Packit(8) Packet analysis and injection tool

SYNOPSIS

Packet capture:
packit -m capture [-cGHnvsX] [-i interface] [-r|-w file] expression

Packet injection:
packit -m inject [-t protocol] [-aAbcCdDeFgGhHjJkKlLmMnNoOpPqQrRsSTuUvwWxXyYzZ] [-i interface]

DESCRIPTION

Packit is a network auditing tool. It's value is derived from its ability to customize, inject, monitor, and manipulate IP traffic. By allowing you to define (spoof) all TCP, UDP, ICMP, IP, ARP, RARP and Ethernet header options, Packit can be useful in testing firewalls, intrusion detection systems, port scanning, simulating network traffic and general TCP/IP auditing. Packit is also an excellent tool for learning TCP/IP.

PACKIT BASE OPTIONS

-m mode

      Select a runtime mode. Currently supported modes

      are capture, inject and trace. The default is inject.

PACKET CAPTURE OPTIONS

Packet capture options are as follows:

-c count

      Specify the number of packets to capture

-e Display link-layer header data.

-G Display the timestamp in GMT rather than localtime.

-i interface

      Listen on interface. If unspecified, packit will use the lowest

      numbered device in the 'up' state (excluding loopback).

-n Don't resolve host addresses to names but resolve ports numbers.

      Disables DNS lookups.

-nn Don't resolve ports numbers to their protocol names but resolve

      host addresses.

-nnn Don't resolve host addresses or port numbers.

-r file

      Read packet data from tcpdump formated binary log file. (example:

      a file created with -w)

-s snaplen

      Read snaplen bytes of data from each packet rather than the

      default of 68.

-v Enables verbose packet capture.

-w file

      Write the raw packets to file rather than displaying time to

      stderr.

-X Display hexadecimal & ascii dump of each packet up to snap

      length bytes.

expression

      selects which packets should be displayed. If no expression is

      given, all packets are displayed. See the tcpdump(1) man page for

      more detailed information.

PACKET INJECTION / TRACE


Packet injection is used to define and inject IP based network traffic onto your network. You have the ability to define essentially any ARP, IP, TCP, UDP, ICMP and Ethernet header value. This can be valuable in a number of ways, including testing firewalls, intrusion detection systems, simulating traffic flow and general TCP/IP auditing.

CHOOSE YOUR PROTOCOL


-t protocol

      Specify the type of packet to inject. Supported values are: ARP,

      RARP, TCP, UDP and ICMP. This option defaults to TCP in inject

      mode and to ICMP in trace mode.

PACKET INJECTION / TRACE GENERAL

This section documents the operational command-line options.

-c count

      The value of count is the total number of packets we would like

      to inject (a count value of 0 means forver).

-b burst rate

      Specifies the number of packets to inject every interval (defined

      by -w). (A burst rate of 0 will send packets as quickly as

      possible)

-h

      Host response mode. Enabling this option will print any packet

      you inject and then wait (see -H for timeout) to see if the remote

      host responds.

-H timeout

      Specify the timeout value (in seconds) to use with '-h'.

      This value defaults to '1' second.

-i interface

      Specify the interface to transmit from, if the machine has

      multiple interfaces.

-v

      Verbose injection mode. Displays each packet you inject. It

      also has the same effect as in capture mode while used with

      the '-h' option.

-p payload

      This option defines the payload portion of the header.

      Hex payload should be prefixed with '0x' with each value

      separated by a whitespace.

      ASCII Example: -p 'hello, this is my packet'

      Hex Example: -p '0x 70 61 63 6B 69 74'

-w interval

      Specify the number of seconds to wait between packet bursts.

      This value defaults to '1' second.

-Z length

      Specify the size of the packet(s) to inject. (Max: 65535)

IP HEADER OPTIONS

This section documents the IP header command-line options.

-s src address

      The IP address the packet will appear to come from. If

      unspecified, packit will default to the IP address of the

      lowest numbered device in the 'up' state (excluding loopback).

-sR Use a random source IP address.

-d dst address

      The IP address of the machine you would like to contact.

-dR Use a random destination IP address.

-o type of service

      TOS values are typically in hexidecimal format, however, packit

      only accepts TOS values as integers.


      Below are the 4 valid TOS bit values:


      - Minimize delay: 16 (0x10)

      - Maximize throughput: 8 (0x08)

      - Maximize reliability: 4 (0x04)

      - Minimize monetary cost: 2 (0x02)

-n ID number

      The ID number is used to identify each datagram sent by a host.

      It generally increments by one with each datagram sent. This

      value is random by default.

-T TTL

      The TTL value defines the upper limit on the number of devices

      through which the datagram may pass to reach it's destination.

      The default value is 128.

-V IP protocol number

      Specify the IP protocol assocated with this packet (RAWIP only).

      The default value is 255.

TCP HEADER OPTIONS

This section documents the TCP header command-line options.

-S src port

      The port from which our source address is communicating from. This

      value is random by default.

-D dst port

      The port on the destination we would like to communicate on. In

      inject mode this value is 0 by default while in trace mode this

      value is random by default. You may also specify a range of

      addresses in the format: -D 1:1024.

-f Do not fragment this packet.

-F tcp flags

      There are 6 TCP header flag bits. They can be used in combination

      with one another and are specified using the following

      identifiers:


      - S : SYN (Synchronization sequence number)

      - F : FIN (Sender is finished)

      - A : ACK (Acknowledgement number is valid)

      - P : PSH (Receiver should push this data to the remote host)

      - U : URG (The urgent pointer is valid)

      - R : RST (Reset this connection)


      As an example, to set the SYN and FIN bits use the

      following: -F SF

-q sequence number

      The sequence number is a 32-bit unsigned (positive) number used

      to identify the byte in a stream of data from the sending TCP

      to the receiving TCP that the first byte of data represents.

-a ack number

      The acknowledgement (ack) number defines the next sequence

      number that the sender of the ack expects to see. It is

      typically the sequence number + 1 during valid TCP

      communication. It is a 32-bit unsigned (positive) number.

-W window size

      The window size provides flow control. It is a 16-bit

      number that defines how many bytes the receiver is willing

      to accept. The default value is 1500.

-u urgent pointer

      In valid TCP communication, the urgent pointer is

      only useful if the URG flag is set. Used with the

      sequence number, it points to the last byte of urgent

      data.

UDP HEADER OPTIONS

This section documents the UDP header command-line options. UDP is the default IP protocol for TRACE mode.

-S src port

      The port from which our source address is communicating from. This

      value is random by default.

-D dst port

      The port on the destination we would like to communicate on. In

      inject mode this value is 0 by default while in trace mode this

      value is random by default. You may also specify a range of

      addresses in the format: -D 1:1024.

ICMP HEADER OPTIONS

This section documents the ICMP header command-line options.

-K type

      Specify the ICMP type. See /usr/share/docs/packit/ICMP.txt for details on types.

-C code

      Specify the ICMP code. See /usr/share/docs/packit/ICMP.txt for details on codes.

ECHO REQUEST / ECHO REPLY OPTIONS

-N id number

      Define the 16-bit ICMP identification number. This value is

      random by default.

-Q sequence number

      Define the 16-bit ICMP sequence number. This value is random

      by default.

UNREACHABLE / REDIRECT / TIME EXCEEDED OPTIONS

-g gateway

      Define the gateway in which to redirect traffic to. This option

      is only used for ICMP redirects (type 5).

-j address

      Define the source address of the original packet.

-J src port

      Define the source port of the original packet.

-l address

      Define the destination address of the original packet.

-L dst port

      Define the destination port of the original packet.

-m time to live

      Define the Time To Live of the original packet. This option

      defaults to 128.

-M id

      Define the IP ID of the original packet. This option defaults

      to random.

-O type of service

      Define the Type of Service of the original packet. See the

      -o option for the possible values.

-P protocol

      Define the protocol of the original packet. This option

      defaults to UDP.

MASK REQUEST / MASK REPLY OPTIONS

-N id number

      Define the 16-bit ICMP identification number. This value is

      random by default.

-Q sequence number

      Define the 16-bit ICMP sequence number. This value is random

      by default.

-G address mask

      Define the address network mask. The default value for this

      option is:255.255.255.0.

TIMESTAMP REQUEST / TIMESTAMP REPLY OPTIONS

-N id number

      Define the 16-bit ICMP identification number. This value is

      random by default.

-Q sequence number

      Define the 16-bit ICMP sequence number. This value is random

      by default.

-U original timestamp

      Define the 32-bit original timestamp. This value is 0

      by default.

-k received timestamp

      Define the 32-bit received timestamp. This value is 0

      by default.

-z transmit timestamp

      Define the 32-bit transmit timestamp. This value is 0

      by default.

ARP AND RARP HEADER OPTIONS

This section documents the ARP/RARP header command-line options. In my opinion, these options have the ability to do the most damage with the least effort, especially on large cable and DSL networks. Use with caution.

Packit only supports ARP/RARP protocol addresses in IPv4 format

-A operation type

      Define the ARP / RARP / IRARP operation type. The valid options

      are as follows:


      - 1 : ARP Request (Default for ARP packages.)

      - 2 : ARP Reply

      - 3 : Reverse ARP Request (Default for RARP packages.)

      - 4 : Reverse ARP Reply

      - 5 : Inverse ARP Request

      - 6 : Inverse ARP Reply

-y target IP address

      The IP address of the target host.

-yR Use a random target host IP address.

-Y target ethernet address

      The ethernet (hardware) address of the target host.

-YR Usage a random target host ethernet address.

-x sender IP address

      The IP address of the sender host.

-xR Use a random sender host IP address.

-X sender ethernet address

      The ethernet (hardware) address of the sender host.

-XR Usage a random sender host ethernet address.

ETHERNET HEADER OPTIONS

This section documents the Ethernet header command-line options.

-e src ethernet address

      The ethernet (hardware) address the packet will appear to come

      from.

-eR Use a random source ethernet address.


      If you define this, you will most likely need to define the
      destination ethernet header value as well. When using either -e or

      -E, you enable link level packet injection and enable link level

      packet injection and the destination cannot be auto-defined while

      injecting in this manner.

-E dst ethernet address

      The ethernet (hardware) of the next routable interface the packet

      will cross while making it's way to the destination.

-ER Use a random destination ethernet address.

The following two rules should be followed if you actually want the destination to receive to receive the packets you're sending:

1) If the destination exists beyond your default route (gateway),

      the destination ethernet address should be set to the default

      routes address should be set to the default routes ethernet

      address. This can typically be found by using the arp(8) command.

  2) If the destination exists on your subnet, the destination

      ethernet address should be set to its ethernet address. This

      can typically be found by using the arp command.

PACKET CAPTURE EXAMPLES

To print all TCP communications that doesn't revolve around SSH (port 22).

      packit -m cap 'tcp and not port 22'

To print the start and end packets (the SYN and FIN pack- ets) of each TCP conversation that involves a non-local host, don't resolve addresses and display hex/ascii dump of the packet.


      packit -m cap -nX 'tcp[tcpflags] & (tcp-syn|tcp-fin) != 0

             and not src and dst net localnet'

To write the first 10 ICMP packets captured to a file.


      packit -m cap -c 10 -w /tmp/mylog 'icmp'

PACKET INJECTION EXAMPLES

IMPORTANT: The ethernet address 'f:00:d:f:00:d' in these examples is a mock representation of the ethernet address of my default route. In order for these examples to work properly, you would need to change it to your correct default route ethernet address.

Inject 10 ICMP type 8 (echo request) packets from host '3.1.33.7' to host '192.168.0.1' and watch for a response.


      packit -t icmp -s 3.1.33.7 -d 192.168.0.1 -c 10 -h

Inject an ICMP type 18 (mask reply) packet with an ICMP id of 211 and an address mask of 255.255.255.0.


      packit -t icmp -K 18 -d 127.0.0.1 -N 211 -G 255.255.255.0

Inject 5 TCP packets from random hosts to 'www.microsoft.com' with the SYN flag set, a window size of 666, a random source ethernet address, a destination ethernet address of f:00:d:f:00:d, with a payload of "HI BILL", displaying each packet injected.

 
      packit -sR -d www.microsoft.com -F S -c 5 -W 666

             -eR -E f:00:d:f:00:d -p 'HI BILL' -v

Inject a total of 1000 TCP packets in 20 packet per second bursts from 192.168.0.1 on port 403 to 192.168.0.20 on port 80 with the SYN and RST flags set, a sequence number of 12345678910 and a source ethernet address of 0:0:0:0:0:0.


      packit -s 192.168.0.1 -d 192.168.0.20 -S 403 -D 80

             -F SR -q 12345678910 -c 1000 -b 20 -e 0:0:0:0:0:0

Inject a TCP packets from 10.22.41.6 to 172.16.1.3 on ports ranging from 1-1024 with the SYN flag set and display each packet we send.


      packit -s 10.22.41.6 -d 172.16.1.3 -D 1-1024 -F S -v

Inject a broadcast ARP reply stating that 4.3.2.1 is at 5:4:3:2:1:0. Also, spoof the source ethernet adddress for a little more authenticity and supply the payload in hex.


      packit -t arp -A 2 -x 4.3.2.1 -X 5:4:3:2:1:0 -e 5:4:3:2:1:0

             -p '0x 70 61 63 6B 69 74'

TRACE ROUTE EXAMPLES

Appear as a DNS response by using a UDP source port of 53 (DNS)


      packit -m trace -t UDP -d 192.168.2.35 -S 53

Appear as HTTP traffic by using TCP port 80


      packit -m trace -t TCP -d www.google.com -S 80 -FS

BUGS

Due to limitations in some versions of *BSD, specifying arbitrary ethernet and/or ARP header data may not be supported.

ARP capture data is incomplete.

Like this man page, packit is still very much a work in progress. Please send bug reports, questions or requests to [email protected].

AUTHOR

Darren Bounds <[email protected]>

The latest version can be found at:

        http://packetfactory.openwall.net/projects/packit