In my production environment I am running NetBackup 6.0 MP6.

The NBU master server is running WIndows 2003 EE R2 and is on an HP SAN with
QLA fiber channel cards and no tape drive.

I also have another server which I will call server "x" that is running WIndows
2003 EE R2 and it is also on the SAN with QLA fiber channel cards and an
EML tape library with 4 LTO2 tape drives. This server acts as the media server
and performs all the read/write tape functions for NBU. The NBU catalog is
also backed up using this media server. This media server is running NBU
6.0 MP6.

I have also have a VMWare ESX box that is running a VMWare Virtual Machine
(Virtual clone of the Netbackup Master server for QA purposes). The VMWare
ESX box is also on the SAN with QLA fiber channel cards. The Virtual Machine
can not see the QLA fiber channel cards but has a SCSI attached tape drive
presented to it. This virtual machine is running WIndows 2003 EE R2 and NBU
6.0 MP6.

For sake of this discussion I will say that I perform a FULL Weekly backup
of the NBU catalog. I took this tape and tried to restore the database in
my QA Virtual Machine. I am successful during phase 1 but phase 2 fails.
Here is the log for phase 2 import:

----------------------------------------------------------------------

"D:\Program Files\VERITAS\NetBackup\bin\admincmd\bpimport.exe" -Bidfile "C:\Documents
and Settings\Administrator\Local Settings\Application Data\VERITAS\NetBackup\20080717104955.bid"
-L "D:\Program Files\VERITAS\NetBackup\Logs\NBimport\Import2-20080717104955.log"
-M csapuslic-003 -l

Import phase 2 started Thu Jul 17 2008 10:49:56
10:49:56 INF - Importing policy Catalog_Backup, schedule MONTHLY_FULL (csapuslic-003_1214661623),
media id VMB673, created on 06/28/2008 10:00:23.
10:50:03 INF - If Media id VMB673 is not in a robotic library administrative
interaction may be required to satisfy this mount request.
10:50:06 INF - Waiting for mount of media id VMB673 on server csapuslic-003
for reading.
10:50:11 INF - Waiting for positioning of media id VMB673 on server csapuslic-003
for reading.
10:50:11 INF - Beginning import on server csapuslic-003 of client csapuslic-003.
10:50:12 /VRTS_IMAGE_SIZE_RECORD
10:50:13 INF - Import of policy Catalog_Backup, schedule MONTHLY_FULL (csapuslic-003_1214661623)
failed, media read error.

10:50:13 INF - Importing policy Catalog_Backup, schedule MONTHLY_FULL (csapuslic-003_1214661734),
media id VMB673, created on 06/28/2008 10:02:14.
10:50:13 INF - Importing by reading INDEX file information only, full list
not available.
10:50:14 INF - If Media id VMB673 is not in a robotic library administrative
interaction may be required to satisfy this mount request.
10:50:56 INF - Waiting for mount of media id VMB673 on server csapuslic-003
for reading.
10:50:58 INF - Waiting for positioning of media id VMB673 on server csapuslic-003
for reading.
10:50:58 INF - Beginning import on server csapuslic-003 of client csapuslic-003.
10:50:59 INF - Import of policy Catalog_Backup, schedule MONTHLY_FULL (csapuslic-003_1214661734)
failed, media read error.

10:50:59 INF - Status = no images were successfully processed.

------------------------------------------------------------------------------------------

Has anyone else experienced or corrected this issue ?

I had read in another forum post that this could have something to do with
SCSI block size issues/differences but I am not sure where to look for this
information.
It gets even more confusing since my restore server is a Virtual Machine
on a VMWare ESX box....the ESX box presents the SCSI tape drive to the VIrtual
Machine...etc.