Information Security News mailing list archives

FTC probes PetCo.com security hole


From: InfoSec News <isn () c4i org>
Date: Mon, 8 Dec 2003 04:28:46 -0600 (CST)

http://www.theregister.co.uk/content/55/34379.html

By Kevin Poulsen
SecurityFocus
Posted: 07/12/2003

Pet supply retailer PetCo disclosed this week that its security and
privacy practices are the target of an investigation by the U.S.  
Federal Trade Commission (FTC), which is following up on an e-commerce
security gaffe that left as many as 500,000 credit card numbers
accessible from the Web earlier this year.

In October the FTC served PetCo with a "Civil Investigative Demand"  
seeking information and documents on how the company protects private
customer information on the PetCo.com e-commerce site, PetCo revealed
in its quarterly report Wednesday. "At the present time, the Company
is unable to determine whether the FTC will initiate any enforcement
action against the Company or the financial impact any such action
might entail," the company wrote.

The probe stems from an incident first reported by SecurityFocus last
June, when then-20 year-old independent programmer Jeremiah Jacks
discovered that PetCo.com suffered from an SQL injection vulnerability
that left its database open to anyone able to construct a
specially-crafted URL.

SecurityFocus notified PetCo of Jacks' discovery, and the company
immediately blocked access to the vulnerable Web page. The company
worked over a weekend to close the hole permanently, and said it had
hired a computer security consultant to assist in an audit of the
site. Jacks also cooperated with PetCo, which said it found no
evidence that anyone prior to Jacks exploited the hole.

The PetCo probe is the second FTC investigation to be sparked by the
young coder. In February, 2002 Jacks discovered a similar SQL
injection hole at the website of fashion-retail Guess that exposed, at
Jacks' count, over 200,000 credit card numbers with corresponding
names and expiration dates.

Consumer Privacy Issues

Jacks, who lives and works in Orange County, California, cooperated
with the FTC as it investigated Guess under its authority to probe
"deceptive trade practices" -- the Guess.com privacy policy had
claimed that credit card numbers were stored in an "unreadable,
encrypted format at all times." The case settled last June, with Guess
agreeing to overhaul its information security practices and promising
not to misrepresent the extent to which it protects the security of
customers' personal information.

The Guess case was only the third time the FTC used its anti-consumer
fraud mandate to crack down on e-commerce cybersecurity gaffes -- last
year it won a consent decree against Eli Lilly for the inadvertent
disclosure of the e-mail addresses of 669 Prozac users, and another
one against Microsoft for inflated security claims about the company's
Passport identity management service.

News media interest in the Guess case prompted Jacks to check a few
other large e-commerce sites for similar bugs, including PetCo.com, he
said at the time. He used Google to find active server pages on
PetCo.com that accepted customer input, then simply tried inputting
SQL database queries into them. "It took me less than a minute to find
a page that was vulnerable," said Jacks. "Any SQL injection hacker
would be able to do the same thing."

Jacks said the database contained 500,000 credit card entries, and
that he could have accessed corresponding customer names and address,
as well as entire orders. A PetCo spokesperson confirmed the hole at
the time, but would not say how many credit card numbers had been at
risk.

In disclosing the FTC probe, the company's quarterly report doesn't
admit to any error, only acknowledging that "a self-proclaimed hacker
purportedly obtained unauthorized access to a portion of the Company's
website."

PetCo's privacy policy assured visitors, "At PETCO.com our customers'
data is strictly protected against any unauthorized access."



-
ISN is currently hosted by Attrition.org

To unsubscribe email majordomo () attrition org with 'unsubscribe isn'
in the BODY of the mail.


Current thread: