redis/redis-py · error · DataError

The requested node does not exist in the cluster.

Error message

The requested node does not exist in the cluster.

What it means

Raised by set_default_node when the supplied node is falsy or is not found in the cluster's node cache (get_node by node_name returns None). Prevents callers from pointing the cluster's default dispatch node at something the client did not discover.

Solutions

1. Obtain the node from the same client via client.get_node(host=, port=) or client.get_primaries() before calling set_default_node.
2. Re-fetch the node after a topology change rather than reusing a stale reference.
3. If you only need a default for command routing, let the client pick its default_node automatically.

Example fix

// before
node = client_other.get_primaries()[0]
client.set_default_node(node)
// after
node = client.get_node(host='10.0.0.5', port=7000)
if node is not None:
    client.set_default_node(node)

node = client.get_node(host=node.host, port=node.port)
if node is None:
    raise ValueError(f'node {node} not in this cluster')
client.set_default_node(node)

def node_belongs_to_cluster(client, node) -> bool:
    return client.get_node(node_name=node.name) is not None

from redis.exceptions import DataError
try:
    client.set_default_node(node)
except DataError as e:
    if 'does not exist' in str(e):
        node = client.get_primaries()[0]
        client.set_default_node(node)
    else:
        raise

When it happens

Trigger: Calling client.set_default_node(stale_node) where stale_node came from a previous topology or a different cluster; passing None; passing a manually constructed ClusterNode whose name does not match any discovered node.

Common situations: Caching a node reference across a failover/resharding; building a ClusterNode by hand instead of via get_node/get_primaries; passing a node from one RedisCluster instance into another.

Related errors

• Failed to subscribe to cluster nodes
• No targets were found to execute
• No targets were found to execute
• No targets were found to execute
• Slot " " is not covered by the cluster.

AI-assisted analysis of redis/redis-py@6a6b581b48 (2026-08-10). Data as JSON: /api/errors/a97001ed55bd8f23. Report an issue: GitHub.

Appendix: source

Thrown at redis/asyncio/cluster.py:816

        """Get the replica nodes of the cluster."""
        return self.nodes_manager.get_nodes_by_server_type(REPLICA)

    def get_random_node(self) -> "ClusterNode":
        """Get a random node of the cluster."""
        return random.choice(list(self.nodes_manager.nodes_cache.values()))

    def get_default_node(self) -> "ClusterNode":
        """Get the default node of the client."""
        return self.nodes_manager.default_node

    def set_default_node(self, node: "ClusterNode") -> None:
        """
        Set the default node of the client.

        :raises DataError: if None is passed or node does not exist in cluster.
        """
        if not node or not self.get_node(node_name=node.name):
            raise DataError("The requested node does not exist in the cluster.")

        self.nodes_manager.default_node = node

    def get_node(
        self,
        host: Optional[str] = None,
        port: Optional[int] = None,
        node_name: Optional[str] = None,
    ) -> Optional["ClusterNode"]:
        """Get node by (host, port) or node_name."""
        return self.nodes_manager.get_node(host, port, node_name)

    def get_node_from_key(
        self, key: str, replica: bool = False
    ) -> Optional["ClusterNode"]:
        """
        Get the cluster node corresponding to the provided key.

View on GitHub (pinned to 6a6b581b48)