The hot swap migration strategy implements the ProcessHub.Strategy.Migration.Base protocol. It provides a migration strategy where the local process is terminated after the new one is started on the remote node and registered.

Hot swap is useful when we want to ensure that there is no downtime when migrating the child process to the remote node. The old process remains alive until the new process is successfully registered, at which point the state is delivered and the old process is terminated.

This is the key difference from ColdSwap, which terminates the local process before starting the remote one.

State Handover

When handover: true is set, HotSwap will:

1. Query state from local processes before sending start requests
2. Store states in ETS with TTL (along with old process reference)
3. Send start requests to new nodes
4. When new processes start on the target node, the post-action callback notifies the originating node
5. Originating node delivers stored state and terminates old local process

To use state handover, your GenServer must use ProcessHub.Strategy.Migration.HotSwap:

defmodule MyServer do
  use GenServer
  use ProcessHub.Strategy.Migration.HotSwap

  # Optionally override these callbacks:
  # def prepare_handover_state(state), do: state
  # def alter_handover_state(_current, handover), do: handover
end

State Handover with Replication

State handover is not supported when using the ProcessHub.Strategy.Redundancy.Replication strategy. With replication, multiple instances of a process run across the cluster, making state handover semantics undefined. If you attempt to use handover: true with replication, the hub will fail to start with {:error, {:invalid_config, :handover_with_replication_not_supported}}.