Andrea Danti - Fotolia

Evaluate Weigh the pros and cons of technologies, products and projects you are considering.

Is the BREACH attack update a threat to Gmail security?

The BREACH attack has been updated to perform faster data theft. Expert Nick Lewis explains the differences in this attack and the threat level for organizations that use Gmail.

The BREACH attack from several years ago has been updated by security researchers. How is this new BREACH attack different, and can it be used to hack Gmail accounts?

As with other areas of applied security research, researchers build on prior exploit research to improve defensive and offensive techniques. Some research is more theoretical or requires researchers to study it in depth, in order to improve the techniques implemented in practical applications. This is particularly true in cryptographic research, where attacks initially categorized as purely theoretical and low risk can become practical attacks, with changes in performance or other breakthroughs.

The updated BREACH attack by Dimitrios Karakostas, a software engineer at Nokia and cryptography researcher, and Dionysis Zindros, a cryptography PhD candidate at the University of Athens, is named Rupture, and it continues to advance the cryptographic attacks on Transport Layer Security (TLS). TLS has incorporated significant cryptographic improvements over its predecessor, SSL, and has undergone significant cryptanalysis since it was introduced. Rupture is a practical implementation of the BREACH attack and it includes an attack framework for attacking other protocols using TLS. Rupture uses HTTP injection to perform a man-in-the-middle attack to analyze the HTTP traffic supported by a client web app running on the target's web browser, which communicates to a server where the data is analyzed.

For organizations that use Gmail or Facebook, this is still low risk. It could be used in targeted attacks, but there are many other faster attacks that would be used to hack Gmail accounts prior to this attack.

The researchers recommend using first-party cookies, but currently first-party cookies have only been proposed in a request for comments. First-party cookies have not been implemented in any web browsers or included in web services. The authors have several other recommendations, with the most practical defense being rate limiting connections. Rate limiting connections could have several other benefits for mitigating other types of attacks. Given the significant number of connections needed as part of a BREACH attack, rules could be set up in the intrusion-detection system or a host-based detection system that is alerted when there are a significant number of connections coming from an individual system.

Next Steps

Learn how to handle press after your enterprise has been hacked

Look into microsegmentation for secure communications

Find out how intrusion detection systems can protect vulnerable assets

This was last published in August 2016

Dig Deeper on Email and Messaging Threats-Information Security Threats