BackupExec 2012: Incremental scheduled to run during Full backup

[David_Grayston]

I'm wondering what is the behavior in BackupExec 2012 (BE) when a full backup is running and a scheduled incremental backup attempts to run?

In my protocol I do a full backup once a month (20 servers) and incremental backups daily. The full backups are scheduled for the first Monday of the month which then overlaps with the scheduled incremental backups. Previously I could exclude specific days from the incremental schedule to avoid this overlap but this scheduling option looks to be removed in BE2012.

Is there logic in BE2012 to cancel the incremental from running if a full backup is currently running even when (as would usually be the case) another full backup exists?

[Mandar_Sahasrab]

Hi David,

When Full and Incremental jobs are scheduled to run at the same time, the full supersedes the incremental and blocks the latter from running. Such an incremental is rescheduled to run on its next recurrence.

Prior to BE 2012, user had to explicitly configure the task relationships between full and incremental within a policy such as "Task A supersedes Task B when start times conflict" where A is full and B is incremental. In BE 2012 user have to configure such relationships. BE 2012 automatically sets these releationships between different tasks in single job family based on task type i.e. , full / incr / diff and the average scheduling frequency. If start times conflict,

1. Fulls always supersedes Incr / Diff.
2. Within a group of tasks of same type i.e. full / incr / diff, task with longer scheduling gap (low frequency) supersedes task with shorter scheduling gap (high frequency). E.g. a full task with schedule 'First Monday of every month' supersedes a full task with schedule 'Monday of every week'.

Regards,
Mandar

[David_Grayston]

Thanks for the great information.

Let me ask though since in my senario the start times aren't the same as in the senario you fully describe. For us its that the full backups take 10-14 hours to complete so they are currently running or queued to be run (waiting for available device) when scheduled incrementals overlaps.

I'm thinking the desirable behavior will still be as you describe but if you or someone else could comment it would be appreciated.

David