Generic notes

• There's no cluster manager, you have to specify all nodes / ports in commands
  • So there's no autoscaling...and if you want to change majority configuration then idk you're just out of luck

Lil' Helpers

cd repos/raft_2025_01/lsprangers-raft/part1 python src/client.py --server-port=54321 --server-host=localhost ps -fa | grep python

Part 3

• Should we do req/resp, or something like actor model?
  • Based on actor model + idempotent RPC + how RAFT paper discusses it, we should use Actor model