Hello Everyone!
We are running for a while ISC DHCP (v4.2.5) on a single linux server and we would like to migrate it a highly available setup using "failover peer".
It is not very clear to me what would be the imapct of adding the failover peer configuration into an exisiting systems with leases that we obvisouly can't lose, I'm hoping that the system will perform the necessary changes but I don't feel confident about this.
Also, our ranges are defined under subnet and it is my understanding that these should be defined under pools where the "failover peer" is to be configured, is that correct? so, I guess same question here - can we move the ranges to "pools" without affecting any of the lease data?
Here's a subnet example in our setup:
subnet xxx.xxx.68.0 netmask 255.255.254.0 {
range xx.xxx.68.50 xx.xxx.68.150;
range xx.xxx.69.50 xx.xxx.69.200;
option routers xx.xxx.68.1;
option subnet-mask 255.255.254.0;
option broadcast-address xx.xxx.69.255;
option domain-search "<domain a>", "<domain b>", "<domain c>";
option domain-name-servers xx.xxx.64.6, xx.xxx.64.7;
default-lease-time 2592000;
max-lease-time 10368000;
}
Can we just change it to:
subnet xxx.xxx.68.0 netmask 255.255.254.0 {
pool {
range xx.xxx.68.50 xx.xxx.68.150;
range xx.xxx.69.50 xx.xxx.69.200;
}
} option routers xx.xxx.68.1;
option subnet-mask 255.255.254.0;
option broadcast-address xx.xxx.69.255;
option domain-search "<domain a>", "<domain b>", "<domain c>";
option domain-name-servers xx.xxx.64.6, xx.xxx.64.7;
default-lease-time 2592000;
max-lease-time 10368000;
}
I couldn't find anything in the documenation that describes how to setup "failover peer" in a pre-exisiting server with leases, so any pointers would be highly welcome.
Thanks!
Danny