I)  Exercise One

Open “Wireshark”, then use the “File” menu and the “Open” command to open the file “Exercise One.pcap”. You should see 26 packets listed.

This set of packets describes a ‘conversation’ between a user’s client and a central server. This entire conversation happens automatically, after a user types something and hits enter. Look at the packets to answer the following questions in relation to this conversation.

In answering the following questions, use brief descriptions. For example, “In frame X, the client requests a web page, and in frame Y, the server delivers the content of the page.”

Hint: See the accompanying document titled the “QuickStart Guide” – Look under the appendix describing “Hits Versus Page Views”.

Hint: a favicon.ico is a small graphic that can be used as an icon to identify a web page. In the following graphic the colorful “G” to the left  is a favicon.ico.

1. a) What is the IP address of the client that initiates the conversation?
2. Use the first two packets to identify the server that is going to be List the common name, and three IP addresses that can be used for the server.
3. What is happening in frames 3, 4, and 5?
4. What is happening in frames 6 and 7?
5. Ignore frame However, for your information, frame eight is used to manage flow control.
6. What is happening in frames nine and ten? How are these two frames related?
7. What happens in packet 11?
8. After the initial set of packets is received, the client sends out a new request in packet This occurs automatically without any action by the user. Why does this occur? See the first “hint” to the left.
9. What is occurring in packets 13 through 22?
10. Explain what happens in packets 23 through See the second “hint” to the left.
11. In one sentence describe what the user was doing (Reading email? Accessing a web page? FTP? Other?).

Review: The TCP/IP network does not know how to route using names, such as www.yahoo.com. It only knows how to route using IP addresses. Therefore, a common name (CNAME), such as www.yahoo.com must be translated to an IP address, like 216.109.117.106

before your computer can request a web page.

Your computer uses a DNS request to lookup a CNAME and it gets back a set of IP addresses (a primary address and one or more backup addresses) that can be used to contact the server.

Hint: See the accompanying document titled the “QuickStart Guide”

– Look under the appendix describing “Hits Versus Page Views”.

II)                Exercise Two

Open “Wireshark”, then use the “File” menu and the “Open” command to open the file “Exercise Two.pcap”. You should see 176 packets listed.

1. In the first few packets, the client machine is looking up the common name (cname) of a web site to find its IP What is the cname of this web site? Give two IP addresses for this web site.
2. How many packets/frames does it take to receive the web page (the answer to the first http get request only)?
3. Does this web site use gzip to compress its data for sending? Does it write cookies? In order to answer these questions, look under the payload for the reassembled packet that represents the web This will be the last packet from question b above. Look to see if it has “Content-Encoding” set to gzip, and to see if it has a “Set-Cookie” to write a cookie.
4. What is happening in packets 26 and 27? Does every component of a web page have to come from the same server? See the Hint to the
5. In packet 37 we see another DNS query, this time for i1.yimg.com. Why does the client need to ask for this IP address? Didn’t we just get this address in packet 26? (This is a trick question; carefully compare the two common names in packet 26 and 37.)
6. In packet 42 we see a HTTP “Get” statement, and in packet 48 a new HTTP “Get” Why didn’t the system need another DNS request before the second get statement? Click on packet 42 and look in the middle window. Expand the line titled “Hypertext Transfer Protocol” and read the “Host:” line. Compare that line to the “Host:” line for packet 48.
7. Examine packet It is one segment of a PDU that is reassembled with several other segments in packet 160. Look at packets 141, 142, and 143. Are these three packets also part of packet 160? What happens if a set of packets that are supposed to be reassembled do not arrive in a continuous stream or do not arrive in the proper order?
8. Return to examine frames 141 and Both of these are graphics (GIF files) from the same source IP address. How does the client know which graphic to match up to each get statement? Hint: Click on each

and look in the middle window for the heading line that starts with “Transmission Control Protocol”. What difference do you see in the heading lines for the two files? Return to the original “Get” statements. Can you see the same difference in the “Get” statements?

III)             Exercise Three

Open “Wireshark”, then use the “File” menu and the “Open” command to open the file “Exercise Three.pcap”. You should see 22 packets listed.

These packets represent two different requests for web pages. Packets 1-7 involve the request for the web page www.yahoo.com. Packets 8-22 involve the request for the web page my.usf.edu.

12. Compare the destination port in the TCP packet in frame 3 with the destination port in the TCP packet in frame What difference do you see? What does this tell you about the difference in the two requests?

The following table compares the two requests for web pages. For example, row i) shows that frames 1-2 and frames 8-9 represent the DNS lookups for each of the web requests.

1. Explain what is happening in row “iii” Why are there no frames listed for yahoo in row “iii”?

6. Look at the “Info” column on frame It says: “GET / HTTP / 1.1. What is the corresponding Info field for the my.usf.com web request (frame 21)? Why doesn’t it read the same as in frame 6?

First, you must decide on a web site to visit. Pick a web site, such as www.yahoo.com, or www.cnn.com that you haven’t visited yet today.

If you pick one that you recently visited, your system may not need to send out a DNS request, since the IP address may be cached (saved temporarily) on your machine. For this exercise, we would prefer to have a DNS request as part of the detail of transactions.

IV)            Exercise Four

In this exercise, you are going to capture live traffic from your computer. Open up Wireshark and use the “Capture” menu to save live traffic. The Wireshark “QuickStart” guide distributed with these exercises contains more instructions on using Wireshark.

Start capturing data, visit a live web site using your standard Internet browser, and stop capturing data.

If you have a large amount of network traffic, the relevant data may be hidden among a lot of broadcast messages. To focus on just the key frames, you can set a display filter like this.

For the IP number enter the IP number of your client machine. Type it as shown (ip.addr==your.ip.address) in the graphic above. Then click on “Apply”.

Using an approach similar to the approach in Exercise One, describe the set of frames that you captured.

• For this description think of this as a conversation – every discussion starts with a question and follows with an

answer.

• For example, two of the frames will contain the DNS request for an IP address for the web site, and the DNS answer with the IP
• Remember that some answers may take several frames if

they need to be reassembled from segmented packets.

V)               Exercise Five

In this exercise, you are going to capture live traffic from your email service. Begin by opening up your email service and preparing an email from you to yourself – you can use a subject line of “Homework” and a body of “Homework” if you chose. Don’t hit the send button yet.

Open up Wireshark, go to the “Capture” menu and tell it to “Start” capturing data. Now return to your email service and hit the send button. After the email is sent, tell Wireshark to “Stop” capturing data.

Look in the protocol column and see what protocol your email server uses. There are several choices.

1. If you use a local client, and your email server does not support secure communications, then you might see something like SMTP is an application running on top of TCP. Wireshark will recognize SMTP and report that the packet is in SMTP format.
2. If you use a local client, and your email server does support secure communications, then you might only see Since communications are secure, Wireshark will not be able to break the encryption. It will not recognize the format of the TCP payload; therefore, it will simply report a TCP message and show the encrypted data.
3. If you use a web browser to access email through a regular http:// web address, you will see commands that look like you are using web For example, you might see a HTTP Post command.
4. If you use a secure web page (https://), you will see TCP with an encrypted payload, similar to item 2) However, the process of setting up the encryption will look different since setting up a secure web page is different from setting up a secure email connection.

What protocol did your system use?

Even if your system does not use a secure protocol for sending and receiving email, it might use a secure protocol for the login. In this case, an eavesdropper on the internet would not be able to see your login, but they would be able to see all of the emails you send and receive.