Question about restore points

[b.russel]

I have a question about the restore points.

Lets assume a simple backup setup for a bunch of vm's:
1) keep a daily backup for 30 days
2) keep a monthly backup for 12 months

In the backup job I have the option of setting the restore points to keep (storage), minimum 1.
In the backup copy job I have the option of setting the restore points to keep (target), minimum 2.

Assuming that the backup job is running daily and the backup copy job is running pr 1 day, should I set the 30 daily backups in the backup job or in the backup copy job?
What is the difference between the 2 restore points to keep? If I set both to 30, won't I be storing the same(comparable) data twice?
Setting the monthly backup in the backup copy job is straight forward.


Now, assuming that I have a single vm with an hourly backup requirement, how would that be combined with the above schedule?
I have a comparable setup at a customers, where I've setup the following to be most space efficient.

1) Hourly backup job for 1 vm, keep 24 restore points
2) Daily backup job for rest of vm's (minus hourly vm), keep 1 restore point
3) 1 pr day backup copy job for 1) and 2), keep 30 points + monthly

The hourly backup vm is around 2.5 TB. Is this an optimal setup size wise?

[PTide]

Hi,

Assuming that the backup job is running daily and the backup copy job is running pr 1 day, should I set the 30 daily backups in the backup job or in the backup copy job? What is the difference between the 2 restore points to keep? If I set both to 30, won't I be storing the same(comparable) data twice?

Backup Copy job synthesizes a restore point from source backup job once in a predefined time period, that is if you configure daily backup job to keep 30 restore points and configure backup copy job to sync once a day then you'll get exactly the same backup chain on your secondary repo. The minimum of 2 restore points is due to how transform works while applying retention settings in backup copy job. Having additional restore point ensures at least one restore point remains valid in case of incremental backup failure.

3) 1 pr day backup copy job for 1) and 2), keep 30 points + monthly

That will give you a chain of 30 daily restore pointsfor each VM + 1 monthly restore point for each VM.

The hourly backup vm is around 2.5 TB. Is this an optimal setup size wise?

The overall size does not matter if you use incremental backup, just stay away from reverse incremental mode for that hourly VM as it is very I/O intensive.

Thank you.

[foggy]

Do both jobs use the same storage as target? Anyway, I'd recommend to have 30 points retention on the original backup job, to avoid possible issues (single restore point is not typically recommended, since you may fail to have a valid backup at all). The secondary backup copy job can then have 2 as regular retention and monthly backups enabled as required.

Backup copy job always copies the latest VM state, so, regardless of the number of restore points you have on the source repository, it will extract all the blocks from hourly backups required to build the VM state corresponding to the latest available restore point in the source chain. So yes, your setup looks good.