about concept "group", "shared-network", and "subnet",

Bruce Hudson Bruce.Hudson at Dal.Ca
Wed Apr 20 15:38:59 UTC 2011

> I'm going to pipe in here because I'm still a bit confused about the 
> proper use of shared-network. What's the use case for it's proper
> use?

    The "shared-network" is meant to handle a case where you have two
or more IP subnets sharing a single broadcast domain (aka VLAN, etc).
In Cisco terms, you want a shared-network when-ever you have an
"ip address ... secondary" on an interface.

    Incoming DHCP requests are mapped to which-ever shared-network,
which may be implicit in the "normal case", that either the gateway
address or the address of the interface (if no gateway) is in. Given
just the basic DHCP protocol, the server has no way to tell what IP
subnet to use to satisfy a given request. You need some additional
configuration to classify the requests. There are a number of ways to
do that: a priori configuration of MAC addresses, relay options,
subnet selection options, etc.

> Does this appear right or wrong? If it's wrong, any explanation of why 
> moving the server subnet into the vlan-2 S-N would cause it to hand
> out IPs from one of the other subnets, instead of ignoring booting?

    Telling the server that you only have one network (broadcast
domain), without anything else, says that all of the networks are
completely equivalent. Any network can be used to satisfy any
request. The "ignore booting" is irrelevant since it will only be
used if the subnet is used and the lack of a range guarentees it
will not be used for dynamic assignments. 
Bruce A. Hudson				| Bruce.Hudson at Dal.CA
ITS, Networks and Systems		|
Dalhousie University			|
Halifax, Nova Scotia, Canada		| (902) 494-3405

More information about the dhcp-users mailing list