Social networking site LinkedIn has revealed details of attempts to increase security in the wake of 6.5 million usernames and hashed passwords being published on a Russian hacker forum.
A blog entry by director Vicente Silveira revealed that LinkedIn has not received any reports of unauthorised account access in the wake of the security breach.
The company also revealed that passwords of the service's users are now salted as well as hashed. "That transition was completed prior to news of the password theft breaking on Wednesday," Silveira wrote.
Ty Miller, chief technology officer of penetration testing firm Pure hacking, said that although the salting of password hashes has been around for a long time, "we find that many Web applications either do not hash their passwords at all, or use common hashing algorithms, such as MD5, without a salt".
"Social and professional networking sites such as LinkedIn are major targets for hackers," Miller said. "Combining this with the complexity of these types of web applications, the chance of a critical vulnerability being present is likely. This means that a defence-in-depth approach should be a necessity for LinkedIn, which includes protecting passwords with strong cryptographic methods."
Miller said that social networking services such as LinkedIn store a wealth of personal information about their users and have a responsibility to implement a very high standard of security, with security measures, such as salting password hashes, implemented as part of application design.
Salting a password makes it less likely an account will vulnerable to hackers using rainbow tables, which are essentially dictionaries of hashes that allow someone to discover what a user's unencrypted password is.
"Salts are designed to ensure that the generated hash is different even if the same password is being hashed," Miller said. "The larger the salt, the more different hashes exist for the same password. This generally means that Rainbow Tables are not a feasible option for cracking salted hashes because there are too many combinations to create.
"This means that attackers have to rely on dictionary-based password attacks, which has to calculate every possible salted hash for each password in the password dictionary. This means that weak passwords will be able to be cracked easily, and stronger passwords are more likely to remain secured."
Sign up for Computerworld eNewsletters.