Backup copy - not running on our main backup Veeam server

[Rich.Philp]

Hi All,
Can I setup the copy job on the backup repository as opposed to the main server doing the backup jobs?

Background:
I have just setup a backup copy job on our "main" Veeam backup server. This is coping data from our backup repository server to a USB drive connected to the backup repository (same) server, not to our Veeam server.

The data we are coping across/off is around the 1.4Tb range, and it isn't completing the copy off over night. There is about a 10 to 12 hour slot between the server we are copying off, last backup job and its first for the day.

What might be causing this, is the backup repository is having all the other Vmware machines having full backups down to it over night as well. This could cause it to slow down as the drive is being blasted. The full backup jobs are staggered.

This so the files come off straight from the server not over the network. Not sure if this will fix my problem, but want to give it a try if it is possible.

Thank you,
Rich

[foggy]

Rich, what are the bottleneck stats for this backup copy job? This could be either due to the source storage being overloaded with tasks or due to USB drive being not capable of providing enough IOPS for a very I/O intensive transform operation.

[Rich.Philp]

Hi Foggy,
Generally the source. Which would point to the repository server drive being overloaded.
What I will try next week, is to separate out this particular backup onto a separate raid and see if that increases performance. I will the job running on my Veeam backup server.

Thanks Foggy. I will come back to this next week once I have tried this, just to update.
Rich

[Rich.Philp]

Got going in the end with no changes to infrasture. Im not sure exactly what changed, but Im getting 1.5 TB over to the ext drive in around 6 hours. Also have changed 2 other servers to reverse incremental and Backup copy to the other USB drive. these also seem to work well.

Seeing an improvement already, and expecting a lot more once I have all jobs migrated across.

One question though:
I have the 2 registry keys for rotated backups, setup on my main Veeam server. under both the 64 area and "normal" area. Problem I have, is that the backups don't seem to be deleting old backups. Looking at the backup copy job, the only retention period I can set is 2 points. I can't fit 2 full backups on a 3 TB drive.

Backup is a reverse incremental, so it has one full backup file.

Is there anyway of making this 1? So the only backup, is the most recent full on the ext drive? I don't want to use a script to do this, would love for Veeam to have a solution.
Any ideas?

Thank you,
Rich

[foggy]

Although minimum retention for the backup copy job is 2, the second restore point is always incremental, so you would never have 2 fulls there. Check the second registry key that manages the old backups deletion (so that it does not have any extra spaces, for example). If everything looks to be set up correctly but backups are not deleted prior creating a new one, please contact support for investigation.

[Rich.Philp]

Thanks Foggy. I'll keep an eye on it over the next week. Will come back if I find it isn't working, also log a support ticket.

[Rich.Philp]

All going very well now, without any changes. Might take a little time for things to settle down, I'm not sure.

I now have around 5 servers all running on staggered schedules copying to their offsite drives. It just works. Its really good.

Got one funny thing I have noticed, can start a new thread if needed.

When I make a change ot the schedule in a job that has run successfully, (IE: ran it dynamically just to get a backup out) the job starts, but starts a new copy interval instead of waiting for a new one to start. Maybe this is by design, and when the time comes, it does a new copy, and comes right.

Rich

[veremin]

When I make a change ot the schedule in a job that has run successfully, (IE: ran it dynamically just to get a backup out) the job starts, but starts a new copy interval instead of waiting for a new one to start.

Can you elaborate on what schedule changes you're talking about? Changes made to synchronization interval, changes made to allowed backup window, etc.? Thanks.

[foggy]

Are you probably talking about starting the new synchronization interval using the Sync Now command? In this case the behavior depends on the actual interval setting, please review the corresponding user guide section.