This chapter covers the critical distinction between Interior Gateway Protocols (IGPs) and Exterior Gateway Protocols (EGPs), focusing on their roles, mechanisms, and deployment in modern networks. Understanding this topic is essential for the CompTIA Network+ N10-009 exam, particularly under Objective 1.5 (Compare and contrast interior and exterior routing protocols). Approximately 5–10% of exam questions will test your knowledge of protocol categories, metrics, administrative distances, and when to use each type. By mastering these concepts, you'll be able to design scalable networks and troubleshoot routing issues effectively.
Jump to a section
Imagine a large corporation with multiple branch offices in the same city (interior) and international partners abroad (exterior). The internal mail system (IGP) is designed for speed and efficiency within the company: it uses standardized routes, knows every hallway and floor, and delivers mail in seconds. The internal mail carriers (routers running OSPF or EIGRP) constantly share updates about new cubicles or closed corridors, ensuring every package takes the fastest path. Because the internal network is trusted, they use simple metrics like hop count or link speed. Now consider international shipping (EGP, specifically BGP). This system must handle different countries with their own postal regulations, political boundaries, and business agreements. International carriers don't trust each other implicitly; they use complex policies to decide which routes are acceptable. They exchange routing information only with explicitly configured neighbors (peers) and apply hundreds of rules (path attributes) to choose paths based on cost, reliability, or legal restrictions. The internal system converges in seconds; the external system may take minutes because it prioritizes stability and policy over raw speed. Just as a company uses both internal couriers and FedEx, an enterprise network uses IGPs for interior routing and BGP for connections to ISPs and other autonomous systems.
What Are Interior and Exterior Routing Protocols?
Routing protocols are categorized by their scope of operation. Interior Gateway Protocols (IGPs) operate within a single Autonomous System (AS)—a network under a single administrative domain. Exterior Gateway Protocols (EGPs) exchange routing information between different Autonomous Systems. The only EGP in modern use is Border Gateway Protocol (BGP). IGPs include RIP, EIGRP, OSPF, and IS-IS. The N10-009 exam expects you to compare these categories across several dimensions: metric type, convergence speed, scalability, administrative distance, and protocol behavior.
Why the Distinction Matters
The Internet is a collection of tens of thousands of ASes. Without a clear boundary between interior and exterior routing, an error inside one company could destabilize the entire Internet. IGPs are designed for fast convergence and simplicity within a trusted environment. EGPs are designed for policy control, scalability, and stability across organizational boundaries. A network engineer must choose the right tool: use an IGP inside your AS, and BGP to connect to ISPs or other ASes.
Key Differences at a Glance
Metric: IGPs use simple metrics—hop count (RIP), bandwidth and delay (EIGRP), or cost based on bandwidth (OSPF). BGP uses a complex path attribute system (e.g., AS_PATH, LOCAL_PREF, MED) with multiple tie-breakers.
Convergence: IGPs converge in seconds or sub-seconds. BGP convergence can take minutes due to path exploration and hold timers.
Scalability: IGPs are designed for hundreds to a few thousand routers. BGP can handle hundreds of thousands of routes.
Administrative Distance (AD): Cisco devices assign default AD values: OSPF (110), EIGRP (90), RIP (120), iBGP (200), eBGP (20). Lower AD means more trustworthy.
Interior Gateway Protocols in Depth
RIP (Routing Information Protocol): The simplest IGP, uses hop count (max 15). Versions: RIPv1 (classful, no authentication), RIPv2 (classless, supports VLSM and authentication), RIPng (for IPv6). Updates every 30 seconds, with hold-down timer 180 seconds, flush timer 240 seconds. Rarely used in modern networks except for educational purposes.
EIGRP (Enhanced Interior Gateway Routing Protocol): Cisco proprietary (though partially open via RFC 7868). Uses composite metric based on bandwidth, delay, reliability, load, and MTU (default: bandwidth + delay). Features Diffusing Update Algorithm (DUAL) for loop-free convergence. Supports unequal-cost load balancing via variance command. Default timers: hello 5 seconds (60 on low-speed links), hold 15 seconds (180).
OSPF (Open Shortest Path First): Link-state protocol using Dijkstra's SPF algorithm. Areas reduce routing table size and LSA flooding. OSPF cost = 10^8 / bandwidth (in bps). Default hello timer 10 seconds (30 on NBMA), dead timer 40 seconds (120 on NBMA). Router ID (RID) is highest loopback IP, then highest physical IP. Supports authentication (type 0, 1, 2).
IS-IS (Intermediate System to Intermediate System): Similar to OSPF but used mainly in ISP backbones. Uses CLNS addressing, not IP. Less common on N10-009 but may appear in comparison questions.
Exterior Gateway Protocol: BGP
BGP (Border Gateway Protocol): Path-vector protocol used between ASes. Current version is BGP-4 (RFC 4271). BGP routers exchange reachability information using TCP port 179. eBGP runs between different ASes; iBGP runs within the same AS. BGP does NOT use a simple metric; it selects the best path based on multiple attributes in a specific order: highest weight (Cisco proprietary), highest LOCAL_PREF, locally originated, shortest AS_PATH, lowest origin type, lowest MED, eBGP over iBGP, lowest IGP metric, lowest router ID.
BGP Timers: Keepalive 60 seconds, hold 180 seconds. BGP convergence is slow because of path exploration and route flap damping. BGP is policy-driven: you can manipulate attributes to influence path selection.
Administrative Distance and Route Selection
When a router learns the same prefix from multiple routing protocols, it uses AD to choose the best source. Default AD values (Cisco): connected (0), static (1), eBGP (20), EIGRP (90), OSPF (110), RIP (120), iBGP (200). Lower is better. This means an eBGP route is preferred over an OSPF route by default, which is important because eBGP routes often represent external connectivity.
Convergence and Scalability
IGPs converge quickly because they use fast hello/dead timers and efficient algorithms (DUAL, SPF). BGP converges slowly because it must process many routes and apply policies. BGP also uses route reflectors and confederations to scale iBGP within large ASes.
Interaction Between IGPs and BGP
In enterprise networks, IGPs (OSPF or EIGRP) run inside the AS for internal connectivity, while BGP connects to ISPs. Redistribution between IGP and BGP is common but must be carefully controlled to avoid routing loops. For example, a default route from BGP can be redistributed into OSPF to provide internet access to internal routers.
Verification Commands
show ip route: View routing table, note the protocol code (O for OSPF, D for EIGRP, R for RIP, B for BGP).
show ip protocols: Display routing protocol parameters, timers, and networks.
show ip ospf neighbor: List OSPF neighbors and their states.
show ip eigrp neighbors: Show EIGRP neighbors.
show ip bgp summary: Display BGP neighbor status and prefix counts.
show ip bgp: Show BGP routing table with path attributes.
Summary of Key Exam Points
IGPs: RIP, EIGRP, OSPF, IS-IS (interior).
EGP: BGP (exterior).
Metrics: Hop count (RIP), composite (EIGRP), cost (OSPF), path attributes (BGP).
AD: eBGP (20), EIGRP (90), OSPF (110), RIP (120), iBGP (200).
Convergence: IGPs fast, BGP slow.
BGP uses TCP port 179, reliable updates.
OSPF uses areas to reduce overhead.
EIGRP supports unequal-cost load balancing.
RIP max hop count 15.
Common Exam Traps
Confusing IGP vs EGP: Remember that BGP is the only EGP.
Mixing up AD values: eBGP is 20, not 200; iBGP is 200.
Thinking BGP uses a simple metric like hop count: It uses multiple attributes.
Assuming all IGPs are distance-vector: OSPF and IS-IS are link-state.
Forgetting that EIGRP is Cisco proprietary (though now partially open).
Conclusion
Interior and exterior routing protocols serve different purposes and must be chosen based on network scope, performance requirements, and administrative boundaries. The N10-009 exam tests your ability to compare these protocols across multiple dimensions. Focus on understanding the underlying mechanisms, not just memorizing facts. Practice with show commands to reinforce your knowledge.
Identify the Autonomous System
Determine whether the routing domain is within a single AS (interior) or spans multiple ASes (exterior). An AS is a collection of networks under a single administrative control with a unique AS number (ASN). For interior routing, you will use an IGP like OSPF or EIGRP. For exterior routing, you must use BGP. This step defines the protocol category and sets expectations for metric complexity, convergence speed, and policy requirements. The exam may present a scenario describing a network topology; you must identify whether the routing is interior or exterior based on whether multiple ASes are involved.
Select the Appropriate Protocol
Choose the specific protocol based on requirements. For interior routing: use OSPF for large, scalable, multi-vendor networks; use EIGRP for Cisco-centric networks needing fast convergence; use RIP only for very small networks or labs. For exterior routing: use BGP. Consider factors like convergence speed (IGPs converge faster), metric complexity (BGP attributes allow policy control), and scalability (BGP handles many routes). The exam may ask which protocol is best for a given scenario, such as a multi-vendor enterprise (OSPF) or connecting to an ISP (BGP).
Configure Neighbor Adjacencies
For IGPs, neighbors are discovered dynamically via multicast or broadcast hello messages. For OSPF, configure the network statement under router ospf to enable interfaces. For EIGRP, use the network command. For BGP, neighbors must be explicitly configured with the neighbor command, specifying the IP address and remote AS number. BGP uses TCP port 179 and requires reachability between peers. The exam may test the difference: IGPs discover neighbors automatically; BGP requires manual configuration.
Exchange Routing Information
IGPs exchange routes using periodic updates (RIP, every 30s) or triggered updates (EIGRP, OSPF via LSAs). OSPF uses link-state advertisements (LSAs) to build a link-state database, then runs SPF to compute routes. EIGRP uses DUAL to maintain loop-free paths. BGP exchanges full routing tables initially, then only incremental updates. BGP routes carry path attributes like AS_PATH and NEXT_HOP. The exam may ask about update mechanisms: link-state vs. distance-vector vs. path-vector.
Select Best Path Using Metrics
Each protocol uses its own metric to choose the best path. RIP uses hop count (max 15). EIGRP uses composite metric (default bandwidth and delay). OSPF uses cost (10^8 / bandwidth). BGP uses a multi-step decision process based on attributes. The router installs the best route into the routing table. If multiple protocols provide routes to the same destination, administrative distance breaks the tie. The exam expects you to know default metrics and AD values.
Converge and Maintain Routing Tables
When topology changes, IGPs converge quickly via triggered updates and SPF recalculation (OSPF) or DUAL (EIGRP). BGP convergence is slower due to route processing and hold timers. BGP also implements route flap damping to suppress unstable routes. The network engineer monitors convergence using show commands and adjusts timers if needed. The exam may test convergence speed comparisons: IGPs converge faster than BGP.
Enterprise Network with Multiple Sites
A large retail chain with 200 stores and a headquarters uses OSPF as its IGP. Each store has a router connecting to the WAN. OSPF provides fast convergence when a WAN link fails; the SPF algorithm recalculates routes within seconds. The network team configured OSPF in a single area for simplicity, but as the network grew, they migrated to a multi-area design to reduce LSA flooding. OSPF costs were tuned by manually setting interface bandwidth to reflect actual link speeds. A common mistake is forgetting to set the router ID, which defaults to the highest loopback IP. In production, they use router-id command to ensure stability.
ISP Peering Using BGP
A medium-sized ISP connects to two upstream providers and peers at an IXP. They run iBGP internally and eBGP externally. BGP allows them to implement policy: they set LOCAL_PREF to prefer one upstream for most traffic, and use AS_PATH prepending to make their routes less attractive to peers. They also use MED to influence inbound traffic from a specific provider. BGP convergence is not a concern because they accept that external routing changes take minutes. A critical misconfiguration is forgetting to set the neighbor next-hop-self on iBGP peers, causing reachability issues. The team uses show ip bgp and show ip route to verify.
Campus Network with Redistribution
A university campus runs EIGRP as its IGP and connects to the internet via BGP to two ISPs. They redistribute a default route from BGP into EIGRP so internal routers can reach the internet. However, they must filter routes to prevent internal EIGRP routes from being advertised to the ISPs. They use route maps and prefix lists. A common issue is accidental redistribution causing routing loops; they use administrative distance adjustments and distribute lists to control the flow. The network team uses show ip protocols to verify redistribution settings.
The N10-009 exam tests this topic under Objective 1.5: "Compare and contrast interior and exterior routing protocols." Expect 3-5 questions. The most common wrong answers stem from confusing protocol categories, metric types, and administrative distance values.
Common Wrong Answers and Why Candidates Choose Them
"RIP is an exterior routing protocol." Wrong because RIP only operates within an AS. Candidates confuse the term "gateway" in IGP/EGP, thinking RIP is used at borders. Remember: RIP is an IGP.
"BGP uses hop count as its metric." Wrong because BGP uses path attributes, not hop count. Candidates familiar with RIP assume all protocols use simple metrics. BGP's AS_PATH length is not a hop count.
"OSPF has an administrative distance of 90." Wrong because OSPF's AD is 110 (Cisco default). Candidates mix up EIGRP (90) and OSPF (110). Memorize the AD values: eBGP 20, EIGRP 90, OSPF 110, RIP 120, iBGP 200.
"EIGRP is a link-state protocol." Wrong because EIGRP is an advanced distance-vector protocol. Candidates see EIGRP's features (DUAL, partial updates) and incorrectly classify it as link-state.
Specific Numbers and Values to Memorize
RIP max hop count: 15
OSPF default cost: 10^8 / bandwidth
OSPF default hello/dead: 10/40 seconds (broadcast), 30/120 (NBMA)
EIGRP default hello/hold: 5/15 seconds (high-speed), 60/180 (low-speed)
BGP default keepalive/hold: 60/180 seconds
Administrative distances: eBGP 20, EIGRP 90, OSPF 110, RIP 120, iBGP 200
BGP uses TCP port 179
Edge Cases and Exceptions
iBGP vs eBGP AD: iBGP has AD 200, eBGP has 20. This means eBGP routes are preferred over iBGP routes even if the iBGP path is shorter. This prevents routing loops.
EIGRP is Cisco proprietary but now partially open via RFC 7868. The exam treats it as Cisco proprietary.
OSPF areas: Area 0 is mandatory; all other areas must connect to area 0.
BGP synchronization: Old rule (no longer default in modern IOS) that iBGP routes must be in IGP to be used. Not tested heavily but may appear.
How to Eliminate Wrong Answers
If the question mentions "between autonomous systems," the answer must involve BGP (or EGP). Eliminate OSPF, EIGRP, RIP.
If the question asks about "fast convergence," IGPs are faster than BGP. Eliminate BGP if speed is critical.
If the question mentions "policy-based routing," BGP is the only protocol with extensive policy control. Eliminate IGPs.
If the question gives a metric like "hop count" or "cost," match to the correct protocol: hop count=RIP, cost=OSPF, composite=EIGRP, attributes=BGP.
If the question lists administrative distances, know the order: eBGP < EIGRP < OSPF < RIP < iBGP.
IGPs include RIP, EIGRP, OSPF, IS-IS; the only EGP is BGP.
RIP uses hop count with a maximum of 15 hops.
OSPF uses cost = 10^8 / bandwidth; default hello/dead timers are 10/40 seconds.
EIGRP uses composite metric (bandwidth and delay); Cisco proprietary.
BGP uses TCP port 179 and path attributes for route selection.
Administrative distance values: eBGP 20, EIGRP 90, OSPF 110, RIP 120, iBGP 200.
IGPs converge faster than BGP; BGP is more scalable and policy-driven.
BGP requires explicit neighbor configuration; IGPs discover neighbors automatically.
These come up on the exam all the time. Here's how to tell them apart.
Interior Gateway Protocols (IGPs)
Operate within a single AS
Use simple metrics (hop count, cost, composite)
Fast convergence (seconds or sub-seconds)
Automatic neighbor discovery via multicast/broadcast
Designed for trust and efficiency
Exterior Gateway Protocols (EGPs)
Operate between different ASes
Use complex path attributes (AS_PATH, LOCAL_PREF, MED)
Slow convergence (minutes due to path exploration)
Explicit neighbor configuration required (TCP peering)
Designed for policy control and scalability
Mistake
BGP is a link-state protocol.
Correct
BGP is a path-vector protocol. It does not maintain a full topology map like OSPF; it only knows the path (list of ASes) to a destination.
Mistake
RIP can be used on the internet.
Correct
RIP's maximum hop count of 15 makes it unsuitable for the internet, where paths often exceed 15 hops. RIP is only used in small, isolated networks.
Mistake
EIGRP is a standard protocol supported by all vendors.
Correct
EIGRP was originally Cisco proprietary. Although Cisco published the basic functionality in RFC 7868, full implementation is still Cisco-centric. Multi-vendor networks typically use OSPF.
Mistake
OSPF and IS-IS are the same protocol.
Correct
While both are link-state protocols, OSPF uses IP to communicate and requires an IP network; IS-IS uses CLNS and can run directly over Layer 2. They are not interchangeable.
Mistake
BGP selects the best path based on the lowest metric.
Correct
BGP uses a multi-attribute decision process. The first tie-breaker is weight (Cisco proprietary), then LOCAL_PREF, then locally originated, then shortest AS_PATH, etc. There is no single metric.
Reveal each answer, then mark whether you got it right. Score 60%+ to unlock the next chapter.
IGP (Interior Gateway Protocol) operates within a single Autonomous System (AS), using simple metrics like hop count or cost for fast convergence. EGP (Exterior Gateway Protocol) operates between ASes, using complex path attributes for policy control. The only EGP in use today is BGP. For the exam, remember that IGPs are used inside your network, while BGP connects your network to others.
BGP convergence is slower because it processes many routes (hundreds of thousands) and applies complex policy decisions. BGP uses keepalive/hold timers (60/180 seconds) and path exploration, where multiple paths may be advertised before the best path stabilizes. OSPF uses fast hello/dead timers (10/40 seconds) and immediate SPF recalculation on topology changes.
Memorize: eBGP 20, EIGRP 90, OSPF 110, RIP 120, iBGP 200. Lower AD means more trustworthy. For example, a route learned via eBGP (AD 20) is preferred over an OSPF route (AD 110) to the same destination.
No. OSPF is an IGP and should not be run between different ASes. To connect to an ISP, you use BGP. Running OSPF with your ISP would violate administrative boundaries and could cause routing instability.
OSPF's default metric is cost, calculated as 10^8 / interface bandwidth (in bps). For example, a FastEthernet interface (100 Mbps) has a cost of 1. You can manually set cost with the `ip ospf cost` command.
EIGRP is classified as an advanced distance-vector protocol. It uses DUAL for loop-free path selection and maintains neighbor tables and topology tables, but it does not maintain a full link-state database like OSPF. The exam considers it a hybrid or advanced distance-vector.
BGP path attributes provide information about routes that is used to implement routing policies. Attributes like AS_PATH (list of ASes traversed), LOCAL_PREF (preference within AS), and MED (metric to influence inbound traffic) allow network administrators to control traffic flow based on business relationships.
You've just covered Interior vs Exterior Routing Protocols — now see how well it sticks with free N10-009 practice questions. Full explanations included, no account needed.
Done with this chapter?