Redis Connection¶

The Redis connection type enables connection to a Redis deployment, either a standalone server or one running in cluster mode.

Default Connection IDs¶

Redis Hook uses parameter redis_conn_id for Connection IDs and the value of the parameter as redis_default by default.

Configuring the Connection¶

Host: The host of the Redis server.
Port: Specify the port to use for connecting the Redis server (Default is 6379).
Login: The user that will be used for authentication against the Redis server (only applicable in Redis 6.0 and above).
Password: The password of the user that will be used for authentication against the Redis server.
DB: The DB number to use in the Redis server (Default is 0). Not supported in cluster mode.
Is cluster: Whether Redis deployment is a cluster or a standalone instance (Default is False). See Redis Cluster Deployment below.
Startup nodes: Extra bootstrap nodes as a comma-separated host:port list. The port may be omitted and defaults to 6379. Only for cluster Redis deployments (Default is None).
Enable SSL: Whether to enable SSL connection to the Redis server (Default is False).
SSL verify mode: Whether to try to verify other peers’ certificates and how to behave if verification fails. For more information, see: Python SSL docs. Allowed values are: required, optional, none.
CA certificate path: The path to a file of concatenated CA certificates in PEM format (Default is None).
Private key path: Path to an ssl private key (Default is None).
Certificate path: Path to an ssl certificate (Default is None).
Enable hostname check: If set, match the hostname during the SSL handshake (Default is False).

Redis Cluster Deployment¶

Redis Cluster spreads the keyspace over 16384 hash slots owned by different masters, and expects the client to route each command to the node owning that key’s slot. A standalone client does not do this: when it asks a node for a key that node does not serve, the node answers MOVED and the standalone client fails.

Enable cluster mode to use a cluster-aware client that follows those redirects:

{
  "cluster": true,
  "startup_nodes": "node-2:6379,node-3:6379" // Connection extra
}

The client discovers the full topology from the first node it reaches. startup_nodes matters for bootstrap resilience: every task builds its own connection, so with a single seed node one unreachable node breaks every task.