CopyJob with 70TB

[jochot]

Hey,

I'm onto copy jobs again, and I have a problem with understanding the intervals behavior.
I need to copy a backup file of ~70TB, the initial copy will create a full backup file, I got that.
Now if I set the copy interval to 1 day (which it should be, as the daily incrementals are max 20GB) it will definitely not finish in the first interval.

Quoting from here: https://helpcenter.veeam.com/docs/backu ... tml?ver=95
"When a new restore point appears on the source backup repository, Veeam Backup & Replication detects what data blocks still need to be copied to make up a full backup file on the target backup repository, and copies these data blocks.
This process continues until there is a full backup file on the target backup repository."

So from my understanding, it will try to create a fullbackup on day1, then the copy interval will end and it will start a new one, it will look for a new incremental (which it will find) and search for all the blocks it needs to create a full backup file.
Then it will try to get a full backup file for this interval.
Will this create a new file or will this be included in the "old" full backup file? Would it make sense to set the interval to 100days to let the first copy finish and then set it to 1day as soon as it is done? Does it matter at all?
What would be the best way to handle that kind of data?

Thanks in advance

[PTide]

Hi,

Will this create a new file or will this be included in the "old" full backup file?

It will create a new incremental backup file which will contain the missing blocks from .vbk + new blocks from the new incremental backup. That is, tpughly speaking, if your .vbk is supposed to be 10GB in size, but BCJ have managed to transfer only 8GB, then the new backup file will contain new blocks + those missing 2GB.

Would it make sense to set the interval to 100days to let the first copy finish and then set it to 1day as soon as it is done? Does it matter at all?

I would recommend to use mapping feature instead.

Thanks

[jochot]

So you say I should manually copy the data to the new location and then map the copy job to the manually copied backup?

[PTide]

Technically, you can increase copy interval so the copy job will finish transfer in the allotted period of time. However, since you've mentioned 100 days, I assume that the job does not make much progress so I suggested to resort to a more convenient way. How much data does the job manage to copy during the first sync?

Thanks

[jochot]

I haven't started the job yet, 100days would be what I set the job to so it can finish the first copy.
The question was if that is the right way to do it, or if there is a better way of copying the data over to a 2nd location.

[PTide]

Both ways are right. If your connection is unstable, copying data to the second location seems to be a more reliable way in your case.

Thanks