Visitors to VTC.com will be able to view all introductory videos for each training course.
Free Trial Members will gain access to first three chapters for each training course.
Full Access Members have full access to VTC.com�s entire library of video tutorials.
So we're coming down to the end of our discussion about BGP and in this last section we're going to talk about Full-Mesh BGP and why it's important to you. Now you'll notice on the Core 1 Router we still have that Static Route pointing everything out to Access 1. And again the more astute among you may have realized that, that might be a bit of a problem, just simply because what if the networks you're trying to access are best accessed or only accessed through Access 2. For example, this 150.10.1.1 which happens to be the Loopback address on this ISP 2 Router. If we look here on Core 1, we notice that if we show his Routing Table he doesn't know anything about a 150. whatever network. All he knows is 192.168.10.20 and a Static Route. So if this truly is the Aggregation Point for our network, if this is truly the core of our network and obviously the core of a really big production network would be more than one Router, but bear with me, I only have so much memory in CPU in my, my lab system over there. If we were to go into Core 1 and we were to try to Ping 150.10.1.1. what would happen? Ping 150.10.1.1, it fails. Why does it fail? Well let's do a Trace Route and see where the breakdown happens. So we'll do Trace 150.10.1.1. It goes to 10.20.1.2. 10.20.1.2 sends it right back to 10.20.1.1 and it just goes in a Routing Loop between these two Routers. The reason why is that Core 1 sends it to Access 1, Access 1 says well I don't have the best routes to get to that network, why don't you go over to Access 2. Your next hop is going to 10.20.2.2. So it sends it right back to Core 1. Core 1 says, no, my Static Route says to go to you, I don't have any, any routes for this 150. whatever network and they just go back and forth, kind of like it did when we were discussing the Weight and Lower Preference Attributes earlier in this section. Now as I've said constantly throughout the course, there's multiple ways to fix this problem. You could get really fancy with your Static Routes and put Static Routes for these individual networks into the Core 1 Router so that it would send the traffic the right direction. That's probably the worst idea you could do, but, it, you could do that. If these Access 1 and Access 2 Routers were sitting in the same data center, you could just connect a cable between the two and that way Core 1 would send it to Access 1. Access 1 would send it right over this other link to Access 2 and the traffic will go out that way, would not be the most efficient path to get out of the network, but it would still work. However whatever if Access 1 is in data center and Access 2 is another data center for Redundancy or Disaster Recovery purposes? You obviously just can't connect them with a big long Ethernet cable, well not without involving a whole bunch of utilities of Rights of Way, a lot more cost prohibitive than you probably want to get into. Well another option is to Redistribute BGP into OSPF. That would get all of these BGP routes and send them into OSPF and then OSPF would determine well, I need to send these routes to one of these two, Access 1 or Access 2 Routers and you'd have to get really funky with the Metrics and the Weights and all the other nice Attributes as you Redistributed them. Also it's generally not suggested by Cisco for the simple matter that if you're getting more than just a Default Route from your ISPs down here, then your IGP in this case, OSPF will have a huge Routing Table. And OSPF is not optimized for scalability. OSPF is optimized for quick convergence. So therefore you're just going to overload this Core 1 Router with this huge IGP Routing Table and you're going to slow down the entire network. The best practice is to set up Core 1 as a BGP Peer so that it gets routes from Access 1 and Access 2. It participates in EBGP and Autonomous System 12 and therefore it knows about all of the best routes and how to get to each of these systems and the networks that are attached to them. And obviously you don't have to run BGP on every Router in the enterprise. You can do it at the Core network where all of your network connections aggregate. In most Wide Area Networks, if you have a redundant Internet connection such as this, all of your Internet traffic will aggregate back at your corporate office or wherever these Internet connections happen to be, if you in, in a DR scenario then it may at your DR location. But it will aggregate at one location and those Core Routers are the ones that need to participate in BGP. So let's fix this problem. So we'll simply go into Core 1, we'll do Show Run and remove the Static Route first off. There's the IP route right there, so we'll just highlight it, highlight all of it and go into Config Mode and do No IP Route that. Now from this point on it's basically BGP and Neighbor Commands, we don't advertise networks here at Core 1 because they're already being Redistributed at Access 1 and Access 2. We could change all of that if we wanted to and Redistribute everything at Core 1 or just not do any Redistribution. But we're going to change the configuration as little as possible so we'll do Router BGP 12 and first thing we'll probably need to do, is do Show CDP Neighbor just so I can get the IP address of the remote system. That doesn't help me. Let's just look back over here at my list, so it's 10.20.1.2 and 10.20.2.2. So I've set both of those up on Core 1. We'll set up the Access 1 and Access 2 Neighborships in the next video.
| Course: | Implementing Cisco IP Routing (642-902 ROUTE) |
| Author: | Greg Dickinson |
| SKU: | 34291 |
| ISBN: | 978-1-61866-028-2 |
| Release Date: | 2011-12-28 |
| Duration: | 10 hrs / 105 lessons |
| Work Files: |
Yes |
| Captions: | No |
| Compatibility: |
Vista/XP/2000, OS X, Linux QuickTime 7, Flash 8 |