Problems w/ Backup to Removable Drives

[trbyrne]

I am running Veritas BUE v10.0 on a MS Server 2003 server and I am experiencing problems backing up to a removable drive. I have a single removable drive bay connected to my server via USB. I cycle out four individual removable drives (take one out and replace it with another one) on a weekly basis (each Friday I change the drive).

This process, when it works, usually lasts about 1.5 hours and backs up about 31GB. The last time it worked successfully was on 1/17. So far I've seen this pattern:

1/17 (Thu) - success - 31GB backed up
1/18 (Fri post drive swap) - failed - 30GB backed up - ERROR: "media mount failed" (also reached four hour time limit)
1/18-1/24 - failed - *no data* - max runtime of four hours reached each night
1/25 (Fri post drive swap) - failed - 8.6GB backed up - ERROR: "media mount failed" (also reached four hour time limit)
1/26 - 1/31 - failed - *no data* - max runtime of four hours reached each night
2/1 (fri post drive swap) - failed - 4.2GB backed up - ERROR: "media mount failed" (also reached four hour time limit)


Apparently each time we swap the removable drive, it attempts to copy some data, and then after reaching the maximum run time errors out with this error text in the log file (this same error message appears each time it has failed since 1/17):
Backup- D: ASCMedia mount failed.
User canceled a Physical Volume Library operation.
V-79-57344-33861 - The media operation was terminated by the user.

When I do a lookup on error code "V-79-57344-33861" (according to

http://support.veritas.com/docs/274516)

it essentially says that the maximum time limit has been reached:
The error occurs because the operation has been canceled manually or a time limit has been set for the backup and that limit has been exceeded. For example, if Enable Automatic Cancellation was selected in the Schedule Options of the Backup Job Properties, the job will be canceled if the run time exceeds the amount of time specified.


I *really* hope someone can help me with this one. I'm losing hair (and sleep) over this.

 

[LawnBoy]

IMO, the media mount fail is the actual problem. It times out waiting for the mount.

You need to look at the job logs and try to determine exactly where the job is failing. The "no data" failures make me suspect that the problem is not with the USB drive, but with the location of the files you're trying to back up. Could be a problem with the snapshot provider you're using. Are you backing up from any remote servers?


"We must fall back upon the old axiom that when all other contingencies fail, whatever remains, however improbable, must be the truth." - Sherlock Holmes

 

[trbyrne]

I have separate backups that backup my files to folders on my hard drive. Then the backup in question backs these backup files to the removable hard drive.

The job logs really don't shed any light on the subject. The ones for the Friday jobs (that partially backup data) only indicate that the time window was exceeded and that they job was canceled. The job logs where no data was backed up also indicate that the time limit was reached. I've copied two logs below for comparison.


==============================================================
==============================================================
JOB LOG FOR PARTIAL BACKUP

Job Log for BACKUP_2_REMOVABLE


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


Completed status: Canceled, timed out Expand AllCollapse All
Job Information
Job server: DBSERVER
Job name: BACKUP_2_REMOVABLE
Job started: Friday, February 01, 2008 at 11:15:03 PM
Job type: Backup
Job Log: BEX_DBSERVER_02245.xml



Drive and media mount requested: 2/1/2008 11:15:03 PM

Device and Media Information
Drive and media information from media mount: 2/1/2008 11:15:05 PM
Drive Name: REMOVABLE DRIVE
Media Label: B2D000035
Media GUID: {673d8c00-558a-4fb2-9948-058dd7a5d11d}
Overwrite Protected Until: 2/2/2008 12:15:05 AM
Appendable Until: 12/30/9999 6:00:00 PM
Targeted Media Set Name: Bromberg's Default Media Set




Job Operation - Backup
Backup Options
Media operation - overwrite.
Hardware compression enabled.





Server - DBSERVER

Set Information - D: ASC
Backup Set Information
Family Name: "Media created 2/1/2008 11:15:03 PM"
Backup of "D: ASC"
Backup set #1 on storage media #1
Backup set description: "BACKUP_2_REMOVABLE"
Backup Type: Full - Back Up Files - Reset Archive Bit




Backup started on 2/1/2008 at 11:15:10 PM.

Backup Set Detail Information
Drive and media mount requested: 2/1/2008 11:20:29 PM
Media mount failed.
User canceled a Physical Volume Library operation.

V-79-57344-33861 - The media operation was terminated by the user.



Backup completed on 2/2/2008 at 3:15:17 AM.

Backup Set Summary
Backed up 1 file in 2 directories.
Processed 4,295,290,214 bytes in 5 minutes and 20 seconds.
Throughput rate: 768 MB/min










Job Completion Status
Job ended: Saturday, February 02, 2008 at 3:15:17 AM
Completed status: Canceled, timed out
The job was canceled by user System.



Errors
Click an error below to locate it in the job log
Backup- D: ASCMedia mount failed.
User canceled a Physical Volume Library operation.
V-79-57344-33861 - The media operation was terminated by the user.





============================================================
============================================================
JOB LOG FOR "NO DATA" BACKUP

Job Log for BACKUP_2_REMOVABLE


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


Completed status: Canceled, timed out Expand AllCollapse All
Job Information
Job server: DBSERVER
Job name: BACKUP_2_REMOVABLE
Job started: Saturday, February 02, 2008 at 11:15:04 PM
Job type: Backup
Job Log: BEX_DBSERVER_02250.xml



Drive and media mount requested: 2/2/2008 11:15:04 PM

Job Completion Status
Job ended: Sunday, February 03, 2008 at 3:15:09 AM
Completed status: Canceled, timed out
The job was canceled by user System.

 

[LawnBoy]

Backup of "D: ASC"

Looks like your selection list is corrupted. There ain't no sech thang as "D: ASC" , there should be a backslash after the colon.

I would completely recreate the jobs. If you're using templates, recreate them as well.




"We must fall back upon the old axiom that when all other contingencies fail, whatever remains, however improbable, must be the truth." - Sherlock Holmes

 

[trbyrne]

The "D: ASC" has always been there, even when it works. The selection criteria is actually the following folders:
"d:\backup_asc_critical"
"d:\backup_data"
"d:\backup_system_states"

I swear that I did not change anything and it worked on 2/8, 2/9 and on 2/10. Backed up 33GB each night in just under 2 hours. The drives were swapped out on 2/8. Do you think that has something to do with it? Perhaps (maybe) that last removable drive was bad? But if it was bad, wouldn't it have prevented me from doing an inventory on the device (which I was able to do)?

Do I need to post the last successful job log (the one that was run on 2/10)?

 

[DTracy]

You should be using duplication rather than backup when going from C: to D: I tried this once and found that I could not restore individual files from the backup of a backup, only the entire B2D folder.

Also, you may need to try a different drive letter and backup job for each external drive. This can be configured through windows Administrative tools, disk manager. That way each job is pre-configured to a drive letter that is already assigned to an external drive, and tested.

 

[willjwill]

I had a similar problem using USB disks. We encountered problems with drive letters but also corrupt data. Apparently the data wasn't writing to the disk properly. We also had slow backup speeds like you are encountering.

Take a look at this product:

http://www.idealstor.com/teralyte.html

We use it in one of our labs where there is only a single server and let our lab techs manage the disks themselves. It should solve most of the issues you are dealing with.