelf with a higher DOT

* I downgraded from DOT 8.0 to 7.3.4

* All 12 disks are Raid 10 and they need to be Raid 9

* I can't load a flexible root volume.

... What is the proper procedure for relabeling all 12 disks for Raid 9 without having a root volume or having any other spare disks?

Any help that anyone cares to give would be most appreciated!

Thanks,

Cory

Bad RaidLabels for FAS 2040

What do you mean exactly with:

I am able to sucessfully zero the filer

Peter

Bad RaidLabels for FAS 2040

What happened when revert_to failed?  That seems like the key issue...the drives zeroed with a 7.3 flashcard boot even without a revert?  You could try "priv set advanced ; disk unfail -s diskid" if the drives show as failed... at another customer we found that disks from an 8.x system in an aggregate do not work in 7.3 without revert_to (to revert the raid label from 10 to 9 as you mentioned) but spare disks in an 8.x system worked fine in a 7.3 system.