Raft was designed by Diego Ongaro and John Ousterhout (2014) with an unusual goal: a consensus algorithm students could actually understand — it trades cleverness for clarity.
Its safety rests on majority overlap: any two majorities of five servers share at least one member, so a new leader is always guaranteed to hold every committed entry.
Liveness is the part you can take away — a partition side without a majority can elect no one and commit nothing; it simply waits, while the majority side never breaks stride.
The protocol you are tormenting here keeps etcd (Kubernetes' control plane), CockroachDB, and Consul consistent in real datacenters.
Something in the simulation stopped unexpectedly — the lesson continues without it. You can move on; nothing you did was wrong.