The FireBrick includes the ability to capture packet dumps for diagnostic purposes. This might typically be used where the behaviour of the FB6000 is not as expected, and can help identify whether other devices are correctly implementing network protocols - if they are, then you should be able to determine whether the FB6000 is responding appropriately. The packet dumping facility may also be of use to you to debug traffic (and thus specific network protocols) between two hosts that the brick is routing traffic between.
This feature is provided via the FB6000's HTTP server and
provides a download of a pcap format file (old format) suitable for use with tcpdump
or Wireshark.
A packet dump can be performed by either of these methods :-
tcpdump
or Wiresharkcurl
)The output is streamed so that, when used with curl
and tcpdump
, you can monitor traffic in real time.
Limited filtering is provided by the FB6000, so you will normally apply any additional filtering you need via tcpdump
.
Table 13.1 lists the parameters you can specify to control what gets dumped. The "Parameter name" column shows the exact parameter name to specify when constructing a URL to use with an HTTP client. The "Web-form field" column shows the label of the equivalent field on the user interface form.
Table 13.1. Packet dump parameters
Parameter name | Web-form field | Function |
interface | Interface | One or more interfaces, as the name of the interface. e.g. interface=WAN, also applies for name of PPPoE on an interface |
snaplen | Snaplen | The maximum capture length for a packet can be specified, in bytes. Default 0 (auto). See notes below. |
timeout | Timeout | The maximum capture time can be specified in seconds. Default 10. |
party | IPs (2-off) | IPs, blocks and ranges can be specified to filter packets. Filling in one field will capture any traffic to/from the given host(s). Filling in a second field will only capture traffic between the specified hosts. |
self | Include this TCP session | By default any TCP traffic to or from the IP and ports of the TCP session used for the dump is excluded. You can opt to include it, but that is usually a bad idea for obvious reasons as you may be dumping the packet dump itself. Note that if dumping tunnel wrappers and dumping via that tunnel then the dump traffic is not normally detected/excluded. |
The following criteria must be met in order to use the packet dump facility :-
level
attribute on the user
object.You may optionally specify up to two IP address to be checked for a match in packets on the interface(s) and/or L2TP session(s) specified. If you do not specify any IP addresses, then all packets are returned. If you specify one IP address then all packets containing that IP address (as source or destination) are returned. If you specify two IP addresses then only those packets containing both addresses (each address being either as source or destination) are returned.
IP matching is only performed against ARP, IPv4 or IPv6 headers and not in encapsulated packets or ICMP payloads.
If capturing too much, some packets may be lost.
The capture can collect different types of packets depending on where the capture is performed. All of these are presented as Ethernet frames, with faked Ethernet headers where the packet type is not Ethernet.
Table 13.2. Packet types that can be captured
Type | Notes |
Ethernet | Interface based capture contains the full Ethernet frame with any VLAN tag removed. |
IP | IP only, currently not possible to capture at this level. An Ethernet header is faked. |
PPP | PPP from the protocol word (HDLC header is ignored if present). An Ethernet header is faked and also a PPPoE header. The PPPoE header has the session PPPoE ID that is the local end L2TP session ID. |
The faked protocol header has target MAC of 00:00:00:00:00:00 and source MAC of 00:00:00:00:00:01 for received packets, and these reversed for sent packets.
The snaplen
argument specifies the maximum length captured, but this applies at the protocol level. As such PPP packets will have up to the snaplen
from the PPP protocol bytes and then have fake PPPoE and Ethernet headers added.
A snaplen
value of 0 has special meaning - it causes logging of just IP, TCP, UDP and ICMP headers as well as headers in ICMP error payloads. This is primarily to avoid logging data carried by these protocols.
The web form is accessed by selecting the "Packet dump" item under the "Diagnostics" main-menu item. Set up the dump parameters with reference to Table 13.1 and click the "Dump" button. Your browser will ask you to save a file, which will take time to save as per the timeout requested.
To perform a packet dump using an HTTP client, you first construct an appropriate URL that contains standard HTTP URL form-style parameters from the list shown in Table 13.1.
Then you retrieve the dump from the FB6000 using a tool such as curl
.
The URL is http://<FB6000 IP address or DNS name>/pcap?parameter_name=value [¶meter_name=value ...]
The URL may include as many parameter name and value pairs as you need to completely specify the dump parameters.
Packet capturing stops if the output stream (HTTP transfer) fails. This is useful if you are unable to determine a suitable timeout period, and would like to run an ongoing capture which you stop manually. This is achieved by specifying a very long duration, and then interrupting execution of the HTTP client using Ctrl+C or similar.
Only one capture can operate at a time. The HTTP access fails if no valid interfaces or sessions etc. are specified or if a capture is currently running.
An example of a simple real-time dump and analysis run on a Linux box is shown below :-
curl --silent --no-buffer --user name:pass 'http://1.2.3.4/pcap?interface=LAN&timeout=300&snaplen=1500' | /usr/sbin/tcpdump -r - -n -v
In this example we have used username name and password pass to log in to a FireBrick on address 1.2.3.4 - obviously you would change the IP address (or host name) and credentials to something suitable for your FB6000.
We have asked for a dump of the interface named LAN
, with a 5 minute timeout and capturing 1500 byte packets. We have then fed the
output in real time (hence specifying --no-buffer
on the curl
command) to tcpdump
, and asked it to take
capture data from the standard input stream (via the -r -
options). We have additionally asked for no DNS resolution (-n
) and verbose output (-v
).
Consult the documentation provided with the client (e.g. Linux box) system for details on the extensive range of tcpdump
options - these
can be used to filter the dump to better locate the packets you are interested in.