AdvancedDatabase & Storage
High-Availability Redis: Sentinel Auto-Failover
Ensure zero downtime for cache and session layers. This recipe walks through setting up Redis replication combined with Sentinel to monitor and promote replicas automatically.
RedisRedis Sentinel
Prerequisites
- 3 Linux VMs (1 Master, 2 Replicas/Sentinels)
Configuration Files
sentinel.conf/etc/redis/sentinel.conf
port 26379\nsentinel monitor mymaster 192.168.1.10 6379 2\nsentinel down-after-milliseconds mymaster 5000\nsentinel failover-timeout mymaster 10000\nsentinel auth-pass mymaster securepasswordExplanation:Configures Sentinel to monitor the master at 192.168.1.10. A quorum of 2 Sentinels must agree the master is down before failover.
Verification Steps
1
Checks the cluster state from Sentinel's perspective.
$redis-cli -p 26379 info sentinel
Expected Outputstatus=ok,address=192.168.1.10:6379,slaves=2,sentinels=3
Production Gotchas
- Clients connecting to Redis must support the Sentinel protocol to ask for the current master address upon failover.
Frequently Asked Questions
What is a split-brain scenario?
When a network partition causes multiple nodes to believe they are the master, corrupting data. Quorum (majority vote) prevents this.