OSPF area configuration best practices

The assignment of router’s interfaces into OSPF areas should be a non-issue these days, but for whatever reason some of the students I’m mentoring still use the ridiculous practice that was promoted in older learning materials: a separate network statement using IP subnet and inverse subnet mask for every single interface. I’ve documented what I consider to be best practices in the “OSPF area configuration best practices” article in the CT3 wiki. If you disagree with my opinion, please feel free to edit my article or share your thoughts in a comment to this post.

Read more in the CT3 wiki

4 comments:

  1. Back when I was teaching CCNP courses some years ago, I had a student who got really upset when I told him there was no good reason to use the matching inverse-mask method. He simply refused to believe that any other method was acceptable, and questioned my competence for even suggesting anything else. I thought he was a little too much emotional investment in IOS configurations. :-)

    ReplyDelete
  2. I just have a slight quibble with one of your recommendations. In the second to last configuration, you recommend making sure that your links are in separate IP ranges, and using the network command to match those ranges, but in the last (complex) example, you recommend using the area command on the individual interfaces. Personally, I have always tried to make sure that I design the network to fit that middle scenario - that way, when I add a new site (and corresponding interface to the HQ/Hub router), the new interface is automatically in the right area. This discipline also helps keep the network summarizable, to keep the size of the routing tables down. There may be some exceptions, but I think it's better to treat those as exceptions, than to abandon good design just because the network is 'complex' (especially since almost everyone thinks their network is complex).

    ReplyDelete
  3. @js: I've also seen students like that :)

    @Anonymous: You're absolutely correct. I've reworded the "complex scenarios" section. This problem might arise if there's no clear IP addressing design (for example, nobody was willing to pay for it :), if the network has grown "organically" or if the network is being migrated from another routing protocol (for example, from EIGRP) to OSPF.

    ReplyDelete
  4. Pls provide the best book to easily understand OSPF,BGP and IP/MPLS.am working with service provider in Dubai

    ReplyDelete

You don't have to log in to post a comment, but please do provide your real name/URL. Anonymous comments might get deleted.

Ivan Pepelnjak, CCIE#1354, is the chief technology advisor for NIL Data Communications. He has been designing and implementing large-scale data communications networks as well as teaching and writing books about advanced technologies since 1990. See his full profile, contact him or follow @ioshints on Twitter.