Reassigning a Disk When ADP is Enabled

[TMADOCTHOMAS]

For the first time, a disk failed on one of my 2220s set up with cDot 8.3.x and ADP enabled. My routine when a disk is replaced is to then assign that disk to one node or the other, however when I checked it appeared that the disk has already been assigned. Anyone here know what happens when a disk is replaced on a 12-disk 2220 with ADP enabled? I've searched NetApp's website, Google, and this message board, and can't find the answer.

[SeanHatfield · Accepted solution]

On the 2000's the default autoassign policy will assign the odd numbered bays to one controllers and the even numbered bays to the other.  

Check the autoassign policy, and if its 'default' or 'bay' you get this behaviour on entry level systems:

Cluster1::> storage disk option show 
Node           BKg. FW. Upd.  Auto Copy     Auto Assign    Auto Assign Policy
-------------  -------------  ------------  -------------  ------------------
Cluster1-01    on             on            on             default
Cluster1-02    on             on            on             default
Cluster1-03    on             on            on             default
Cluster1-04    on             on            on             default
4 entries were displayed.

Cluster1::> 

On ADP enabled systems you can juggle ownership at the partition level with the -root or -data options from the advanced priviledge level, if needed.

Here's a link to some documentation on disk auto assignment:

https://library.netapp.com/ecmdocs/ECMP1636022/html/GUID-03041BA5-F93B-434D-A0FC-E74C44C4D739.html

[TMADOCTHOMAS]

Sean,

Thanks very much. We never used autoassign on our 7-mode systems and it didn't occur to me that it was enabled by default in cdot.