1.1 What is a "packet sniffer"?
However, computer conversations consist of apparently random binary data. Therefore, network wiretap programs also come with a feature known as "protocol analysis", which allow them to "decode" the computer traffic and make sense of it.
Sniffing also has one advantage over telephone wiretaps: many networks use "shared media". This means that you don't need to break into a wiring closet to install your wiretap, you can do it from almost any network connection to eavesdrop on your neighbors. This is called a "promiscuous mode" sniffer. However, this "shared" technology is moving quickly toward "switched" technology where this will no longer be possible, which means you will have to actually tap into the wire.
1.2 What is it used for?
Sniffing programs have been around for a long time in two forms. Commercial packet sniffers are used to help maintain networks. Underground packet sniffers are used to break into computers.
Typical uses of such wiretap programs include:
1.3 Is there a single point on the Internet I can plug into in order to see all the traffic?
Think of it this: you have two machines in your own office talking to each other, and both are on the Internet. They take a direct route of communication, and the traffic never goes across the outside public portion of the Internet. Any communication anywhere in the net follows a similar "least-cost-path" principle.
1.4 How does sniffing/wiretap work?
This implies that all machines are able to "see" all the traffic on the same wire.
Thus, Ethernet hardware is built with a "filter" that ignores all traffic that doesn't belong to it. It does this by ignoring all frames whose MAC address doesn't match.
A wiretap program turns off this filter, puttin the Ethernet hardware into "promiscuous mode". Thus, Mark can see all the traffic between Alice and Bob, as long as they are on the same Ethernet wire.
This is accomplished by putting a unique 12-digit hex number in every piece of Ethernet hardware. Section 1.5.4 explains how to discover the Ethernet MAC address of your own machine.
To really understand why this is so important, you might want to review the information in section 5.4 below. Ethernet was designed to carry other traffic than just TCP/IP, and TCP/IP was designed to run over other wires (such as dial-up lines, which use no Ethernet). For example, many home users install "NetBEUI" for File and Print Sharing because it is unrelated to TCP/IP, and therefore hackers from across the Internet can't get at their hard-drives.
Raw transmission and reception on Ethernet is governed by the Ethernet equipment. You just can't send data raw over the wire, you must first do something to it that Ethernet understands. In much the same way, you can't stick a letter in a mailbox, you must first wrap it in an envelope with an address and stamp.
Following a is a brief explanation how this works:
_________
/.........\
/..Internet.\
+-----+ +------+.........+---+
|Alice|-----|ROUTER|.........|Bob|
+-----+ ^ +------+.........+---+
| \.........../
| \---------/
+-------+
|wiretap|
+-------+
Alice has IP address: 10.0.0.23In order to talk to Bob, Alice needs to create an IP packet of the form 10.0.0.23-->192.168.100.54
As the packet traverses the Internet, it will be passed from router-to-router. Therefore, Alice must first hand off the packet to the first router. Each router along the way will examine the destination IP address (192.168.100.54) and decide the correct path it should take.
In the able diagram, we draw the Internet as a "cloud". All Alice knows about is the local connection to the first router, and Bob's eventual IP address. Alice knows nothing about the structure of the Internet and the route that packet will take.
Alice must talk to the router in order to send the packet. She uses the Ethernet to do so. An Ethernet frame looks like the following:
+--+--+--+--+--+--+ | destination MAC | +--+--+--+--+--+--+ | source MAC | +--+--+--+--+--+--+ |08 00| +--+--+-----------+ | | . IP . . packet . . . | | +--+--+--+--+-----+ | CRC | +--+--+--+--+What this means is that the TCP/IP stack in Alice's machine might create a packet that is 100 bytes long (let's say 20 bytes for the IP info, 20 bytes for the TCP info, and 60 bytes of data). The TCP/IP stack then sends it to the Ethernet module, which puts 14 bytes on the front for the destination MAC address, source MAC address, and the ethertype 0x0800 to indicate that the other end's TCP/IP stack should process the frame. It also attaches 4-bytes on the end with a checksum/CRC (a validator to see if the frame gets corrupted as it goes across the wire).
The adapter then sends the bits out onto the wire.
All hardware adapters on the wire see the frame, including the ROUTER's adapter, the packet sniffer, and any other machines. Proper adapters, however, have a hardware chip that compares the frame's "destination MAC" with its own MAC address. If they don't match, then it discards the frame. This is done at the hardware level, so the machine the adapter is attached to is completely unaware of this process.
When the ROUTER ethernet adapeter sees this frame, it reads it off the wire and removes the leading 14-bytes and the trailing 4-bytes. It looks at the 0x0800 ethertype and decides to send it to the TCP/IP stack for processing (which will presumably forward it to the next router in the chain toward the destination).
In the above scenario, only the ROUTER machine is supposed to see the Ethernet frame, and all other machines are supposed to ignore it. The wiretap, however, breaks the rules and copies the frame off the network, too.
See Charles Spurgeon's Ethernet website at: http://wwwhost.ots.utexas.edu/ethernet/ethernet-home.html
1.5.2 What does "MAC" stand for?
The logic behind this is that the Ethernet has multiple sublayers, PHY (physical), MAC, LLC (logical). The Ethernet address is considered part of the MAC sublayer. The physical layer is responsible for the wire, the MAC is responsible for formatting the data that goes on the wire, and the logical is responisble for things such as retransmitting the data on the wire.
However, the OUI is really only 22-bits long, two of the bits in that field are used for other purposes. One bit indicates if the address is a "broadcast/multicast" address, the other bit indicates if the adapter has been reassigned a "locally administered address" (where a network administrator reassigns the MAC address to fit some local policy).
For example, you will commonly see the MAC address 03
00 00 00 00 01 on the wire. The first byte contains the binary
representation of 00000011 where both these special bits
are set (and the rest are zero). This is a special multicast packet that
is sent to all machines that run the "NetBEUI" protocol (which
is commonly installed on Windows machines to share files locally without
using TCP/IP as the transport).
The IEEE maintains the list of vendor/OUI codes at http://standards.ieee.org/regauth/oui/.
Windows NT IP Configuration
Host Name . . . . . . . . . : sample.robertgraham.com
DNS Servers . . . . . . . . : 192.0.2.254
Node Type . . . . . . . . . : Hybrid
NetBIOS Scope ID. . . . . . :
IP Routing Enabled. . . . . : No
WINS Proxy Enabled. . . . . : No
NetBIOS Resolution Uses DNS : No
Ethernet adapter SC12001:
Description . . . . . . . . : DEC DC21140 PCI Fast Ethernet Adapter
Physical Address. . . . . . : 00-40-05-A5-4F-9D
DHCP Enabled. . . . . . . . : No
IP Address. . . . . . . . . : 192.0.2.160
Subnet Mask . . . . . . . . : 255.255.255.0
Default Gateway . . . . . . : 192.0.2.1
Primary WINS Server . . . . : 192.0.2.253
eth0 Link encap:Ethernet HWaddr 08:00:17:0A:36:3E
inet addr:192.0.2.161 Bcast:192.0.2.255 Mask:255.255.255.0
UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
RX packets:1137249 errors:0 dropped:0 overruns:0
TX packets:994976 errors:0 dropped:0 overruns:0
Interrupt:5 Base address:0x300
1.5.5 What are the Ethernet addresses of machines I'm talking to?
1.5.6 Can I change my MAC address?
First, you can spoof your address. Remember that the MAC address is just part of the frame data. Therefore, when you send an Ethernet frame on the wire, you can overwrite your own contents in those frames. Of course, you have to be running a program that does this for some reason.
Second, most adapters allow you to reconfigure the runtime MAC address. For example, some cards allow you to reconfigure the address within the Windows control panel.
Third, you can re-burn (i.e. reprogram the EEPROM) the address in the card. You need a program/hardware that knows the specifics of the chipset being used by the card. This changes the card forever to have the new address.
In other words, you are asking about this scenario:
The answer is of course "no", it isn't even vaguely possible. You have to have access to the wire that the communication is going across in order to eavesdrop. Same as with telephones, same as everywhere.
Remote access to the wire
However, if you are a really, really good cracker/hacker, there are ways of getting access to those lines. Typical examples are:
Close to the wire
In some situations, like cable-modems, DSL, Ethernet VLANs, etc., you can redirect traffic between two people to go through your own machine. This is because while you are not directly in the path of communication, you can sometimes move that path to flow past your own computer. It's much like the concept that you can divert a stream slightly, though not very far. See the "Redirect section under Cable-Modems.
Rootkits and Remote Admin Trojans
Another possibility is to break into a person's machine and install a sniffing program. (Exactly how to break into someone's machine is, of course, beyond the scope of this document). On UNIX, sniffing programs are part of most "rootkits". On Windows, sniffing is part of some RATs (Remote Admin Trojans, e.g. BackOrifice).
In theory, these programs can be used to sniff traffic in general, but usually they are configured to simply sniff for passwords and e-mail them back to the hacker.
Section 2.5 contains information on how to detect when this has happened to you, especially how to detect sniffers installed on a UNIX host.
"Secure Sockets Layer", SSL is built into all popular web browsers and web servers. It allows encrypted web surfing, and is almost always used in e-commerce when users enter their credit card information.
This site for Apache SSL describes this: http://www.modssl.org/
E-mail can be sniffed in many alternative ways. It passes through corporate firewalls, which may monitor the traffic. It often gets logged and saved for extended periods of time. It may get accidentally misdirected, and end up in somebody else's mailbox. The best way to keep such e-mail secret is to encrypt it. The two common ways of doing this are with PGP (Pretty Good Privacy) and S/MIME (Secure MIME). PGP can be purchased as an add-on to many products. S/MIME is built into e-mail programs by Netscape and Microsoft.
"Secure Shell", ssh has become the de facto standard for logging into UNIX machines from the Internet. You should immediately replace telnet with this service. Numerous other protocols can be tunneled through ssh connections (i.e. file copy). The product was originally developed by a Finish company http://www.ssh.fi/ but many open-source/freeware implementations also exist.
VPNs provide encrypted traffic across the Internet. However, if a hacker compromises the end-nodes of a VPN connection, they can still sniff the traffic. A typical scenario is an end-user who surfs the Internet normally and gets compromised with a Remote Access Trojan (RAT) that contains a sniffing plug-in. When the user establishes the VPN connection, the sniffing program is able to see not only the encrypted traffic that can be seen on the Internet, but also the unencrypted traffic before it gets sent through the stack to the VPN.
In the Windows/SAMBA environment, make sure that you have the older LanManager authentication turned off. This requires SAMBA v2 or later, WinNT SP3 or later, and so on.
Both Windows 2000 and UNIX provide support for Kerberos authentication. This is one of the strongest generic mechanisms available. ftp://aeneas.mit.edu/pub/kerberos/doc/KERBEROS.FAQ
There are numerous smart card implementations around providing one-time passwords. These are often used when connecting remotely, either dial-in or VPN across the Internet.
Enhancements to Telnet and FTP for UNIX and Windows. http://srp.stanford.edu/srp/
Replacing your hub with a switch will provide a simple, yet effective defense against casual sniffing.
While this solution is extremely effective in practice (and should be strongly considered), it shouldn't be relied upon as a complete defense against sniffing. A switch still creates a "broadcast domain", providing an attacker the ability to spoof ARP packets.
The easiest such exploit is the "router redirection". ARP queries contain the correct IP-to-MAC mapping for the sender. In order to reduce ARP traffic, most machines will cache this information that they read from the query broadcasts. Therefore, a malicious attacker can redirect nearby machines to forward traffic through it by sending out regular ARP packets containing the router's IP address mapped to its own MAC address. All the machines on the local wire will believe the hacker is the router, and therefore pass their traffic through him/her.
A similar attack would be to DoS a target victim and force it off the network, then begin using its IP address. If a hacker does this carefully, s/he can inherit connections already established without dropping them. Windows machines are even so polite that when they come onto the network and see someone else using their address, they will kindly shut down their own TCP/IP stacks and allow this to continue. SMB (the Windows file sharing protocol) is also kind enough to allow predictable identifiers, allowing cr/hackers to predict enough information to keep the connection going.
Most intrusion detection systems and even network management tools like the Expert Sniffer(tm) will detect these shenanigans. For example, putting the BlackICE IDS on all the Windows end-nodes or hooked to a normal port (to receive broadcasts) will alert the security admin that such things are taking place (but, will generate false positives when DHCP reassigns addresses. Sigh.)
Most Ethernet adapters allow the MAC address to be manually configured. Thus a hacker can spoof MAC addresses by reassigning the address on the adapter, or by bypassing the built-in stack and hand-crafting frames. The hacker must maintain a a constant stream of outgoing frames in order to convince the auto-learning switch that they are the legitimate owner of the MAC address.
Many (most??) switches allow MAC addresses to be configured statically in order to prevent this sort of thing. While it may be a difficult management burden to do this for all end-nodes, it may prove useful for the router, restricting the hacker to wiretapping individual end-nodes instead of everyone all at once.
Some switches can be kicked out of "bridging" mode into "repeating" mode where all frames are broadcast on all ports all the time. This is done by overflowing the address tables with lots of false MAC addresses. This can be done with a simple traffic generation phase, or by sending a continual stream of random garbage through the switch.
The real answer is "yes", there are some older adapters that do not support promiscuos mode. In particular, the original IBM Token Ring adapters (TROPIC chipset) were not able to run in promiscuous mode. There are also a few Ethernet where promiscuous mode is broken, either in the hardware or in the driver. Actually, there are far more adapters who simply lack the functionality in the driver in order to turn on promiscuous mode, meaning all programs that attempt to put them into promiscuous mode will fail even though the hardware supports the mode in theory. If you really must have one, then call technical support for a sniffing product vendor (such as NAI) and ask them which cards they DON'T support. For Windows, you might check with Microsoft support to see which cards do not support NetMon (I remember there are a few, but I can't find the documentation for it).
Note that in the Intel/Microsoft "PC99" guidelines, promiscuous mode is a "required" feature.
If this is a concern, it will be cheaper in the long run simply to upgrade to switching hubs, which basically does the same thing. An Ethernet switch moves the "address match" upstream, so that the switch does it rather than the adapter.
Finally, it should be noted that most new networks are switched in some fashion. Even though hackers cannot sniff an entire Ethernet segment, they still install sniffers on machines in order to see the incoming/outgoing traffic. A non-promiscuous adapter won't help defend against this.
A stand-alone packet sniffer doesn't transmit any packets, but when installed non-standalone on a normal computer, the sniffing program will often generate traffic. For example, it might send out DNS reverse lookups in order to find names associated with IP addresses.
Non-standalone packet sniffers are indeed what you want to detect. When crackers/hackers invade machines, they often install sniffing programs. You want to be able to detect this happening.
General Overview of Detection Methods
Most "packet sniffers" run on normal machines with a normal TCP/IP stack. This means that if you send a request to these machines, they will respond. The trick is to send a request to IP address of the machine, but not to its Ethernet adapter.
To illustrate:
There are ways defending against this. Now that this
technique is widely publicized, newer hackers will enabled a virtual MAC
address filter in their code. Many machines (notably Windows) have MAC
filtering in drivers. (There is a hack for Windows: most drivers just
check the first byte, so a MAC address of
This technique will usually work on switched/bridged Ethernets. When switches see an unknown MAC address for the first time, they will "flood" the frame to all segments.
The ping method can be enhanced in a number of ways:
The ARP method is similar to the ping method, but an ARP packet is used instead. An explanation (in Spanish) is given at http://www.apostols.org/projectz/neped/ which includes a program called neped to do this detection.
The simplest ARP method transmits an ARP to a non-broadcast address. If a machine responds to such an ARP of its IP address, then it must be in promiscuous mode.
A variation of this technique takes advantage of the fact that machines "cache" ARPs. Each ARP contains the complete information of both the sender as well as the desired target information. In other words, when I send out a single ARP to the broadcast address, I include my own IP-to-Ethernet address mapping. Everyone else on the wire remembers this information for the next few minutes. Therefore, you could do something like sending out a non-broadcast ARP, then a broadcast ping. Anybody who responds to your ping without ARPing you could only have gotten the MAC address from a sniffed ARP frame. (To make double-sure, use a different source MAC address in the ping).
Many sniffing programs do automatic reverse-DNS lookups on the IP addresses they see. Therefore, a promiscuous mode can be detected by watching for the DNS traffic that it generates.
This method can detect dual-homed machines and can work remotely. You need to monitor incoming inverse-DNS lookups on the DNS server in your organization. Simply do a ping sweep throughout the company against machines that are known not to exist. Anybody doing reverse DNS lookups on those addresses are attempting to lookup the IP addresses seen in ARP packets, which only sniffing programs do.
This same technique works locally. Configure the detector in promiscuous mode itself, then send out IP datagrams to bad addresses and watch for the DNS lookups.
One interesting issue with this technique is that hacker-based sniffing programs tend to resolve IP addresses as soon as they are found, whereas commercial programs tend to delay resolution until the point where the packet sniffer user views the protocol decodes.
Another technique involves configuring the source-route information inside the IP header. This can be used to detect packet sniffers on other, nearby segments.
Details:
In loose source-routing, an option is added to the IP header. Routers will ignore the destination IP address and instead forward to the next IP address in the source-route option. This means when you send the packet, you can say "please send packet to Bob, but route it through Anne first".
In this scenario, both "Anne" and "Bob" are on the segment. Anne does not route, and therefore will drop the packet when received. Therefore, "Bob" will only respond if he has sniffed the packet from the wire.
On the off chance that Anne does indeed route (in which case Bob will respond), then the TTL field can be used to verify that Bob responded from routing through Anne, or answering directly.
Whereas the ping and ARP methods only work on the local network, the decoy method works everywhere.
Since so many protocols allow "plain text" passwords, and hackers run sifters looking for those passwords, the decoy method simply satisfies that need. It consists simply of setting up a client and a serve on either side of the network, which the client runs a script to logon to the server using Telnet, POP, IMAP, or some other plain-text protocol. The server is configured with special accounts that have no real rights, or the server is completely virtual (in which case, the accounts don't really exist).
Once a hacker sifts the usernames/passwords from the wire, he/she will then attempt to log on using this information. Standard intrusion detection systems or audit trails can be configured to log this occurance, alerting the fact that a sniffing hacker has found the traffic and attempted to use the information.
http://www.zurich.ibm.com/~dac/Prog_RAID98/Full_Papers/sniffer_detector.html/index.htm
When hackers break into your systems, they will often leave behind wiretap programs running in the background in order to sniff passwords and user accounts off the wire. These are often imbedded (as a trojan) in other programs, so the only way to find if something like this is running is to query the interfaces to see if they are running in promiscuous mode.
The most technique is to run the program "ifconfig -a". On my computer (Solaris 2.6) the output looks like:
# ifconfig -a lo0: flags=849<UP,LOOPBACK,RUNNING,MULTICAST> mtu 8232 inet 127.0.0.1 netmask ff000000 hme0: flags=863<UP,BROADCAST,NOTRAILERS,RUNNING,PROMISC,MULTICAST> mtu 1500 inet 192.0.2.99 netmask ffffff00 broadcast 192.0.2.255 ether 8:0:20:9c:a2:98Of course, the first thing a hacker will do is replace the 'ifconfig' program to hide this. There are other utilities you can download from the net that will query the hardware directly in order to discover this information, or you could run the 'ifconfig' program directly from a CD-ROM distribution.
This is a more evil method. On one hand, it can significantly degrade network performance. On the other hand, it can 'blind' packet sniffers by sending too much traffic.
This method functions by sending huge quantities of network traffic on the wire. This has no effect on non-promiscuous machines, but has a huge effect on sniffing machines, especially those parsing application layer protocols for passwords. Simply ping the machine before the load and during the load and testing the difference in response time can indicate if the machine is under load.
One problem with this technique is that packets can be delayed simply because of the load on the wire, which may case timeouts and therefore false positives. On the other hand, many sniffing programs are "user mode" whereas pings are responded to in "kernel mode", and are therefore independent of CPU load on a machine, thereby causing false negatives.
A TDR is basically RADAR for the wire. It sends a pulse down the wire, then graphs the reflections that come back. An expert can look at the graph of the response and figure out if any devices are attached to the wire that shouldn't be. They also roughly tell where, in terms of distance along the wire, the tap is located.
This can detect hardware packet sniffers that might be attached to the wire, but which are completely silent otherwise.
TDRs used to be used a lot in the old days of coax Ethernet in order to detect vampire taps, but these days with star topologies, they are used very rarely.
There also exist OTDR equipment, but this is really only for the truely paranoid.
You can manually check hub-lights to see if there are any connections you don't expect. It helps to have labeled cables to figure out where (physically) a packet sniffer might be located.
Smart hubs with SNMP management can provide automated monitroning of Ethernet (and other) hubs. Some management consoles will even let you log connections/disconnections to all your ports. If you've configured the system with the information where all the cables terminate, you can sometimes track down where a packet sniffer might be hiding.
Tools to detect packet sniffers
The most comprehensive sniffer-detection tool.
A tool from Carnegie-Mellon that checks to see if promiscuous mode is enabled on a UNIX machine.
A tool from The Apostols that detects packet sniffers running on the local segment.
A UNIX tool for checking the promiscuous-mode status of adapters.
Another UNIX utility that should be run from crontab in order to monitor when adapters are put into promiscuous mode.
Other Sniffing Detection Resources
http://www.securiteam.com/unixfocus/Detecting_sniffers_on_your_network.html
Reference:
http://www.robertgraham.com/pubs/sniffing-faq.html