The Details in an Nmap Protocol Decode

In the last mini-training course lesson, we learned about the different methods of gathering the network traffic created during our Nmap scans. In this lesson, we’ll go through our trace file and see exactly what happens when Nmap hits the network.

An Overview of a Protocol Decode

In this lesson, we’ll use Wireshark to provide a visual explanation of Nmap traffic information. In the Wireshark protocol decode screen, there are three display panes. The top pane describes a summary of each network packet, the middle pane shows details associated with the current selected frame, and the bottom pane displays a hexadecimal representation of the packet contents.

For our purposes during this lesson, we’ll focus on the contents of the top pane only. Although there’s some interesting information in the details and hex, our concern is the basics of the Nmap scan at a packet level.

The Nmap Output

Before we examine the trace file, let’s review the information from our original Nmap scan:


Starting Nmap 4.21ALPHA4 ( ) at 2007-04-10 17:26 Eastern Daylight Time
Interesting ports on (
Not shown: 1698 filtered ports
22/tcp open ssh
25/tcp closed smtp
53/tcp open domain
70/tcp closed gopher
80/tcp open http
113/tcp closed auth

Nmap finished: 1 IP address (1 host up) scanned in 94.062 seconds

In our example, we scanned the host with the default Nmap SYN scan. Three open ports were identified, three closed ports were found, and 1,698 ports were identified as filtered. In our trace file, we should be able to easily identify each of the open and closed port conversations and verify that the large number of filtered ports provided no response to Nmap’s probes.

The DNS Lookup

We began our trace by specifying the host Since our Nmap station doesn’t know the IP address of this device, the first thing that occurs is a normal DNS lookup for

Frame 1 is the DNS request for, sent from to the default DNS server of Frame 2 shows the response to this query with the IP address of the host of

The Nmap Ping

Before Nmap will begin scanning our newly-resolved IP address, it first needs to confirm that the remote device is active on the network. To complete this process, Nmap sends two frames to the target device to help determine the disposition. Frame 3 contains a traditional ICMP echo request and frame 4 is a not-so traditional out-of-order TCP ACK to port 80 of the remote device.

The goal is to obtain an answer of some kind to either of these probes. As we can see in our trace, frame 5 includes the normal ICMP echo reply and frame 6 is a RST frame in response to our out-of-order TCP ACK. Although we only needed one of these responses, both of them in return ensure us that the remote device is absolutely available.

The Reverse DNS

Although we began this scan with a normal DNS query, Nmap also performs a reverse DNS to determine what the “real” name of the remote device might be. Often, the name used in the original query will be a different name configured in the remote DNS servers.

Frame 7 includes the reverse DNS query, and frame 8 includes the reply. In this example, the answer to the reverse DNS is the same name as the original query:

The SYN Scan

The rest of the trace file is associated with the actual Nmap TCP SYN scan. As we can see from this example, Nmap performs scans to devices in groups instead individually. This allows an efficient scanning process that does not overly tax the remote device.

Frames 9 through 18 begin the scan with a block of TCP SYN queries to random port numbers. Frame 19 is the first response from these probes, and the RST frame received in response confirms that port 25 is closed to all TCP traffic.

Frame 23 is the first positive response with a SYN/ACK to the original SYN frame on port 80, and it’s followed immediately by another positive “open” port SYN/ACK for port 22 in frame 24. Later on in the trace, our Nmap station sends a RST frame to both of these ports to close the connection before the TCP three-way handshake can be established.

Do It Yourself

Every scan has different characteristics and traffic patterns. Some scans send thousands of frames, and others just send a few. To see what’s really happening, perform your own experiments with Wireshark and Nmap and you’ll be able to change the speed and tempo of your own Nmap scans!