I'm exercising my Ansible playbooks for creating my microk8s cluster in Triton and deleted the server I was working with to recreate and rebuild. That's a bhyve VM using a PCI passthrough network interface. The delete operation caused the CN to crash and reboot.

Is there a process I should go through to guard against that in the future?

I didn't do anything special - it was running and a I just issued a "delete" command. I imagine if I shut it down nicely, maybe that would have been better?

the hard reboot also left the VM in a weird state ("incomplete") and it took a couple of tries from the command line using vmadm to delete it. The first try timed out and left the vm in a "configured" stated. The second try deleted it successfully.

Can you share the kernel dump?

You shouldn't be able to panic a machine doing that.

I do have global core dumps enabled, but it looks like my /cores (linked to /zones/global/cores) directory is empty.

a kernel dump would be in /var/crash/volatile

see if there's anything in there

I do have a vmdump.0 file there from today.

(and METRICS.csv and bounds)

that would be it

Cool. Let me know where to send it and I'll pass it along. The dump file is 2.61G

i don't have a good place to receive it, but hopefully since dan,cd was interested he'll be able to provide something for you once he's back around

err danmcd

sounds good.