Deploying Redis Sentinel for High Availability in a Master‑Slave Setup

This article explains how to deploy Redis Sentinel to provide high availability for a Redis cluster consisting of one master and two slave nodes.

Prerequisite IP addresses:

master: 192.168.20.24

slave1: 192.168.20.25

slave2: 192.168.20.8

On each node, create /usr/local/redis/conf/sentinel.conf with the following content:

bind 0.0.0.0

port 26379

daemonize yes

logfile "/usr/local/redis/log/sentinel.log"

pidfile /var/run/sentinel.pid

sentinel monitor mymaster 192.168.20.24 6379 2

sentinel auth-pass mymaster 123456

sentinel down-after-milliseconds mymaster 10000

sentinel failover-timeout mymaster 180000

Start the three Redis server instances first, for example:

/usr/local/redis/bin/redis-cli -h 192.168.20.8 -p 6379 -a 123456

Then launch the three Sentinel instances. Two common ways are:

/usr/local/redis/bin/redis-server /usr/local/redis/conf/sentinel.conf --sentinel

or

/usr/local/redis/bin/redis-sentinel /usr/local/redis/conf/sentinel.conf

After the services are up, monitor the Sentinel log to confirm proper operation: tail -fn 1000 /usr/local/redis/log/sentinel.log Typical log output includes messages such as:

8181:X 17 Jun 2022 14:22:12.734 # Redis is starting

8181:X 17 Jun 2022 14:22:12.734 # Redis version=6.2.4, bits=64, commit=00000000, modified=0, pid=8181, just started

8181:X 17 Jun 2022 14:22:12.734 # Configuration loaded

8181:X 17 Jun 2022 14:22:12.735 * monotonic clock: POSIX clock_gettime

8181:X 17 Jun 2022 14:22:12.736 * Running mode=sentinel, port=26379.

8181:X 17 Jun 2022 14:22:12.737 # Sentinel ID is 988662ff1703bd362453265278cabf459dddd051

8181:X 17 Jun 2022 14:22:12.737 # +monitor master mymaster 192.168.20.24 6379 quorum 2

8181:X 17 Jun 2022 14:22:12.738 * +slave slave 192.168.20.25:6379 192.168.20.25 6379 @ mymaster 192.168.20.24 6379

8181:X 17 Jun 2022 14:22:12.739 * +slave slave 192.168.20.8:6379 192.168.20.8 6379 @ mymaster 192.168.20.24 6379