11-08-2011 02:53 AM
Master: 7.1.0.1
Clients: 7.1.0.1
Have SQL backups with following schedules:
Full: Once a Week - Windows sat 23 - sun 23
Diff: Once a day - Windows weekdays 22 - 04
Log: Once an hour - Windows weekdays 06 - 19
If for some reason the logfilebackup failes NBU doesn't start logfilebackup until next day after the next Diff is run
Maybee this by design but can't seem to figure out why?
Solved! Go to Solution.
11-08-2011 04:00 AM
I have seen problem since NBU 5.x days and honestly thought it would go away once the bpsched process in NBU 5 got replaced with the 3 new processes in post-6.x. Seems NOT....
The problem seems to be with the global setting of " # tries every ## hours" (default - 2 tries every 12 hours; yours probably 1 try every 24 hours?):
Scenario that I've seen in version 5.1:
Backup is scheduled to run every hour.
If a failure occurs on 1st and 2nd attempt, scheduler 'ignores' policy for 12 hours due to global setting of 2 tries per 12 hours.
After 12 hours, backup runs again every hour as per schedule.
The workaround that I've suggested back then:
Try a setting of 2 tries per 3 hours (although this will also cause 2 hourly attempts to be skipped after a failed backup...)
The other suggestion is to troubleshoot the failed backup and then manually kick it off. After a successful backup, the scheduler should submit it again hourly.
Same problem described back in 2003: http://adsm.org//lists/html/Veritas-bu/2003-12/msg00195.html
11-08-2011 04:00 AM
I have seen problem since NBU 5.x days and honestly thought it would go away once the bpsched process in NBU 5 got replaced with the 3 new processes in post-6.x. Seems NOT....
The problem seems to be with the global setting of " # tries every ## hours" (default - 2 tries every 12 hours; yours probably 1 try every 24 hours?):
Scenario that I've seen in version 5.1:
Backup is scheduled to run every hour.
If a failure occurs on 1st and 2nd attempt, scheduler 'ignores' policy for 12 hours due to global setting of 2 tries per 12 hours.
After 12 hours, backup runs again every hour as per schedule.
The workaround that I've suggested back then:
Try a setting of 2 tries per 3 hours (although this will also cause 2 hourly attempts to be skipped after a failed backup...)
The other suggestion is to troubleshoot the failed backup and then manually kick it off. After a successful backup, the scheduler should submit it again hourly.
Same problem described back in 2003: http://adsm.org//lists/html/Veritas-bu/2003-12/msg00195.html