Corrupt Files In Storage Pool

Hi everyone,

We had a major hardware failure on one TSM server and I ended up needing
to restore the TSM database. Since then, I've been having issues with a
few of the disk storage pool volumes. When I try to back them up or
migrate them I get numerous errors like the following.

ANR1331E Invalid frame detected.  Expected magic 53454652 sequence number
00000001 server id
00000000 segment id 0000000467149962.
ANR1330E The server has detected possible corruption in an object being
restored or moved. The
actual values for the incorrect frame are: magic 53454652 hdr version 0002
hdr length  0032
sequence number 00000001 data length 0003FFE0 server id 00000000 segment
id 0000000473989839 crc
00000000.

I've been able to clear most of them out by issuing move data commands,
then deleting the volume altogether with "discard=yes". The data wasn't
copied to DRM before the server crash occured or I would have attempted to
restore them first. These storage pool volumes, and the TSM DB are on Raw
Volumes.  The TSM DB is on ESS disk, and the storage pool volumes are
local SCS JBOD. I copy them to DRM then migrate them to 0% utilized daily.


So, after first running a move data, I run an audit and it says it's
empty, but a delete volume discard=yes fails with the following in the
actlog.

ANR2017I Administrator 7760 issued command: DELETE VOLUME /dev/rlv_rawsp24
discard=yes
ANR2017I Administrator 7760 issued command: DELETE VOLUME /dev/rlv_rawsp24
discard=yes
ANR2406E DELETE VOLUME: Volume /dev/rlv_rawsp24 still contains data.


The move data, audit volume, delete volume routine worked for about 10
other raw storage pool volumes, but not these two or three. Anyone have
any tips?

TSM Server              5.2.2.3
Server OS               AIX 5.2ML4

Nothing interseting in the AIX errpt since the TSM restores.


Nothing interseting in the AIX errpt.

Curtis