Basic question about GFS

[LaurieS]

Hi
In the Veeam 9.5 documentation it states that (See https://helpcenter.veeam.com/docs/backu ... tml?ver=95)

At 00:00 of the day set in the GFS job schedule the GFS job starts and checks periodically if the source job created a backup:
• If the source job creates a full backup, the GFS job copies it.

So, I setup a backup job to run on Saturday creating a synthetic full backup and scheduled the GFS job to start on Sunday. I expected the synthetic fulls to be copied to tape. Instead I got the following message
"No restore points found for date 8/07/2018 Synthesized full backup was not created because the most recent restore point is a full backup" and the GFS job showed Pending. To kick off the GFS job I had to manually create a restore point.

I am obviously doing something wrong here. Can someone explain what I'm doing wrong. All I am trying to achieve is a GFS backup of VMs to tape that runs once a week.
I've read all the Veeam doco that I can find on GFS backups but I don't see what is causing this problem.

Cheers
Laurie

[Dima P.]

Hello Laurie.

GFS job has detected that you do not have a restore point for Sunday (since that's the scheduled day for GFS restore point the priority is given to Sunday restore points) and started to wait for new restore point to be created. It waits till the end on the day and if there is no new restore point it processes the restore point for the previous day (in your case synthetic full for Saturday). Makes sense? Cheers!

[aich365]

Hi Laurie
You can set a timeout to force the job to start as advised by Veeam:

Also, you may implement the following registry keys, both should be created on the Veeam server in this registry hive:
[HKEY_LOCAL_MACHINE\SOFTWARE\Veeam\Veeam Backup and Replication\]
Name: TapeGFSBackupWaitTimeout
Type: REG_DWORD
Value: 0

Description: Specify a value in minutes that GFS jobs should wait for new restore points. By default GFS job starts at 00:00 and waits for the new restore points for up to 24h before failing over to taking the previous restore point.
If the value is 0, a GFS job will start at 12AM using an existing restore point.

Name: TapeWaitForActiveStorageTimeoutInHours
Type: REG_DWORD
Value: 0

Description: Specify a value in hours that GFS jobs should wait for the active restore point (locked by the source job). By default GFS job waits for up to 7 days till the lock is released.
If the value is 0, GFS job will immediately switch to the previous storage.

Tape Timeout
It seems that the cause of the issue, is actually a workflow of the GFS tape job: By default GFS job starts at 00:00 and waits for the new restore points for up to 24h before failing over to taking the previous restore point.

We could make the waiting value less (10 minutes as example), so the job would not wait for the whole day.

Please add the following registry key to your VeeamServer:

HKEY_LOCAL_MACHINE\SOFTWARE\Veeam\Veeam Backup and Replication\

TapeGFSBackupWaitTimeout

Type: REG_DWORD
value: 10

Regards

Clive

[LaurieS]

Hi

Thanks very much for this information. I'll give it try next month, when I return to work, however from what you've written I believe this explains what I am seeing.