Chapter 19 - Understanding OSPF Concepts

Question 1

What are some examples of routing protocols.

Answer 1

• RIP (Routing Information Protocol)
• EIGRP (Enhanced Interior Gateway Routing Protocol) - Previously known as the Cisco proprietary IGRP
• OSPF (Open Shortest Path First)
• BGP (Border Gateway Protocol)

Question 2

What is a Routed/Routable Protcol?

Answer 2

These terms refer to a protocol that defines a packet structure and logical addressing (e.g. IPv4 and IPv6).

Question 3

What is Path Selection?

Answer 3

The process that a routing protocol uses to choose the best route to a destination.

Question 4

What is the relationship between Routing Protocols and Routed/Routable Protocols?

Answer 4

Routing protocols forward packets with address defined by routed protocols. For example, OSPF defines a process for routing IPv4 packets.

Question 5

What are the general functions of a routing protocol?

Answer 5

• Learn routing information about subnets from neighboring routers.
• Advertise routing information about subnets to neighboring routers.
• Choose the best route to a destination based on a certain metric. The lower the metric, the better.
• React to changes in the network topology. For example by advertising that some routes have failed and a new best route (known as convergence).

Question 6

What is an IGP?

Answer 6

An Interior Gateway Protocol is a routing protocol that was designed to be used inside a single AS (Autonomous System). Examples of these would be OSPF, RIP, EIGRP (Enchanced Interior Gateway Routing Protocol)

Question 7

What is an EGP?

Answer 7

An Exterior Gateway Protocol is a routing protocol that was designed to be used between multiple AS (Autonomous System). An example of one is BGP.

Question 8

What is an AS (Autonomous System)?

Answer 8

• An Autonomous System is a network or group of networks under the administrative control of a single organisation.
• For example a network(s) paid for and managed by a school or an ISP could be an AS.

Question 9

What organisation assigns ASNs (Autonomous System Numbers)?

Answer 9

IANA (Internet Assigned Numbers Authority)

Question 10

What is a Routing Protocol Algorithm?

Answer 10

The logic that routing protocols use to learn routes, choose the best route for a destination, and converge in reaction to network topology changes. The three main ones are:
- Distance Vector (Sometimes called Bellman-Ford after its creator) (IGP)
- Path Vector (EGP)
- Link-State (IGP)
- There is also Advanced Distance Vector (Sometimes called Balanced Hybrid)

Question 11

What are the three IGP Routing Protocol Algorithms?

Answer 11

• Distance Vector (Sometimes called Bellman-Ford after its creator)
• Advanced Distance Vector (Sometimes called Balanced Hybrid)
• Link-State

Question 12

What routing protocols use Link-State algorithms?

Answer 12

• OSPF
• IS-IS (Intermediate System to Intermediate System)

Question 13

What routing protocols use Advanced Distance Vector algorithms?

Answer 13

• EIGRP (Enhanced Interior Gateway Routing Protocol)

Question 14

What metrics are used by RIP, OSPF, IS-IS, and EIGRP?

Answer 14

• OSPF - Cost - The cost of each link is determined by its bandwidth (ref bandwidth/interface bandwidth). The total metric is the cumulative cost of all links in the route.
• IS-IS - Cost - The cost of each link is determined by its bandwidth. Unlike OSPF, the cost of each link is not automatically calculated, and the cost of every link is 10 by default. So if this is not changed, IS-IS is indifferent to RIP.
• RIP - Hop Count - The number of outgoing interfaces between a router and a destination subnet
• EIGRP - Bandwidth and Delay Calculation (by default) - Calculated based on the route’s slowest link and the cumulative delay associated with each interface along the route. The delay is a metric assigned to each interface based on its bandwidth.

Question 15

Describe a benefit of OSPF over RIP?

Answer 15

RIP will always take the shortest route regardless of how slow the links are. OSPF may take a longer route but overall it will be faster as its metric uses interface bandwidth. (i.e. 3 gigabit links are better than a single serial link of 1000kbps.)

Question 16

What does the ‘bandwidth’ command do?

Answer 16

• It tells higher level protocols what the available bandwidth on an interface is. This can be used to force OSPF routes since, without this configured, OSPF will just take the interface’s speed as its bandwidth.
• Doesn’t affect the actual speed of an interface.
• Not recommended to change this specifically for OSPF, as this setting is used in calculations done by other protocols (e.g. EIGRP).
• Entered in Kilobits
• Example: Would be useful if you were using a 10Mbps Ethernet cable between two Gigabit interfaces as the router would assume that the bandwidth is Gigabit even though the link would never reach this speed.

Question 17

What routing protocols use Distance Vector algorithms?

Answer 17

• RIPv2

Question 18

True or False. OSPF is a classful routing protocol.

Answer 18

False. OSPF is a classless routing protocol like RIPv2 and EIGRP. RIP and IGRP are classful.

Question 19

True or False. EIGRP is a Cisco proprietary protocol.

Answer 19

True. Although they publish it as a public RFC for vendors to use.

Question 20

What is Route Redistribution?

Answer 20

The ability for a router to advertise routes learned in one routing protocol using a different routing protocol.

Question 21

What is Administrative Distance?

Answer 21

• A number that defines how believable a routing protocol is on a single router.
• The reason this is necessary is in scenarios where route redistribution is being utilised and a route to a single subnet is learned by both protocols. Both protocols use different metrics and therefore cannot be compared.
• The administrative distance will essentially act as a tie breaker.
• EIGRP has a lower administrative distance than OSPF so regardless of the metric for the OSPF route, the EIGRP route will be seen as best. This can be changed in Cisco routers.

Question 22

Routing Protocol Administrative Distances

Answer 22

Connected - 0
Static - 1
eBGP - 20
EIGRP (internal routes) - 90
IGRP - 100
OSPF - 110
IS-IS - 115
RIP - 120
EIGRP (external routes) -170
iBGP - 200
DHCP Default Route - 254
Unusable - 255

Question 23

What is Dijkstra SPF?

Answer 23

Dijkstra Shortest Path First is a mathematical algorithm that allows a router using a link-state routing protocol to query its LSDB and build the shortest route to a destination based on that information.

Question 24

What are OSPF Neighbors?

Answer 24

OSPF enabled routers that are connected to the same network and are part of the same area. They have also gone through the Hello process to decide that they will become neighbors.

Question 25

What is an OSPF RID?

Answer 25

- 32-bit number that is used as the unique identifier of a router for OSPF communication. - A router will choose its RID from one of the following: - Manual configuration - Highest loopback interface IP - Highest physical interface IP

Question 26

True or False. OSPF neighbors must be part of the same subnet in order to become neighbors.

Answer 26

True. However it is possible to also achieve a neighbor relationship over PPP using two /32 addresses.

Question 27

What address are OSPF Hello messages sent to?

Answer 27

- 224.0.0.5 which is a multicast address designated for all OSPF speaking routers (known as the "All SPF Routers" multicast address) - The Hello messages are encapsulated in an IP header with a value of 89 in the Protocol field

Question 28

What are the 7 OSPF states in order?

Answer 28

- Down - No Hellos have been sent or received to or from OSPF neighbors but they can be. (Optional) - Attempt - Only used in NBMA (Non-Broadcast Multi Access) environments. - Init - A Hello has been received from another router but the local router's RID was not included. The local router must then respond with its RID in a Hello message. - Two-Way - when DR and BDR are elected. Indicates that both neighbors have received Hellos with their own RID included along wth the neighboring router's RID. Each router only enters this state in the other's neighbors table if it has sent a Hello message with both RIDs in. All conditions have now been met for the routers to become neighbors. - ExStart - The neighbors negotiate a master and slave neighbor by exchanging DBD (Databse Description) packets to decide who will begin the LSA exchange. The router with the higher RID becomes the master and defines the sequence numbers that are listed against each LSA (for LS Request purposes). - Exchange - The routers provide a list of LSAs (Using DBDs) that they are going to send information for. Not the routing information included within them. They compare the information in the DBDs to their own LSDBs to see what LSAs they need. - Loading - Neighbors send LSR (Link State Request) messages to request LSAs they don't already have. The LSAs are then sent in LSU (Link State Update) messages. To acknowledge receipt of the LSAs, each router sends LSAck messages. - Full - Both neighbors have fully exchanged the contents of their LSDB. They then send Hello messages on the Hello interval to ensure the adjacency continues and the Dead timer is not reached. If it is not then they will continue to share LSAs as the network changes.

Question 29

What is the OSPF Hello Interval?

Answer 29

- This is the interval at which a router will send a Hello message to its neighbor(s) to advise them that they want the relationship to continue. Also to form new neighbor relationships. - Default is 10 seconds on a Broadcast and Point to Point OSPF network.

Question 30

What is the OSPF Dead Interval?

Answer 30

- This is the interval at which a router will consider it's adjacency with another router as failed. - Default is 40 seconds Broadcast and Point to Point OSPF networks. - It counts down from 40 and if a Hello is not received, the adjacency is removed. If a Hello is received, the interval is reset.

Question 31

True or False. When something changes in a network topology, the OSPF enabled routers will flood updated LSAs with the new topology information to all of their OSPF neighbors.

Answer 31

True. However if they are in a broadcast network, routing information will only be flooded to DRs and BDRs.

Question 32

What are the responsibilites of each router in an OSPF neighbor relationship?

Answer 32

- Maintain neighbor state by sending Hello messages at the Hello Interval and listening for Hello messages before the Dead Interval is reached. - Flood any changed LSAs to each neighbor. - Reflood unchanged LSAs once their lifetime expires (30 minutes by default from the moment the LSA is created).

Question 33

What is multicast address 224.0.0.6?

Answer 33

- "All SPF DRs" multicast address. - Allows OSPF enabled routers to send routing updates to only DRs and BDRs on the link.

Question 34

What is the technical term for Broadcast type OSPF network routers that are neither DR or BDR?

Answer 34

- DROthers. - These will not reach a Full state with any router on the network, apart from the DR and BDR, and will remain in a two-way state with other DROthers. - The DR and BDR will reach a Full state with all other OSPF neighbors.

Question 35

What is the difference between a Neighbor and an Adjacent/Fully Adjacent Neighbor?

Answer 35

- A Neighbor is a router that has only achieved a two-way state. - An Adjacent/Fully Adjacent Neighbor has reached Full state with its neighbor(s). This relationship is called an Adjacency.

Question 36

What is the formula for working out OSPF interface cost?

Answer 36

Interface Cost= Reference bandwidth/interface bandwidth - By default reference bandwidth is 100mbps (10^8) for any interfaces that are 100mbps or faster. - The answer must be positive and a whole number - Any number lower than 1 will be 1 by default

Question 37

What issues can be caused by implementing single area OSPF designs on larger networks?

Answer 37

- Larger LSDB - This requires more storage on each router - The SPF algorithm requires processing power that grows the larger the LSDB gets - A state change on a single interface causes every router to run SPF again so it is slower - SPF takes more time to calculate routes

Question 38

What is the general consideration for when you should start dividing your OSPF network into areas?

Answer 38

It isn't a set rule but as a generalisation it is 50 routers. You should then create a new area to add extra neighbors.

Question 39

What is an OSPF area?

Answer 39

- A logical collection of OSPF networks, routers, and links that share the same LSDB and area Id. - A router within an area only maintains an LSDB for routers in its area. This makes it more efficient in larger networks.

Question 40

List rules for an OSPF area

Answer 40

- OSPF areas are designated per interface. - All interfaces connected to the same subnet should be in the same area - Areas should be contiguous meaning each area should be connected to the next either directly for via the backbone - All non-backbone areas must be directly connected to the backbone area (area 0) by having at least one ABR (Area Border Router) connected to both the backbone area and the relevant non-backbone area

Question 41

True or False. The backbone OSPF area is always 0.

Answer 41

True.

Question 42

What is an ABR (Area Border Router)?

Answer 42

An OSPF enabled router with interfaces connected multiple areas.

Question 43

What is a Backbone router?

Answer 43

A router connected to the backbone area. This includes ABRs.

Question 44

What is an Internal router?

Answer 44

A router with interfaces only in one non-backbone area.

Question 45

What is the Backbone area?

Answer 45

Area 0. This is the area which all other areas must connect with.

Question 46

What is an intra-area route?

Answer 46

A route to a destination inside the same area as the local router. Also learned in the same area.

Question 47

What is an inter-area route?

Answer 47

A route to a destination which is in a different area to the local router.

Question 48

What are the four main types of OSPF LSAs?

Answer 48

- Router (Type 1) - Sent by all routers - Describes the RID of a router and what neighbors can be reached via its interfaces. - Also describes the networks attached to the routers OSPF enabled interfaces. - Doesn't leave the area. - Network (Type 2) - Sent by the DR of each multi-access (broadcast) OSPF network - Lists the routers that are part of the broadcast network - Doesn't leave the area - Summary (Type 3) - Sent by ABRs - Describe a subnet in another area - Flooded to all areas via the backbone area - AS External (Type 5) - Sent by ASBRs to advertise routes to destinations outside of the AS.

Question 49

True or False. Routers in an OSPF area will contain full LSDB information for all subnets including inter-area subnets.

Answer 49

False. Routers in an area will only have very basic information about subnets in other areas so that they can build routes but also do not become overwhelmed. This information will include the subnet ID, mask, and RID of the ABR that advertised the LSA.

Question 50

What are examples of interim OSPF states?

Answer 50

- ExStart - Exchange - Loading

Question 51

What are examples of stable OSPF states?

Answer 51

- Full - Two-Way (applicable in OSPF networks where DR and BDRs are present)

Question 52

True or False. DR and BDR selection happen per area.

Answer 52

False. DR and BDR selection happens per network segment so you can have multiple DRs and BDRs in an area.

Question 53

What Routing Protocol Algorithm is used by EGPs?

Answer 53

- Path Vector

Question 54

What Routing Protocol uses Path Vector algorithms?

Answer 54

- BGP (External to be specific)

Question 55

How does a router react when the link between itsself and a router that it has learned routes from goes down?

Answer 55

- The routes learned from this router are removed from its routing table.

Question 56

What are characteristics of the Distance Vector Routing algorithm?

Answer 56

- Developed prior to Link State - Used by RIP and IGRP in the early days - Protocols that use this algorithm share their known destination networks and metrics to these networks with their neighbors. This process is known as 'routing by rumor' because the router doesn't specifically know about the network beyond its neighbors. Only what it has been told. - The name of the algorithm comes from the fact that a router learns the 'distance' (metric) and 'vector' (direction - next hop) of each route.

Question 57

What are characteristics of the Link State Routing algorithm?

Answer 57

- Every router creates a 'connectivity map' of the network. They do this by advertising information about their interfaces (and connected networks) with their neighbors, they do this until each router has the same 'connectivity map'. - Each router then uses this map to determine the best path to each network. - More CPU and memory intensive as more information is shared than in Distance Vector. - Faster at reacting to changes in the network than Distance Vector.

Question 58

Define Metric

Answer 58

The value that a routing protocol uses to determine the best route to a specific destination. The lower the metric, the better the route.

Question 59

What would happen if a router learns multiple routes to the same destination, via the same routing protocol, with the same metric?

Answer 59

All are added to the routing table and traffic is load balanced between them. This is known as ECMP (Equal Cost Multi-Path)

Question 60

True or False. Administrative distance and metric take presidence over route specificity. Why?

Answer 60

False. Administrative distance and metric are only there to determine which routes get added to the routing table. Route specificity is used to determine which route traffic will take.

Question 61

What is OSPF?

Answer 61

- Open Shortest Path First - Uses Dijkstra's algorithm (Shortest Path First algorithm)

Question 62

What is an LSA?

Answer 62

- Link State Advertisement - A data unit (not a packet) that contains routing information about different networks from a router that is using a Link-State routing protocol (e.g. OSPF). - All routers in an area will flood LSAs to other routers in the same area until all routers have built their LSDB. This should be the same for all OSPF routers in the area.

Question 63

What is an LSDB?

Answer 63

- Link State Database - A data structure made up of LSAs - All OSPF enabled routers in an area should have the same LSDB

Question 64

True or False. OSPF will be enabled on an interface that is not in an up/up state.

Answer 64

False.

Question 65

What is the default aging timer value for an LSA?

Answer 65

- 30 minutes. - Once this has elapsed, the LSA will be flooded again even if the LSA has already been sent previously.

Question 66

What are the steps that the overall OSPF process takes to learn the best routes?

Answer 66

- Become neighbors with other routers connected on the same segment - Exchange LSAs with neighbors - Use Dijkstra SPF to calculate the best routes

Question 67

True or False. ABRs maintain a separate LSDB for each area they are connected to.

Answer 67

True. This is why it is recommended to have only upto 2 areas on a single router as to not over burden the router.

Question 68

What does the 'clear ip ospf process' command do?

Answer 68

- Reloads the OSPF process on the local router.

Question 69

What is an ASBR?

Answer 69

- Autonomous System Boundary Router - A router that connects and OSPF area to an external network (e.g. The Internet) - Using the 'default-information originate' command on a router will make it become an ASBR

Question 70

What is the OSPF cost of Loopback interfaces?

Answer 70

1

Question 71

Good way of remembering OSFP neighbor states.

Answer 71

Demons - Down In - Init Texas - Two-way Eat - Exstart Eels - Exchange Like - Loading Fries - Full

Question 72

What are the different OSPF message types?

Answer 72

- Hello - Used for neighbor discovery and maintenance. - DBD - Database Description - Summary of the LSDB of the router that neighbors use to compare LSDBs - LSR - Link State Request - Used to request specific LSAs - LSU - Link State Update - Used to send specific LSAs - LSAck - Link State Acknowledgement - Used to acknowledge LSUs