Redis with replication and failover (keepalived)

This project help you to deploy redis With HA and automatic recovery. A basic cluster is a redis master and a redis slave, slave sync data from master. Only a redis master and a redis slave based on this arch. The project only for redis cache service not support transaction managemant. This program is for Ubuntu 14.04 or later, use redis_ha_installer_el6.py or redis_ha_installer_el7.py instead of CentOS 6/7 if you are using CentOS/RHEL.

• Clone this repo on all of the node:

git clone https://github.com/nofdev/redis-ha.git

• According to the need to edit the configuration file. The configuration files stored in the following location:

  • redis-ha/conf/keepalived.conf.master
    • This is the keepalived master configuration, you have to edit the VIP.
  • redis-ha/conf/keepalived.conf.backup
    • This is the keepalived backup configuration, you have to edit the VIP.
  • redis-ha/conf/redis.conf.master
    • This is the redis master configuration.
  • redis-ha/conf/redis.conf.slave
    • This is the redis slave configuration.
  • redis-ha/tools/redis.sh
    • This is the fail-over script invoked by keepalived, you have to edit the redis master ip.

• To deploy the master node:

sudo python redis_ha_installer.py master

• To deploy the backup node(s):

sudo python redis_ha_installer.py backup

• Please use the stable version.

###About the redis.sh script

• On the master node, set the REDIS_MASTER_IP as slave IP.
• On the slave node, set the REDIS_MASTER_IP as master IP.

• Please deploy docker daemon on your server.
• Choose your correct Docker file and rename it to Dockerfile.
• Using docker build -t user/repo:tag to build your image.

jiasir (Taio Jia) [email protected]

2014.12.20 - Add support for the docker. 2014.10.28 - Change the redis configuration settings, compatible with the version 2.8.17 and resoled the soft and hard limits.