There are countless reasons for configuring routers with loopback interfaces. What is the best way

There are countless reasons for configuring routers with loopback interfaces. What is the best way to address the loopbacks? What sort of addressing scheme would produce the least amount of routing overhead, but still be scalable to grow to a 200-300 node network. I have heard of people allocating an entire class c address space per loopback, all the way down to addressing them with a 32 bit mask. What is the best way?

    Requires Free Membership to View

    Login

You can use whatever addressing policy you are following in your organization. Loopback is mainly used for some processes where an IP is needed for administrative purposes. Like your need to have a router-ID or an IP for IPSEC tunnel association.

A good practice is to have a block and assign IP addresses from that with /32 subnet mask. Care should be taken in assigning those IPs, because if you assign 192.168.1.1/32 to one router and 192.168.1.2 to another, and you have OSPF so 192.168.1.2 will become DR. So, much thought should be given before assigning any IP addresses to loopback interfaces.

This was first published in October 2004

Join the conversationComment

Share
Comments

    Results

    Contribute to the conversation

    All fields are required. Comments will appear at the bottom of the article.
    Back to top