VAW backup fails with "Timeout to start agent"

[AlexLeadingEdge]

Support ID # 04257067

We are in the middle of migrating the client from SBS 2010 to a new Server 2019. For some reason the backups have stopped, they just timeout without any information as to why.

I have tried:

- Rebooting the server
- Rebuilding the database because we changed drive letters of the data drive to match the old server
- Uninstalled Shadow Protect just in case it was causing a conflict
- Uninstalled VAW v4 and installed VAW v3
- Turned the Shadow Copies on, for some reason they were turned off

No change after any of these steps.

I'm at a loss as to why the backup just times out. Where do I find the logs, and what am I looking for to figure out the cause?

[Egor Yakovlev]

Logs are under C:\ProgramData\Veeam\Endpoint and you are looking for errors or timeouts in the first place - check \JobName\Job.xxx.xxx.log to start with.
Also worth checking VSS Providers leftover in your system, as sometimes deleting 3rd party tools will not cleanup their VSS Providers and keep causing conflicts(cmd > vssadmin list providers) as well as VSS Writers and their status(all should be in "[1] - Stable" statem as reported by cmd > vssadmin list writers). Never hurts to perform VSS Snapshot manually using Diskshadow.
/Cheers!

[AlexLeadingEdge]

Thanks Egor. I'm not sure but I found this in the logs:

[30.06.2020 11:09:34] <01> Error Failed to connect to agent 'SERVER', EP '127.0.0.1:2500' at Veeam.Backup.AgentProvider.CSocketAgentService.Start(Boolean redirectOutput)

When I run the command netstat -ao |find /i "listening" I get this:

TCP 127.0.0.1:2500 KW-SVR:0 LISTENING 25308

PID 25308 is the EdgeTransport.exe, the new Exchange Server. I can't easily change the Exchange Server ports, so how can I get Veeam Agent to use another port?

[AlexLeadingEdge]

Come to think of it, another problem I've seen with this new server is that after a reboot of the server the Exchange Server blocks connections to the Outlook clients on the end user's machines. I found that restarting the Edge Transport fixes the connection issue.

If Veeam Agent is grabbing Port 2500 before Exchange can this would explain why I have to keep restarting the Edge Transport service to allow the Outlook it to work again.

[AlexLeadingEdge]

Eureka! I've found it!

We installed Hexamail the day before yesterday, which uses Port 2500 as the Receive Connector.

I will change the Hexamail port and see how Veeam Agent works after that.

[Egor Yakovlev]

Actually port range 2500-3300 is a default data transfer range, and if first in line is taken, we should just pick next one free available.
Since you have many apps there, it will be cool to check other ports we use too, just in case..
/Cheers!

[AlexLeadingEdge]

Thanks Egor.

You're right, it "should" but it doesn't Not sure why. A programming issue?

[Dima P.]

Alex,

If I am not mistaken the failover to the next available port should be performed during the job start. I'll ask QA folks to check your case logs. Cheers!