This topic is to discuss the following lesson:
So…administrative distance is a static number? EIGRP is always 90 and OSPF is always 110? Where are the numbers derived?
Those are the default values yes. You can change them globally so that the router prefers one routing protocol over another, for example by changing the AD of OSPF to 80. It’s also possible to change the AD on a prefix level.
videos are excellent. i can understand topics very easily. I hope Rene Molenaar updates network lessons with more videos as new features are introduced in cisco .
why would u need to run more than 1 routing protocol in your network. wouldn’t you either run in todays world purely eigrp or ospf? why both?
If you design a network, you will always use one routing protocol (if possible). Here are some scenarios where you might have to use two routing protocols:
- You are migrating your company network with the network from another company. Maybe you are running OSPF and they are running EIGRP. As a temporary solution you could use redistribution and run both protocols.
- You run OSPF on your internal network and are installing a new site. Your SP offers you a MPLS VPN connection but only supports BGP as the routing protocol.
- You use EIGRP on your network but you want a backup static route. In this case you can play with the administrative distance so that the static route is not installed until the EIGRP route dissapears.
Is there a reason why the administrative distances are chosen to be 110, 120 etc. Any number can be used right?
While you can choose to make the Administrative Distance to almost any number you choose, you should have a good reason for changing, and you should make the change across all routers under your control that use the protocol in question. There are legitimate circumstances where you need to change an AD (often times with multiple redistribution points), but in general this is not done.
The AD number acts as a sort of ranking on how trustworthy the information is from the routing protocol. For example, OSPF routes (AD = 110) are considered far more trustworthy than Internal BGP routes (AD = 200).
But I still don’t understand why OSPF(AD：110) is better than RIP(AD：120) by default, any good reason ?
The default values of the Administrative Distances have been assigned based on the reliability characteristics of the specific protocols. In general, within a network, OSPF is considered more trustworthy than RIP. Why? For several reasons. It is a more efficient protocol, it can deal with a larger number of devices, it converges faster… Similarly, EIGRP is better than OSPF in that it converges even faster having a feasible successor (backup route) immediately available if a primary route goes down etc… In most cases, the actual values of the Administrative Distances won’t have such a great impact on your network. It’s just that a mechanism had to be created that helps determine which route will be used when two routing protocols suggest different paths to the same destination. And this mechanism requires defaults for each routing protocol. And the defaults have been determined based on general reliability and effectiveness of each protocol.
I hope this has been helpful!
Thanks Laz, it kind of helps by giving me a basic concept.
what do you mean by convergence. also under what circumstance should we be using EIGRP. if EIGRP is better then why we do not use EIGRP always?
“Convergence” means the state of a routing domain where stability has occurred. In other words, routes have been learned by the devices that need them, and there are no outstanding requests to learn new information.
Using EIGRP as a routing protocol is a good idea if you have a pure Cisco environment for routing nodes, as EIGRP is a Cisco proprietary protocol. This is also the reason why wouldn’t be able to use EIGRP exclusively. Let’s say you had one business purchased by another. One business might use EIGRP while the other uses OSPF. It isn’t uncommon for multiple routing protocols to be present on a single network.
I would add when it comes to more specific routes, for example a router learns 192.168.1.0/24 from OSPF (AD 110) and 192.168.1.0/25 from RIP (AD 120), the router will choose the more specific route, so the RIP one, despite it has a higher (worse) AD .
So, if we want to reach the host 192.168.1.1 from this router, it will use the /25 route from RIP.
The key thing to remember is that AD only applies when you have two exact routes.
192.168.1.0/24 and 192.168.1.0/25 are not the same routes.
Our network uses static routes and VLANs to direct and separate traffic. I do not believe we have a routing protocol enabled. If I enable EIGRP will my network stop working correctly?
A straight answer is NO, EIGRP or any other dynamic routing protocol won’t break your network and/or change your routing table.
This lesson can be helpful!
@sales2161 is guiding you in the right direction. Whenever you enable a routing protocol on any network where static routes are functioning, even if you start advertising EIGRP routes and destinations between EIGRP routers, no EIGRP routes will enter the routing table. This is because static routes have a smaller Administrative Distance (AD) than EIGRP routes, so static routes will always be preferred and installed instead of EIGRP routes.
Now this is very helpful because you can create your EIGRP topology, put in all your routes and have destinations be shared among EIGRP routers without that affecting your network. Then you can gradually remove static routes from your devices and verify that the correct routes are indeed being installed in the routing table. If anything goes wrong, just revert to your saved config and you’re back to your original state.
Just keep in mind that if there is a route that is not statically configured in the routing table, and EIGRP does share it with other routers, only that route will be added to the routing table. All static routes will always supersede EIGRP routes to the same destinations.
I hope this has been helpful!