Connection keeps getting lost between proxy and repository.

[jonrees73]

Hi

Thought I would ask here before raising a ticket in case anyone has come across this.

Environment-
Virtual Veeam management server running on ESXi 6.0
Virtual vCenter 6.0
Physical Proxy dual 16GB FC to NetApp storage and 2x10GBE LACP to network.
Physical Repository 2x10GBE LACP to network.

When backing up a test VM located on the NetApp using Physical Proxy->Repository it fails with the unable to connect to port 2500 error.
Netstat shows port 2500 being opened by the service and Veeam logs show communication starts. It successfully copies the .vmx .vmxf .nvram files successfully.

If the same job is run with storage snapshot disabled using network mode instead it copies over 2GB data before it encounters the port 2500 error.

To Check port 2500 on proxy and repository created two test jobs. One using Virtual Veeam management server as proxy sending to physical repository and no port 2500 errors job runs fine. Second using physical proxy sending to Veeam management server dump collector volume as repository that also runs fine.
These two test jobs show no errors with either sending to or from port 2500 on either physical server.

to check communications between the physical servers created a windows share on the repository server and added as a repository to Veeam. Backing up using Storage Snapshot from Physical Proxy to share on Physical Repository works fine. Should i take it that the network is fine or is SMB transport just more robust than Veeam transport service?

Any suggestions before ticket gets raised tomorrow?

Thanks
Jon

[Andreas Neufert]

Port 2500 and higher are used from our data mover (VeeamAgents) on the Proxy and Repsoitory/Gateway Server.

If you see errors communication between Port 2500 there is something in the network that blocks this or an antivirus service that "active" monitors the Agent and lead into drops as it can not keep up with all the data traffic.
Another option is to check the network and MTU size settings are correct.

In general I would add the Repository as Windows Repository and configure Antivirus like described in KB1999. If you see the error collect logs and open a ticket at our support. Maybe they can find our what´s wrong.
As well a good idea is to update to latest version or Update (v9.5 U1).

[larry]

I had an issue that sounds like this when first installed Netapp. On the repository with the 10g nic’s. Disable tcp checksum offload under the advance settings of the nic. Mine issue only occurred with large VMs and only from snapshots.