How to fix CLUSTERDOWN Hash slot not served in Redis

First, verify your cluster status and see which slots are missing:

redis-cli -c cluster nodes
redis-cli -c cluster slots
redis-cli -c cluster info

Look for nodes marked as "fail" or "disconnected". The cluster slots output should show all 16384 slots assigned. If slots are missing or nodes report "unknown" slots, that's your problem.

If you're not actually running Redis in cluster mode and this error appeared unexpectedly, disable clustering:

1. Edit your Redis config file (usually /etc/redis/redis.conf or /etc/redis/redis-cluster.conf):

cluster-enabled no

2. Restart Redis:

redis-cli shutdown
# then restart redis-server

This resolves the error if you're not intentionally using cluster mode.

If a node is disconnected but still running, force it back into the cluster:

# From the disconnected node, explicitly rejoin the cluster
redis-cli CLUSTER MEET {master-ip} {master-port}

Replace {master-ip} and {master-port} with a known working cluster node. The node should re-synchronize slot assignments.

Then verify slot coverage:

redis-cli CLUSTER SLOTS | grep -c "^[0-9]"

All 16384 slots should eventually be assigned.

If a resharding operation was interrupted, slots might be stuck. Check:

redis-cli CLUSTER NODES | grep -i "migrating|importing"

If slots are stuck, use redis-cli to stabilize them:

# For each stuck slot number (e.g., 5000)
redis-cli CLUSTER SETSLOT 5000 STABLE

# Or on the target node during migration
redis-cli CLUSTER SETSLOT 5000 STABLE

After stabilizing, verify with CLUSTER SLOTS again.

Redis provides an automated repair tool:

redis-cli --cluster fix {any-node-ip}:{port}

This command connects to the cluster and attempts to:
- Detect nodes that are missing or in a bad state
- Reassign unserved slots to available nodes
- Correct inconsistent slot assignments

Example:

redis-cli --cluster fix 192.168.1.10:6379

Review the output carefully before confirming any changes.

If more than one master crashed and no replicas exist, you may need to rebuild:

1. Ensure all nodes are running and accessible
2. Flush data if acceptable:

redis-cli FLUSHALL

3. Recreate the cluster with proper slot distribution:

redis-cli --cluster create \
  192.168.1.10:6379 \
  192.168.1.11:6379 \
  192.168.1.12:6379 \
  --cluster-replicas 1

Use --cluster-replicas 1 to ensure each master has one replica for failover protection.