Читать книгу Wireshark for Security Professionals - Parker Jeff T. - Страница 5

We start with the busy Wireshark GUI, which is packed with features. We provide a high-level overview of where you need to look to start seeing some packet data. With packet capturing covered, we then discuss the more powerful features of Wireshark, starting with dissectors. In Wireshark, dissectors are what parse a protocol and decode it for presenting on the interface. They enable Wireshark to give the raw bits and bytes streaming across the wire some context by displaying them into something more meaningful to the human analyst. We then round off the chapter by covering the various filters available to help limit and zero in on just the network data you are interested in.

The home screen appears when you open Wireshark. On this screen are shortcuts you can use to start a new capture or open a previous capture file. For most newcomers to Wireshark, the brightly colored Capture button is the most attractive option. Starting a capture leads to a flurry of scrolling packets, which for the newcomer then leads to overwhelm. But let's go back to the home screen. There are also links to online documentation that you can use to figure out how to accomplish a certain task.

On the top of the screen, as shown in Figure 1-1, is the menu bar in the classic format you are probably familiar with. These menus have settings and other features like statistics that can be accessed when needed. (Don't worry – we aren't really worried about statistics.) Below these menus is the Main toolbar, which has quick access icons for the functionality you will use most while analyzing network traffic. These icons include things like starting or stopping a capture, and the various navigation buttons for finding your way around captured packets. Icon buttons are typically grayed if not applicable or usable – for example, without a capture yet.

Figure 1-1: The Wireshark home screen

Icons change over time from version to version. At the time this book was written, the blue shark fin starts a capture and the red square stops a capture. The shark fin is gray until the network interface is chosen, and we cover that soon. Also note that this toolbar area gives you a visual indication of the capture process. Again, many options are grayed out in Figure 1-1 because we are not yet capturing or don't have a capture completed. As you go through this chapter, pay attention to this area to understand how it changes and how it reflects the various capture states. In many respects, Wireshark has an intuitive user experience.

The Filter toolbar, which is below the Main toolbar, is a vital part of the Wireshark UI. You will soon fall in love with this little box, as you often find yourself drowning in a torrent of traffic. The Filter toolbar lets you remove whatever is uninteresting to the task at hand and presents just what you're looking for (or takes out what you're not looking for). You can enter display filters in the Filter text box that help you drill down what packets you see in the Packet List pane. We discuss filters in detail later in this chapter, but for now just trust me: They will be your new best friends.

Packet List Pane

The largest portion in the middle of the interface is reserved for the packet list. This list shows all the packets captured along with useful information, such as source and destination IP, and the time difference between when the packets were received. Wireshark supports color coding various packets to make sorting of traffic and troubleshooting easier. You can add custom colors for packets of interest, and the columns within the Packet List pane display useful information such as the protocol, packet length, and other protocol-specific information (see Figure 1-2).

Figure 1-2: The Packet List pane

This window is the bird's-eye view into the network you are sniffing or the packet capture you have loaded into Wireshark. The last column, by default labeled “Info,” offers a quick summary of what that packet contains. Of course, it depends on the packet, but it might be the URL for an HTTP request or the contents of a DNS query, which is really useful for getting a quick handle on important traffic in your capture.

Packet Details Pane

Below the Packet List pane is the Packet Details pane. The Packet Details pane shows information for the selected packet in the Packet List pane. This pane contains a ton of information, down to what the various bytes are within the packet. Information such as the source and destination MAC address is included here. The next row contains IP information. The next row reveals the packet is sending to UDP port 58351. The next row reveals what information is contained in that UDP packet.

These rows are ordered by the headers as they are ordered when sending data on the network. That means they are subject to change if you are capturing on a different type of network, such as a wireless network, that has different headers. The DNS column, which is the application data encapsulated within UDP, is expanded in Figure 1-3. Notice how Wireshark allows you to easily pull out information, such as the actual DNS query that was made within this DNS packet. This is what makes Wireshark the powerful network analysis tool that it is. You don't have to memorize the DNS protocol to know which bits and bytes at what offset translate into a DNS query.

Figure 1-3: The Packet Details pane

Subtrees

Because the details would be overwhelming if shown all at once, the information is organized and collapsed into sections. The sections, called subtrees, can be collapsed and expanded to display only what you need. (In Figure 1-2, the subtrees are collapsed; in Figure 1-3, they are expanded.)

NOTE

You might hear the message sent between devices referred to as a data frame or a packet. But what's the difference? When referring to the message at the OSI layer 2 (the data link layer, where the MAC address is used), the whole message is called a frame. When referring to the message at OSI model layer 3 (the network layer, for example, using the IP address), then the message is called a packet.

If you're already familiar with how a data frame is structured, you recognize how the packet details subtrees are divided. Details are structured into subtrees along the lines of the data frame's headers. You can collapse/expand a subtree by clicking the arrow sign next to the relevant section. The arrow is pointing to the right if the subtree is collapsed. Once you click on the arrow to expand that subtree, you'll see the arrow points down (refer to Figure 1-3). And, of course, you'll always have the option to expand or collapse all subtrees by right-clicking anywhere in the Packet Details pane to launch its pop-up menu.

In Figures 1-2 and 1-3, packet number 7 is selected. Whatever packet is selected in the Packet List pane is the packet presented in the panes below it. In this case, it's packet number 7 showing within the Packet Details pane.

NOTE

Packets are usually numbered based on the time they are received, although this isn't guaranteed. The packet capture (pcap) library determines how to order the packets.

If you double-click this packet, a separate window appears, to open the packet details. This is useful when you want to visually compare two different packets quickly. The Packet Details area in Figure 1-3 shows various rows of information that can be expanded or collapsed.

Capturing Enough Detail

The first row contains metadata regarding the packet, such as the number of the packet, when it was captured, on what interface it was captured, and the number of bytes captured versus the number of bytes that were on the wire. That last part might sound a little strange. Wouldn't you always capture all the bytes that go across the wire? Not necessarily. Some network capture tools allow you to capture only a subset of the bytes that are actually transmitted across the wire. This is useful if you only want to get an idea of the type of packets that are going across the wire but not what actual data those packets have, which can greatly reduce the size of the packet capture. The downside, of course, is that you get only a limited amount of information. If disk space is not an issue, feel free to capture it all. Just be mindful that you are capturing and storing all traffic traversing that network cable, which can quickly become a significant amount.

There are ways to limit the size of the capture. For example, instead of truncated packet data, capture only specific packet types and not all traffic. If someone wants to send you a capture, or if you want to see specific traffic, you can have Wireshark capture only the traffic you want, saving space. Everything is done using the right filters – and that section is coming soon enough!

Packet Bytes Pane

What follows the Packet Details pane is the Packet Bytes pane. This pane is at the bottom of the screen and wins the award for least intuitive. At first glance, it simply looks like gibberish. Bear with me for a couple of paragraphs; it will all make sense soon.

Offsets, Hex, and ASCII

You can see the Packet Bytes pane is divided into three columns. The first, left-most column simply counts incrementally: 0000, 0010, 0020, and so on. That's the offset (in hexadecimal) of the selected packet. Here, offset simply means the number of bits off from the beginning – again, counting in hexadecimal (where 0x0010 = 16 in decimal). The middle column shows information, in hexadecimal, at that offset. The right-hand column shows the same information, but in ASCII. For example, the total amount of information from the very beginning (offset 0000) to offset 0010 is 16 bytes. The middle column shows each of the 16 bytes in hex. The right-hand column shows each of the 16 bytes in ASCII characters. When a hexadecimal value doesn't translate to a printable ASCII character, only a “.” (period), is shown. So the Packet Bytes pane is actually the raw packet data as seen by Wireshark. By default, it is displayed in hex bytes.

Right-clicking the pane gives you the option to convert the hex bytes into bits, which is the purest representation of the data, though often this might not be as intuitive as the hex representation. Another neat feature is that any row you highlight within the Packet Details pane causes the corresponding data within the Packet Bytes pane to be highlighted. This can be helpful when troubleshooting Wireshark's dissection, as it allows you to see exactly which packet bytes the dissector is looking at.