From: Gene_Thorne@doh.state.fl.us
Date: Sat Oct 25 2003 - 11:01:36 GMT-3
Yesterday I posted a question regarding RIP on several interfaces. Thanks to
all who replied, but it appears that I did not do a good job of wording my
question. I will try again.
When you enable RIPv2 on an interface, 3 things happen by default:
1. The subnet associated with that interface enters the RIP database and
becomes eligible to be advertised by RIP interfaces.
2. The interface sends out RIP updates with the routes in the database
(subject to split-horizon, etc.)
3. The interfaces listens to and processes incoming RIP updates.
Now suppose I have 3 interfaces on different subnets of a class B. The lab
requirements say to enable RIPv2 on one of the interfaces. Of course, when I
enter the network statement RIP is enabled on all 3 interfaces. But there is
another requirement -- "Do not enable RIP on any other interface". The obvious
answer is to make the other 2 interfaces passive. But that only stops #2 on my
list. What about #1 and #3? I know of no way to stop the subnet on a passive
interface from entering the database, but I could stop it from being
advertised using route filtering on the active interface. Similarly, I could
use route filtering on the passive interfaces to stop routes learned on those
interfaces from entering the RIP database.
So the real question is-- what does it mean to "not enable RIP" on an
interface? Is passive-interface enough? Is route-filtering (or something else)
also needed? Or, as I am beginning to suspect, am I simply reading too much
into this requirement?
-Gene
This archive was generated by hypermail 2.1.4 : Mon Nov 24 2003 - 07:53:08 GMT-3