Veeam v10 immediate copy does not copy synthetic full

[cupra1980]

My primary backup job is configured as follow:
- daily backup at 20.00
- forward incremental
- synthetic full backup every saturday (ReFs)

Now I created new backup copy jobs because I wanted to use the "immediate copy" function.
It works so far for standard incremental backup restore points. Immediately after the restore point is created, a copy job is started and copies it over to my data domain.

In my copy job I have configured that I want to keep 4 weekly and 12 monthly restore points as full backups for archival.
But this does not happen. The copy job only copies the incremental backup to the data domain. In the same time, the synthetic backup tasks starts to run.
After the synthetic backup is finished, nothing happens. Even it is Saturday, which is in my case the day I wanted to have the additional GFS copies for the full backup archival.

With the traditional copy mode, this worked and the synthetic full backup was copied over to the data domain appliance.

Best regards, Christoph

[Gostev]

Hello! Please include the support case ID for this issue, as requested when you click New Topic. Thanks!

[cupra1980]

Here is the support case id: 04053054

[foggy]

Hi Christoph, most likely the backup copy chain hasn't reached the configured number of restore points of recent backups. Once this happens, GFS retention will come into action.

[DominikM]

I have the same question but the answer of foggy isn't satisfying enough for me. I'd like to now how exactly the mechanism works and how can I achieve that from week 1 on the weekly full backups are made. Neither the GUI nor the manual is a reasonable explanation for the behavior that foggy described. My case ID: 01960263

[foggy]

Hi Dominik, first of all 01960263 is not the correct case ID number, I believe the valid number is #04291053 while the number you've mentioned is the support contract ID. As far as I can see from the case notes, you've already discussed the issue with our engineers and was provided with the explanation of the observed behavior and how GFS works in case of Immediate backup copy. Feel free to ask any additional questions, if required. Thanks!

[TheDrew2012]

I would like to know what the observed behavior is as the documents are not clear, and we're running v10 w/ Immediate Copy mode and need GFS retention on the offsite copies.

[foggy]

Hi Andrew, what specific behavior you're asking about? GFS should come into action when the number of restore points in the regular part of the backup chain reaches the number specified in the job settings.

[TheDrew2012]

What I'm observing on a new backup copy job I setup to take advantage of immediate copy in v10 (previously ran regular continous runs in v9). Started a new backup copy job, setup sufficient retention points that we can generate a GFS weekly copy every Saturday. When I went into the list of restore points I can see every backup restore point from the start last week to today's most recent one, but I do not see any Weekly GFS copies created on the backup copy job.

We do have "Use per-VM backup files" enabled in our repository setting but that shouldn't impact GFS as far as I know.

Just to add,

The Primary backup jobs run every six hours, 21 days worth of retention, synthetic full every Saturday morning. We've not enabled GFS on the primary back jobs.

[foggy]

Does the job already reached its retention, i.e. do you see the specified number of restore points on disk (what is it, btw?)?

[TheDrew2012]

It had been set to 21 restore points, which given it was started on a Tuesday, and we run four backups a day, should have been enough to reach Saturday. I did increase that on Monday as that won't be enough, moved it to 84 restore points, which I'll adjust downwards if need be.

[foggy]

The first GFS restore point is offloaded once the regular part of the chain reaches the specified retention and the full backup reaches the specified GFS creation date after that. Seems that this hasn't happened yet.

[TheDrew2012]

Okay. So basically in this mode, once the restore point count reaches the specified limit (originally 21, adjusted up to 25) and the vbk file that represents the oldest available restore point reaches the point a weekly copy should be retained, that vbk file is duplicated & tagged as a GFS weekly copy, and the backup cycle continues.

I will note that overall I've been quite happy with v10 so far. The new backup copy mode (immediate copy) has made my boss happier as he can now prove to his bosses, that not only are we achieving an RPO of 6hrs on our backups, but with the maybe 1hr window from initial backup run to the backup copy placing it on disk at our offsite recovery site, we are able to assure senior management that our data is offsite within 60-90min of each backup run.

[foggy]

Yes, your understanding is correct. And thanks for the good feedback!