File copy question

[sweepsd]

Hi
I'm trying to get a backup copy of our exchange server to a remote site (2.5TB). I'm struggling to find the best way to get it there. I have created a local backup copy here and using file copy with a schedule to send it to the remote site so the normal backup isn't interfered with and then re-point the backup copy to it. But I'm not sure if I going about it the right way. I had got the file copy running continuously but with non permitted windows (so that the other backup copies are sent through) but not sure if it will continue when the permitted period starts again or just start a fresh.

TIA

[wishr]

Hi Sweepsd,

Could you please let us know why the use of backup copy for backup data transfer to the remote location is not acceptable in your case?

Thanks

[sweepsd]

if it is still copying the data using backup copy wont the normal backup not work when it is time to run?

[Shestakov]

Backup copy job doesn't mess primary backup jobs, so backup copy is indeed the recommended way in your case.
Since you have large backups, I would also consider leveraging backup seeding.

[sweepsd]

That's exactly what I am doing - I have created a local backup copy and using filecopy to copy the files.

[Shestakov]

Using file copy to seed backups for backup copy job doesn't seem as a great idea since the backup copy job transfers data with the same speed or even faster if you leverage WAN acceleration.
Seeding is used when you either have faster channel for data transmission or move the seed to secondary site on the physical drive.
Thanks

[sweepsd]

To create a seed for the primary backup copy job, do the following:

Create a backup copy job. Add machines whose restore points you want to copy to this backup copy job. Target the backup copy job to some backup repository on the source side. This backup repository will be used as an intermediate one.

Run the backup copy job to create a full backup file (VBK) in the intermediate backup repository.
Transfer the created VBK file and VBM file from the intermediate backup repository to the target backup repository. - I assumed I could use File copy for this?

[wishr]

Correct, but what's the point of doing that? You'll get the same as when you simply perform a backup copy to the remote site, but it may be even slower.

If there is no possibility to transfer VBK and VBMs to the remote site physically/over a faster network you may want to use the backup copy alone and accept the data transmission times.

[Shestakov]

You can use File Copy, but it doesn't work faster than Backup Copy itself. The idea of seeding is to move the full backup or chain to target repository in alternative faster way, than backup copy itself.

I had got the file copy running continuously but with non permitted windows (so that the other backup copies are sent through) but not sure if it will continue when the permitted period starts again or just start a fresh

By the way, what do you mean by the permitted period?
Thanks!

[sweepsd]

the permitted time during the schedule

[Shestakov]

File copy job doesn't have the backup window option. Could you specify where did you set the option? Or you mean backup window of backup copy job?
Thanks

[sweepsd]

So at the moment my backup for this job is merging the incremental into the full, what will happen if I create a backup copy that is going to take longer to replicate to the remote site when the next incremental backup \ merge into the full happens?

[wishr]

A backup job always has higher priority than a backup copy job. Thus, if during a backup copy run it's time to perform a merge process, the priority will be given to the merge itself, and the backup copy will be paused in order to release the backup files for the backup job operations. Once the merge has been done, the backup copy will resume and start processing the files it used to process before the interruption. Note, that changes made during the merge process will be only copied during the next interval, but only if a backup copy has already started processing the corresponding backup files before the pause caused by the merge process.

Hope it helps!