Backup copy job to remote Wan Accelator hangs

[stevenrodenburg1]

Hello,

I'm puzzled by the following problem in our Lab. In Datacenter A, in City A, we have the main Veeam Backup infrastructure which happily makes backups, all good. Here we also have the source WAN accelerator.
In Datacenter B, in city B, we have two repositories. 1 Windows Repo which also has the target Wan Accelerator on it. And 1 NAS Repo via a Linux Gateway server (it mounts the NAS over NFS3).

We have two test copy-jobs, each having a different backup as source.
We run a backup job in City A, then the corresponding copy-job kicks in and starts replicating that data via the source wanaccel to the target wanaccel.
Copy job #1 has the Windows based Repo in City B as the destination. This Windows Server in City B also has the WanAccel software on it. The traffic goes "source wanaccel -> target wanaccel -> Windows Repo on same server as wanaccel".
Copy job #1 works perfectly.

Copy job #2 has the NAS based Repo (via that linux Gateway) in City B as the destination. So the traffic goes "source wanaccel -> target wanaccel -> Linux Repo -> NFS NAS".
Copy job #2 starts succesfully, I see it creates a new folder (with the job name) plus a .vmb file and a 22MB .vbk file in the NAS and then the job hangs. It starts with the first VM in the list and the last message it shows is "storage initialized" with a green checkmark. This is the moment where the .vmb and .vbk file got created.
After this, nothing. Just nothing. No errors, just nothing. Just sits there until the end of time.
Disabling the copy job #2 results in a "stopping" status which never goes away. I really have to kill the job in taskmgr and unlock the source files or else the backup job cannot run in the evening due to the files on the source storage being locked by the copy-job.

This is a lab, so support won't care (they have better things to do). I want to learn how to troubleshoot these sort of issues. In al my years with Veeam i've not come across something like this (copy-job starts, then hangs without ever timing out and/or showing errors, just sits there like a dead dodo).

The "Linux Gateway + NAS via NFS" construct is exactly the same as we use at our customers. It is proven to work correctly.

Where do I start looking? The target wanaccell works fine with the repository when that repo is on "itself" (same host). The target wanaccel box and the Linux Gateway are in the same subnet and can communicate freely. I even disabled the firewall on the Linux gateway in datacenter B (even though the correct ports are open).
The fact that .vbk and .vmb files get created on the NAS in City B must mean that the target wanaccel can talk/write to the Linux Repo and do it's thing (that is, until the job hangs)
Where do I start? Which logfiles do I check?

[HannesK]

Hello,

(they have better things to do).

they support everything with an active support contract

well, if you lab has no license with active support, then you can dive into c:\programdata\veeam\... and just go the the folder with the corresponding job name. Notepad++ with search in files&folders is also a good helper tool

Best regards,
Hannes

PS: the Linux gateway is not required anymore for NFS shares with V10. that might solve your problem without digging into why it fails