eigrp clarification

Unanswered Question
Aug 13th, 2007

This is a question about eigrp network statements . Say you have networks , and . What would you use as a network statement ? Can you just use network or does it have to be network , network and network . Under what circumstances do you use a mask or inverted mask with the network statement , trying to compare this to say ospf . Thanks for the clarification , its not real clear even under cisco's config docs.

I have this problem too.
0 votes
  • 1
  • 2
  • 3
  • 4
  • 5
Overall Rating: 0 (0 ratings)
mohammedmahmoud Mon, 08/13/2007 - 14:41

Hi Glen,

With the network command you only control which interface should EIGRP run on, and has nothing to do with the subnet mask of the routes advertised (the routes advertised inherit the subnet mask from the interfaces), and thus its your call how to control the wildcard mask of the network command to control which interface(s) should run EIGRP, personally i prefer running the network command with a wildcard of to indicate certain specific interfaces and to be clear in the configuration.


Mohammed Mahmoud.

Edison Ortiz Mon, 08/13/2007 - 14:51

If you have few interfaces needing EIGRP enable, I recommend using as the network mask. If you have many interfaces, then try to summarize the mask as close as possible, in your case

If you don't use a network mask in the EIGRP process, the command will enter the IP under its class boundary. For instance, if you enter

router eigrp 100


and then you show run, you will see that it will activate EIGRP in all interfaces running under the subnet, not just the subnet. The network mask will aid you on specifying the correct interface to enable EIGRP on, so the correct commands would be

router eigrp 100



Richard Burts Tue, 08/14/2007 - 03:11


Mohammed and Edison have given good answers about use of the wildcard mask with EIGRP. I would like to add just a little to their answers.

First I would observe that with EIGRP use of the mask is optional where with OSPF it is required. In EIGRP the fact that the mask is optional reflects the classful background of the routing protocol. If you enter an address in the network statement without specifying a mask then EIGRP will assume the classful mask of the address.

Second I would attempt to answer your question about advice on how to use the mask that the decision of what degree of specificity to use is quite similar between EIGRP and OSPF. With both protocols you can get very specific and granular (specifying an exact match to an address) using a network statement for each interface, to less specific where you might use a mask that included all interfaces within a particular network, to very general where you have a single network statement that includes all interfaces (

My personal preference tends to be more like Mohammed and to favor more specific masks usage (true for both EIGRP and OSPF). It gives us more control over what is happening, it makes it more obvious in reading the config what is included (and what is not), it gives more flexibility for later when specifications might need to change, but it does make the config longer and requires more typing. Various people have differing opinions on this and sometimes I believe that what is the best choice will vary with the individual circumstances of the project. So if you have principles that you are comfortable with for mask usage in OSPF, then I believe that you can pretty much use the same principles in EIGRP.



mmohanni1981 Tue, 08/14/2007 - 04:41

you can specify the following:



the eigrp will know that is subneted.you would add a wild card mask if you are using different routing protocols for different subnets.


This Discussion