This article is more than 1 year old

Rainbow warriors crack password hashes

Is it safe? Nope

A trio of entrepreneurial hackers hope to do for the business of password cracking what Google did for search and, in the process, may remove the last vestiges of security from many password systems.

Over the past two years, three security enthusiasts from the United States and Europe set a host of computers to the task of creating eleven enormous tables of data that can be used to look up common passwords. The tables - totaling 500GB - form the core data of a technique known as rainbow cracking, which uses vast dictionaries of data to let anyone reverse the process of creating hashes - the statistically unique codes that, among other duties, are used to obfuscate a user's password.

Last week, the trio went public with their service. Called RainbowCrack Online, the site allows anyone to pay a subscription fee and submit password hashes for cracking.

"Usually people think that a complex, but short, password is very secure, something like $FT%_3^," said Travis, one of the founders of RainbowCrack Online, who asked that his last name not be used. "However, you will find that our tables handle that password quite easily."

While security professionals have questions whether a business can be created by offering access to rainbow tables, the endeavor does highlight the weaknesses in security of password-only authentication. History has shown that password systems are imminently breakable.

In August, a group of Chinese researchers found further breaks in a common hash function, the Secure Hash Algorithm or SHA-1, used by the U.S. government. In September, researchers from the University of California at Berkeley published a paper that demonstrated that the sound of a person typing can reveal the content, including passwords. Those technical breaks do not even account for the human factor: People tend to pick simple passwords and disclose them frequently. In fact, many viruses and worms have successfully spread by trying to log into administrator accounts using a small list of common passwords.

Because of the problems, the U.S. government is requiring that banks move towards two-factor authentication, where the typical password security is augmented by a biometric or a physical security device. Some security researchers maintain that even adding a second type of security check is not enough.

The latest attack focuses on the hash functions used to verify passwords. Because operating systems cannot keep a copy of the password on the disk without weakening system security, the software instead saves a statistically unique code generated from the pasword. While the code, or hash, is computationally easy to create, reversing the process to recover the password is nearly impossible, given a correctly implemented hash function.

Rainbow tables side step the difficulty in cracking a single password by instead creating a large data set of hashes from nearly every possible password. To break a password, the attacker merely looks up the hash to find the password that produces that code.

"Creating the tables takes much more time than cracking a single hash, but then you can use the tables over and over again," said Philippe Oechslin, CEO of Swiss information-technology firm Objectif Sécurité and the inventor of rainbow tables. "The advantage of rainbow tables is that once you have the tables it is faster than a brute force (attack) and it needs less memory than a full dictionary (attack) of the function."

More about

TIP US OFF

Send us news


Other stories you might like