This is my first real HA implementation – The aim is to provide redundancy in the event of hardware or software fail. Put simply: If one router dies, another will automatically take its place.

Each workstation on the 192.168.1.0/24 network has an internet gateway of ’192.168.1.253′ set. This is not a real IP, but rather points at either 192.168.1.254 (Eskan) or 192.168.1.1 (Krenim) depending on the circumstance. In the background, the keepalived daemons on Eskan and Krenim are constantly checking on each other for who should own 192.168.1.253.
In a neutral state, Eskan will receive the 192.168.1.253 IP address as it is set to be the ‘Master’. However, if Krenim (the ‘Backup’) detects that Eskan has become unreachable, it will claim the 192.168.1.253 address, and become the gateway without the workstations even noticing that router just died (hopefully). Once the master returns, the backup will hand the 192.168.1.253 address back to it and happily wait for it to happen again.
Read more to see how you can do this for your network.









