From: Marvin Greenlee (marvingreenlee@yahoo.com)
Date: Fri Sep 22 2006 - 16:16:05 ART
Basic MPLS may be another option in this case.
R1 (global)
mpls ip
R1 (interface to R2)
mpls ip
R2 (global)
mpls ip
R2 (interfaces to R1, R3)
mpls ip
R3 (global)
mpls ip
R3 (int to R2)
mpls ip
In a basic setup like this, you don't necessarily need
to worry about configuring other things like hardcoded
tag IDs, etc.
Looking at the various options (MPLS, PBR, tunnel,
redist bgp into IGP), I think that MPLS is the easiest
to configure as long as you understand the basic
concept. R2 won't have direct reachability for the
BGP networks but will pass the traffic from 1 to 3
instead of dropping it. Other than making sure CEF is
on, it's pretty straightforward.
For those that want to complain about MPLS not being
on the R&S blueprint, tunnels and PBR aren't on the
blueprint either, but are covered by the line "other
relevant or related topics may also appear".
Read the lab very carefully, especially the first page
with regards to whether or not adding addresses is
permissible for that particular lab.
Thanks,
Marvin Greenlee
--- Arun Arumuganainar <aarumuga@hotmail.com> wrote:
> Hi Group ,
>
> And I just wanted to get a couple of quick
> clarification. I would be glad if
> you can help me with this questions.
>
> 1) I ofter come across a situation where in a BGP is
> configured across non
> BGP routers . A scenario like this
>
> R1---------R2--------R3------BackBONE
>
> Here ibgp is configured between R1 and R3 and R3 is
> connected to Backbone via
> eBGP. R2 is a non BGP router .
>
> This is will result in black holing of BGP traffic
> at R2 . What is the best
> way of handling the situation . Actually I am
> thinking of two solution .
>
> a) PBR on R2's Interface to R1 ==> Match bgp
> networks and set ip next hop R3
>
> b) Redistribute BGP in to IGP ( probably a summary )
>
>
> Are my approaches right !!! I have seen a RULES in
> first sheet which says do
> not use PBR unless other wise it is specified . Also
> Redistributing bgp in to
> IGP is not recommended by Standard Network design
> practice . What would you
> recommend for me ( are there any other ways to avoid
> this Black Hole )
>
> 2) Is creation of Tunnel interface without addition
> of a new IP address is
> allowed in LAB . I saw a question in Check IT lab
> where in ODR needs to
> configured over routers that are connected across a
> CAT switch . I also
> happened to see this in many of my practice
> Multicast labs . But not sure if
> creation of tunnel interface is legal from CCIE LAB
> Perspective ??? Any
> comments...
>
> 3) With out violating NDA , can you pls tell me
> ...Whether the Goals and
> Restrictions Section in the commercially available
> and popular mock labs such
> as CHECKIT LABs contains all the restrictions that
> would be specified in the
> LABs !!!
>
> Pls. Note : You can ignore this question if you feel
> this will violate NDA .
> But essentially I wanted get my self familiarized
> with LAB environment ...
> Here My goal is to avoid spending much time in
> reading these general
> guidelines rather than the actual questions provided
> in the LAB Question Paper
> .
>
> I would greatly appreciate your earliest response .
>
> Thanks and Regards
> Arun
>
>
This archive was generated by hypermail 2.1.4 : Sun Oct 01 2006 - 16:55:41 ART