re: RaidFrame with unequal size replacements

[matthew green <mrg%eterna.com.au@localhost>]

      I worry I may be "in a pickle" here.  I have a sparc64, that has a  
   raidframe raid1.  I lost a disk recently, and the replacement has  
   arrived, but appears to be a few sectors (143363688 vs 143538120)  
   smaller than the one it's replacing. and the pair in the raidframe.
   
      So, the first question is, does raidframe know how to deal with  
   this?  I doubt it, given that dynamically sized filesystems don't, as  
   far as I know, exist in NetBSD.  The next question would be, any  
   suggestions on how to fix it?  Only idea I have is to build a  
   secondary RAID1, with only one disk in it (the new smaller one), then  
   build a new system onto that, and then kill the old RAID making the  
   one remaining disk in it be the partner to the new disk in the new  
   RAID.  But, I don't want to walk down that path without asking for  
   help.  Maybe there's something easier I could do.


raidframe usually skips a few sectors for some alignment i'm not
sure about, but the above numbers are in the 200,000 sector range
and that's really significant.

i've done the create a new raid and copy over a bunch of times.
unless someone else has other ideas, i think this is all you can
really do at tihs point...


.mrg.