On Thu May 02, 2013 at 03:58:29PM +0100, Robin Hill wrote:

On Thu May 02, 2013 at 09:26:55AM -0500, Barrett Lewis wrote:

quoted

That is a really good idea.  At minimum I will ddrescue the failed drives.
But even if I do that, is there a way to reassemble the array (even with
new drives) now that those components are marked failed?

Yes, "mdadm -Af /dev/mdX" should reassemble the array, making use of the
best set of drives available to get the array running - i.e. all the
working drives plus the most-recently failed drive(s). It'll still be in
a degraded state, but hopefully you can then get the other disk added
(though you'll still have to keep your fingers crossed that there's no
read error of any kind during the several days required to rebuild a 4T
drive).

Sorry, got you mixed up with someone else there. Anyway, the command
should reassemble the array and allow you to get the rest of the data
off. You might want/need to also specify the disks/partitions it should
be using for the assembly.

Cheers,
    Robin
-- 
     ___        
    ( ' }     |       Robin Hill        [off-list ref] |
   / / )      | Little Jim says ....                            |
  // !!       |      "He fallen in de water !!"                 |