IBGP peering getting loop

[libra_ali786]

I have 4 routers connected each other by IBGP neighbor ship. Neighborship up with no issue but  each router form loops with indirect connected router.

Between R2 and R4

R4#
*Feb 7 13:14:52.267: %BGP-3-NOTIFICATION: received from neighbor 2.2.2.1 4/0 (hold time expired) 0 bytes
R4#
*Feb 7 13:14:52.271: %BGP-5-ADJCHANGE: neighbor 2.2.2.1 Down BGP Notification received
*Feb 7 13:14:52.279: %BGP_SESSION-5-ADJCHANGE: neighbor 2.2.2.1 IPv4 Unicast topology base removed from session BGP Notification received
*Feb 7 13:14:52.723: %BGP-5-ADJCHANGE: neighbor 2.2.2.1 Up
R4#
*Feb 7 13:15:54.107: %IPRT-3-RIB_LOOP: Resolution loop formed by routes in RIB

R4#show ip route loop
->default:ipv4:base 2.2.2.0/24 -> base 2.2.2.1 bgp 00:13:16 N

Between R1 and R3

R3#
*Feb 7 13:14:16.891: %IPRT-3-RIB_LOOP: Resolution loop formed by routes in RIB
R3#
*Feb 7 13:14:18.611: %BGP-3-NOTIFICATION: received from neighbor 1.1.1.1 4/0 (hold time expired) 0 bytes
*Feb 7 13:14:18.611: %BGP-5-ADJCHANGE: neighbor 1.1.1.1 Down BGP Notification received
*Feb 7 13:14:18.619: %BGP_SESSION-5-ADJCHANGE: neighbor 1.1.1.1 IPv4 Unicast topology base removed from session BGP Notification received

[Georg Pauwen]

Hello,

your design is flawed, as iBGP requires a full mesh between all iBGP routers. You are trying to accomplish that by peering with indirect links, this won't work. You need to have a physical link between each router (meaning each router needs 3 links, one to each of the three other routers). Or use a route reflector.

[libra_ali786]

Sir please look that youtube video , which i apply same topology for my understanding but i am getting routing loop.

https://www.youtube.com/watch?v=X9b0VtSlKwk

[paul driver]

Hello
So you have an IGP running or have you just tried to apply ibgp on all 4 rtrs.

sh ip protocols

[libra_ali786]

Only apply IBGP on all 4 routers.

R1#show ip protocols
*** IP Routing is NSF aware ***

Routing Protocol is "bgp 500"
Outgoing update filter list for all interfaces is not set
Incoming update filter list for all interfaces is not set
IGP synchronization is disabled
Automatic route summarization is disabled
Neighbor(s):
Address FiltIn FiltOut DistIn DistOut Weight RouteMap
1.1.1.2
2.2.2.2
4.4.4.1
Maximum path: 1
Routing Information Sources:
Gateway Distance Last Update
2.2.2.2 200 00:00:08
1.1.1.2 200 00:00:11
4.4.4.1 200 00:00:10
Distance: external 20 internal 200 local 200

[paul driver]

Hello

All 4 rtrs
router eigrp 100
no auto-summary
network 0.0.0.0

If however you not using any igp check you ibgp peerings are correct?

[libra_ali786]

Sir please look that youtube video , which i apply same topology for my understanding but i am getting routing loop. Above I Mentioned the link. 

[MHM Cisco World]

Friend It is so common issue

R1-subnetA-R2

Router ospf 100

Net subnetA

Router bgp 100

Neighbor R2 

You see we not add command net subnetA under bgp this make flap issue.

[libra_ali786]

Hi

Thanks for the reply. Can you check above I Mentioned youtube video link which I trying to simulate. They did not used any IGP protocol.

Only IBGP with network commands. 

[MHM Cisco World]

I am out now later i will check link,

But if you not use routing protocl then use static route carefully.

[libra_ali786]

OK. Problem is not a reachability. When you get time please check the link and reply back. Thanks 

[paul driver]

Hello
Please post the ouput below from each router in a file and attach to your post:

Also make sure you have advertised all interfaces in bgp and each rtr has ibgp peering to all other ibgp rtrs via their physical interfaces

sh ip int brief
sh run | sec router

[libra_ali786]

Please see attachment of  my routers configuration in my post.

Thanks 

[MHM Cisco World]

No need direct link between each peer,

For full mesh run routing protocol like ospf, this make each router know other router ip.

Note:- do not advertise the subnet of direct link connect by bgp, ospf will do that

[Dan Frey · Accepted solution]

R3:  I did a clear ip BGP and when BGP starts to establish BGP sessions with the directly connected neighbors network 1.1.1.0/24 has a NLRI of a directly connected subnet (2.2.2.1).   This enables the BGP establishment to the non directly connected neighor R1.

R3#sh ip bgp
BGP table version is 8, local router ID is 30.1.1.1
Status codes: s suppressed, d damped, h history, * valid, > best, i - internal, 
              r RIB-failure, S Stale, m multipath, b backup-path, f RT-Filter, 
              x best-external, a additional-path, c RIB-compressed, 
              t secondary path, L long-lived-stale,
Origin codes: i - IGP, e - EGP, ? - incomplete
RPKI validation codes: V valid, I invalid, N Not found

     Network          Next Hop            Metric LocPrf Weight Path
 *>i  1.1.1.0/24       2.2.2.1                  0    100      0 i
 * i  2.2.2.0/24       2.2.2.1                  0    100      0 i
 *>                    0.0.0.0                  0         32768 i
 *>   3.3.3.0/24       0.0.0.0                  0         32768 i

Neighbor R1 comes up and changes the next hop IP to 1.1.1.1 to reach network 1.1.1.0/24.   

*Feb  7 14:50:05.196: %BGP-5-ADJCHANGE: neighbor 1.1.1.1 Up 
R3#sh ip bgp
BGP table version is 13, local router ID is 30.1.1.1
Status codes: s suppressed, d damped, h history, * valid, > best, i - internal, 
              r RIB-failure, S Stale, m multipath, b backup-path, f RT-Filter, 
              x best-external, a additional-path, c RIB-compressed, 
              t secondary path, L long-lived-stale,
Origin codes: i - IGP, e - EGP, ? - incomplete
RPKI validation codes: V valid, I invalid, N Not found

     Network          Next Hop            Metric LocPrf Weight Path
 *>i  1.1.1.0/24       1.1.1.1                  0    100      0 i
 * i                   2.2.2.1                  0    100      0 i
 * i  2.2.2.0/24       2.2.2.1                  0    100      0 i
 *>                    0.0.0.0                  0         32768 i
 *>   3.3.3.0/24       0.0.0.0                  0         32768 i

R3 does not have a connected route on 1.1.1.0/24 network so it can not maintain the BGP adj and the holdtimer expires and BGP session goes down.  

*Feb  7 14:53:05.410: %BGP-3-NOTIFICATION: received from neighbor 1.1.1.1 4/0 (hold time expired) 0 bytes 
*Feb  7 14:53:05.411: %BGP-5-NBR_RESET: Neighbor 1.1.1.1 reset (BGP Notification received)
*Feb  7 14:53:05.414: %BGP-5-ADJCHANGE: neighbor 1.1.1.1 Down BGP Notification received
*Feb  7 14:53:05.414: %BGP_SESSION-5-ADJCHANGE: neighbor 1.1.1.1 IPv4 Unicast topology base removed from session  BGP Notification received
*Feb  7 14:53:05.484: %BGP-5-ADJCHANGE: neighbor 1.1.1.1 Up 
*Feb  7 14:56:10.792: %BGP-3-NOTIFICATION: received from neighbor 1.1.1.1 4/0 (hold time expired) 0 bytes 
*Feb  7 14:56:10.793: %BGP-5-NBR_RESET: Neighbor 1.1.1.1 reset (BGP Notification received)
*Feb  7 14:56:10.795: %BGP-5-ADJCHANGE: neighbor 1.1.1.1 Down BGP Notification received
*Feb  7 14:56:10.795: %BGP_SESSION-5-ADJCHANGE: neighbor 1.1.1.1 IPv4 Unicast topology base removed from session  BGP Notification received
*Feb  7 14:56:10.894: %BGP-5-ADJCHANGE: neighbor 1.1.1.1 Up