Module 4: Implementing OSPF

1
Module 4 Implementing OSPF
2
Lessons

• OSPF
• OSPF Areas and Hierarchical Routing
• OSPF Operation
• OSPF Routing Tables
• Designing an OSPF Network

3
Lesson 4-1 OSPF
4
RIP vs. OSPF

• RIP Problems
• Limited metric
• Next-hop view
• Limited network diameter
• Slow convergence
• Non-hierarchical routing

• OSPF Solutions
• Arbitrary, 16-bit metric
• Complete map
• Theoretical unlimited network diameter
• Fast convergence
• Hierarchical routing structure

5
RIP vs. OSPF

• Figure 4-1

6
OSPF Features

• Authentication
• Classless Routing
• Arbitrary Metric
• Hierarchical Routing Structure
• Equal-Load Balancing
• Multicast/Unicast Packets

7
OSPF and Link State Routing

• Neighbor Discovery/Maintenance
• Virtual Links
• Link State Advertisements (LSAs)
• Link State Database (LSDB)
• Adjacencies
• Route Generation Algorithm
• Routing Tables

8
Adjacent Routers

• Figure 4-2

9
OSPF Sub-Protocols
Table 4-1
10
Non-linear Phases of OSPF Operation

• Figure 4-3

11
RIP vs. OSPF
Table 4-2
12
Lesson 4-2 OSPF Areas and Hierarchical Routing
13
Concepts

• Types of OSPF Areas
• Types of OSPF Routers
• Types of Traffic
• Route Summarization
• Area Partitioning
• Virtual Links

14
OSPF Areas, Routers, and Traffic

• Figure 4-4

15
Route Summarization

• Figure 4-5

16
Sample Area
Figure 4-6
17
Router X Fails and Area X is Partitioned
Figure 4-7
18
Connecting an Area Partition to the Backbone
through a Normal Area
Figure 4-8
19
Connecting a Partitioned Backbone through a
Normal Area
Figure 4-9
20
Lesson 4-3 OSPF Operation
21
Format of an OSPF Packet
Figure 4-10
22
Types of OSPF Packets
Table 4-3
23
Function of Hello Packets

• Neighbor Discovery / Maintenance
• announce the presence of a router
• act as keepalives to verify the continued
  participation of a router
• Election Process
• determine the designated router (DR) and backup
  designated router (BDR) of a multi-access network
  segment

24
Only Router A Forwards LSAs from the Multi-Access
Network
Figure 4-11
25
Exchange ProtocolPhase 2 Operation

• Exchange DD Packets
• establish master-slave relationship
• Exchange DD Packets
• exchange information about LSDBs to build a link
  state request list
• Transmit LSRequest Packets
• request specific LSAs that are missing from the
  LSDB of a router

26
OSPF LSA Packet Format
Figure 4-12
27
Flooding Protocol Concepts

• 6 Types of LSAs
• Sequencing of LSAs
• Aging of LSAs
• Guaranteed Delivery of LSAs

28
Types of LSAs
Table 4-4
29
Sequence Numbers
Figure 4-13
30
Link State Acknowledgements

• Implicit Acknowledgement
• router sends back a copy of the LSA in an
  LSUpdate packet
• Explicit Acknowledgement
• router sends back an LSAck packet that contains
  the 20-byte header of the LSA

31
Lesson 4-4 OSPF Routing Tables
32
OSPF Metrics

• OSPF uses a 16-bit, arbitrary metric
• Metrics usually based on bandwidth
• Metric of a route sum of all outgoing
  interfaces to the destination
• OSPF provides equal-load balancing

33
OSPF Path Types

• Intra-Area Paths
• Inter-Area Paths
• E1 Paths
• E2 Paths

34
Dijkstra Algorithm

• Also called the Shortest Path First (SPF)
  algorithm
• Converts the LSDB of a router into a shortest
  path tree (the router is the root of the tree)
• Run twice
• first to create the internal routing table
• second to create the standard routing table

35
Sample Network
Figure 4-14
36
Shortest Path Tree of Router J as Determined by
the First SPF Calculation
Figure 4-15
37
Shortest Path Tree of Router J as Determined by
the Second SPF Calculation
Figure 4-16
38
Lesson 4-5 Designing an OSPF Network
39
OSPF Routing Concepts

• Hierarchical Routing Structure
• three-tiered model
• Route Summarization
• summarize at area boundaries
• conserve bandwidth and router resources

40
Figure 4-17
41
Route Summarization
Figure 4-18
42
Topology Considerations

• Minimum/Maximum Routers per OSPF Network
• Minimum/Maximum Routers per OSPF Area
• Maximum Number of OSPF Areas

43
Scalability Considerations

• Plan for growth
• Ensure routers have the appropriate memory and
  processing power
• Place routers appropriately in the network

44
Area Design Considerations

• Designing the Backbone Area
• Designing Stub Areas
• normal stub areas
• not-so-stubby areas
• Avoiding Partitions and Virtual Links
• Providing for Route Summarization

45
Commonly Configurable OSPF Parameters
Table 4-5
46
Commonly Configurable OSPF Parameters (cont.)
Table 4-5
47
Module 4 Lab Exercise

• Designing OSPF Solutions

48
Lab Exercise 4-1Designing OSPF Solutions
49
Lab Exercise 4-1 Overview

• Setup
• discussion lab
• Format
• answer questions as a group
• share proposed solutions with class
• discuss the different strengths and weaknesses of
  each proposed solution

50
Module 4Review Questions...Summary...
51
1. Which of the following are link state routing
protocols?

• A. Link Service Protocol
• B. Open Shortest Path First (OSPF)
• C. Routing Information Protocol (RIP)
• D. Intermediate System to Intermediate System
  (IS-IS)

(Choose three.)
52
1. Which of the following are link state routing
protocols?

• A. Link Service Protocol
• B. Open Shortest Path First (OSPF)
• C. Routing Information Protocol (RIP)
• D. Intermediate System to Intermediate System
  (IS-IS)

(Choose three.)
53
2. What piece of information regarding network
topology is available to a link state router, but
not to a distance vector router?

• A. the network ID of all reachable destinations
• B. the next hop along the path to each
  destination
• C. the distance (or metric) from the router to
  the destination
• D. the status of the links between the router and
  any router in the network

(Choose one.)
54
2. What piece of information regarding network
topology is available to a link state router, but
not to a distance vector router?

• A. the network ID of all reachable destinations
• B. the next hop along the path to each
  destination
• C. the distance (or metric) from the router to
  the destination
• D. the status of the links between the router and
  any router in the network

(Choose one.)
55
3. Why do link state routers require more
processing power than distance vector routers?

• A. because link state routers use the Dijkstra
  algorithm to compute paths
• B. because link state routers maintain more
  complex routing tables than distance vector
  routers
• C. because link state routers can make
  connections to more destination networks and
  other routers than distance vector routers
• D. because link state routers utilize a flooding
  process that requires them to transmit
  information about themselves and their links to
  every other router in their routing domains

(Choose one.)
56
3. Why do link state routers require more
processing power than distance vector routers?

• A. because link state routers use the Dijkstra
  algorithm to compute paths
• B. because link state routers maintain more
  complex routing tables than distance vector
  routers
• C. because link state routers can make
  connections to more destination networks and
  other routers than distance vector routers
• D. because link state routers utilize a flooding
  process that requires them to transmit
  information about themselves and their links to
  every other router in their routing domains

(Choose one.)
57
4. In terms of link state routing protocols,
what are neighboring routers?

• A. routers that share a common link
• B. routers located on adjacent subnets of an IP
  network
• C. routers that communicate with low latency
  because of their physical proximity
• D. link state routers that can communicate
  without routing their packets through any
  distance vector routers

(Choose one.)
58
4. In terms of link state routing protocols,
what are neighboring routers?

• A. routers that share a common link
• B. routers located on adjacent subnets of an IP
  network
• C. routers that communicate with low latency
  because of their physical proximity
• D. link state routers that can communicate
  without routing their packets through any
  distance vector routers

(Choose one.)
59
5. What feature of link state routing protocols
enables link state routing domains to converge
more quickly than distance vector routing
domains?

• A. Hello packets
• B. Dijkstra algorithm
• C. link state database
• D. link state flooding

(Choose one.)
60
5. What feature of link state routing protocols
enables link state routing domains to converge
more quickly than distance vector routing
domains?

• A. Hello packets
• B. Dijkstra algorithm
• C. link state database
• D. link state flooding

(Choose one.)
61
6. What are the advantages of the hierarchical
routing structure used by link state routers?

• A. reduces amount of time necessary to build
  adjacencies
• B. reduces load on router memory, router
  processors, and network bandwidth
• C. reduces number of fields required in routers'
  link state databases (LSDBs)
• D. reduces number of link state advertisements
  (LSAs) that must be flooded to a routing domain

(Choose two.)
62
6. What are the advantages of the hierarchical
routing structure used by link state routers?

• A. reduces amount of time necessary to build
  adjacencies
• B. reduces load on router memory, router
  processors, and network bandwidth
• C. reduces number of fields required in routers'
  link state databases (LSDBs)
• D. reduces number of link state advertisements
  (LSAs) that must be flooded to a routing domain

(Choose two.)
63
7. What are the three categories of OSPF design
considerations?

• A. topology considerations
• B. reliability considerations
• C. scalability considerations
• D. bandwidth considerations
• E. availability considerations
• F. area design considerations

(Choose three.)
64
7. What are the three categories of OSPF design
considerations?

• A. topology considerations
• B. reliability considerations
• C. scalability considerations
• D. bandwidth considerations
• E. availability considerations
• F. area design considerations

(Choose three.)
65
8. What is the Internet Engineering Task Force
(IETF) recommendation for the maximum number of
routers in an OSPF network?

• A. 200
• B. 700
• C. 1000
• D. 1200

(Choose one.)
66
8. What is the Internet Engineering Task Force
(IETF) recommendation for the maximum number of
routers in an OSPF network?

• A. 200
• B. 700
• C. 1000
• D. 1200

(Choose one.)
67
9. In an OSPF network, what is the best location
for routers that are relatively low in memory and
processing power?

• A. backbone
• B. stub areas
• C. ISP interfaces
• D. subnet interfaces

(Choose one.)
68
9. In an OSPF network, what is the best location
for routers that are relatively low in memory and
processing power?

• A. backbone
• B. stub areas
• C. ISP interfaces
• D. subnet interfaces

(Choose one.)
69
10. Why should virtual links be reserved for
emergencies in an OSPF network and not used as a
permanent part of the network's topology?

• A. Virtual links are prone to errors.
• B. Virtual links require extra bandwidth.
• C. Virtual links are difficult to configure.
• D. Virtual links are slower than physical links.
• E. Virtual links place increased loads on
  routers.

(Choose two.)
70
10. Why should virtual links be reserved for
emergencies in an OSPF network and not used as a
permanent part of the network's topology?

• A. Virtual links are prone to errors.
• B. Virtual links require extra bandwidth.
• C. Virtual links are difficult to configure.
• D. Virtual links are slower than physical links.
• E. Virtual links place increased loads on
  routers.

(Choose two.)
71

72
Assumptions
Module 4 Implementing OSPF

• You understand the distance vector algorithm.
• You know RIP.
• Routing tables
• Next-hop router
• Convergence process