Re: EIGRP - Split Horizon

The nuance I was referring to was more the "interface >that the router
itself uses< to reach the destination." A traditional definition of
split horizon would hold that a router is not to advertise a route out
of an interface through with it learned of of that same route.
According to that definition, depending on the timing of things, R3
would advertise R1's Lo0 to R4 - or the other way around - and then
the learning router would not advertise R1's Lo0 back to the other
neighbor. What Dave noticed was that this wasn't actually what was
happening: both R3 and R4 were advertising R1's Lo0 to each other.
This is all perfectly fine and acceptable, because both R3 and R4 are
presumably using a path directly via R2. Now if you go and jack up
the cost on R3 or R4 via R2, you might see a different result...

On Jan 8, 2011, at 3:55 , jules NYA BAWEU wrote:

> I believe the key word is out of the "interface" the route was
> learned from. R4 sure receives the route from R2 also - same route
> but different metric and path - check the topology table, you will
> sure see that they have different metrics
>
> On Sat, Jan 8, 2011 at 10:05 AM, Scott M Vermillion <scott_ccie_list_at_it-ag.com
> > wrote:
> Hey Dave,
>
> There's a nuance in the description of EIGRP split horizon in the
> command ref:
>
> http://www.cisco.com/en/US/docs/ios/iproute_eigrp/command/reference/ire_s1.html#wp1058799
>
> "The split-horizon rule prohibits a router from advertising a route
> through an interface that the router itself uses to reach the
> destination."
>
> I'm guessing neither R3 nor R4 use the other to reach R1's Lo0,
> correct?
>
> Cheers,
>
> Scott
>
>
>
> On Jan 8, 2011, at 10:54 , Dave Serra wrote:
>
> > Hi Guys,
> >
> > I thought I understood the split horizon rule until I built the
> > following topology in GNS:
> >
> > R3---------R4
> > | ----R2----|
> > |
> > R1
> >
> > What I'm trying to depict in the above diagram is a triangle
> > topology between
> > R2, R3, and R4. R1 hanging off of R2 outside the triangle.
> > All routers are running EIGRP on all interfaces. I create a
> > loopback of
> > 1.1.1.1/32 on R1. I then see the route travel from R1 to R2, from
> > R2 to R3
> > and
> > R4 and finally (and most confusingly) from R3 to R4 and from R4 to
> R3.
> > It is this last part that I am having trouble with. When R3 learns
> > the
> > route of
> > 1.1.1.1/32 from R2 and sends it to R4, shouldn't R4 NOT send that
> > same route
> > back to R3 due to split horizon???
> >
> > I show in the 'show ip eigrp
> > top all' on both R3 and R4 that this route is
> > learned from each other.
> >
> > Can
> > someone help me to better understand this?
> >
> > Thanks in advance :)
> >
> > Dave
> > Make a small loan, Make a big difference - Kiva.org
> > ________________________________
> > From: Juan <fferrer10_at_gmail.com>
> > To: Cisco
> > certification <ccielab_at_groupstudy.com>
> > Sent: Fri, January 7, 2011 1:35:34 AM
> > Subject: Test, Please ignore
> >
> > Test
> >
> >
> > Blogs and organic groups at
> > http://www.ccie.net
> >
> _______________________________________________________________________
> > Subscription information may be found at:
> > http://www.groupstudy.com/list/CCIELab.html
> >
> >
> > Blogs and organic groups at http://www.ccie.net
> >
> >
> _______________________________________________________________________
> > Subscription information may be found at:
> > http://www.groupstudy.com/list/CCIELab.html
>
>
> Blogs and organic groups at http://www.ccie.net
>
> _______________________________________________________________________
> Subscription information may be found at:
> http://www.groupstudy.com/list/CCIELab.html

Blogs and organic groups at http://www.ccie.net
Received on Sat Jan 08 2011 - 16:04:37 ART