SOLVED Scale (Bluefin) host hard resets shortly after boot, after I created a Win10 VM.

[o1982]

Where to start. Hardware is in the signature. Relatively new install, but had alot of config issues, like trying to change interface IP not working, SMB Shares and Apps permissions. But this one is a MUCH bigger problem: I attempted to make a Win10 VM, it took ages to create after clicking go at the bottom of the config side panel, which I thought was weird - I had just made a pfsense VM and that was quick (it was turned off at this point). I went and had dinner. Came back and everything is unresponsive, no webui and no input possible through IPMI console. I hard resetted. I checked temps, everything fine.

I was watching the boot up screen and saw this before it reset itself:

I am slowly losing my patience with Scale, I want to like it, but I just don't get the feeling it's ready.

Has anyone any idea how to solve this? I looked through the manual to try and find a CLI command to disable startup of that VM but didn't find anything.

 

[morganL]

Which version of SCALE are you using?

 

[o1982]

Hi, It was 22.10 and there were no remaining updates available.
Thank you.

 

[morganL]

Hi, It was 22.10 and there were no remaining updates available.
Thank you.

There is no 22.10..... 22.12.0??

 

[o1982]

There is no 22.10..... 22.12.0??

Ah, my bad it was 22.12.0. I just upgraded via USB to 22.12.1 - issue still persists - truenas boots and starts to load the Win10 VM(*), I could manage to get into the web console and navigate to the Virtual Machines section, it was loading but did not actually get to show/list the VM, then I got the "waiting for truenas controller..." I switched back to the IPMI console and it was already restarting.

I can boot into the 22.12.1 recovery mode - is there a command I can run that will either delete the VM or remove the "start after boot" flag/setting? Unfortunately the webui doesn't work in this mode.

The Win10 installed completed on the VM but I could never configure it, so just totally removing it is no problem.

(*) - I'm guessing it was the VM that's causing the issues - what are the logs I should look at to confirm/deny?

 

[o1982]

Ok, seemed to have fixed it. Well not really, I found the commands in the truenas cli to delete all the VMs and the VLAN & bridge they were connected to and rebooted now it doesn't reboot by itself.

For others that are in a similar position where the config is borked (I assume this is listed elsewhere, but I include it here for convenience):

1. When TN is booting and you get the blue menu with the bootable environments, go to Advanced Options and select the recovery mode.
2. Wait for it to boot. For me, I had to press enter half way through for it to carry on booting (I think it stops just before loading the middleware...)
3. You will eventually be greeted with a console with your hostname.
4. Type "cli" (FYI: typing "menu" will bring up the normal config menu.)
5. Type "service vm get_instance id=" and then enter "1" to see details of the first VM. I used this to make sure I got the correct VM id for the VM I wanted to delete.
6. Type "service vm delete id=" and then type the ID number of the VM you want to delete.
7. I then went to the menu and rebooted and it booted without issue.