Troubleshooting BGP Route Advertisement

(Rene Molenaar) #1

This topic is to discuss the following lesson:

(vijitharan v) #2

thank you for your lesson, can you explain why we use command bgp log-neighbor-changes no synchronization

(Andrew P) #3

bgp log-neighbor-changes
This command simply causes a message to displayed to the console or in the event log that a status has changed with one of your established BGP neighbors. This is very useful for an administrator to know.

no synchronization
This can be considered a legacy command now, because in modern IOS the “no synchronization” is on by default. The explanation of this is a bit long. Your best bet is to review the synchronization lesson.

In a nutshell, the purpose for Synchronization rule was back when many internal routers didn’t have the CPU and RAM capable of running BGP. If you have a BGP router advertising a network, but the less powerful routers inside your network don’t know about the route being advertised, if traffic needs to cross your internal network, it would be dropped. The Synchronization rule would only allow BGP to advertise a network if that network was already known by an internal routing protocol (like RIP, OSPF, EIGRP, etc). The idea was that if the route is known by an IGP, you wouldn’t act like a “black hole” for the traffic.

Most modern routers can easily run BGP now (so long as they don’t get the full Internet table!), so the idea of a “Synchronization Rule” is outdated.

--Andrew

(Ajay Krishna M) #4

Hi Rene,

How to advertise 10 interface IP’s in BGP using single command in BGP. do we have any syntax for this.

Regards,
Ajay

(Rene Molenaar) #5

Hi Ajay,

Not really, unlike RIP, OSPF or EIGRP…BGP requires specific network commands. If it’s only for 10 network addresses then it’s best to use notepad and some copy/pasting or use the “up arrow” on the console for the previous command and change the address.

Rene

(Mark R) #6

Hello,

I think the sentence above “Lesson learned: You can’t advertise what you don’t have. Create a static route and point it to the null0 interface to create a loopback interface that has a prefix that falls within the summary address range” should read:

“Lesson learned: You can’t advertise what you don’t have. Create a static route and point it to the null0 interface OR create a loopback interface that has a prefix that falls within the summary address range”.

Or am I still missing something?

Great lesson as always!

Warm regards

Mark

(rohit v) #7

Yes, i think you are right Mark.

(Andrew P) #8

Thanks Mark and Rohit. I will ask Rene to get that corrected.

(rohit v) #9

Hi Rene,
I really appreciate your nice explanation. I love the way you explained the things. I have one confusion in below statement.
"When two EBGP routers that are directly connected do not form a working BGP neighbor adjacency there could be a number of things that are wrong:

_Layer 2 down preventing us from reaching the other side._
_Layer 3 issue: wrong IP address on one of the routers._
_Access-list blocking TCP port 179 (BGP)._
_Wrong IP address configured for BGP neighbor router."

What do you mean by layer 2 down preventing us from reaching the other side? What could be the layer 2 issues except interface down?

(Rene Molenaar) #10

It could be any issue on layer 2…an interface that goes down but a MAC address filter that prevents two routers from communicating with each other is also a layer 2 issue :slight_smile:

(sachin v) #11

Hello
I have a doubt,will it successfully advertise, if we only have a default ROUTE for any network,without the above mentioned static route to null0 or same network loopback creation.

(Lazaros Agapides) #12

Hello sachin

You could add the network 0.0.0.0/0 command on the BGP configuration, however this would only advertise the default route to the other routers. It would not allow R1 to advertise 172.16.0.0/16 to R2. You need the specific destination and subnet mask in the routing table using the two options that are described in the lesson. It would probably be very beneficial if you tried it out in a lab to see first hand how BGP behaves with these configuration commands. If you do try it, let us know your results!

I hope this has been helpful!

Laz

(sachin v) #13

hi laz
In case if we want to use bgp for outbound traffic,what will be the difference in bgp config(for normal internet browsing)?is it practical for customer routers ?

(Lazaros Agapides) #14

Hello sachin

BGP is very specific in what it advertises. it will only advertise what is in the routing table exactly as it is stated in the routing table including prefix.

If you want to use BGP to advertise a default route for outbound traffic, there are several ways to do this:

**network 0.0.0.0 command** - this will inject the default route into BGP only if this same default route is currently present in the routing table.

**redistribution from another routing protocol** will also inject the default route into BGP, again, only if it is currently in the routing table and if it has been learned from the specific routing protocol you are redistributing from.

using the default-information originate command which causes the default route to be artifically generated and injected into the BGP RIB even if it is not in the routing table. It will be advertised because it is in the RIB.

finally you can use the neighbor X.X.X.X defaut-originate command which is similar to the previous command except that the default route will be advertised only to this specific BGP neighbour and not to all BGP neighbours.

I hope this has been helpful!

Laz

(Brian C) #15

I was reading through this and then got to the trouble section on BGP Summarization. I had never really thought about it before as I have not had to configure it in the real world though I am sure I have came across it but was no problems with it so did not take any notice.

I studied CCNP Switch and Route and I don’t ever remember coming across summarization for BGP. I then did a DDG (DuckDuckGo my google replacement search engine) search on a line of code you had in there called aggregate-address.

Now normally when I thought about aggregate I thought about LACP or interfaces aggregated together so this was a bit confusing except the trouble shooting portion said summary so I had a bit of an inkling of where this was going.

I had to search on this site and found the following link: https://networklessons.com/bgp/bgp-aggregate-as-set/

is this not a CCNP or a CCIE concept do you have to study BGP specifically to pick this up academically?

Anyway this snuck up on me but figured I would make a post about it because I am always thinking out loud.

(Lazaros Agapides) #16

Hello Brian

According to the exam topics for the CCNP ROUTE exam, BGP route summarization is not explicitly mentioned. However, it does affect individual route BGP attributes, and this is important to know. For example, when the aggregate-address command is used, the AS_PATH among other attributes, is lost.

Ultimately, it is great to know because when you work with BGP it just add one more parameter that you can “play” with when configuring a production network. It may not be that important for certification purposes, but more for the real world usage of the protocol.

I hope this has been helpful!

Laz

1 Like
(Regis jean T) #17

It is so easy when you have videos supporting you courses that is what attracted me here in the first place.
this bgp tshoot needs a video in CLI showing all the steps

(Rene Molenaar) #18

Hi Regis,

I agree, I hope to add a video to each lesson sometime. That’ll take quite some work though.

Rene

(Regis jean T) #19

Thank you for replying Rene.
I understand how time consuming this could be to make videos for great topics.
Your website has a lot of good contents.
The videos are easy to follow and allow me to follow your labs.
Reading the content can be challenging and that is why I sent my last comment.
Great work on all your topic but the videos are what brought me here in the first place.

1 Like