NAS backup fails if running longer than 24 hours

[milosk]

Hello,

we have a problem with backing up large NAS share (32.2 TB and over 19 millions files). It's an initial full backup and it fails regularly after 24 hours of run. Then Veeam retries, repairing backup (3-4 hours), calculating share size (slowing down backup data throughput) for hours and backup again fails when running longer than 24 hours.

Veeam Backup server is 11a on Windows Server 2019; 4 vCPU with 12 GB RAM
File proxy is on Windows Server 2019; 6 vCPU with 16 GB RAM
NAS job has only this one share using VSS snapshot; share added manually (not using NAS integration)
NAS filer is NetApp FAS 2720 with ONTAP 9.8P6
Network is 1Gbit

Since backup fails quite regularly after 24 hours and based on error my suspicion is that VSS snapshot times out or something like that:

"Failed to complete backup items The network name cannot be found. Failed to read data from the file"

Has anybody idea what is the problem?

Veeam support case #05595164

Thank you.

[Mildur]

Hi Milo

Thanks for the case number. I see, that you requested an update on the case.
I suggest also to use our "Talk to a manager" option if you need to escalate the case or you don't get an answer:
https://www.veeam.com/kb2320

In the Case, backups of NFS Share is discussed. Here in this topic, you mentioned NAS Job with VSS snapshot.
Doing VSS Snapshots for File Share Backup Jobs is only possible for Managed Windows File Server and SMB Shares.
NFS File Shares only support backups from storage snapshots.
Specify Advanced NFS File Share Settings

Can you confirm what sort of File Share you have added? SMB, NFS or Managed Server?

Thanks
Fabian

[milosk]

Hi Mildur,

yes, case was opened initially for different issue with NFS shares (access denied but that was already fixed on our side). Then we found more issues with NAS backup and one of them is this large SMB share so I updated case accordingly.

SMB share was added to Veeam manually. I like when I can browse share in Veeam console which is not possible when whole filer is added (as managed server/filer).

Thanks.

[Mildur]

Ah, I understand.
Do you see anything regarding the vss snapshot in the netapp logs?
Could you create the initial backup with a storage snapshot and then change it to VSS snapshot for the incremental runs? Could be worth a try.

Thanks
Fabian

[milosk]

Yep, that's a good idea. Now third retry of job is running, so when it fails tomorrow I will try it without VSS for initial backup.

Thank you.

[milosk]

So, workaround works fine (disable VSS for initial full backup and enable it only for incremental backups). I also checked NetApp logs and I found that VSS snapshot was aborted due to:

“No activity has been seen for at least 1464 minutes”

Question now is, if Veeam is not sending keep alive "signal" to NetApp or NetApp ignores it . Since workaround is just one-time action I decided to not dig deeper and asked for case closure.

Thanks Fabian for hint.