Dragonfish k3s certificate expired (+ k3s corruption?)

I’m feeling slightly panicky right now, so please bear with me.
It appears that the k3s certificates on my Dragonfish-24.04.2.5 just expired a few hours ago. Symptoms: The app list is empty, “k3s kubectl” complains it can’t authenticate, and the k3s log is full of “certificate expired or not yet valid” errors. The apps are still running. (This system has been up for 206 days.)

My /var/lib/rancher/k3s seems have been wiped; it contains nothing but a hash-directory with bin and an empty etc. So “just restart k3s” probably won’t work.

How do I get out of this? I’m still on Dragonfish for fair reasons, so “upgrade to EE” isn’t it.

FWIW, chatgpt advises me stop k3s, move /var/lib/rancher/k3s away, restart k3s and hope for magic repairs. Is that a good path, or what should I do?

Thanks
– perry

I am not sure about that ChatGPT advice, but I do know that you should save a copy of your config (System → General Settings → Manage Configuration) if you haven’t already, so that you can easily reinstall if everything goes sideways.

Turning off the apps pool and turning it back to the original did manage to reset k3s, including making new certificates. I had to bounce the middleware service to get the apps UI to work again. Emergency over (I think).