What we're dealing with here is a problem with Cisco documentation that goes
back centuries, and has never been corrected.
Split Horizon and Poison Reverse are TWO DIFFERENT THINGS. They do not work
together, they are *alternatives*, similar to how ISL and 802.1Q are
alternatives which do pretty much the same thing.
With split horizon, if a router learns about a route from a neighbor on a
certain interface, it will not advertise that same route out that same
interface to that same neighbor.
In poison reverse, the router *will* advertise that same route out that same
interface to that same neighbor, but will jack up its hop count to 15 -
making it automatically unreachable.
Which is better? Not sure. I've heard that Poison Reverse is *slightly*
more efficient than Split Horizon, simply because poison reverse still
forces the receiving router to allocate memory space for the route it
receives, even if its hop count is 15. When the topology changes, all it
needs to do is change the hop count and the next hop address - but the route
itself is already in memory. Here's the funny thing: split horizon is the
default on Cisco routers. Bay routers use Poison Reverse by default.
----- Original Message -----
From: Cisco Boy
To: [EMAIL PROTECTED]
Sent: Friday, May 25, 2001 5:17 AM
Subject: Split Horizon & Poison Reverse [7:5887]
Can someone help explain how Split Horizon and Split Horizon with Poison
Reverse works in correlation with RIP?
---------------------------------
Do You Yahoo!?
Yahoo! Auctions $2 Million Sweepstakes - Got something to sell?
FAQ, list archives, and subscription info:
http://www.groupstudy.com/list/cisco.html
Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
Message Posted at:
http://www.groupstudy.com/form/read.php?f=7&i=5889&t=5887
--------------------------------------------------
FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html
Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
