[Feature Request] - Synthetic Backup every X restore points

[davemann]

Hello,

Having run into and been pained by the 7 restore point limitation while using StoreOnce Catalyst stores, I think it would be prudent to have an option that would require a job to create a Synthetic Full on every x'th restore point, rather than using a daily/weekly schedule. This would prevent folks from experiencing the pain of corrupted backup chains on StoreOnce Catalyst if the number of restore points kept was set to a value greater than 7, and manual jobs were run during the week before the Synthetic full creation is run.

It's frustrating that Veeam documentation states that there is a limitation of 7 restore points using StoreOnce, but there is no safeguard in the application that stops you from trying to do otherwise, and corrupt your backup chain.

We need to be able to retain more than 7 days on our dedupe device, and need to ensure that our chains do not get corrupted.

I have a ticket into HP to get an answer as to when they may be releasing an updated OS version which deals with this issue, but I fear it will be quite some time.

Also - Having an option to manually invoke creation of a Synthetic Full at any time would be welcomed.

[foggy]

Could you please clarify what kind of corruption are you referring to? If, for some reason, the chain gets the higher number of restore points and synthetic full fails on reaching the device's open data sessions limit, all existing backups are left intact, it is just the new full restore point that is unusable (and that will be removed on the next run).

There're a couple of options for that, please review these two topics and there's also a ForceTransform registry value that being set to 1 will trigger synthetic full on every job run.

[Gostev]

This would prevent folks from experiencing the pain of corrupted backup chains on StoreOnce Catalyst if the number of restore points kept was set to a value greater than 7

This limitation most definitely does not result in "corrupted backup chains" as soon as you cross 7 restore points. Besides, 7 is a very "soft" limitation anyway, so don't worry about manual job runs. Moreover, 7 is for the lowest end StoreOnce offering only - you will be totally safe with much larger numbers on beefier models. Our next version, 9.5 will be determining the recommended number automatically by querying the given StoreOnce model's capabilities.

Anyway, the only potential issue with a very long incremental chain can be at restore while StoreOnce is busy processing a lot of backup or restore jobs at the same time, and has a lot of backup files open - so many that there are not enough slots to load a long incremental chain of backup files. The workaround is to simply stop all other jobs and try the restore again.