failover for mixed subnets

[Simon Hobson]

Luc T. wrote:

>  pool{
>      deny dynamic bootp clients;
>       failover peer "dhcp-failover-1";
>     range 192.168.1.101 192.168.1.199;
>}
>         
>         host 0014389390E2 { hardware ethernet 00:14:38:93:90:E2; 
>fixed-address 192.168.1.12;}
>         host 0014389390E3 { hardware ethernet 00:14:38:93:90:E2; 
>fixed-address 192.168.1.13;}
>         host 0014389390E4 { hardware ethernet 00:14:38:93:90:E2; 
>fixed-address 192.168.1.14;}
>         host 0014389390E5 { hardware ethernet 00:14:38:93:90:E2; 
>fixed-address 192.168.1.15;}
>         host 0014389390E6 { hardware ethernet 00:14:38:93:90:E2; 
>fixed-address 192.168.1.16;}
>}
>   
>
>   My questions are:
>   1. for the static part, namely 192.168.1.12 to 192.168.1.16, how 
>can I set up failover?

Just create the same host statements. There will be no co-ordination 
between the servers, and there is no need for any - the clients will 
simply get the lease from whichever server answers first.

>   2. do I have to create a pool? Can I use following configuration 
>for the subnet:
>   
>   subnet 192.168.1.0 netmask 255.255.255.0 {
>     deny dynamic bootp clients;
>       failover peer "dhcp-failover-1";
>     range 192.168.1.101 192.168.1.199;
>}

No, you MUST use a pool, that is part of the syntax.