6.8 cannot disable VMs

[Tom3]

Unraid 6.8.

I have docker disabled. All VMs are stopped.   When I select

Settings -> VM Manager -> Enable VMs -> select No

then click APPLY, nothing happens and the VMs remain enabled.

 

I recall this used to work, but has been several versions ago that I last tried.

 

-- Tom

 

 

 

 

[Tom3]

Decided to revert Unraid to 6.7.2 and the VM problems seem to have cleared up.

 

The VM Manger can now disable / enable VMs.

The images shutdown fully and the VMs stop (as they used to).

Not sure I'm brave enough to try installing 19.10 yet. Maybe later.
 

-- Tom

 

This is the last issue I had under Unraid 6.8, trying to start a previously working VM caused a noVNC exception:

noVNC encountered an error:

TypeError: ({_listeners:null, addEventListener:(function(type, callback) {
      if (!this._listeners) {
         this._listeners = new Map();
      }
      if (!this._listeners.has(type)) {
         this._listeners.set(type, new Set());
      }
      this._listeners.get(type).add(callback);
   }), removeEventListener:(function(type, callback) {
      if (!this._listeners || !this._listeners.has(type)) {
         return;
      }
      this._listeners.get(type).delete(callback);
   }), dispatchEvent:(function(event) {
      if (!this._listeners || !this._listeners.has(event.type)) {
         return true;
      }
      this._listeners.get(event.type).forEach(function (callback) {
         callback.call(this, event);
      }, this);
      return !event.defaultPrevented;
   })}) is not a constructor
http://192.168.0.250/plugins/dynamix.vm.manager/novnc/core/rfb.js:36:16
@http://192.168.0.250/plugins/dynamix.vm.manager/novnc/core/rfb.js:36:16

 

 

[LintHart]

I just had a similar but opposite issue. I had a crash under 6.8 and lost all my VMs. Couldn’t re-enable vms in settings so also decided to downgrade to get the switch to work 

[bonienl]

On 12/12/2019 at 10:44 PM, Tom3 said:

This is the last issue I had under Unraid 6.8, trying to start a previously working VM caused a noVNC exception

Can you open a bug report under Stable Releases.

This would allow further tracking and investigation.

(and diagnostics are required)

Edited December 15, 2019 by bonienl

[Tom3]

I changed back to 6.8 and reproduced the problem. Created diagnostics and submitted a report.

I have upgrade / downgraded multiple times in the process, and now it is not possible to install VMs

of any vintage in either 6.7.2 or 6.8 .  I setup in 6.7.2, disabled VMs, deleted libvirt (using

the VM manger GUI).  Then rebooted.

 

Now in either OS version, when I start to install a VM I get the grub menu, select Install Ubuntu,

Ubuntu starts up and goes to the first menu (Installation, Select Language) and the assigned CPU cores

go to 100% and stay there.  At this point the VM GUI is unresponsive.  I left it for an hour with CPU cores

pegged at 100% but it made no further progress.  

 

Perhaps the downgrade from 6.8 left some file changes unreverted?

 

So the next question - how do I restore the flash from the backup created before starting this adventure?

It is as easy as just copying the files from the ZIP over on top of the flash, or is something more needed?

Diagnostics attached.

 

-- Tom

 

 

 

 

 

tower-diagnostics-20191216-2350.zip

Edited December 16, 2019 by Tom3
post diagnostics

[Tom3]

Update - was able to install and run Ubuntu 18.04 Server VM with no issues both in 6.7.2 and 6.8.  There is

no gui, all command line and text menus based.  

 

-- Tom

 

Edited December 17, 2019 by Tom3
mention both versions

[Tom3]

Solved.

 

Finally got VMs (including 19.10) installed under UNRAID 6.8.  My Solution:

1. Assign all 16 cores to the VM for initial install.

2. Assign 8G minimum RAM.

3. Keep the VNC display window very small during installation.

It requires scrolling through all subsequent windows.

4. Very quickly attach VNC to the VM when first starting so that the

short grub timeout does not try to do a 'Try Ubuntu'.  There are only a few seconds

to get to the grub menu and select Install Ubuntu using up and down arrow keys.

5. Very quickly navigate to the installation buttons after the graphics comes up

and be quick to click the buttons.  Waiting just 2-3 seconds causes the CPU to go to 100%

and the display, mouse, and keyboard goes unresponsive for all time (well an hour at least).

This is the select language 1st page.

6. Similarly, go through pages 2,3, and 4 fast. Don't read anything as any delay will cause

the whole thing to go unresponsive and lock up.

7. Once at about the 5th page of the Ubuntu install menu it seems to be better behaved, and one

can take enough time to enter a name, password for Ubuntu, etc.

8. Complete the install with very small VNC window size. Once through the power down and

reboot cycle, then the VNC window can be enlarged.

9. After the VM normal shutdown 2nd time, the VM can be edited to select fewer cores, less 

memory, etc. and it seems to run OK tried with 1920x1080 Ubuntu display size.  

NOTE: I successfully edits 3 VMs, but the fourth won't let me edit the VM.. Clicking

apply after changing the number of cores just causes the UNRAID GUI to spin for a long time.

 

My server does not have a dedicated graphics card, just a VGA card enough to configure the BIOS.

It runs headless.

 

I don't know what is going on here, but the appearance is that during the VM startup/install phase

that the graphics/mouse/keyboard is consuming huge amounts of server resources enough to cause

it to lock up.

 

-- Tom

 

 

Edited December 17, 2019 by Tom3
add note about cant edit #cores on one VM

[joshbgosh10592]

I also have the same issue with clicking "Apply" when either enabling or disabling VM manager.

Did anyone ever submit a bug report for 6.8 for that?

[JorgeB]

Yes, it fixed on v6.8.1.rc1