Advanced BGP Interview Questions: Mastering Border Gateway Protocol

This guide delves into advanced BGP concepts, ideal for experienced network engineers. We explore BGP path attributes (MED, AS Path, Local Preference), route selection preferences, and the use of community attributes for route filtering and manipulation. This resource provides in-depth explanations of advanced topics such as BGP authentication, route reflection, and the role of route reflectors in optimizing iBGP communication. We also cover key differences between eBGP and iBGP, various BGP timers, and techniques for preventing routing loops. Prepare for challenging BGP-related questions with this comprehensive guide.

Top BGP Interview Questions and Answers

What is BGP?

BGP (Border Gateway Protocol) is a standardized exterior gateway protocol (EGP) used to exchange routing information between autonomous systems (ASes) on the internet. An AS is a collection of networks under a single administrative domain. BGP is a path-vector protocol, meaning it uses paths (sequences of ASes) as its routing metric. It's crucial for routing traffic across the internet, enabling different network operators (ISPs) to connect and exchange routing information.

BGP and Subnets

BGP neighbors (routers exchanging BGP information) do not need to be on the same subnet. They communicate using TCP, allowing for connections across different subnets and networks. The key is that they must be directly connected.

Main Purpose of BGP

BGP's primary function is to exchange routing information between autonomous systems. Unlike interior gateway protocols (IGPs) that exchange individual routes, BGP typically exchanges route summaries, making it efficient for large-scale networks.

Key Characteristics of BGP

• Uses TCP port 179 for reliable communication.
• Sends updates only when changes occur in the network topology (not periodic updates).
• Uses keepalive messages to maintain connections.
• Uses path attributes (like AS path length, local preference, etc.) as its metric.
• Is an interdomain routing protocol (used between different ASes).
• Supports route reflectors and confederations to reduce overhead.
• Uses eBGP for external routing and iBGP for internal routing.
• Uses MED (Multi-Exit Discriminator) for influencing inbound traffic.
• Uses weight to influence outbound traffic.

BGP TCP Port Number

BGP uses TCP port 179 for communication between BGP peers.

eBGP vs. iBGP

Running Multiple BGP Processes on a Single Router

Generally, only one BGP process can run per router. Running multiple instances requires specific configuration and might not be supported by all router vendors.

BGP and IGPs (Interior Gateway Protocols)

BGP is an exterior gateway protocol (EGP) used for inter-AS routing, while IGPs (like OSPF, RIP, EIGRP) are used for intra-AS routing. They serve different purposes and cannot directly replace one another.

Checking BGP Status

Use commands like show ip bgp summary (or vendor-specific equivalents) to check the BGP status and neighbor relationships. You may also use configuration commands to view the BGP configuration itself.

BGP Peers (Neighbors)

BGP peers are routers that have established a TCP connection to exchange routing information. They are also referred to as BGP neighbors.

Poison Reverse in BGP

Poison reverse is a technique used in distance-vector routing protocols (like RIP) to prevent routing loops. It's not directly used in BGP, which employs other loop prevention mechanisms.

BGP Timers

• Keepalive timer: Specifies the interval between keepalive messages (default: 60 seconds).
• Hold-down timer: Specifies how long to wait for a keepalive before declaring a neighbor down (default: 180 seconds).

Loop Prevention in BGP

BGP prevents routing loops through:

• iBGP Route Filtering: Prevents internal BGP peers from redistributing routes to each other.
• AS_PATH Attribute: When a route is advertised, its AS path is added to show that the Autonomous System has processed it. If a router sees its own ASN in an AS path, it rejects the route to avoid loops.

Hard Reset vs. Soft Reset in BGP

BGP Message Types

• OPEN: Establishes a BGP session.
• KEEPALIVE: Maintains the BGP session.
• UPDATE: Exchanges routing information.
• NOTIFICATION: Reports errors and terminates the session.

Well-Known BGP Path Attributes

BGP uses various path attributes to make routing decisions. Key attributes include:

• Weight (Cisco proprietary)
• Local Preference
• AS Path Length
• Origin Code
• Multi-Exit Discriminator (MED)
• Next Hop

RIP (Routing Information Protocol)

RIP is a distance-vector routing protocol that uses hop count as its metric. It has a hop count limit to prevent routing loops.

Purpose of BGP MED

BGP MED (Multi-Exit Discriminator) influences how external BGP (eBGP) traffic enters an autonomous system. A lower MED value indicates a preferred path.

BGP Path Attributes (Details)

• Next Hop: The IP address of the next router in the path.
• Weight: A locally configured value influencing route selection (not advertised to neighbors).
• Local Preference: Used within an AS to prefer certain paths.
• Multi-Exit Discriminator (MED): Used to influence traffic entering an AS from external neighbors.
• AS Path: A list of AS numbers traversed by a route.
• Origin: Indicates the source of a route (IGP, EGP, etc.).

BGP Path Attribute Preferences

BGP uses path attributes to select the best route. Several attributes influence this selection, and the order of preference can vary depending on the specific attributes involved. Generally, a lower value is preferred for attributes like MED and AS path length.

BGP Route Selection Order (Inbound vs. Outbound)

The order of preference for route selection differs between inbound and outbound updates. This is due to how route filters and policies are applied within the BGP routing table.

BGP Communities

BGP communities are used for route filtering and policy control. They are essentially labels (numeric values) that can be added to routes. These labels help the routers make decisions and control route dissemination.

• no-export: Prevents a route from being advertised to external BGP (eBGP) peers.
• no-advertise: Prevents a route from being advertised to any peers (internal or external).
• internet: Allows a route to be advertised externally.
• local-AS: Prevents a route from being advertised externally or to confederate peers.

BGP Authentication

BGP supports MD5 authentication to secure the communication between peers.

BGP and Full Mesh Topologies

Full mesh topologies (where each router has a direct connection to every other router) are not required for eBGP but are generally needed for iBGP to avoid routing loops.

Route Reflectors in BGP

Route reflectors are BGP routers that can relay routing information between internal BGP (iBGP) peers without requiring a full mesh topology. This simplifies the design of large BGP networks.

Administrative Distance (AD) of eBGP and iBGP

The administrative distance (AD) determines which routing protocol's route is preferred when multiple routes exist for the same destination. eBGP has an AD of 20, and iBGP has an AD of 200.