1

CHAPTER 22
Network Layer:
Delivery, Forwarding, and Routing
Solutions to Review Questions and Exercises
Review Questions
1. We discussed two different methods of delivery: direct and indirect. In a direct
delivery, the final destination of the packet is a host connected to the same physical
network as the deliverer. In an indirect delivery the packet goes from router to
router until it reaches the one connected to the same physical network as its final
destination.
2. The three common forwarding methods used today are: next-hop, network-spe-
cific, and default methods. In the next-hop method, the routing table holds only
the address of the next hop for each destination. In the network-specific method,
the routing table holds only one entry that defines the address of the destination
network instead of all hosts on that network. In the default method, a host sends all
packets that are going out of the network to a specific router called the default
router.
3. A routing table can be either static or dynamic. A static routing table contains
information entered manually. A dynamic routing table is updated periodically by
using one of the dynamic routing protocols such as RIP, OSPF, or BGP.
4. RIP is an intradomain routing protocol that enables routers to update their routing
tables within an autonomous system.
5. A RIP message is used by a router to request and receive routing information
about an autonomous system or to periodically share its knowledge with its neigh-
bors.
6. The time-out for the expiration time is 6 times that of the periodic timer to allow
for some missed communication between routers.
7. The hop count limit helps RIP instability by limiting the number of times a mes-
sage can be sent through the routers, thereby limiting the back and forth updating
that may occur if part of a network goes down.
8. The two major shortcomings are two-node instability and three-node instability.
For the former, infinity can be re-defined as a number such as 20. Another solution
is the split horizon strategy or split horizon combined with poison reverse. These
methods do not work for three-node instability.
2
9. In OSPF, four types of links have been defined: point-to-point, transient, stub, and
virtual. A point-to-point link connects two routers without any other host or router
in between. A transient link is a network with several routers attached to it. The
packets can enter and leave through any of the routers. A stub link is a network
that is connected to only one router. The data packets enter the network through
this single router and leave the network through this same router. This is a special
case of the transient network. When the link between two routers is broken, the
administrator may create a virtual link between them, using a longer path that
probably goes through several routers.
10. OSPF messages are propagated immediately because a router using OSPF will
immediately flood the network with news of any changes to its neighborhood. RIP
messages are distributed slowly because a network using RIP relies on the periodic
updates that occur every 30 seconds to carry any news from one router to the next
and to the next.
11. BGP is an interdomain routing protocol using path vector routing.
12. We mentioned two groups of multicast routing protocols: the source-based tree and
the group-shared tree. In a source-based tree protocol, each router needs to have
one shortest path tree for each group. The shortest path tree for a group defines the
next hop for each network that has loyal member(s) for that group. In a group-
shared tree protocol, only one designated router takes the responsibility of distrib-
uting multicast traffic. The designated router has m shortest path trees in its routing
table. The rest of the routers in the domain have none.
Exercises
13. A host that is totally isolated needs no routing information. The routing table has
no entries.
14. A routing table for a LAN not connected to the Internet and with no subnets can
have a routing table with host-specific addresses. There is no next-hop address
since all packets remain within the network.
15. See Figure 22.1.
16. If the packet with destination address 140.24.7.194 arrives at R3, it gets sent to
interface m0. If it arrives at R2, it gets sent to interface m1 and then to router R3.
Figure 22.1 Solution to Exercise 15
m1
m2
m0
130.56.12.4
R1
Unknown Network
145.23.192.0
202.14.17.224
Rest of the
Internet
3
The only way R1 can receive the packet is if the packet comes from organization 1,
2, or 3; it goes to R1 and is sent out from interface m3.
17. R1 cannot receive a packet with this destination from m0 because if any host in
Organization 1 sends a packet with this destination address, the delivery is direct
and does not go through R1. R1 can receive a packet with this destination from
interfaces m1 or m2. This can happen when any host in Organization 2 or 3 sends a
packet with this destination address. The packet arrives at R1 and is sent out
through m0. R1 can also receive a packet with this destination from interface m3.
This happens in two cases. First, if R2 receives such a packet, the /24 is applied
The packet is sent out from interface m0, which arrives at interface m3 of R1. Sec-
ond, if R3 receives such a packet, it applies the default mask and sends the packet
from its interface m2 to R2, which, in turn, applies the mask (/24) and sends it out
from its interface m0 to the interface m3 of R1.
18. See Table 22.1.
19. See Table 22.2.
20. See Table 22.3.
Table 22.1 Solution to Exercise 18: Routing table for regional ISP
Mask Network address Next-hop address Interface
/20 120.14.64.0 --- m0
/20 120.14.96.0 --- m2
/20 120.14.112.0 --- m3
/0 0.0.0.0 default router m4
Table 22.2 Solution to Exercise 19: Routing table for local ISP 1
Mask Network address Next-hop address Interface
/23 120.14.64.0 --- m0
/23 120.14.66.0 --- m1
/23 120.14.68.0 --- m2
/23 120.14.70.0 --- m3
/23 120.14.72.0 --- m4
/23 120.14.74.0 --- m5
/23 120.14.76.0 --- m6
/23 120.14.78.0 --- m7
/0 0.0.0.0 default router m8
Table 22.3 Solution to Exercise 20: Routing table for local ISP 2
Mask Network address Next-hop address Interface
/22 120.14.96.0 --- m0
/22 120.14.100.0 --- m1
4
21. See Table 22.4.
22. See Table 22.5.
/22 120.14.104.0 --- m2
/22 120.14.108.0 --- m3
/0 0.0.0.0 default router m4
Table 22.4 Solution to Exercise 21: Routing table for local ISP 3
Mask Network address Next-hop address Interface
/24 120.14.112.0 --- m0
/24 120.14.113.0 --- m1
/24 120.14.114.0 --- m2
/24 120.14.115.0 --- m3
/24 120.14.116.0 --- m4
/24 120.14.117.0 --- m5
/24 120.14.118.0 --- m6
/24 120.14.119.0 --- m7
/24 120.14.120.0 --- m8
/24 120.14.121.0 --- m9
/24 120.14.122.0 --- m10
/24 120.14.123.0 --- m11
/24 120.14.124.0 --- m12
/24 120.14.125.0 --- m13
/24 120.14.126.0 --- m14
/24 120.14.127.0 --- m15
/0 0.0.0.0 default router m16
Table 22.5 Solution to Exercise 22: Routing table for small ISP 1
Mask Network address Next-hop address Interface
/30 120.14.64.0 ---- m0
/30 120.14.64.4 ---- m1
/30 120.14.64.8 ---- m2
/30 120.14.64.12 ---- m3
.
.
.
.
.
.
.
.
.
.
.
.
/30 120.14.65.252 ---- m127
/0 0.0.0.0 default router m128
Table 22.3 Solution to Exercise 20: Routing table for local ISP 2
Mask Network address Next-hop address Interface
5
23. In distance vector routing each router sends all of its knowledge about an autono-
mous system to all of the routers on its neighboring networks at regular inter-
vals. It uses a fairly simple algorithm to update the routing tables but results in a lot
of unneeded network traffic. In link state routing a router floods an autonomous
system with information about changes in a network only when changes occur.
It uses less network resources than distance vector routing in that it sends less traf-
fic over the network but it uses the much more complex Dijkstra Algorithm to cal-
culate routing tables from the link state database.
24. We assume that router C is one hop away. Then the modified table from C is Table
22.6:
Comparing this to the old table, we get Table 22.7:
25. There are 2 + (10 N) = Empty bytes in a message advertising N networks
26. See Figure 22.2.
Table 22.6 Solution to Exercise 24
Network Hops
Net1 3
Net2 2
Net3 4
Net4 8
Table 22.7 Solution to Exercise 24
Network Hops
Net1 3 C
Net2 2 C
Net3 1 F
Net4 5 G
Figure 22.2 Solution to Exercise 26
Com: 2 Version Reserved
Family: 2 All 0s
net 1
All 0s
All 0s
4
Family: 2 All 0s
net 2
All 0s
All 0s
2
Family: 2 All 0s
net 3
All 0s
All 0s
1
Family: 2 All 0s
net 4
All 0s
All 0s
5
6
27. See Figure 22.3.
28. See Figure 22.4.
29. Transient networks: N1, N2, N5, and N6. Stub networks: N3 and N4
30. See Table 22.8.
31. No, RPF does not create a shortest path tree because a network can receive more
than one copy of the same multicast packet. RPF creates a graph instead of a tree.
32. Yes, RPB creates a shortest path tree and its leaves are networks. However, the
delivery of the packets are based on broadcasting instead of multicasting.
33. Yes, RPM creates a shortest path tree because it is actually RPB (see previous
answer) with pruning and grafting features. The leaves of the tree are the networks.
Figure 22.3 Solution to Exercise 27
Figure 22.4 Solution to Exercise 28
Table 22.8 Solution to Exercise 30
Destination Interface
--- ---
10.0.0.0 2
--- ---
N1
N2
N3
N4
N5
N6
N7
R1
R2
R3
R4
R5
R6
R8
R7
N8
R1
R2
R3
R4
R5
R6
R7
R8
N1
N2
N3
N5
N6 N7
N8
N4