An existing connection was forcibly closed by the remote host

[eternal_noob]

Veeam Support - Case # 04001255

Backup Server: WS2012R2 (physical)
Backup Target: WS2012R2 (physical)
Backup Repository: Synology NAS

Current Veeam error message: Error: An existing connection was forcibly closed by the remote host
Error: Storage '\\YORSYNBAK01\Veeam01\NTFS2-VID - NTFS2-VID\NTFS2-VID - NTFS2-VIDD2020-02-11T173102_B319.vbk' is closed. Failed to download disk. Reconnectable protocol device was closed. Failed to upload disk. Agent failed to process method {DataTransfer.SyncDisk}.

Two other agent-based jobs on W2008R2 physical servers are working fine.

This failing job used to work fine for several weeks before I tried to add another volume to the job a couple weeks ago. It has been fail city ever since.

I sent logs to Veeam today and did hear back earlier. The support technician said that the connection to the backup target is "timing out while attempting to run the tasls to process the VM". They asked about my proxies which is set to all Veeam defaults.

Since I haven't heard back now in a few hours, I continued my self-investigation and the backup target shows to errors in the Server Manager events field.

NTFS2-VID Error 8194 Error VSS Application 02-11-2020 @ 5:31:02PM
NTFS2-VID Error 8194 Error VSS Application 02-12-2020 @ 5:07:14AM

Those timeframes correspond directly with the Veeam backup job on 02-11-2020 which was set to start at 5:30PM, as well as Retry #1 which began at 5:06AM on 02-12-2020.

I checked the server for “vssadmin list writers” and found none in an error state.

This is urgent since we don't have a backup solution in place and this target is our primary file server. I just started working here recently and have been tasked to get a working backup solution in place. I've heard wonderful things about Veeam and decided to give it a try. Was thrilled with it when it was working, and nearly had the CEO's sign-off to buy including cloud-based storage, etc... Now CEO is losing confidence. I know this can be fixed and haven't given up hope. Meanwhile, my only "backups" are presumably no good and I am very uneasy now. Any and all help is greatly appreciated.

Thanks everyone!

[Gostev]

Hello! Please note that using SMB protocol for backing up to non-CA file shares is specifically not recommended due to SMB stack reliability considerations. There's a note about this right in the user interface, when you register an SMB repository. Consider mounting your Synology via iSCSI protocol instead. Thanks!

[eternal_noob]

OK, thank you Gostev for this response. I don't recall seeing that warning, but I believe you entirely. Sadly I have never worked with iSCSI before so this is going to set me back while I research and test and implement. I will definitely start looking into it.

Still open to any and all other suggestions that can help me get limping by until iSCSI can be rolled-out.

Thanks everyone.

[eternal_noob]

After weeks of researching, I ended up finding a fix to my problem here:

https://blog.ryanbetts.co.uk/2014/07/ve ... m.html?m=0

“The cause of this issue is that packets are being dropped. The fix is to create a new DWORD key in the Registry of the VM that is failing to backup. The Registry path is \HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Services\Tcpip\Parameters, create a new DWORD with the value of SynAttackProtect and set the value to 0.”

Making this registry change has fixed my backups.

I hope this will help others to find this fix where necessary, in the future.

[dalbertson]

Thanks for posting this. I would recommend others read about what this feature is and the caveats of disabling it.

https://docs.microsoft.com/en-us/previo ... dfrom=MSDN

An alternate solution would be to use the windows firewall to see which of this packets are being blocked and then create a rule to allow just this traffic instead of turning the protection all the way off.