Router gets pissed off

I tried creating an aggregate summary route just for kicks and this is what I got from the router. Man IOS is getting very bossy these days! Read more

BGP Finished! Still need to work on REGEXP

After spending a week on BGP, I think I have a good understanding of the basic concepts of BGP. There have been a few gotchas and ohsnaps but over all I am satisfied by the way Narbik’s workbook has carried me through. I am still looking at IE Vol 1 but I think I am going to get done with these workbooks before starting on a different topology.

Read more

Cisco Regular Expression Characters

I found this great list of all the characters that we can use. I think this might be helpful for everyone and it’s right from Cisco’s website. Read more

BGP Attributes

AS-Path

We can use AS-Path attribute to influence choice of paths in another AS. Since routers chose the path with the shortest AS-Path, using set as-path prepend command can be used to add additional values of originating AS so the receiving router will choose the router with the least AS-path value.

Read more

BGP Communities

The community attribute is a numerical value that can be assigned to a given prefix using a route-map/access-list combination and then can be advertised to a neighbor. By default community attributes are NOT sent to the neighbors and you must use the “send-community” option with the neighbor statement to enable this behavior. Once the neighbor receives the community, it will either filter that prefix or use the value in route selection process.

Read more

BGP Route Aggregation

We can use route aggregation in BGP to decrease the size of the BGP routing tables. This helps in speeding up the convergence time and improves network performance. Just keep in mind that aggregate route gets forwarded ONLY if there is atleast ONE specific route in the routing table. This is why it is recommend that we always advertise the networks using the network statements. There are a few ways to implement route aggregation in BGP so after reading through documents, I will try to summarize them all. Please read full Cisco Documentation for your benefit. Read more

BGP Route Dampening

Route dampening is a BGP feature designed to minimize the propagation of flapping routes across an internetwork. A route is considered to be flapping when its availability alternates repeatedly. Since BGP routing tables are huge, you don’t want that many routing updates to be traveling all over the place every time a route flaps.

Read more

BGP Backdoor Routes and Conditional Advertisement

We often run into situations where we have two different routes to a network and we know one is faster than the other. However, a router will only keep one route in the routing table and that will be the one with the shortest administrative distance. Now if two routers (R1 and R2) in an AS are directly connected to each other via a 10meg ethernet running EIGRP (Admin distance 90) but they also have a T1 connection to a service provider running EBGP (Admin distance 20). Now, as human beings, we are smart enough to see that if we are R1 and have to reach a network located behind R2, we should use the EIGRP route via 10meg connection. Now think of this from a router’s perspective, R1 sees two routes to that network, via EIGRP and EBGP. R1 is going to take the slower path via EBGP because EBGP’s admin distance is 20. Changing the default administrative distances is not recommended because that may lead to routing loops. So this is where we can use the network backdoor command so router will prefer EIGRP route over EBGP route. BGP treats the network specified by the network backdoor command as a locally assigned network but it doesn’t advertise the network in BGP updates. In short, BGP network backdoor command changes the admin distance of that network to 200 thus making it a worst route compared to EIGRP (90). Read more

BGP Route Reflectors

One of the requirements of BGP is that all IBGP peers MUST be fully meshed. This is not a big deal when you are dealing with 4-5 routers, but in larger networks, this can be an administrator’s nightmare. Route-reflectors take care of this problem. The only requirement is that each IBGP speaker MUST establish a peer session with the route-reflector. It is the route-reflector’s job to pass routes between the route-reflector-clients.

Read more

BGP Peer Groups & Peer Templates

BGP Peer-Groups are used to optimize BGP Networks. If you have a number of BGP peers, Peer-Groups can simplify configuration and updates can be more efficient. Key is that all those peers share the same update policies such as outbound route-maps, distribute-lists etc.

Some of the benefits of BGP Peer-groups:

• Peer-groups provide optimization of BGP convergence. For example, if a BGP speaker has 10 IBGP peers that exchange full BGP routing (100,000 prefixes),  without the use of peer-groups, local router will go through 1 million prefixes. Now if you configure peer-groups, the router has to go through only 100,000 prefixes.
• It provides a mechanism for peers that have an identical outbound policy.
• It can reduce the administrative overhead by cutting down redundant configuration.

Example of Peer-group:

router bgp 100
 no synchronization
 bgp log-neighbor-changes
 network 192.168.1.0
 neighbor TST peer-group
 neighbor TST remote-as 100
 neighbor TST password cisco
 neighbor TST update-source Loopback0
 neighbor TST version 4
 neighbor 2.2.2.2 peer-group TST
 neighbor 3.3.3.3 peer-group TST
 neighbor 4.4.4.4 peer-group TST

As you can see, we defined a peer-group “TST” and then set the parameters for “TST” such as remote-as, version etc. Finally we can use the neighbor statement to apply peer-group TST to all the peers. Read more

Next Page »