<Mic92>
mdlayher: do you know, how I could test my watchdog?
<mdlayher>
Mic92: in my case I wrote a Go program to ping it once and then killed the program without disabling it, then waited for the reboot.
<Mic92>
mdlayher: but this does not trigger the hardware watchdog, does it?
<mdlayher>
What are you trying to test? In my case I just wanted to verify that my system did reboot when the device wasn't pinged
<Mic92>
mdlayher: A full test if my system reboots, when system/the kernel no longer respond.
<mdlayher>
Ah I suppose I didn't go that far. This week has been my first time looking into watchdog devices.
infinisil has quit [*.net *.split]
makefu has quit [*.net *.split]
lukegb has quit [*.net *.split]
makefu has joined #nixos-systemd
infinisil has joined #nixos-systemd
lukegb has joined #nixos-systemd
<flokli>
Mic92: I assume you could echo 1 > /proc/sys/kernel/sysrq, touch /dev/watchdog (or already have systemd done that), then echo c > /proc/sysrq-trigger to trigger a crash, and see if the watchdog eventually reboots it
<mdlayher>
TIL, makes sense though
<arianvp>
Now can we do this in a VM test??
<Mic92>
I think there is a softdog.
<Mic92>
But I could imagine there is also virtio watchdog as well.