From: David Graves (dsgraves@gmail.com)
Date: Tue Feb 21 2006 - 22:26:43 EST
I have an Ultra 30 connected to a D1000 with 2 controllers (D1000 is split
in 2, 6 disks to a controller). All the disks are configured as one Raid 5
metadevice.
I experienced what appears to be a power glitch: enough to power down the
D1000, but not powering down the server.
This is my guess what happened next: The server tried to write to the first
disk, and, being unable to, marked it 'maintenance' . The next disk write
produced a 'last erred' error, and took the metadevice (Raid 5) offline.
I powered back up the d1000 and it came back to life.
My _guess_ is that the data is intact.
A metadb shows the following:
flags first blk block cnt
Wm pc l 16 1034 /dev/dsk/c3t4d0s0
W pc l 16 1034 /dev/dsk/c4t1d0s0
a pc luo 16 1034 /dev/dsk/c0t1d0s7
a pc luo 1050 1034 /dev/dsk/c0d0s7
a pc luo 2084 1034 /dev/dsk/c0t1d0s7
and a metastat shows:
c3t1d0s0 okay
c3t3d0s0 Maintenance
c3t4d0s0 okay
cdt5d0s0 okay
c4t0d0s0 okay
c4t1d0s0 okay
c4t3d0s0 last erred
c4t4d0s0 okay
QUESTION: While I have backups, they're older than I'd like them to be, and
I have reason to believe that the only thing wrong here is that the D1000
powered down, and the data on the array is good. WHAT is the best way to
attempt to recover? Is using the metainit -k command a safe way to
proceed? Is that the best way to proceed?
Will post all answers in summary.
TIA
-dave
_______________________________________________
sunmanagers mailing list
sunmanagers@sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers
This archive was generated by hypermail 2.1.7 : Wed Apr 09 2008 - 23:39:02 EDT