Weekly Full, Daily Incremental, 30 Restore Points

[intekular]

Hi,

Just want to make sure I have this set correctly before kicking off into production.

I currently am backing up to Exagrid units with plenty of space, and here is what I have set for my backup jobs-

No synthetics, forward incrementals on M, T, W, Th, and a true full every Friday. No backups on Sat or Sun. I have retention points set to 30.

My Veeam infrastructure is all virtual so I will be using an external app to backup to tape for the time being.

With this said, is the 30 retention points enough to keep a months worth of backups in place? At the end of each month I will be backing up the Exagrid CIFs share to tape with an external app.

Any suggestions or comments would be appreciated.

[Vitaliy S.]

Hello,

Yes, with this job and retention configuration you will have restore points for 1 month. Keep in mind though, that 30 restore points to keep is probably a bit more than you need. At the time when retention policy kicks in you will have 7 VBKs (full backups) and 23 VIBs (incremental files). You can save some space on the repository if you configure your jobs in the following way:

1. 20 restore points to keep (this will cover 4 business weeks in the month). With this retention policy setting, restore points (for the first week) will be deleted at the end of the 5th business week. This should be enough to offload all backups to external device.

2. If you have plenty space on the repository, you can set 25 restore points to keep. This will store backup chain for 5 business weeks.

P.S. not sure what the backup volume you're going to have, but I believe it might make sense to offload backup to tapes on weekly basis. Should take less time compared to transferring the entire backup chain for the the month.

Let me know of that helps! Thank you.

[chrisdearden]

if the OP is using Exagrid units , I would expect those full backups to deduplicate quite well. From a performance perspective if the Exagrid landing zone is large enough to hold the last full backup ( and associated increments ) then instant restores should be served form that faster disk zone.

[intekular]

Thanks Vitaliy and Chrisdearden. The Exagrids are sized in such a way that the landing zone is twice the data size with all of the VM disks fully provisioned and the included swap files etc even though this will be less when Veeam does the actual backup. I just wanted to be sure I was not whacking myself off at the knees.

As for the tape backups, weekly is not a bad idea, but I need to wait for the new year to roll around to upgrade the existing libraries so for now once a month will have to work. Would I be better served to do a backup copy job that consolidates things down to a new share on the Exagrid for the tape process?

Thanks,

Intekular

[Vitaliy S.]

If you need to offload the entire backup chain that will have restore points for each day, then backup copy jobs should be configured with "1 day" as a sync interval (copy every). This will basically re-create the same files on the Exagrid appliance, not sure if you need that or not (I wouldn't do it ), pretty much depends on what time will be required for the backup to tape jobs to transfer all VM data to tapes.

[veremin]

Would I be better served to do a backup copy job that consolidates things down to a new share on the Exagrid for the tape process?

If I were you, I wouldn't do that, as this approach will add unnecessary complexity for no obvious reason. Thanks.

[intekular]

Thanks for the input. I will stay away from the copy job.