Keepalived Guide for High Availability on Linux

Keepalived is a daemon that provides simple and robust high-availability (HA) for Linux systems, mainly by managing Virtual IPs (VIPs) and failover using VRRP (Virtual Router Redundancy Protocol).

---

1. Install Keepalived

On Ubuntu/Debian:

sudo apt update
sudo apt install keepalived -y

On CentOS/RHEL:

sudo yum install epel-release -y
sudo yum install keepalived -y

Check version:

keepalived -v

---

2. Basic Concept

• VRRP: Protocol to assign a Virtual IP to a master server. Backup servers monitor the master and take over the VIP if the master fails.
• VIP: Virtual IP shared among servers. Clients connect to VIP, not the real server IP.
• Priority: Determines master (higher priority wins).
• State: MASTER or BACKUP.

---

3. Simple Keepalived Configuration

Edit /etc/keepalived/keepalived.conf on MASTER:

vrrp_instance VI_1 {
    state MASTER
    interface eth0
    virtual_router_id 51
    priority 100
    advert_int 1
    authentication {
        auth_type PASS
        auth_pass mysecret
    }
    virtual_ipaddress {
        192.168.1.100
    }
}

On BACKUP server:

vrrp_instance VI_1 {
    state BACKUP
    interface eth0
    virtual_router_id 51
    priority 90
    advert_int 1
    authentication {
        auth_type PASS
        auth_pass mysecret
    }
    virtual_ipaddress {
        192.168.1.100
    }
}

Explanation:

• interface: network interface to bind VIP.
• virtual_router_id: unique VRRP ID (1–255) shared by the group.
• priority: higher = master.
• advert_int: advertisement interval in seconds.
• authentication: simple password auth.
• virtual_ipaddress: IP that floats between servers.

---

4. Advanced Configuration

4.1 Track Scripts

Run a script to check service health and adjust priority dynamically.

vrrp_script chk_nginx {
    script "pidof nginx"
    interval 2
    weight 20
}

vrrp_instance VI_1 {
    state MASTER
    interface eth0
    virtual_router_id 51
    priority 100
    advert_int 1
    authentication {
        auth_type PASS
        auth_pass mysecret
    }
    virtual_ipaddress {
        192.168.1.100
    }
    track_script {
        chk_nginx
    }
}

• weight: decreases priority if the script fails, triggering failover.

---

4.2 Multiple VIPs

You can assign multiple VIPs:

virtual_ipaddress {
    192.168.1.100
    192.168.1.101
}

---

4.3 Notifications

Keepalived can trigger scripts on state changes:

notify /etc/keepalived/notify.sh

Example notify.sh:

#!/bin/bash
TYPE=$1
NAME=$2
STATE=$3

echo "$(date) - $TYPE $NAME changed state to $STATE" >> /var/log/keepalived.log

---

5. Start & Enable Keepalived

sudo systemctl start keepalived
sudo systemctl enable keepalived
sudo systemctl status keepalived

Check VIP:

ip addr show eth0

---

6. Troubleshooting

• Logs: /var/log/syslog (Ubuntu/Debian) or /var/log/messages (CentOS/RHEL).
• Test failover: stop keepalived on master, backup should take VIP.
• Verify priority and state:

sudo ip addr show
sudo systemctl status keepalived

---

7. Practical Tips

• Always test failover in a safe environment.
• Keep VRRP virtual_router_id unique per VIP group.
• Set higher priority for your main server.
• Use track_script for service-aware failover (Nginx, HAProxy, MySQL, etc.).
• Combine with HAProxy for full HA load balancing.