A remote operation center is deploying a set of I-BGP and E-BGP connections for multiple IOS-XR platforms using the same template.
The I-BGP sessions exchange prefixes with no apparent issues, but the E-BGP sessions do not exchange routes.
What causes this issue?
Click on the arrows to vote for the correct answer
A. B. C. D.B.
The issue described in the question is that the E-BGP sessions are not exchanging routes, while the I-BGP sessions are working fine. This indicates that the problem is specific to the E-BGP sessions and not a general routing issue.
Answer A suggests that mistyped I-BGP neighbors could be the cause, but this is unlikely as the I-BGP sessions are working fine. Answer B suggests that the customer platform's default policy is blocking E-BGP information exchange, but this is also unlikely as the problem is occurring on multiple IOS-XR platforms using the same template.
Answer C suggests that a "PASS ALL" policy is missing for the I-BGP neighbors, but this is not relevant to the problem at hand, which is related to the E-BGP sessions.
Answer D suggests that the "next-hop-self" command is not implemented on both E-BGP neighbors, which is the most likely cause of the issue. When BGP routes are advertised between different Autonomous Systems (AS), the next-hop attribute in the BGP updates must be changed to the IP address of the egress interface on the advertising router. If the next-hop-self command is not used, the next-hop attribute will still point to the IP address of the egress interface of the advertising router, which may not be reachable by the receiving router. This will prevent the E-BGP sessions from exchanging routes.
Therefore, the most likely cause of the problem described in the question is that the "next-hop-self" command is not implemented on both E-BGP neighbors.