What causes IBGP route flapping in enterprise networks?
In enterprise networks utilizing Junos OS, IBGP (Internal BGP) route flapping is frequently caused by recursive routing failures within the underlying IGP (Interior Gateway Protocol), such as OSPF or IS-IS.
Recursive Next-Hop Resolution: Unlike EBGP, where the neighbor is typically directly connected, IBGP neighbors are often several hops away. Consequently, an IBGP route's protocol next-hop is usually a loopback IP address that requires a recursive lookup to resolve. The router must consult its unicast routing table (typically inet.0 or inet.3) to find a physical path to that next-hop.
The Flapping Mechanism: A 'flap' occurs when a route is repeatedly withdrawn and readvertised. If the IGP path used to resolve the BGP next-hop becomes unstable or if the BGP route itself is inadvertently used to resolve its own next-hop (a recursive loop), the BGP next-hop becomes unreachable.
Impact on Routing Table: When the next-hop cannot be resolved, the BGP route is considered invalid and is withdrawn from the routing table. Once the route is withdrawn, the original IGP path to the next-hop often becomes valid again, causing the BGP route to be readvertised. This cycle of withdrawal and readvertisement creates the instability known as flapping.
Why other options are incorrect: Packet misordering (Option A) and community tagging (Option B) may affect performance or policy but do not cause a route to be withdrawn and readvertised in this cyclical manner. While update intervals (Option C) control how quickly changes are sent, they are not the root cause of the reachability failure that triggers the flap itself.
Currently there are no comments in this discussion, be the first to comment!