<mike_sw294503>
Okay, I removed everything related to k8s, deleted those network interfaces, restarted, still nothing. I'm currently running Sandstorm off the workstation host OS, I'm going to make a fresh Debian VM, move the install in there, update my router port forwarding, and try again. If that still doesn't work I'll come back with more information.
mike_sw294503 has quit [Quit: Connection closed]
griff_ has joined #sandstorm
griff_ has quit [Ping timeout: 264 seconds]
griff_ has joined #sandstorm
mike_sw294503 has joined #sandstorm
<mike_sw294503>
Putting it into a Debian 10.8 (latest) VM fixed my certificate renewal problem, so it must have been something specific to Ubuntu 20.04 or my particular setup. Thanks for the help, sorry for the hassle.
dustyweb has quit [Read error: Connection reset by peer]
<mike_sw294503>
Well ()$&%($%& I figured it out - I replaced my Comcast router, the new one messed up my home network DNS, and I had fiddled with dnsmasq to get it working again. That's what broke certificate renewal. Sorry for wasting people's time.
<kentonv>
anything that is worth capturing in the docs in case someone else runs into the same problem?
<mike_sw294503>
Good question. I had used dnsmasq to hard-code *.swierczek.sandcats.io to 10.0.0.11 for machines in my home network, including the one hosting Sandstorm. That's what botched the LetsEncrypt renewal. So this kind of mistake could happen to someone else, but probably not often.
<mike_sw294503>
What would be a coherent way to express that in the docs?
<mike_sw294503>
Nothing like 6 hours of work to track down your own foot-gun. Hooray for technology. :D