cancel
Showing results for 
Search instead for 
Did you mean: 

freeze media when OPEN_ERROR encountered

deamon
Level 4
Partner
Hi,

Backep failed with an 83 after tape could not be opend to write data

20:04:26.785 [13469] <16> io_open: cannot open file /usr/openv/netbackup/db/media/tpreq/drive_HPUltrium3-SCSI0, Permission denied
20:04:26.786 [13469] <2> send_MDS_msg: DEVICE_STATUS 1 2235 unac1 000232 4000085 HPUltrium3-SCSI0 2000005 OPEN_ERROR 0 0
20:04:26.893 [13469] <2> log_media_error: successfully wrote to error file - 08/04/09 20:04:26 000232 2 OPEN_ERROR HPUltrium3-SCSI0
20:04:26.893 [13469] <2> check_error_history: just tpunmount: called from bptm line 23508, EXIT_Status = 83

Netbackup then tries to use the same tape for the next backup and fails in exactly the same way, all backups failed this overnight. The tape is in the scratch pool and there are other tapes available but it always try to use the same one and fails in the same way. (tape was not write protected)

The following setting are in place

MEDIA_ERROR_THRESHOLD = 1
TIME_WINDOW = 24
Does the MEDIA_ERROR_THRESHOLD parameter not include OPEN_ERROR as a valid error? 

What i want is for Netbackup to freeze the tape if this error is encountered

TIA
1 ACCEPTED SOLUTION

Accepted Solutions

Nicolai
Moderator
Moderator
Partner    VIP   

You probably have a defective media that  tape drive can't open. But since it first on the scratch list, it will be picked over and over again.

You can freeze the media or move it to the "None" pool.

Re-labeling will not help - LTO medias has reserved servo track and if they get damaged, you have no other choice than discarding the media.

Best Regards

Nicolai


View solution in original post

7 REPLIES 7

Marianne
Level 6
Partner    VIP    Accredited Certified
Which version of NBU?
If 6.x, have you verified that these settings exist in EMM database?
nbemmcmd -listsettings -machinename ....

deamon
Level 4
Partner
All servers are at 6.5.2, and nbemmcmd -listsettings reflect the settings as above,

It does not look like the MEDIA_ERROR_THRESHOLD takes account of when a OPEN_ERROR's, can someone confirm?

they are logged in the errors log
......
08/04/09 20:04:26 000232 2 OPEN_ERROR HPUltrium3-SCSI0
08/04/09 20:06:10 000232 2 OPEN_ERROR HPUltrium3-SCSI0
08/04/09 20:08:21 000232 2 OPEN_ERROR HPUltrium3-SCSI0
08/04/09 20:10:07 000232 2 OPEN_ERROR HPUltrium3-SCSI0
08/04/09 20:11:51 000232 2 OPEN_ERROR HPUltrium3-SCSI0
08/04/09 20:13:36 000232 2 OPEN_ERROR HPUltrium3-SCSI0
08/04/09 22:40:56 000232 2 OPEN_ERROR HPUltrium3-SCSI0
08/04/09 22:42:40 000232 2 OPEN_ERROR HPUltrium3-SCSI0
08/04/09 22:44:24 000232 2 OPEN_ERROR HPUltrium3-SCSI0
08/04/09 22:46:07 000232 2 OPEN_ERROR HPUltrium3-SCSI0
08/04/09 22:47:51 000232 2 OPEN_ERROR HPUltrium3-SCSI0
08/05/09 01:01:22 000232 2 OPEN_ERROR HPUltrium3-SCSI0
08/05/09 02:16:06 000232 2 OPEN_ERROR HPUltrium3-SCSI0
08/05/09 04:02:40 000232 2 OPEN_ERROR HPUltrium3-SCSI0
08/05/09 08:29:01 000232 2 OPEN_ERROR HPUltrium3-SCSI0

and for some reason NBU continues to try and use the same tape (scratch pool) when there are others available, causing every backup to fail, is this normal behaviour?

Marianne
Level 6
Partner    VIP    Accredited Certified
I agree with you - I would also expect it to freeze the media. My recommendation is to log a call with Symantec.
Send them nbemmcmd -listsettings as well as errors log.

zippy
Level 6
 

Nicolai
Moderator
Moderator
Partner    VIP   

You probably have a defective media that  tape drive can't open. But since it first on the scratch list, it will be picked over and over again.

You can freeze the media or move it to the "None" pool.

Re-labeling will not help - LTO medias has reserved servo track and if they get damaged, you have no other choice than discarding the media.

Best Regards

Nicolai


deamon
Level 4
Partner
when the tape was removed from library, the backups ran fine using other tapes

Nicolai
Moderator
Moderator
Partner    VIP   

Good you resolved the issue.  Have a great weekend.