Fixed as of 03:34 UTC. The Traversal server bugged out and hung in an infinite loop. Ideally our monitoring would have caught that -- it was reflected in metrics but no alerting configured for it. I'm hopefully fixing that for next time.
Even with alerting, note that I can't guarantee fixing it quickly when it fails. If you want to make this better for everyone, go and implement systemd heartbeat support (or something like that) in the Traversal server code
Even with alerting, note that I can't guarantee fixing it quickly when it fails. If you want to make this better for everyone, go and implement systemd heartbeat support (or something like that) in the Traversal server code