I would say it is based by scenario but general rules will be:
1. Split horizon is only possible with distance-vector protocols, as far as
RS LAB is concerned it is only RIP and EIGRP.
2. OSPF has no concept of split horizon.
3. Depends of default split horizon setting of the interface. By default:
Serial int (physical): disabled, this is why we do not disable split h when
physical interface is used in FR.
Serial p2p sub int: enabled (OK, no need to change it)
Serial multipoint sub int: enabled (in most cases needs to be disabled)
In your topology there is no need to disable split h.
------------------------------------
PS : I reserve the right to be wrong.
From: nobody_at_groupstudy.com [mailto:nobody_at_groupstudy.com] On Behalf Of
--Hammer--
Sent: 19-Oct-10 11:14
To: 'Cisco certification'
Subject: Frame Relay and Split Horizon rule of thumb for routing protocols
Struggling here. Is there a rule of thumb for this? Given the below topology
where R2 is the hub
R3 (spoke) -----R2 (hub) ----- R4 (spoke)
Is there a rule of thumb for when I should consider disabling split horizon?
Like, if it's distance vector you disable split, if it's link state you
don't. Or is it unique to each routing protocol and how it's applied? So
you'll have to make a judgement call based on the lab scenario.
--Hammer
"I was a normal American nerd."
-Jack Herer
Blogs and organic groups at http://www.ccie.net
Received on Tue Oct 19 2010 - 17:55:42 ART
This archive was generated by hypermail 2.2.0 : Mon Nov 01 2010 - 06:42:06 ART