Anyone else playing around with propolis?

propolis-cli -s 127.0.0.1 state run just reboots the host

It generated a panic?

elwisp: What host are you running on? The version of propolis/bhyve from illumos generally need to be fairly in sync.

i run it on helios which runs in a VM

Did the helios vm panic or the guest inside?

Hard to say what happens because as soon as i run the "state run" it just reboots the machine

Does the hypervisor you're running helios in provide nested guest?

*nested virt

ah, that might be it!

lemme check

And what's rebooting? The VM or the hypervisor?

I've got to step out for a while, but knowing which of those is rebooting is going to change our next steps a lot. Specifically not your propolis guest, but the helios VM or the hypervisor running the helios VM.

the VM

i enabled nesting now, going to try again

but doing a get of state shows that the vm is just "creating"

nesting wasnt the issue :)

cheers, got a question on SMF, in case I missed something in the past year or two: did it by chance grow any way to execute installable late-shutdown logic? (like rc.6/* scripts allowed, or systemd-shutdown hooks)?

namely I want to integrate the NUT ability to call the UPS driver when the system is otherwise quiesced, to tell the device to cut the power (when shutdown conditions call for that)

which needs a way to either let certain daemons to survive past the killall phase, or for the shutdown scripts to

call the nutshutdown script perhaps after they remounted R/O and lost most of the filesystem that way