I always get questions on bgp active state incorrect. I get a quesion on BGP asking why a neighbor is in the Active state, with the answers of:
A. The router will not accept connections from the peer.
B. The router has sent out an active TCP connection request to the peer
C. The router is listening on its server port for connection requests from the peer
D. BGP can exchange routing information in this state.
According to the answer book, the answer is C. What I would like to know is why B is incorrect? In the official CCIE R&S guide, there is a table on page 378 showing that in the Active state, the router listens for a TCP session & Initiates a TCP session, making both B and C correct. So I don't understand.
I think you forgot two states.
a) idle state
b) connect state
c) active state
d) opensent state
e) openconfirm state
f) established state
In regards to the answer I agree with "C". Per the RFC,
In this state, BGP FSM is trying to acquire a peer by listening for, and accepting, a TCP connection.
Hope this helps.
would you please clarify me that if we are using BGP in our enterprise network then is there necessary to configure the default route ???? I am so confused..!
And is there any condition that we are using BGP and still we need to configure static route for the outside network ??
And if there is any documentation that could explain the "next-hop self" concept then plzzz share
There are 2 types of BGP peerings: eBGP and iBGP, the former is between routers in the same AS and the latter is between routers in different ASes.
By default eBGP peers when advertising a prefix always change the next-hop to their update-source address but iBGP peers do not.
So if your edge router receives an eBGP update, the next-hop is the IP address of the other BGP peer which is connected to your router.Then when your border router advertises this same prefix to its iBGP peer it doesn't change the next-hop but this next-hop is not known by any IGP in your AS so it is unreachable.
The first condition for a BGP prefix to be advertised is the next-hop must be reachable otherwise the prefix is not considered as a best route and not advertised by BGP and not installed in the routing table.
To remedy to this situation there is many solutions including the next-hop-self command telling the router to set the next-hop as its update-source towards the iBGP peer and so this next-hop is then reachable in the AS.
Hope this answers your question.
For default-route question we need more infos: iBGP only or also eBGP, which other IGP if any.
You probably should have started a new thread. My apologies to Stephen. However, since you already received one reply, let me try and answer your questions.
(Q1) Would you please clarify me that if we are using BGP in our enterprise network then is there necessary to configure the default route ????
Not always, that depends on if the network is directly connected to you or not. In the case when it is not directly connected to you, then you must have either static route to this network or be learned via an IGP.
When you advertise a network in BGP using the "network" command under the BGP process, you must have an exact match for the network in the IP routing table for BGP to advertise the network. This can be learned via any IGP (OSPF, EIGRP, RIP, etc) or be a static route. One of the easiest methods is to use a static route to null 0. For example, you wish to advertise the following network under BGP,
description EBGP PEER TO AS-65002
ip add 192.168.1.1 255.255.255.252
router bgp 65001
network 172.16.25.0 mask 255.255.255.0
neighbor 192.168.1.2 remote-as 65002
ip route 172.16.25.0 255.255.255.0 null 0
(Q2) And is there any condition that we are using BGP and still we need to configure static route for the outside network ??
Yes, when you are acting as a "transit AS" for a third party AS, then you will need to have a static route to the external outside AS or it must redistribute the external BGP routes into IGP. Not recommended for the complete BGP table.
(Q3) And if there is any documentation that could explain the "next-hop self" concept then plzzz share.
IMHO the best source is the Cisco Press book, Routing TCP/IP Volume II, by Jeff Doyle. It covers BGP really well and you will find some very nice examples. You can also search Cisco's website for "BGP next-hop self command" . As cadetalain has pointed out, you use "next-hop-self" command on iBGP peers only, as this is the "default" behavior for eBGP peers.
Hope this helps.
Message was edited by: Brian