[c-nsp] mpls and BGP question

[Darryl Dunkin]

This may be obvious, but have you fully meshed BGP between all of your
routers? Is PE1 peering with PE3? If fully meshed, you should not have
to bother with route reflectors in this setup.

-----Original Message-----
From: cisco-nsp-bounces at puck.nether.net
[mailto:cisco-nsp-bounces at puck.nether.net] On Behalf Of Don Hickey
Sent: Tuesday, October 23, 2007 10:56
To: cisco-nsp at puck.nether.net
Subject: Re: [c-nsp] mpls and BGP question

Well,

The reasons the pings were timing out was due to the load on the
computer 
running dynamips. I moved it to a quicker computer and my pings go
through.

I still have questions about need a Route Reflector, or maybe someone
could 
suggest a better way to accomplish this.

Thanks

Don Hickey


----- Original Message ----- 
From: "netman" <netman at oneidatel.net>
To: <cisco-nsp at puck.nether.net>
Sent: Monday, October 22, 2007 8:16 PM
Subject: [c-nsp] mpls and BGP question


> Hi,
>
> I have the following situation I am trying to configure.
>
> We have eight 7604 routers all connected point-to-point via 10 gig
links
>
> so something like this (10 gig network)
>
> R1--R2--R3--R4--R5--R6--R7--R8
>
> Hanging off of each router is a companies cisco or foundry router
(CE). 
> The
> current 10 gig network supports IPTV (multicast traffic) using private
> addresses. OSPF routing is configured on the 10 gig network
>
> Right now there are multiple T1's coming from the CE's to the Internet
> Gateway. All routes are handled by static routes  (not my doing)
>
> I need to move the intenet traffic off of the T1's and over to the 10
gig
> network.
>
> I connected an interface on the Internet gateway router (IGW) to R5. I
> cannot use sub-interfaces on the connection between R5 and the IGW. In

> fact
> they do no want me messing with the configuration on the IGW.
>
> R1--R2--R3--R4--R5--R6--R7--R8
>                                |
>                            IGW
>
> I want to use MPLS to carry our Internet traffic from all 8 companies.
I 
> was
> thinking of one vrf for all eight companies)
>
> I have been playing around with dynamips, since I cant experiment on a
> production network. I simulated what I wanted to do with only 5
routers
> (CE--PE1--PE2--PE3--IGW)
>
> I was having problems getting the default route from PE3-IGW through
BGP 
> to
> PE1. It was showing up on PE2.
>
> on CE I created a loopback interface of 2.2.2.2/24
> on IGW I created a loopback interface of 1.1.1.1/24
>
> on IGW I had a static route to 2.2.2.2 via the interface on PE3 facing
the
> IGW.
> on CE I had a default route pointing to the interface on PE1 connect
to CE
> on Pe1 I had ip route vrf INTERNET 2.2.2.2 255.255.255.0 to IP address
of 
> CE
> interface
> on PE3 I had the default route ip route vrf INTERNET 0.0.0.0 0.0.0.0
> serial1/0 IGW's ip address
>
> I could see the default route on PE2 when I did a sho ip route vrf 
> INTERNET.
>
> However I could not see it on PE1 issuing the same command.
>
> Reading somewhere on the Internet I made PE1 and PE3 route reflector 
> clients
> for PE2.
>
> I now had the route at PE1 .
> I turned on debug ip icmp on all routers
>
> I did an extended ping on CE to IGW using the ip address to 1.1.1.1
(IGW
> loopback) and from 2.2.2.2 (CE loopback)
>
> The debug on router IGW - showed it sent a reply src 1.1.1.1 dest
2.2.2.2
> the debug on router CE - showed it received a reply from src 1.1.1.1
for
> dest 2.2.2.2
>
> However the success rate on CE was 0 %. (even though the debug showed
it
> received the reply)
>
> So here is my question.
>
> On the example above do I need to have a router reflector for  this to

> work?
>
> If so, if PE5 was the RR, would I need to include all the other PE
routers
> as clients?
>
> Finally, is there an easier way to accomplish this task. All 8
companies
> share the costs of the 10 gig network and bandwdith to the Internet.
So it
> is a privately owned network.
>
> I was wanting to use MPLS for our Internet traffic, because in the
future
> their might be other companies that use this network as a transit from
one
> location to another. I would like to keep it seperate from our
traffic.
>
> I have no problems reading if someone were to have a nice link to send
me. 
> I
> have been working on this for too long and would like to knwo if I am
on 
> the
> right track or if there is a better way to implement a solution.
>
> Thanks
>
> Don Hickey
>
>
> _______________________________________________
> cisco-nsp mailing list  cisco-nsp at puck.nether.net
> https://puck.nether.net/mailman/listinfo/cisco-nsp
> archive at http://puck.nether.net/pipermail/cisco-nsp/
> 

_______________________________________________
cisco-nsp mailing list  cisco-nsp at puck.nether.net
https://puck.nether.net/mailman/listinfo/cisco-nsp
archive at http://puck.nether.net/pipermail/cisco-nsp/