Re: Moving disks to another controller (head)

Hi Folks,

Thanks so much for all of your responses.  It looks like my plan below will
work fine.  However, I didn't make it clear that I am moving some (but not
all) disks/aggregates from the source filer to the destination filer. The
source and destination filers will both remain in production after the
move.  I don't think that this matters, but just thought I would check.

We are moving some of our equipment to a new building and for DR we need to
keep primary volumes and their snapmirror copies in different buildings.

A more complete picture of the plan:

1) upgrade DOT to the same version on both heads (8.0.1)

2) shut down and power off both heads

3) physically move the disk shelves

4) boot the destination head in maintenance mode and use
   "disk assign" to take ownership of the moved disks

5) boot the destination head normally

6) aggr online the foreign aggregates

7) aggr rename the foreign aggregates to get rid of aggrX(N) names

8) boot the source head normally, which now has fewer aggregates.


At step 4) when booted in maintenance mode, I believe that all
aggrs are offline so it is safe to assign disk ownership
without risk of data corruption.  The disks need not be unowned.

At step 8) I presume the source filer may notice that some
of its former aggregates/volumes are missing, but it will
soldier on.

Anyone see any obvious trouble?

Thanks again!




Steve Losen   scl@...    phone: 434-924-0640

University of Virginia               ITC Unix Support