03-25-2013 04:28 PM
I had an issue where a selection list was changed and almost 1TB of data which was not supposed to be backed up was included in backups of this file server for a couple of weeks. Caught this after noticing the the B2D volume was almost out of space and the duplicate to tape jobs were using much more space than a normal increase in data would account for. As I found this at the time weekly full backups were being run, and the volume was almost out of space, I put all jobs on hold and removed the B2D media that this job had used (moved media to scratch then deleted from backup exec and disk).
The following week the duplicate job would not run as it was likely still trying to duplicate the previous weeks backup set which no longer exists on disk, but must still exist in the catalogs. I've created one-off duplicate jobs to get the data to tape for the last couple of weeks, but I need a permenant solution to get the duplicate job running again.
I've tried deleting and recreating the jobs associated with the policy but that hasn't resolved the issue.
Is there a way to purge this single backup set from the catalogs? Would an inventory of the B2D media accomplish this?
My next step was to delete the policy and backup selection list, then recreate them and the jobs. Would these jobs then ignore backup sets created with the older backup selection list and policy?
Here are the errors from the duplicate jobs
Last week, after deleting the B2D media:
Unable to determine ADAMM media identification. Catalog query failed.
Job Completion Status
Job ended: Monday, March 18, 2013 at 3:17:21 PM
Completed status: Failed
Final error: 0xe0009444 - The requested source duplicate backup sets catalog record could not be found. Perhaps the media containing the source backup sets was previously deleted.
Final error category: Job Errors
For additional information regarding this error refer to link V-79-57344-37956
Errors
Click an error below to locate it in the job log
Unable to determine ADAMM media identification. Catalog query failed.
This week, after recreating the jobs using the original policy and selection list:
Job type: Duplicate
There are no selections for this job.
Job Completion Status
Job ended: Monday, March 25, 2013 at 1:19:40 PM
Completed status: Failed
Final error: 0xe00081d2 - The script associated with this job contained no valid selections.
Final error category: Job Errors
For additional information regarding this error refer to link V-79-57344-33234
Errors
Click an error below to locate it in the job log
There are no selections for this job.
03-25-2013 09:37 PM
Deleting the parent job and recreating that and the duplicate job should fix things. If you know the catalog used for that particular job/s where the incorrect data was backed up, you could try deleting those.
Alternatively, because this information will sit in the BEDB, if you have support with Symantec, contact them to edit it out of the BEDB. They'd have the knowledge of the table structure to do so.
Thanks!
03-26-2013 02:02 AM
As you have posted against BE 2010
Right click the failed job history the last time the job ran and select the Edit Selection List option. - Then use the details tab to view the selections as a list of text entries and remove any catlog lines that are invalid.
Then save those changes and retry that job immediately (right click the failed job history again.
Sorry I don't know the exact on screen text for some of what I mentioend above as it is 2010 (and earlier) specific and you can't do it in 2012 (we are aware this might cause issues)
03-26-2013 06:28 PM
It may be easier to create a new policy with the same settings as the existing one and then generate a new set of jobs.
03-27-2013 07:59 AM
03-27-2013 08:01 AM
09-16-2013 11:44 PM
youhou
ca c'est de la reponse.
supprimer pour recrée des job de sauvegarde. Non mais serieux!!!!
09-17-2013 12:06 AM
We are planning a change to the behaviour (probably in the next complete Backup Exec release) that will remove any selections in duplicate jobs that no longer exist automatcially. At this current time however exactly when this may appear in the prodct remains subject to change and therefore you either have to:
BE 2010 R3: edit the selections (right click on failed job history) once the duplicate has failed and delete the text based / details view of the selections that reference the invalid catalogs and then re-run the duplicate with the remaining valid selectiosn in it. (again right click job history and use retry job nowe option)
BE 2012: delete the duplicate stage from the job definition, save the job without the stage, then edit the job and recreate the duplicate stage. Run the origfinal backup job to restart this sequence , starting with the full if you have incrementals or differentials involved.