05.12
So I ran into an issue today where I had a backup server using Backup Exec 12.5 where the backup job would just do nothing but sit in queue. The IBM Ultrium TD4 drive was connected to NetApp via SCSI on adapter 1a. I shut off the IBM Tape drive off and restarted the backup server. Once booted back up, the job was still there. Backup Exec still thought it had a tape in the IBM. So after pulling my hair out I jumped on the Filer to do some troubleshooting.
Ran command:
filer01> sysconfig -t
Tape drive (1a.2) IBM ULTRIUM-TD4
rst0l - rewind device, format is: LTO 2 ro 400GB cmp
nrst0l - no rewind device, format is: LTO 2 ro 400GB cmp
urst0l - unload/reload device, format is: LTO 2 ro 400GB cmp
rst0m - rewind device, format is: LTO 3 800GB cmp
nrst0m - no rewind device, format is: LTO 3 800GB cmp
urst0m - unload/reload device, format is: LTO 3 800GB cmp
rst0h - rewind device, format is: LTO 4 800GB
nrst0h - no rewind device, format is: LTO 4 800GB
urst0h - unload/reload device, format is: LTO 4 800GB
rst0a - rewind device, format is: LTO 4 1600GB cmp
nrst0a - no rewind device, format is: LTO 4 1600GB cmp
urst0a - unload/reload device, format is: LTO 4 1600GB cmp
Well NetApp is talking to my Tape Drive. Next I am going to see if I can bounce the SCSI port.
filer01> storage disable adapter 1a
Adapter 1a is part of a parallel SCSI bus dual-channel
adapter so adapter 1b will also be disabled.
Tape drive 1a.2 IBM ULTRIUM-TD4 is in use.
Adapter ports 1a and 1b cannot be disabled
Well it looks like the tape drive is in use, but its not, its actually turned off.
This command will try to force to disable the adapter.
filer01> storage disable -f adapter 1a
Adapter 1a is part of a parallel SCSI bus dual-channel
adapter so adapter 1b will also be disabled.
Tape drive 1a.2 IBM ULTRIUM-TD4 is in use.
Adapter ports 1a and 1b cannot be disabled
Same error, so I check to see if any backups were occurning.
filer01> backup status
No backup activity.
So then I check the ndmpd status
filer01> ndmpd status
ndmpd ON.
Session: 200
Active
version: 4
Operating on behalf of primary host.
tape device: nrst0a
mover state: Idle
data state: Idle
data operation: None
Bingo, it looks like the ndmpd still thought it was backing up to the tape drive. So the next command we will kill that session.
filer01> ndmpd kill 200
filer01> ndmpd status
ndmpd ON.
No ndmpd sessions active.
Now lets try to disable the adapter again.
filer01> storage disable adapter 1a
Adapter 1a is part of a parallel SCSI bus dual-channel
adapter so adapter 1b will also be disabled.
Disabling adapter ports 1a and 1b…
Host adapter 1a and 1b disable succeeded
Please follow the instructions described in the man page to add or
remove tape drives or medium changers from the SCSI bus.
BINGO! We have finally disabled the adapter.
Now lets re-enable the adapter.
filer01> storage enable adapter 1a
[filer01: scsiAdapter.onlining:info]: Bringing SCSI host adapter 1a online.
Wed May 12 18:27:34 EDT [cefs02: scsiAdapter.onlining:info]: Bringing SCSI host adapter 1b online.
Wed May 12 18:27:43 EDT [cefs02: scsiAdapter.online:info]: SCSI adapter 1b is now online.
Wed May 12 18:27:43 EDT [cefs02: scsiAdapter.online:info]: SCSI adapter 1a is now online.
After that I went to my backup server and was able to restart the Backup Exec services and BOOM! My backup jobs were running successful. I hope this was some what informal.
No Comment.
Add Your Comment