All VMs immediately fail with Exit Code 11 when connecting noVNC after 1 day

[Stage5-F100]

SETUP:

• 11.3-U2 install, upgraded from a fresh 11.3-U1 install
• New Windows 10 VM and new CentOS 8 VM
• (Can't test with Ubuntu, since FreeNAS x.x no longer supports Ubuntu per https://jira.ixsystems.com/browse/NAS-105649 )
• No jails. No custom/advanced parameters.

~ ~ ~ ~ ~ ~

SYMPTOMS:

All virtual machines will start and run fine initially, but when left overnight, will immediately crash/fail when attempting to connect over noVNC.

Windows 10 VM, immediately upon clicking "VNC" button after running overnight:
Apr 21 12:27:33 [host] kernel: pid 33026 (bhyve), jid 0, uid 0: exited on signal 11
Apr 21 12:27:33 [host] kernel: tap0: link state changed to DOWN
Apr 21 12:27:33 [host] kernel: tap0: link state changed to DOWN

CentOS VM, immediately upon clicking "VNC" button after running for two days:
Apr 21 12:28:04 [host] kernel: pid 33258 (bhyve), jid 0, uid 0: exited on signal 11
Apr 21 12:28:04 [host] kernel: tap1: link state changed to DOWN
Apr 21 12:28:04 [host] kernel: tap1: link state changed to DOWN

~ ~ ~ ~ ~ ~

TROUBLESHOOTING:

These same systems are running fine immediately before clicking "VNC" in the webUI, as-shown by CPU load, disk activity, TeamViewer (to the VM), and RealVNC (also to the VM).

The problem begins, every time, when clicking the "VNC" button to launch noVNC.

 

[Stage5-F100]

Just another attachment; this is what is seen at the exact same time the VM goes down (and the error is logged as posted above).

NOTE: if trying to access the VM through any other means (RDP, RealVNC, TeamViewer, Linux-standard VNCs, etc), no issues are noted. The crash happens every time with noVNC/clicking the VNC button in the FreeNAS webUI.

 

[Visseroth]

I'm experiencing a similar issue...

Apr 25 21:49:43 storage kernel: pid 55611 (bhyve), jid 0, uid 0: exited on signal 11
Apr 25 21:49:43 storage kernel: tap0: link state changed to DOWN
Apr 25 21:49:43 storage kernel: tap0: link state changed to DOWN

I just setup a Ubuntu server for the first time, haven't even made it past the setup screen for languages yet

 

[fastleo63]

Same story happened to me, since I upgraded my rig to FreeNAS-11.3-U2.1.
It happens with a preexistent VM (Ubuntu Server 18.04 LTS), but with a brand new "Focal Fossa" VM (Ubuntu Server 20.04 LTS), too.

 

[Visseroth]

I finally gave up.
All I could get was the VM to boot for a while, couldn't even get the installer to complete and then CRASH.
VM Deleted, guess I didn't need it.

 

[fastleo63]

Seems that I found a solution, at least for my case (all VNC clients, from UltraVNC to TightVNC, including HTTP-based ones, made my Ubuntu Server machines crashing with exit code 11).
I changed the NIC "Adapter Type" from "Intel e82545" to "VirtIO", and started the VMs.

 

[Visseroth]

Interesting.
Give me a few and I'll give it a try as well and I'll confirm if I get the same results or not.

 

[Visseroth]

Well I couldn't even get through a installer with the Intel NIC. With the VirtIO NIC it seems to be working just fine and is now finally installing!
Nice Find!!

 

[quiet-truck]

Seems that I found a solution, at least for my case (all VNC clients, from UltraVNC to TightVNC, including HTTP-based ones, made my Ubuntu Server machines crashing with exit code 11).
I changed the NIC "Adapter Type" from "Intel e82545" to "VirtIO", and started the VMs.

same issue for me on a fresh install of FreeNAS-11.3-U3.2

just changed all my VM NIC to VirtIO, will see.

 

[diskdiddler]

I just discovered my DietPiVM decided to drop out with this 11 fault today, no idea why - was this rolled in, as a bug to FreeNAS recently? 11.3-U3?


EDIT: confirmed bug with VNC connections, soon as I connect, she crashes- has anyone reported this to IX?

 

[quiet-truck]

I just discovered my DietPiVM decided to drop out with this 11 fault today, no idea why - was this rolled in, as a bug to FreeNAS recently? 11.3-U3?


EDIT: confirmed bug with VNC connections, soon as I connect, she crashes- has anyone reported this to IX?

The VirtIO fix worked for me.

 

[diskdiddler]

Sure, but this shouldn't be crashing a VM either.

 

[diskdiddler]

Can confirm VirtIO fix worked for me tonight with this. Kind of annoying I had to do it though. Default new VM creation should at least go to the one which isn't going to crash the VM.

 

[quiet-truck]

Can confirm VirtIO fix worked for me tonight with this. Kind of annoying I had to do it though. Default new VM creation should at least go to the one which isn't going to crash the VM.

which processor you are on?
I encounter this issue after I switch from Ryzen 3200G to E3 1220 v3

 

[diskdiddler]

I have the Denverton C3758 I think. - I got the impression from this thread, that the problem was due to a recent patch?