Skip to main content
Solved

Need Help/Advice for setting RPO correctly

  • November 7, 2024
  • 4 replies
  • 38 views

Forum|alt.badge.img+6

Hi,

 

im still trying to get the RPO settings right for the following scenario (we are currently on V 11.32.69):

 

Incremental Backup Mo-Fr 09:00 PM

Transaction for Database every 4h

Full Backup of Databases every Day 09:00 PM

Full Backup (of all) Saturday 09:00 PM (for Tape export)

 

We have some flexability with the start time, but for the first try we started with the rule set above:

The outcome was mixed, the SQL Backups all had some Jobs with an “Failed to start” Error, also some of the Job there startet as Differential (the SQL option to incr. Backup) for other its started the Full Job, depending what job was “faster” to start.

 

Isn’t there some intelligence that priorities the “higher value” Backup Job (Full > Diff > Incr.)?

 

I could manually move the Database Fulls a few minutes eralier, that would at least make sure, the Full would be ran and the Incr. would “Fail to start” but is there a way, i can get rid of attempt to start the job at all?

In the RPO you can select “all Agent” and “Database Agent”, but not “all Agent expect Database”. But without this option, it seems we are forced to create an additional plan just for Databases.

But our environment has a few satilite offices, so instead of adding just 1 plan we are talking about a dozen more.

 

So i would like to hear your ideas and imput.

 

Thank you for reading the wall of Text and have a good Day

Joerg

Best answer by Onno van den Berg

You could also consider creating a plan using the solution filter and as Mohammed already pointed out staggering your jobs is better. Maybe consider trying out by using SLA-based scheduling combined with a backup & blackout window and a solutions filter. I b.t.w. wouldn't create traditional schedule policy via the legacy console any longer. If you have plans to move to FR34 or FR36 you will find more fine-grained filtering possibilities and scheduling options in the protection plan aks server plan config. 

View original
Did this answer your question?

4 replies

Mohamed Ramadan
Forum|alt.badge.img+11

hello Joerg

1st It seems that all of your backup jobs are currently scheduled at the same time (9:00 PM) i dont know why Joerg ?! I would recommend optimizing the job schedules by staggering their start times

2nd regarding the “Failed to Start” jobs, you can set the jobs to be queued rather than failing immediately. This way, jobs will remain in a queued state in the Job Controller and will automatically resume once the condition causing the queue clears. You can find more information on job queuing here:
https://documentation.commvault.com/11.20/queuing_jobs.html

3rd if you have access to the CommCell Console, you can create a new schedule policy specifically for database backups. This will help ensure that the backup jobs for your databases are handled according to your requirements with out need for create new SP.
 

I’m happy to help. Wishing you a great day as well !

Best Regards,
Mohamed Ramadan
Data Protection Specialist


sbhatia
Vaulter
Forum|alt.badge.img+4
  • Vaulter
  • 22 replies
  • November 10, 2024

backing up on what Ramadan mentioned, if you are okay to stagger the schedule, that would be a viable option, if not, then the only way forward is that you avoid “Failed to start “ by queuing the jobs :https://documentation.commvault.com/2023e/expert/queuing_jobs.html


Onno van den Berg
Commvault Certified Expert
Forum|alt.badge.img+19
  • Commvault Certified Expert
  • 1228 replies
  • Answer
  • November 10, 2024

You could also consider creating a plan using the solution filter and as Mohammed already pointed out staggering your jobs is better. Maybe consider trying out by using SLA-based scheduling combined with a backup & blackout window and a solutions filter. I b.t.w. wouldn't create traditional schedule policy via the legacy console any longer. If you have plans to move to FR34 or FR36 you will find more fine-grained filtering possibilities and scheduling options in the protection plan aks server plan config. 


Forum|alt.badge.img+6
  • Author
  • Byte
  • 20 replies
  • November 11, 2024

Thank you all for your inputs.

Active Job Queue would solve the issue with the “fail to start” but it would also run “unneccesary jobs”, we there hope that Commvault would see “at 9:00 PM there should run a Full and a Incr. Job” Conclusion = Run the Full Job, because it contains all the Data that a Incr. Job would cover anyway.

Im with Onno, that we don’t want to modify the existing Server Plans via the Java Console, that sounds like a maintaining hazard in the making.

After looking at the current data we can see, that the Databases in the satilite offices are not that big, so we will collect the data in a central server plan. for the 2 DBs that are bigger, we will create a sepparte Server Plan and also make an alert in the monitoring system, if any other DB will grow.

We are currently planing to switch to 11.36.x in the spring and we will take another look at the options we will get with the new version.

Once more, thank you for your Infos and Opinions,

Have a great week

Joerg


Reply


Cookie policy

We use cookies to enhance and personalize your experience. If you accept you agree to our full cookie policy. Learn more about our cookies.

 
Cookie settings