View Source ProcessHub.Strategy.Redundancy.Base protocol (ProcessHub v0.5.0-beta)

The redundancy protocol relies on the HashRing library to distribute processes across the cluster and determine which node should be responsible for a given process by its child_id key.

It is possible to start the same process on multiple nodes in the cluster.

Summary

Types

t()

All the types that implement this protocol.

Functions

handle_redundancy(strategy, hub, calculated_cids, nodes)

Handles redundancy when nodes join or leave the cluster.

init(strategy, hub)

Triggered when coordinator is initialized.

master_node(strategy, hub, child_id, child_nodes)

Returns the master node that the given child_id belongs to.

replication_factor(strategy)

Returns the replication factor for the given strategy struct. This is the number of replicas that the process will be started with.

Types

t()

@type t() :: term()

All the types that implement this protocol.

Functions

handle_redundancy(strategy, hub, calculated_cids, nodes)

@spec handle_redundancy(
  struct(),
  ProcessHub.Hub.t(),
  calculated_cids :: %{required(ProcessHub.child_id()) => [node()]},
  nodes :: [node()]
) :: :ok

Handles redundancy when nodes join or leave the cluster.

For Replication strategy:

Starts/stops replicas locally (indices 1+ from calculated_cids)
Sends mode signals (active/passive transitions)

Note: Primary process starting/stopping is handled by migration strategy.

Parameters

strategy - the redundancy strategy struct
hub - the hub state

calculated_cids - map of child_id => [node()] from migration strategy.

                Index 0 = primary, indices 1+ = replicas

nodes - list of nodes that triggered redistribution