IBGP-EBGP confusion simple network

karanc

as according to the diagram R3 and R4 are IBGP peers and R4 and R5 are EBGP peers. the advertise route of 50.1.1.0 /24 network is reaching
R3 and even choosen as best route and so in its routing table ..
R3#show ip bgp
BGP table version is 4, local router ID is 10.1.34.1
Status codes: s suppressed, d damped, h history, * valid, > best, i - internal,
r RIB-failure, S Stale
Origin codes: i - IGP, e - EGP, ? - incomplete


Network Next Hop Metric LocPrf Weight Path
*>i50.1.1.0/24 10.1.34.2 0 100 0 6500 i
R3#


R3#show ip route
Codes: C - connected, S - static, R - RIP, M - mobile, B - BGP
D - EIGRP, EX - EIGRP external, O - OSPF, IA - OSPF inter area
N1 - OSPF NSSA external type 1, N2 - OSPF NSSA external type 2
E1 - OSPF external type 1, E2 - OSPF external type 2
i - IS-IS, su - IS-IS summary, L1 - IS-IS level-1, L2 - IS-IS level-2
ia - IS-IS inter area, * - candidate default, U - per-user static route
o - ODR, P - periodic downloaded static route


Gateway of last resort is not set


50.0.0.0/24 is subnetted, 1 subnets
B 50.1.1.0 200/0 via 10.1.34.2, 00:15:39
10.0.0.0/24 is subnetted, 1 subnets
C 10.1.34.0 is directly connected, Serial0/0




But when i am tring to ping 50.1.1.1 my pings are dying
R3#ping 50.1.1.1


Type escape sequence to abort.
Sending 5, 100-byte ICMP Echos to 50.1.1.1, timeout is 2 seconds:
.....
Success rate is 0 percent (0/5)


since R3 is an IBGP ....do i have to advertise the link between R3 & R4(IBGP) to R5 router ?????
or i am missing something with BGP configuration.....


i have setup R4 as next-hop self ......no synronization on all routers
configs of R3 bgp
router bgp 5500
no synchronization
bgp log-neighbor-changes
neighbor 10.1.34.2 remote-as 5500
no auto-summary



config of R4
router bgp 5500
no synchronization
bgp log-neighbor-changes
neighbor 10.1.34.1 remote-as 5500
neighbor 10.1.34.1 next-hop-self
neighbor 10.1.45.2 remote-as 6500
no auto-summary


config of R5
router bgp 6500
no synchronization
bgp log-neighbor-changes
network 50.1.1.0 mask 255.255.255.0
neighbor 10.1.45.1 remote-as 5500
no auto-summary


i just want to confirm is it some configuration mistake ....or R5 should know about the 10.1.34.0 network ....to reach R3???


i am just confuse on the IBGP - EBGP relations........do i have to advertise all my IBGP links...to other AS .....to reach them ????? shouldnt it be taken care by
R4 next hop self command...i did a degub on R5 and saw that pings are reaching R5 with source ip 10.1.34.1.....and this network is not know to R5....shouldnt it be coming from source 10.1.45.1 ???? coz thats the edge router for AS 5500



plz plz clear my confusion ......

networker050184

The pings are not working because R5 has not route back. You will need to advertise all networks you want reachable. From your configs posted below you are not advertising any networks to R5 from R4. BGP is still a routing protocol and you need to tell it what networks to advertise.

kryolla

Usually for BGP you dont advertise the transit link just networks. Put a network in R3 and advertise it in BGP and ping the 50. network source from R3 network and it will work. Another solution is run IGP between R3 & R4 and then redistribute IGP into BGP and R5 will have the transit links. HTH

karanc

thnx for the replies guys.....ok so u have to have the link know to outside AS system ......i just got confuse ....that why i have to advertise my .....internal links to other AS.......

kryolla

you don't have to advertise your internal links to other AS. Put a 30. network in R3 and advertise it into BGP and ping 50. network source from 30. network and it will work

networker050184

kryolla wrote:

you don't have to advertise your internal links to other AS. Put a 30. network in R3 and advertise it into BGP and ping 50. network source from 30. network and it will work

This is more of a work around then actually getting to the cause of why these pings are not working. IMO its important to get the fundamentals down rather than adding more to the equation and not having a base to stand on.

kryolla

networker050184 wrote:

kryolla wrote:

you don't have to advertise your internal links to other AS. Put a 30. network in R3 and advertise it into BGP and ping 50. network source from 30. network and it will work

This is more of a work around then actually getting to the cause of why these pings are not working. IMO its important to get the fundamentals down rather than adding more to the equation and not having a base to stand on.

It is not a work around but how it is done in the real world. In the real world you don't advertise you internal links out your AS to other companies. I do agree with the you that he need to know the fundamentals on why his ping is failing.

networker050184

I understand its how its done in the real world and why (I'm sure you know too) but I don't think the OP understands the fundamentals of BGP and just adding more networks and redistributing an IGP isn't going to help the OP with the basics.

I was going to say just use NAT on R4 which would have fixed the issue as well, but wouldn't have helped in understanding.

tech-airman

karanc,

What network environment are you using? Actual Cisco routers? Some kind of network simulation program? If so, which one?

karanc wrote:

as according to the diagram R3 and R4 are IBGP peers and R4 and R5 are EBGP peers. the advertise route of 50.1.1.0 /24 network is reaching
R3 and even choosen as best route and so in its routing table ..

Which router is network 50.1.1.0/24 connected to?

karanc wrote:

R3#show ip bgp
BGP table version is 4, local router ID is 10.1.34.1
Status codes: s suppressed, d damped, h history, * valid, > best, i - internal,
r RIB-failure, S Stale
Origin codes: i - IGP, e - EGP, ? - incomplete


Network Next Hop Metric LocPrf Weight Path
*>i50.1.1.0/24 10.1.34.2 0 100 0 6500 i
R3#

For IBGP, what is the logical topology requirement? How is this logical topology requirement met by using a route reflector?

karanc wrote:

R3#show ip route
Codes: C - connected, S - static, R - RIP, M - mobile, B - BGP
D - EIGRP, EX - EIGRP external, O - OSPF, IA - OSPF inter area
N1 - OSPF NSSA external type 1, N2 - OSPF NSSA external type 2
E1 - OSPF external type 1, E2 - OSPF external type 2
i - IS-IS, su - IS-IS summary, L1 - IS-IS level-1, L2 - IS-IS level-2
ia - IS-IS inter area, * - candidate default, U - per-user static route
o - ODR, P - periodic downloaded static route


Gateway of last resort is not set


50.0.0.0/24 is subnetted, 1 subnets
B 50.1.1.0 200/0 via 10.1.34.2, 00:15:39
10.0.0.0/24 is subnetted, 1 subnets
C 10.1.34.0 is directly connected, Serial0/0




But when i am tring to ping 50.1.1.1 my pings are dying
R3#ping 50.1.1.1


Type escape sequence to abort.
Sending 5, 100-byte ICMP Echos to 50.1.1.1, timeout is 2 seconds:
.....
Success rate is 0 percent (0/5)


since R3 is an IBGP ....do i have to advertise the link between R3 & R4(IBGP) to R5 router ?????

For IBGP, what is the recommended neighbor IP address that you should use?

karanc wrote:

or i am missing something with BGP configuration.....


i have setup R4 as next-hop self ......no synronization on all routers

In which case do you use "next-hop self" IBGP or EBGP? Why do you need to use "next-hop self?"

karanc wrote:

configs of R3 bgp
router bgp 5500
no synchronization
bgp log-neighbor-changes
neighbor 10.1.34.2 remote-as 5500
no auto-summary



config of R4
router bgp 5500
no synchronization
bgp log-neighbor-changes
neighbor 10.1.34.1 remote-as 5500
neighbor 10.1.34.1 next-hop-self
neighbor 10.1.45.2 remote-as 6500
no auto-summary


config of R5
router bgp 6500
no synchronization
bgp log-neighbor-changes
network 50.1.1.0 mask 255.255.255.0
neighbor 10.1.45.1 remote-as 5500
no auto-summary


i just want to confirm is it some configuration mistake ....or R5 should know about the 10.1.34.0 network ....to reach R3???

Which BGP attribute is responsible for routing packets between and within BGP Autonomous System?

karanc wrote:

i am just confuse on the IBGP - EBGP relations........do i have to advertise all my IBGP links...to other AS .....to reach them ?????

You have to understand the architecture of BGP. BGP speakers belong entirely within Autonomous Systems. The links between the routers are just a transit network between Autonomous Systems (EBGP) or between routers within an Autonomous System (IBGP). So your "dying ping" problem to the serial links is a direct feature of BGP and not a problem that it seems to you.

karanc wrote:

shouldnt it be taken care by
R4 next hop self command...i did a degub on R5 and saw that pings are reaching R5 with source ip 10.1.34.1.....and this network is not know to R5....shouldnt it be coming from source 10.1.45.1 ???? coz thats the edge router for AS 5500

If you initiate a ping from R3 trying to reach R5, then it makes perfect sense that the source IP address would be 10.1.34.1 and NOT 10.1.45.1 because even though the packet is being routed by R4 doesn't mean it suddenly changes the original source IP address, unless something like NAT was involved which would actually change the source IP address. Since NAT is NOT (pun intended) involved, then a ping packet from R3 is going to retain a source IP address of 10.1.34.1. Now, this is also related to how BGP works. What is the relationship between the source IP address of a packet and the neighbor statement in the BGP routing protocol configuration?

karanc wrote:

plz plz clear my confusion ......

dtlokee

debug ip packet on R5 and ping again from R3 to R5.

karanc

to tech-airman:

this is the topology i created on GNS3 ......50.x.x.x network is simply a loopback on R5 which i have redistributed in BGP .....

my question here is ....in BSCI level ...thier is no route reflector ...so i have no idea about it ....all what i want to ask was ....do we have to advertise our internal networks to other AS to be able to reach them ???

or its only by something call route reflector this thing is possible ...otherwise we have to have advertise the local AS link .

networker050184

Like I sated before, any networks you want reachable you must advertise. This is basic low level routing. If a router does not have a route for a network the packet will be sent to the bit bucket.