SSL_ERROR_SYSCALL Error observed by underlying SSL/TLS BIO: Connection reset by peer

[cjanson]

Hi,

we have recently installed a new server (Dell PowerEdge™ R6515) with  ESXi 8.01.

Everything is working as expected but the backup software (Acronis Cyber Backup 12.5). Most of the times when the backup of a VM starts, it gets a connection error (first part is translated):

The connection to the VMware ESX(i) or vCenter server failed due to a network error: 'SOAP 1.1 fault SOAP-ENV:Client[no subcode].
"SSL_ERROR_SYSCALL
Error observed by underlying SSL/TLS BIO: Connection reset by peer".
Detail: SSL/TLS handshake failed

We don't use VCenter but a direct connection to the ESXi server.

Sometimes it works as expected so I don't think it's a configuration problem.

The same setup is working with ESXi 7.0 with no problems.

My questions:

What could be the reason? Has anyone ever had this or a similar problem?

 

Regards,

Christian

[Saschljk]

I have the same problem for about a month. The only difference is, you got a "Connection reset by peer", I got an "Unknown error".
Hardware is a Fujitsu Primergy RX2540 M5. My ESXi are on version 7.0U3g. No vCenter. My Acronis is Cyber Protect 15.0.31791. As I write this it occurs to me... I use GDATA Business for antivirus. But I use it for jears with acronis. The "SSL_ERROR_SYSCALL" occurs now and then. "Unknown errors" are the best!

I have a second ESXi 7.0U3g in the same net and backed it up with Veeam. Everything is fine here.

 

Edith:
One backupjob failed with the message "EOF was observed that violates the SSL/TLS protocol. The client probably provided invalid authentication information." An EOF error means the server is closing its end of the TCP connection during the TLS handshake. Fits into the scheme. Now I need to find out why.

[Byteshunter]

Hi, I also have this situation with ESXi 7.x

Have you solved it?

Thank you

[Saschljk]

My solution was to install a new Acronis Management Server on a Linux CentOS.

First i installed a new Acronis MMS on a Windows VM. In the first days, everything worked like a charm. But after one week, the same errors.

[cjanson]

We never got it solved. The Acronis support wasn't able to help us. We have switched to Nakivo, which is working very well for ESXi 8.0 and 8.01 (with 8.02 there was another problem which could be solved also, but that is a different story).

 

[Byteshunter]

Thanks, we have the exact same error with Veeam. I'm investigating the I/O problem.

[Syedrali1]

 
Did you get to know the Root cause for the issue associated with the Veeam Backup.  

The SSL_ERROR_SYSCALL error typically occurs when the TCP three-way handshake between client and server completes but then a TCP reset packet (often written as “RST”) is received by the client, terminating the connection during the SSL phase.

[BarryGrowler]

Noticed your switch to Nakivo after facing unresolved issues with Acronis on ESXi 8.0 and 8.01. But you told mentioned some story... Curious, did you have a chance to reach nakivo's support team? Your experience would be valuable to understand everything