From: John (jgarrison1@austin.rr.com)
Date: Tue Mar 04 2008 - 14:31:43 ARST
For the third time the rip network is a stub. Stub in the networking
nomenclture means that there is only one path in and out of that routing
domain. from the perspective of my ospf domain. That means I'm not
concerned with routing loops caused by routing information entering my ospf
domain at another point, because there is no other point of redistribution
for rip into the ospf domain.
I also stated that at this point I am not concerned with the rip routes that
are being redistributed into the ospf domain. That means we are talking
about a single point of redistribution into the rip domain. I don't know
how much clearer I could be about the topology
----- Original Message -----
From: "Hash Aminu" <hashng@gmail.com>
To: "John" <jgarrison1@austin.rr.com>; "Carlos Alberto Trujillo Jimenez"
<carlos.trujillo.jimenez@gmail.com>; <ccielab@groupstudy.com>
Sent: Tuesday, March 04, 2008 4:48 AM
Subject: Re: redistribution
> Hi,
> your situation will not be clearly understood without the
> topology...i would figure out that u have mutual redistribution in
> more than one point ...in that sense i would have to play with the AD
> not to have the redistributed routes from one end gets back and cause
> routing loop......just a general view though
> HTH
>
> Hash
>
> On 3/4/08, John <jgarrison1@austin.rr.com> wrote:
>> As I said the rip is a stub network. Now when rip is applying a default
>> AD
>> of 109 to all redistributed routes it means that the ospf routes that
>> have
>> been redistributed into rip will have a lower AD than the original ospf
>> routes on the ABSR. For all intents and purposes I am disregarding any
>> issues with rip beig redistributed into ospf at this point. I'm trying
>> to
>> see the logic of making the AD of the redistributed routes lower than
>> that
>> of the originating routing protocol
>>
>> I do get the fact that different topologies will require different
>> methods.
>> I'm trying to understand the logic this specific "tool"
>> ----- Original Message -----
>> From: "Carlos Alberto Trujillo Jimenez"
>> <carlos.trujillo.jimenez@gmail.com>
>> To: "John" <jgarrison1@austin.rr.com>
>> Cc: <ccielab@groupstudy.com>
>> Sent: Monday, March 03, 2008 2:20 PM
>> Subject: Re: redistribution
>>
>>
>> > as you mention "On a couple of occasions" but not in all, why not in
>> > all??
>> > because every topology is different.
>> > Im trying to figure your topology scenario and what I understand is you
>> > have
>> > one rip domain considered STUB (only one exit point). One router is
>> > redistributing rip into ospf.
>> > The decision of changing the administrative distance in rip depends
>> > specifically in the ospf topology.
>> > For example, if you follow the path of a native rip route redistributed
>> > into
>> > ospf and then that same "initially rip native route" come back to the
>> > router
>> > doing the redistribution but from another ospf router there is a
>> > possibility
>> > you HAVE A routing loop, because you know that the TRUE PATH towards
>> > the
>> > native rip route is following the rip domain, not the OSPf domain, but
>> > if
>> > the router doing the redistribution recibes the same prefix by two
>> > rouring
>> > protocols (RIP AND OSPF) it prefers the ospf one, because of its
>> > administrative distance.
>> >
>> > Again, this could happen depending in the physical topology, I assume
>> > in
>> > your phisical topology diagram the OSPF ASBR ROUTER has two paths
>> > towards
>> > the OSPF DOMAIN, and it makes sense, because if it sends and
>> > advertisement
>> > to one path it could recibe that same advertisement from another path
>> > (sourced from another neighbor), and it could belive it has a better
>> > path
>> > to
>> > the rip prefix than the rip path. as a conclusion the LOOP is formed.
>> >
>> > So to prevent that looping you must decrease the A.D in rip in the
>> > router
>> > doing redistribution
>> >
>> > It is better to understand looking at the diagram, following the path
>> > of
>> > the
>> > redistributed routes.
>> > remeber, there are some topologies where a redistributed route maybe
>> > come
>> > back to the router doing redistribution via different path. In that
>> > scenarios, and depending the routing protocols, you may need to change
>> > the
>> > A.D. to prevent a loop.
>> >
>> >
>> >
>> > You are redistributing rip into ospf,
>> >
>> > 2008/3/3, John <jgarrison1@austin.rr.com>:
>> >>
>> >> I seem to have started to get a handle on route redistribution.
>> >> Theres
>> >> one
>> >> thing I keep seeing that I can't make sense of. is a router
>> >> redistributing
>> >> ospf and ripv2. On a couple of occasions I've seen ospf redistributed
>> >> into
>> >> rip with a distance of 109. This means that rip routes will have a
>> >> lower
>> >> AD
>> >> thasn the OSPF routes. Therip network is a stub. what am I missing.
>> >> I
>> >> kind
>> >> of get tags, route-maps and the the AD on external routes, bhut I
>> >> don't
>> >> get
>> >> this.
>> >>
>> >> _______________________________________________________________________
>> >> Subscription information may be found at:
>> >> http://www.groupstudy.com/list/CCIELab.html
>> >
>> > _______________________________________________________________________
>> > Subscription information may be found at:
>> > http://www.groupstudy.com/list/CCIELab.html
>>
>> _______________________________________________________________________
>> Subscription information may be found at:
>> http://www.groupstudy.com/list/CCIELab.html
>>
>
> --
> Sent from Gmail for mobile | mobile.google.com
This archive was generated by hypermail 2.1.4 : Tue Apr 01 2008 - 07:53:52 ART