We help IT Professionals succeed at work.

Solaris: tape drive error "LUN communication failure"

Williams225
Williams225 asked
on
Hello experts,

 please I am having the  error  bellow everytime i try to do a backup, the os is solaris 10

bash-3.00# exit
  DUMP: SIGTERM()  try rewriting
  DUMP: The ENTIRE dump is aborted.
  DUMP: SIGTERM()  try rewriting
  DUMP: Rewriting attempted as response to unknown signal.
  DUMP: Rewriting attempted as response to unknown signal.
Nov  3 15:17:41 ACHQSDP01a scsi: WARNING: /pci@0/pci@0/pci@8/pci@0/pci@8/pci@0/scsi@8/st@0,0 (st0):
Nov  3 15:17:41 ACHQa      Error for Command: release                 Error Level: Fatal
you have mail
# Nov  3 15:17:41 ACHQa scsi:      Requested Block: 0                         Error Block: 0
Nov  3 15:17:41 ACHQa scsi:        Vendor: HP                                 Serial Number:    9   $DR-1
Nov  3 15:17:41 ACHQa scsi:        Sense Key: Aborted Command
Nov  3 15:17:41 ACHQa scsi:        ASC: 0x8 (LUN communication failure), ASCQ: 0x0, FRU: 0x0

Open in new window

Comment
Watch Question

Brian UtterbackPrinciple Software Engineer
CERTIFIED EXPERT

Commented:
Is the dump hanging? From the data above, it looks like the dump was manually aborted with the kill command.
Williams225System Administator

Author

Commented:
it wasnt manually aborted. The dump stops each time after a few secondes
Brian UtterbackPrinciple Software Engineer
CERTIFIED EXPERT

Commented:
In the output above, the first line is "exit". Are you putting the dump into the background and then exiting? That might cause the process group to get sent a TERM signal. Try putting the command "nohup" as the first item on the command line you use. That is, if the command were

dump xyx  &

try this instead

nohup dump xyz &
Williams225System Administator

Author

Commented:
actually,

the unix engineer told me that he is not able to do backups (with backup scripts).  So I have tried to do a backup test


ufsdump 0ucf /dev/rmt/0 /export/home
  DUMP: Date of this level 0 dump: November  3, 2011  3:16:35 PM GMT
  DUMP: Date of last level 0 dump: the epoch
  DUMP: Dumping /dev/md/rdsk/d30 (ACHQSDP01a:/) to /dev/rmt/0.
  DUMP: Mapping (Pass I) [regular files]
  DUMP: Mapping (Pass II) [directories]
  DUMP: Writing 63 Kilobyte records
  DUMP: Estimated 7652 blocks (3.74MB).
  DUMP: Dumping (Pass III) [directories]
Nov  3 15:16:50 ACHQa scsi: WARNING: /pci@0/pci@0/pci@8/pci@0/pci@8/pci@0/scsi@8 (mpt1):
Nov  3 15:16:50 ACHQa      Target 0 reverting to async. mode
Nov  3 15:16:50 ACGQa scsi: WARNING: /pci@0/pci@0/pci@8/pci@0/pci@8/pci@0/scsi@8/st@0,0 (st0):
Nov  3 15:16:50 ACHQa      Error for Command: write                   Error Level: Fatal
Nov  3 15:16:50 ACHQa scsi:        Requested Block: 0                         Error Block: 0
  DUMP: Write error 0 feet into tape 1
Nov  3 15:16:50 ACHQa scsi:        Vendor: HP                                 Serial Number:    9   $DR-1
  DUMP: NEEDS ATTENTION: Do you want to restart?: ("yes" or "no") Nov  3 15:16:50 ACHQSDP01a scsi:      Sense Key: Aborted Command
Nov  3 15:16:50 ACHQa scsi:        ASC: 0x47 (data phase crc error detected), ASCQ: 0x1, FRU: 0x0





bash-3.00# exit
  DUMP: SIGTERM()  try rewriting
  DUMP: The ENTIRE dump is aborted.
  DUMP: SIGTERM()  try rewriting
  DUMP: Rewriting attempted as response to unknown signal.
  DUMP: Rewriting attempted as response to unknown signal.
Nov  3 15:17:41 ACHQa scsi: WARNING: /pci@0/pci@0/pci@8/pci@0/pci@8/pci@0/scsi@8/st@0,0 (st0):
Nov  3 15:17:41 ACHQa      Error for Command: release                 Error Level: Fatal
you have mail
# Nov  3 15:17:41 ACHQa scsi:      Requested Block: 0                         Error Block: 0
Nov  3 15:17:41 ACHQa scsi:        Vendor: HP                                 Serial Number:    9   $DR-1
Nov  3 15:17:41 ACHQa scsi:        Sense Key: Aborted Command
Nov  3 15:17:41 ACHQa scsi:        ASC: 0x8 (LUN communication failure), ASCQ: 0x0, FRU: 0x0

Open in new window

CERTIFIED EXPERT
Top Expert 2007

Commented:
Please verify that you did not change the tape drive recently? it could be that you have wrong scsi id

Also, are you using the right tape media / density?

can you show the command you used?
Williams225System Administator

Author

Commented:
i didn't change the tape drive recently. I am using the right media.

The command used is
ufsdump 0ucf /dev/rmt/0 /export/home

Open in new window

, its a command to do a backup test
CERTIFIED EXPERT
Top Expert 2007

Commented:
why do you want to use the c option?

can you use the mt command to test if you can move forward / backward the media in the tape drive?
Williams225System Administator

Author

Commented:
<i have tried it, I am able to rewind and move forward
CERTIFIED EXPERT
Top Expert 2007

Commented:
what about the c option?

Do you have other media to test with?
CERTIFIED EXPERT
Top Expert 2007

Commented:
if the tape media already have previous backups then can you try listing them?
Williams225System Administator

Author

Commented:
y es i can list the previous backups
CERTIFIED EXPERT
Top Expert 2007

Commented:
is it by any chance that the tape is full?
Williams225System Administator

Author

Commented:
tape was empty before I did this test
CERTIFIED EXPERT
Top Expert 2007
Commented:
you are able to:

use mt command with the tape drive and media which shows that tape dev name is ok

list tape content which means you have no h/w problem

Now, either the new tape (the empty one) is faulty or have wrong density

can you try to use other backup command like

tar cvf /dev/rmt/0 /export/home

and see if  it succeed or fail
Williams225System Administator

Author

Commented:
tar cvf /dev/rmt/0 /export/home
a /export/home/ 0 tape blocks
a /export/home/user/ 0 tape blocks
a /export/home/user/.profile 1 tape blocks
a /export/home/user/local.cshrc 1 tape blocks
a /export/home/user/local.login 1 tape blocks
a /export/home/user/local.profile 1 tape blocks
a /export/home/user/.hushlogin 0 tape blocks
a /export/home/user/.ssh/ 0 tape blocks
a /export/home/user/.ssh/authorized_keys 4 tape blocks
a /export/home/user/.ssh/id_dsa 2 tape blocks
tar: write error: I/O error

Open in new window

System Administator
Commented:
Its a new tape, it seems that the tape drive is faulty
CERTIFIED EXPERT
Top Expert 2007

Commented:
This is possible, try to clean it
Williams225System Administator

Author

Commented:
i have cleaned it many times
Williams225System Administator

Author

Commented:
We had to replace the tape drive