Problem with ibgp configuration

Question

I'm trying to set up this configuration for testing purposes, but it's not working at all and I don't know why:

The problem is that R3 is not receiving any learned route from bgp neigbors

The config:

Router R1

interface FastEthernet0/0
 ip address 192.168.35.209 255.255.255.240
 speed 100
 full-duplex
!
interface FastEthernet0/1
 ip address 192.168.34.10 255.255.255.128
 duplex auto
 speed auto
!
router bgp 35
 no synchronization
 bgp log-neighbor-changes
 redistribute connected
 neighbor 192.168.34.3 remote-as 33
 neighbor 192.168.35.210 remote-as 35
 no auto-summary
!

Router R2

 interface GigabitEthernet0/0
     ip address 192.168.35.225 255.255.255.240
     duplex full
     speed 100
    !
    interface GigabitEthernet0/1
     ip address 192.168.35.210 255.255.255.240
     duplex full
     speed 100
    !
    router bgp 35
     bgp log-neighbor-changes
     redistribute connected
     neighbor 192.168.35.209 remote-as 35
     neighbor 192.168.35.226 remote-as 35
     no auto-summary
    !

Router R3

interface FastEthernet0/0
 ip address 192.168.35.241 255.255.255.240
 speed 100
 full-duplex
!
interface FastEthernet0/1
 ip address 192.168.35.226 255.255.255.240
 speed 100
 full-duplex
!
router ospf 1
 no compatible rfc1583
 log-adjacency-changes
 network 0.0.0.0 255.255.255.255 area 0
!
router bgp 35
 no synchronization
 bgp log-neighbor-changes
 redistribute connected
 neighbor 192.168.35.225 remote-as 35
 no auto-summary
!

As I said before, R3 is not receiving routes from BGP neigbors, for example, the route to 192.168.34.0/25 from F0/1 of R1.

ROUTER3#sh ip bgp
BGP table version is 14, local router ID is 192.168.35.241
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
*>i192.168.35.208/28
                    192.168.35.225           0    100      0 ?
*> 192.168.35.224/28
                    0.0.0.0                  0         32768 ?
* i                 192.168.35.225           0    100      0 ?
*> 192.168.35.240/28
                    0.0.0.0                  0         32768 ?
ROUTER_OP1.3#

However, R2 and R1 are receiving all routes:

ROUTER1#sh ip bgp
BGP table version is 292, local router ID is 1.1.35.1
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
*> 10.194.0.0/20    192.168.34.3                           0 33 i
*> 10.197.48.0/20   192.168.34.3                           0 33 i
*> 172.31.48.0/24   192.168.34.3                           0 33 i
*> 172.31.49.0/24   192.168.34.3                           0 33 i
*> 192.168.34.0/25  0.0.0.0                  0         32768 ?
* i192.168.35.208/28
                    192.168.35.210           0    100      0 ?
*>                  0.0.0.0                  0         32768 ?
*>i192.168.35.224/28
                    192.168.35.210           0    100      0 ?
ROUTER1#

ROUTER2#sh ip bgp
BGP table version is 343, local router ID is 1.1.35.2
Status codes: s suppressed, d damped, h history, * valid, > best, i - internal,
              r RIB-failure, S Stale, m multipath, b backup-path, x best-externa
l
Origin codes: i - IGP, e - EGP, ? - incomplete

   Network          Next Hop            Metric LocPrf Weight Path
*>i10.194.0.0/20    192.168.34.3             0    100      0 33 i
*>i10.197.48.0/20   192.168.34.3             0    100      0 33 i
*>i172.31.48.0/24   192.168.34.3             0    100      0 33 i
*>i172.31.49.0/24   192.168.34.3             0    100      0 33 i
*>i192.168.34.0/25  192.168.35.209           0    100      0 ?
*> 192.168.35.208/28
                    0.0.0.0                  0         32768 ?
* i                 192.168.35.209           0    100      0 ?
* i192.168.35.224/28
                    192.168.35.226           0    100      0 ?
*>                  0.0.0.0                  0         32768 ?
*>i192.168.35.240/28
                    192.168.35.226           0    100      0 ?

If you need additional info, please ask for more

Answer 1

The rule for iBGP is that an iBGP peer cannot advertise iBGP routes learned from another iBGP peer. That means that each iBGP peer must be in a full mesh with every other iBGP peer, or you must use a mitigation, e.g. route reflectors or confederations. This is a loop prevention mechanism.

R3 can only learn eBGP routes from R1 because R2 cannot pass on iBGP routes originated by R1 to R3. R3 would also need to directly peer with R1, and that usually involves an IGP. Alternatively, you could configure R2 as a route reflector.