Warewolf Skips DHCP config (on a different network)

[Arash Nemati Hayati]

One of the problems I am facing is to setup the new dhcp server on the cluster. Currently we have a Rock's master's node serving as the DHCP server with certain ip address coverage. We cannot turn down the old master's node or shut down the dhcp server on it before the transition is complete. This limits the leases on certain ip ranges.

I am setting the master's node with the following config:

Ethernet port for communication:

ens2f1: flags=4163<UP,BROADCAST,RUNNING,MULTICAST>  mtu 1500

        inet 10.2.1.1  netmask 255.255.0.0  broadcast 10.2.255.255

One of the new nodes to be added:

eth0      Link encap:Ethernet  HWaddr A0:42:3F:2D:86:98  

          inet addr:10.1.255.60  Bcast:10.1.255.255  Mask:255.255.0.0

provision file on warewolf:

subnet 10.2.0.0 netmask 255.255.0.0 {

   not authoritative;

   # option interface-mtu 9000;

   option subnet-mask 255.255.0.0;

}

As you see, the new node is from a different network (10.1...) and therefore warewolf skips adding that node to the dhcp list. 

How should I rectify this? Should I change the netmask on both master's node and the client's node to 255.0.0.0 to make this work? If so, what else should be changed? 

Any advice on this would be greatly appreciated..

[Arash Nemati Hayati]

Sorry for the long email. I think changing the static ip address of the client's node to 10.2... may resolve the issue.