I am trying to simulate retention for Veeam Agent Linux with this information:
So it's a “Forever Forward Incremental” retention mode. The problem is that there is no periodic full backup with this retention, and it seems that the calculator does it by default:
I find 1.3TB, whereas if I calculate manually:
1 full backup (300GB source, 50% compression) = 150GB
Is it possible to simulate this type of retention?
Best regards,
Thomas
Best answer by Michael Melter
Your estimation is storage but depending on the repository type there is also working space required as well which is what I believe the 1.3TB is giving you.
In the second screenshot, we can see that there is 0.16TB (160GB) of working space, so there is still 500GB too much in the calculation.
If I check the “ReFS/XFS” box, I get: 0.6TB (600GB), so that seems to match.
Does this mean that to simulate “Forever Forward Incremental,” I have to check the “ReFS/XFS” (fast clone) box?
Exactly. If you select “ReFS/XFS” you get synthetic fulls of only the size of increments (fast-cloning). This would consume exactly the same storage amount (at least in the primary chain) as a forever forward incremental chain with only increments.
On top it is still compatible with immutability. Therefore, this is my best recommendation anyways.
Working space is only a small addition compared to non-synthetic fulls without fast cloning.
Your estimation is storage but depending on the repository type there is also working space required as well which is what I believe the 1.3TB is giving you.
Your estimation is storage but depending on the repository type there is also working space required as well which is what I believe the 1.3TB is giving you.
In the second screenshot, we can see that there is 0.16TB (160GB) of working space, so there is still 500GB too much in the calculation.
If I check the “ReFS/XFS” box, I get: 0.6TB (600GB), so that seems to match.
Does this mean that to simulate “Forever Forward Incremental,” I have to check the “ReFS/XFS” (fast clone) box?
Your estimation is storage but depending on the repository type there is also working space required as well which is what I believe the 1.3TB is giving you.
In the second screenshot, we can see that there is 0.16TB (160GB) of working space, so there is still 500GB too much in the calculation.
If I check the “ReFS/XFS” box, I get: 0.6TB (600GB), so that seems to match.
Does this mean that to simulate “Forever Forward Incremental,” I have to check the “ReFS/XFS” (fast clone) box?
I cannot say for sure on that one but if your repo uses this type of file system you need to check it. If not then unsure why there is a mismatch. Hopefully someone from Veeam will chime in here.
Your estimation is storage but depending on the repository type there is also working space required as well which is what I believe the 1.3TB is giving you.
In the second screenshot, we can see that there is 0.16TB (160GB) of working space, so there is still 500GB too much in the calculation.
If I check the “ReFS/XFS” box, I get: 0.6TB (600GB), so that seems to match.
Does this mean that to simulate “Forever Forward Incremental,” I have to check the “ReFS/XFS” (fast clone) box?
Exactly. If you select “ReFS/XFS” you get synthetic fulls of only the size of increments (fast-cloning). This would consume exactly the same storage amount (at least in the primary chain) as a forever forward incremental chain with only increments.
On top it is still compatible with immutability. Therefore, this is my best recommendation anyways.
Working space is only a small addition compared to non-synthetic fulls without fast cloning.
