Timing side channel attacks are non-intrusive attacks that are still widely ignored in day-to-day penetration testing, although they allow attackers to breach the confidentiality of sensitive information. The reason for this is, that timing attacks are still widely considered to be theoretical. In this talk, I present a toolkit for performing practical timing side channel attacks and showcase several timing attacks against real-world systems. Timing side channels are vulnerabilities in software applications that leak sensitive information about secret values such as cryptographic keys. They differ from common intrusive vulnerabilities such as Buffer Overflows or SQL-Injection because the attacker sends normally looking requests to the server and infers secret information just from the time it took to process the request. In academia, timing side channel attacks are well researched, especially against cryptographic hardware, but in day-to-day penetration testing, they are still widely ignored. One reason for this is that the timing differences are often small compared to the jitter introduced in networked environments. This makes practical timing side channel attacks challenging, because the actual timing differences blend with the jitter. In this talk, I will present methods and tools to accurately measure response times despite the jitter in networked environments. I will introduce a programming library that enables penetration testers to measure accurate response times of requests send over networks. Furthermore, I will describe algorithms and statistical filters to reduce the jitter from measurements. For this, I will introduce a reporting tool that takes a dataset with network measurements as input, automatically applies the algorithms and filters, and produces a report with the results. This report enables even novice penetration testers to analyze a response time dataset for timing side channel vulnerabilities. In the end, I will show that timing side channels are practical by showing several attacks. First, I show how to determine if a given user name is an administrative user in a productive installation of the popular CMS Typo3. Second, I show how to determine how many pictures are hidden in a private album of an online gallery. Third, I show how to perform an adaptive chosen cipher text attack against implementations of the XML Encryption standard. This attack allows to decrypt any Web Service message whose body was encrypted using XML Encryption only by measuring the response time of the Web Service.
Secdocs is a project aimed to index high-quality IT security and hacking documents. These are fetched from multiple data sources: events, conferences and generally from interwebs.
Serving 8166 documents and 531.0 GB of hacking knowledge, indexed from 2419 authors from 163 security conferences.