Configured ISC DHCPd failover this morning.

Setup was  pretty smooth overall, although you have to take care that when you restart your (before “only”) dhcpd, it will go into recovery mode, and doesn’t seem to come out of that until it can reach its peer. I had only started one, but this meant no DHCP service:

DHCPREQUEST for 1.2.3.4 from 00:24:e8:41:55:2d via 
 vlan123: not responding (recovering)

 
Configuration is reasonably easy:

failover peer "dhcp-failover" {
  primary;
  address 192.168.1.1;
  port 647;
  peer address 192.168.1.2;
  peer port 647;
  max-response-delay 30;
  max-unacked-updates 10;
  load balance max seconds 3;
  mclt 1800;
  split 128;
}

 
The same block goes on the secondary, but replace “primary” by “secondary”, swap the 2 peer addresses, and remove mclt and split parameters.

After that, you can add this to your subnet blocks:

pool {
  failover peer "dhcp-failover";
  range 192.168.2.91 192.168.2.200;
}

 
Restart the servers and everything should start to work. Don’t forget to keep your configs in sync!

I took most of my notes from Paul Heinlein’s document, which while over 4 years old seems to contain everything to get it to work.