ECE 4110 Internetwork Programming Lab 9: OSPF and Fixing Broken Networks

ginglyformweekNetworking and Communications

Oct 29, 2013 (3 years and 7 months ago)

65 views

1






ECE 4110 Internetwork Programming

Lab 9: OSPF and Fixing Broken Networks


Group Number: _________________________


Member Names: _________________________ _________________________


Date Issued: November 28, 2005 (Do NOT start before this date. The la
b will not
work.)

Turn
-
in Due: December 6, 2005

Last Edited: November 29
, 2005



This lab requires that you use one of three setups. If a signup sheet has been
posted, you must sign up in advance on the lab door. You may reserve each setup for no
more than

2 hours

at a time. You must use the same setup each time you work on this lab.
A TA must be present for two checkoffs of this lab.
Also, it would be useful to bring
your ECE4110 Lab 6 documentation to help you in troubleshooting the network.


Goal



When

you begin this lab, there will be several problems in the OSPF network.
Your job will be to find and repair the problems. Once you believe you have correctly
repaired the network, a TA will check you off after you reach the designated check
-
off
point (the
re are two check
-
off points in this lab).


Prelab Questions



Answer the following questions on a separate sheet of paper, and include your
answers with your lab report.


1) What are the differences between a distance vector and a link
-
state routing
pro
tocol? What kind of routing protocol is OSPF?



2) What IP protocol number does OSPF use?


3) What are the mechanisms used by OSPF routers to exchange routing
information? Describe them.



4) What is VLSM? Does OSPF support it? Justify your ans
wer.






2






5) How does OSPF determine the metric/cost for a route?







6) Why does not OSPF need any mechanism (e.g., split horizon for RIP) to prevent
the occurence of rouitng loops?








7) Compared to OSPF, what is the limitation of RIP in terms
of the size of
networks?








8) How does OSPF achieve routing scalability to reduce LSAs traffic in a large
network?



3






Lab Scenario



At the end of this lab, your netwo
rk should look like the one p
ictured below:


:


OSPF1


OSPF3


OSPF2


Ethernet1


FastEthernet0


Ethernet0/0


10.c.3.1


10.c.6.2


10.c.7.1


10.c.6.1


10.c.8.1


10.c.2.2


10.c.2.1


10.c.3.
2


10.c.7.2


Ethernet1/0



Ethernet1


Ethernet0


Ethernet0


FastEthernet0


FastEthernet0/0


FastEthernet0


10.c.1.1


10.c.5.1


10.c.8.2


10.c.9.1


OSPF4


OSPF5


Ethernet0


10.c.5.2



Linux Computer 1



Linux Computer 2


Ethernet1


10.c.
9.2


10.c.1.2


Ethernet0


Ethernet1


4






Section 1: Troubleshooting a Network


For this lab we have three hardware setups. When you signed up for time in the
lab you should have signed up for a particular playstation: 1, 2, or 3. For the entirety of
this lab,
c

denotes your pl
aystation number. As with the previous lab,
make certain that
no one else is using your setup before beginning work on the lab
.


In a network, there are often several points of possible failures. Three of the most
common reasons why a network is not workin
g properly are:



The physical connection between two nodes in the network has been
compromised, such as the cable being damaged or one of the nodes being
unplugged. You will not encounter this problem in this lab.




The routing table in one or more of the c
omputers or routers is misconfigured.



One or more of the interfaces on one or more computers or routers has been
misconfigured or is down due to any reason.



Unlike the last lab, we are now using OSPF to route our packets. In OSPF we do
not need to expl
icitly define our neighbors in order to route; the
hello

protocol will
perform automatic discovery.




Use the following table to identify your routers and switch.



Playstation 1

Playstation 2

Playstation 3


Switch

5

14

23

OSPF 1 Router

7

16

25

OSPF
2 Router

8

17

26

OSPF 3 Router

9

18

27

OSPF 4 Router

10

19

28

OSPF 5 Router

11

20

29




Similarly to your Labs 6
-
8, there are three files of scripts, which you will need to reset
the routers, download and upload the configuration files of routers. You

will need to run
those scripts on your hard drive where you installed them.


To begin the lab, you need to get a TA to upload the broken network to your playstation.
They will sign off that it is indeed broken.


1st TA Signoff


Broken Network:
TA has t
o make sure that routers are loaded properly
broken.


5







Log into the computer marked
ECE 4110 Labs 7,8 Setup c
. This is the machine
marked
Linux Computer 1

in our network diagram on page 2. Your username and
password are both group
#
.


Telnet into OSPF1 by
typing:

telnet 10.c.1.1


The telnet password is
owen
. Enable, then type
show ip route.
Notice that not all
of the subnets on our diagram are present in the output. This indicates that there is
something wrong with the network.
Save your output from the
sho
w ip route

command and include it in your lab report.




Log into
ECE 4110 Lab 8 Setup c Computer 2.

This is the machine marked
Linux
Computer 2

on our network diagram on page 2. Your username and password are again
both group#. Try to ping
Linux Computer
1

from
Linux Computer 2
. This ping should
fail.



There are several problems in the network, although none of them involve
physical connection issues. Your job is to use the commands you have learned in
previous labs to fix the network so that it function
s like the network diagramed on page 2.
Once you are done fixing the errors, you should be able to ping
Linux Computer 1

from
Linux Computer 2

and vice versa.



In addition to being able to ping the opposite station, you must make sure that all
links in t
he given network diagram are active.
You will not get full credit on the lab
report if the network does not entirely match the diagram, even if you receive a
check off.


Hints:



From each of the Linux machines try pinging every router. Once you have noted
w
hich routers cannot be pinged, you should have a better idea of which links may
be down.



Remember, there are several possible sources of failures as mentioned above. Log
onto each of the routers and verify that the routing tables are correct. Run
show ip
r
oute

and
show running
-
config

on each of the routers and compare the shown
routing tables with what should be expected based on the network diagram.



You may ask TAs about the function of specific router configuration commands.
You may not ask TAs how to sol
ve specific routing problems.



There may also be problems with the way the switch is configured. Nothing is
guarnteed to work correctly.



Once you have fixed the network,
screenshot the routing table on OSPF1 again
and include it in your lab report.

6






Se
ction 2: Observing OSPF



Now that we have fixed our network, we will look specifically at how OSPF
responds to changes in the network. Connect to OSPF1 and observe the output to the
following command



1.

show ip route

a) Are all the networks indicated i
n the diagram in the routing table?



b) How many routes are there for the network 10.c.8.0/24? Why?




c) Comment on one difference between where packets are routed in this table
and where they would be routed if you were running RIP.





Telnet to 10.c.
3.2 (OSPF3), then shutdown the interface FastEthernet0/0. Keep
track of when you typed
shutdown
(start a stopwatch if you have one). Note that you have
simulated a network failure: interface 10.c.7.2 on OSPF3 no longer works.



Return to OSPF1, and look at

the routing table by typing
show ip route.

If your
routing table is unchanged, keep trying until it does change.


2.

a)What routes do you see? b) Is the routing table different from before? If so,
how? c) How long did it take your routing table to chang
e? d) Why? How does this
compare to RIP?





Telnet back into OSPF3 and Re
-
enable FastEthernet0/0 with the
no shutdown
command.

Ping
Linux Computer 1

from
Linux Computer 2
and simultaneously ping
Linux Computer
2
from
Linux Computer 1
. Show this ping to t
he TA.


2nd TA Signoff


Fixed Network:
TA has to make sure that ping works properly.






Answer the following questions:


7






3.a) What was wrong with the network? Include the network drawing on page 3
with X’s over the links/interfaces that were broken.





3.b) How did you go about fixing the problem?






4. a)What are the differences in the routing table for OSPF1 as compared to
when the network was broken? b) How did OSPF1 learn about the other
subnets?





c) How are packets routed from
Linux Compute
r 2
to
10.c.3.2
? Why?





d) Do you prefer OSPF or RIP? Why?






e) Observe your routing table for OSPF1. Write the path taken to each of the
eight networks, and state why each cost is its current value.

(Ex. To go to 10.1.10.0/24, the path is 10.1.11.2

(cost 3) to 10.1.23.2 (cost 13) to
10.1.10.0 (cost 2) for a total cost of 18.)








Turn
-
In Checklist



Write a report that includes answers to all the bold questions in this lab. Explain
in detail exactly what was wrong with the network and
how you dis
covered and fixed

the problems. Include your answers to the pre
-
lab questions with this report. Finally,
8






write a one to two page summary of what you have learned. Include the routing tables
from OSPF1, the router configurations from all four routers, and a
ny other data that you
captured.




Prelab questions and answers.



All lab questions and answers. Include any requested screenshots.



Router Configuration Files (You get these by running the download script.
Just print out the switch and 5 routers ones and t
urn them in.)



A one to two
-
page summary of the lab.



Group number

on first page