11.3 Bug? - VMs crashing when initiating a VNC session

[gsrcrxsi]

FreeNAS 11.3
Supermicro X9DRi-LN4F+
2x E5-2680v2
128GB Reg ECC DDR3-1600

Don't know where this should go. please advise if this is better posted somewhere else.

I just upgraded my install from 11.2-U8 to 11.3 via the manual upgrade option from 11.3 installation media (USB stick). Neither system recognized the 11.3 train (missing from the dropdown list) when trying to do an in-place upgrade. But I'm experiencing the same thing on a totally separate system with the same 11.3 upgrade from 11.2-U6.

First off, this issue NEVER happened on any version of 11.2

after upgrading to 11.3, whenever I have a VM that's been running for a little while whenever I try to VNC into it, either via webVNC built in or via the VNCViewer client, the VM most of the time will crash and VNC says that it is unable to connect. I have verified that the VM is not down or crashed prior to attempting the VNC connection. the FreeNAS GUI shows that it's up and I can SSH into it fine. but once I try VNC, it crashes and the GUI changes it to a down state.

all VMs are standard Ubuntu 18.04 installs with nothing crazy going on and plenty of resources allocated to them for what they are doing. some are even 8thread VMs with 8GB memory.

has anyone noticed this?

 

[gsrcrxsi]

this seems like the same issue:

https://jira.ixsystems.com/browse/NAS-104926

 

[gsrcrxsi]

issue is still present after upgrading to 11.3-U1

 

[thepixelgeek]

My VM wasn't working right after the 11.3 updates and before I tried using VNC or PuTTY.

Ubuntu VM installed on 11.2-U7 was working well, but after I upgrade to 11.3-STABLE, I can no longer connect via VNC or PuTTY.

Freenas > Virtual Machines, shows it to be on.

I've restarted the VM, powered off then on again. I also upgraded my pools which had no effect.

Seeing this in the logs:
[2020-02-29 08:39:13,784] (ERROR) VMService.vm_5.running():488 - ===> VMM zen is running without bhyve process.
And this
[2020-02-29 08:39:13,803] (WARNING) VMService.vm_5.destroy_vm():351 - ===> Destroying VM: zen ID: 5 BHYVE_CODE: 1

Seeing this after running ps aux
root 81583 1.6 0.6 4242556 419904 - SC 09:22 1:01.24 bhyve: 5

 

[Versadyne]

This is an issue I am also having since upgading to 11.3 (now 11.3-U3.1). When I use VNC to log into either of my 2 VM's, one of two things happen. 1) if the VM has been running for more than 24hrs the VM stops and I get an 'unable to connect' on the VNC. 2) if the VM has been running less than 24hrs the VNC connects without problem. One VM is a mailserver so I know that is working when not logged into it. The other is FAMP with contact managment software. Works from ip on port 80 no problem but when you log in with VNC it stops. Again with the above two conditions.