BGP Third party next hop

One of my buddies got confused with this concept when we had this configured ! So I thought i will blog !!

Assume 3 router R1, R2, R3 are in the common subnet. R1 is in AS1001 R2 in AS2002 and R3 in AS3003. R1 peers only with R2 and R3 only peers with R2. So any routes advertised by R1 or R3 will be learnt via R2 in AS2002. But R2 will set the next hop to the router advertising the prefix(because they are in the common subnet).

Lets check R2 bgp table

R2#sh ip bgp
BGP table version is 4, local router ID is 2.2.2.2
Status codes: s suppressed, d damped, h history, * valid, > best, i - internal,
r RIB-failure, S Stale
Origin codes: i - IGP, e - EGP, ? - incomplete

Network Next Hop Metric LocPrf Weight Path
*> 1.1.1.0/24 192.168.0.1 0 0 1001 i
*> 2.2.2.0/24 0.0.0.0 0 32768 i
*> 3.3.3.0/24 192.168.0.3 0 0 3003 i
R2#

1.1.1.0/24 is advertised by R1 and 3.3.3.0/24 bu R3.

Lets check the bgp tables of R1 and R3 now;

R1#sh ip bgp
BGP table version is 4, local router ID is 1.1.1.1
Status codes: s suppressed, d damped, h history, * valid, > best, i - internal,
r RIB-failure, S Stale
Origin codes: i - IGP, e - EGP, ? - incomplete

Network Next Hop Metric LocPrf Weight Path
*> 1.1.1.0/24 0.0.0.0 0 32768 i
*> 2.2.2.0/24 192.168.0.2 0 0 2002 i
*> 3.3.3.0/24 192.168.0.3 0 2002 3003 i

We see 3.3.3.0/24 next hop hbeing directly pointed to R3 instead R2.

R3#sh ip bgp
BGP table version is 4, local router ID is 3.3.3.3
Status codes: s suppressed, d damped, h history, * valid, > best, i - internal,
r RIB-failure, S Stale
Origin codes: i - IGP, e - EGP, ? - incomplete

Network Next Hop Metric LocPrf Weight Path
*> 1.1.1.0/24 192.168.0.1 0 2002 1001 i
*> 2.2.2.0/24 192.168.0.2 0 0 2002 i
*> 3.3.3.0/24 0.0.0.0 0 32768 i

We see 1.1.1.0/24 next hop hbeing directly pointed to R1 instead R2.
To override this we need to issue neighbor next-hop-self command.
You can refer to this RFC for more information

Done playing around with MPLS-VPN

Troubleshooting was the focus of my preparation. I wanted to build the network from scratch and troubleshoot it as I went along. I dint know how to put the faults in place and see what happened. SO I decided to build a network and keep varying routing protocols for the same topology !! I must say I am very satisfied with what I had to do :-)

The topologies i used were pretty simple and they looked something like this;




Right from static routes to rip, ospf, eigrp & bgp were configured. I kept the Provide protocl the standard and moved the customer protocol around and again did the other wayt around.

It was very interesting right before my eyes I made some mistake and hit some config errors. Its was nice to get me up to speed. Well I can tell one thing for sure .. I am ready for the lab as far as this topic goes :-)

Although I dint dwell too much with the mpls vpns I think i know enough to deal with stuff that the lab might throw at me.

All said and done there is one topic that i still haven't understood well. Its the SOO(site of origin ) attribute to prevent loops. I did configure it but still I don't get how it works.

If you have the same value configured on both PE the mpls backbone is not used and only the backdoor is used. But when we have different values configured I still hit the same problem. Not sure if I configured it wrong.

Well now mpls is under my belt, other two topic that i will complete over the week include qos and security. Once I am done with that the fun with full scale labs can start again.

changing cciedash ...

This is my blog and pretty much I call it my diary..I was planning to do a overhaul to the blog and or create a website to start fresh and then changed my mind to keep it.

However I have planned to make it more useful for me and readers alike !! I was going through multiple blogs and planned to include their links in mine. One that caught my eye was that of jennifer huber who is pursuing ccie wireless and one of the first woman(AFAIK) who is blogging.

Apart from that almost all of the other sites listed in network words top15 list are already listed here.

I will be starting to work on the nexus platforms, so personally i have got a lot of studying and learning to do !! Its becoming increasingly difficult to manage time and sometimes 24hours is just not enough !!

Back to the routers now ..

2010 -- A year to forget

Suddenly i feel like i have come out of a coma .... when i turn back and see what happened in 2010 ... well nothing much .. everything left hanging halfway... Failed my ccie yet again... deteriorated health wise ... clearly a year to forget.

2011 is here and things are looking much better !! I am able to think much better ... visualize where and what i want to do in the year ahead ...

I will be moving into a better project so work wise also things will be better and most importantly i will be working better to get my ccie number :-)

First off I am revisiting and playing around with mpls at a larger level so that i can do my troubleshooting faster.

I went through the mpls fundamentals to refresh my concepts, took a paper and drew some topologies that i believe will help me learn, understand and troubleshoot mpls better.