Networks are Insecure?
Let's begin with the obvious: wireless technology is here to stay. What
is not so obvious is that for the foreseeable future it will be risky to
deploy. In this column we'll see why.
NEED TO KNOW ABOUT WiFi IN 400 WORDS OR LESS
Well, here's a
tad of trivia that will move any high-tech wallflower to the center of
attention at any elite gathering. Question: When was the first Wireless
Network deployed? 1995 (WRONG). 1987 (NOPE). 1982? (huh-uh). The answer
is 1970. It was called ALOHANET (
), and it anticipated many of the core network
protocols in use today, including Ethernet and Wireless Fidelity (aka
If that piece of
intoxicating trivia won't produce a collective drool of golden sterlet
caviar, nothing will. If you can recite the trivia I provide in these
columns without making yourself look like you once had a bit part in a
Cheech and Chong movie, you'll make every A-list from Wall Street to
Hollywood. Trust me on this. Jeannie Caruso and I are even talking about
a converting this concept into a mini-series for the Discovery Channel:
“Survival IT: last nerd standing.” But I digress.
Basically, the flavor
of digital wireless technologies that we're likely to be most concerned
about in our enterprise are Personal Area Networks (PANs) and Wireless
Local Area Networks (WLANS). Both are widely used in business. Both are
saddled with insecurities.
STONE AND WEP ENCRYPTION
Champollion showed that
comparing fragments of an understood plaintext (Ancient Greek) with an
encoded text (demotic or hieroglyphic) could be used to reveal the
correspondences . Hmmm, I wonder if there isn't an analogy somewhere out
there in cyberspace.
How's this? The WiFi
standard for encryption is called Wired Equivalent Privacy (WEP). Since
it's built into the WiFi standard, it comes free with WiFi appliances.
(Even though it's free, it's overpriced.)
There are two common
varieties of WEP based on key length: 40-bit (standard) and 104-bit
(extended). However they're called 64-bit and 128-bit encryption because
the vendors want us to think that WEP is more secure than it is. The
additional 24 bits in both case comes from a 3-byte sequence that is
prepended to the key. This sequence is called an initialization vector,
or IV for short. I tell my clients that IV stands for “invasive vermin.”
The core algorithm of
WEP is RC4, but the implementation of WEP is fundamentally flawed: it's
both poorly designed and feebly implemented, but other than that it's a
nice piece of work☺. WEP is to wireless security what the Tacoma Narrows
Bridge is to landmark engineering failures. Both are case studies in how
not to do things.
The essence of the
weakness is the feeble way that the WEP designers approached the age-old
key distribution problem. Encryption requires the use of keys to obscure
the message. But somehow the keys, or means to re-generate the keys,
must be shared by sender and receiver. The WEP designers decided to
handle key management by sending the initialization vector and the key
ID in plaintext in the management frame of the message sequence (Figure
Figure 1: The Management Frame of an Encrypted WiFi Message. Note that
the WEP initialization vector, the WEP key ID and the 40-bit encryption
format are all broadcast in plaintext for any hacker sniffing the
wireless traffic to intercept.
Not content to leave
good enough alone, the WEP designers implemented a version of RC4 that
is hobbled. Whenever the middle byte of the initialization vector is all
ones (0xff), the byte of ciphertext pointed to by the first byte of the
initialization vector is exactly the same as in the message text - it's
just a matter of comparing the two (pieces of text (ala Champollion and
the Rosetta Stone). This is called a weak IV. In 'geek speak' we say
that a weak IV has a format of B+3::ff::X (where B is the byte of the
key to be found, ff is the constant 255, and X is irrelevant).
IS WiFi READY
FOR PRIME TIME?
Take a look at Figure
2. This is a screen shot of casual sniffing of WiFi traffic from one of
my offices on Las Vegas Blvd. The column “BSSID” is the internal ID of
the wireless appliance. The columns of greatest relevance are “WEP” and
“Interesting.” When the entry for WEP is not “Y,” it means that the
wireless network isn't using WEP encryption, so everything is in
plaintext - email, Web pages, database commands, --- everything!
Interesting packets are those with “weak” IVs discussed above. The PW
columns would be passwords that are disclosed during the normal
interflow of packets. Bear in mind that this is only 45 seconds of
traffic that wafted into my office. What is more important, every
encrypted message from these wireless sites is vulnerable to attack -
and most aren't even encrypted!
Figure 2: Sniffing WiFi on Las Vegas Blvd with AirSnort. The column
marked “interesting” betrays the weak keys. A rule-of-thumb is that
wireless hacking tools require only a few megabytes of “interesting”
packets in order to break the key.
AND DIGITAL RISK MANAGEMENT?
Wireless is here to
stay. At the moment, however, it's radically over-deployed. This is in
equal parts a result of convenience and a desire to be perceived as
“current.” Unfortunately, recent legislation like Sarbanes-Oxley, HIPAA
and Gramm-Leach-Bliley presents a real-and-present-danger to executives
who underestimate the potential threat of insecure wireless. Given the
inherent vulnerabilities in WiFi, a good starting strategy is to assume
that all wireless traffic is printed and left in public areas for all to
see. If your organization doesn't mind if that traffic is read, there's
no problem. On the other hand...
This actually leads me
to a topic that comes up a lot in my consulting work. Wireless security
has less to do with technology than it does with risk management. CIOs
and IT CSOs typically understand this point. It's lost on most CFOs and
CEOs that I've worked with.
The fault actually lies
on the IT side of the ledger. IT executives and managers have
all-too-often tried to justify WiFi security (for that matter, all
computer and network security) to CFOs and CEOs by means of a technology
mandate. That's the wrong way to look at it. WiFi security is best
justified as a risk management mandate. Conceptually, digital security
of digital assets is no different than physical security of physical
Now that we've closed
most of our modem banks, WiFi has become the greatest security breach in
our organization. Eventually, WiFi will become as secure as LAN-based
communications. But until that time, it's incumbent on the CIO to focus
on the risk, not the glitz, if we're to safely control the growth of
WiFi so that the organization and it's customers remain both well-served
So anyone may be
listening to your WiFi traffic. Remember, it's not paranoia if it's
by Hal Berghel