Packet capture via CLI command

To use the built-in sniffer, connect to the CLI and enter the following command:

diagnose network sniffer [{any | <interface_name>} [{none | '<filter_str>'} [{1 | 2 | 3 | 4 | 5 | 6} [<count_int> <tsformat>]]]]

where:

• <interface_name> is either the name of a network interface, such as port1, or enter any for all interfaces.
• '<filter_str>' is the sniffer filter that specifies which protocols and port numbers that you do or do not want to capture, such as 'tcp port 80', or enter none for no filters. Filters use tcpdump (http://www.tcpdump.org) syntax.
• {1 | 2 | 3} is an integer indicating whether to display the network interface names, packet headers, and/or payloads for each packet that the network interface sends, receives, or sees:
• 1—Display the packet capture timestamp, plus basic fields of the IP header: the source IP address, the destination IP address, protocol name, and destination port number.
  

  Does not display all fields of the IP header; it omits:

  • IP version number bits
  • Internet header length (ihl)
  • Type of service/differentiated services code point (tos)
  • Explicit congestion notification
  • Total packet or fragment length
  • Packet ID
  • IP header checksum
  • Time to live (TTL)
  • IP flag
  • Fragment offset
  • Options bits
  • For example:

  
  
  

• 2—All of the output from 1, plus the packet payload in both hexadecimal and ASCII. For example:
  

• 3—All of the output from 2, plus the link layer (Ethernet) header. e.g.:

• 4—All of the output from 2, plus the ingress or egress interface.
  

• 5—All of the output from 2, plus the ingress or egress interface.
  

• 6—All of the output from 3, plus the ingress or egress interface.
  

• <count_int> is the number of packets the sniffer reads before stopping. Packet capture output is printed to your CLI display until you stop it by pressing Ctrl+C, or until it reaches the number of packets that you have specified to capture.

Packet capture can be very resource intensive. To minimize the performance impact on your FortiWeb appliance, use packet capture only during periods of minimal traffic, with a local console CLI connection rather than a Telnet or SSH CLI connection, and be sure to stop the command when you are finished.

For example, you might capture all TCP port 443 (typically HTTPS) traffic occurring through port1, regardless of its source or destination IP address. The capture uses a high level of verbosity (indicated by 3).

A specific number of packets to capture is not specified. As a result, the packet capture continues until the administrator presses Ctrl+C. The sniffer then confirms that five packets were seen by that network interface.

(Verbose output can be very long. As a result, output shown below is truncated after only one packet.)

• <tsformat> is the format of timestamp.

  • a: absolute UTC time, yyyy-mm-dd hh:mm:ss.ms

  • otherwise: relative to the start of sniffing, ss.ms

FortiWeb# FortiWeb# diagnose network sniffer port1 'tcp port 443' 3

interfaces=[port1]

filters=[tcp port 443]

10.651905 192.168.0.1.50242 -> 192.168.0.2.443: syn 761714898

0x0000 0009 0f09 0001 0009 0f89 2914 0800 4500 ..........)...E.

0x0010 003c 73d1 4000 4006 3bc6 d157 fede ac16 .<s.@.@.;..W....

0x0020 0ed8 c442 01bb 2d66 d8d2 0000 0000 a002 ...B..-f........

0x0030 16d0 4f72 0000 0204 05b4 0402 080a 03ab ..Or............

0x0040 86bb 0000 0000 0103 0303 ..........

Instead of reading packet capture output directly in your CLI display, you usually should save the output to a plain text file using your CLI client. Saving the output provides several advantages. Packets can arrive more rapidly than you may be able to read them in the buffer of your CLI display, and many protocols transfer data using encodings other than US-ASCII. It is often, but not always, preferable to analyze the output by loading it into in a network protocol analyzer application such as Wireshark (http://www.wireshark.org).

For example, you could use PuTTY or Microsoft HyperTerminal to save the sniffer output to a file. Methods may vary. See the documentation for your CLI client.

Requirements

• Terminal emulation software such as PuTTY (http://www.chiark.greenend.org.uk/~sgtatham/putty/download.html)
• A plain text editor such as Notepad
• A Perl interpreter (http://www.perl.org/get.html)
• Network protocol analyzer software such as Wireshark (http://www.wireshark.org)

To view packet capture output using PuTTY and Wireshark

1. On your management computer, start PuTTY.
2. Use PuTTY to connect to the FortiWeb appliance using either a local console, SSH, or Telnet connection. For details, see the FortiWeb CLI Reference:

HTTPS://docs.fortinet.com/product/fortiweb/

3. Type the packet capture command, such as:

diagnose network sniffer port1 'tcp port 443' 3

but do not press Enter yet.

4. In the upper left corner of the window, click the PuTTY icon to open its drop-down menu, then select Change Settings. A dialog appears where you can configure PuTTY to save output to a plain text file.
5. In the Category tree on the left, go to Session > Logging.
6. In Session logging, select Printable output.
7. In Log file name, click the Browse button, then choose a directory path and file name such as C:\Users\MyAccount\packet_capture.txt to save the packet capture to a plain text file. (You do not need to save it with the .log file extension.)
8. Click Apply.
9. Press Enter to send the CLI command to the FortiWeb appliance, beginning packet capture.
10. If you have not specified a number of packets to capture, when you have captured all packets that you want to analyze, press Ctrl + C to stop the capture.
11. Close the PuTTY window.
12. Open the packet capture file using a plain text editor such as Notepad.
13. Delete the first and last lines, which look like this:

=~=~=~=~=~=~=~=~=~=~=~= PuTTY log 10/6/2023.07.25 11:34:40 =~=~=~=~=~=~=~=~=~=~=~=

FortiWeb-2000 #



These lines are a PuTTY timestamp and a command prompt, which are not part of the packet capture. If you do not delete them, they could interfere with the script in the next step.

14. Convert the plain text file to a format recognizable by your network protocol analyzer application.

You can convert the plain text file to a format (.pcap) recognizable by Wireshark (formerly called Ethereal) using the fgt2eth.pl Perl script. To download fgt2eth.pl, see the Fortinet Knowledge Base article "Troubleshooting Tool: Using the FortiOS built-in packet sniffer (http://kb.fortinet.com/kb/documentLink.do?externalId=11186).

The fgt2eth.pl script is provided as-is, without any implied warranty or technical support, and requires that you first install a Perl module compatible with your operating system.

To use fgt2eth.pl, open a command prompt, then enter a command such as the following:

fgt2eth.pl -in packet_capture.txt -out packet_capture.pcap

where:

• fgt2eth.pl is the name of the conversion script; include the path relative to the current directory, which is indicated by the command prompt
• packet_capture.txt is the name of the packet capture’s output file; include the directory path relative to your current directory
• packet_capture.pcap is the name of the conversion script’s output file; include the directory path relative to your current directory where you want the converted output to be saved

Open the converted file in your network protocol analyzer application. For further instructions, see the documentation for that application.

Viewing sniffer output in Wireshark

For additional information on packet capture, see the Fortinet Knowledge Base article "Troubleshooting Tool: Using the FortiOS built-in packet sniffer (http://kb.fortinet.com/kb/documentLink.do?externalId=11186).

For more information on CLI commands, see the FortiWeb CLI Reference:

HTTPS://docs.fortinet.com/product/fortiweb/