OWASP Belgium Chapter September 2012 Wrap-Up

Steven Van Acker on stage

The holidays are gone, kids are back to school. For the security landscape, it means that security meetings are also back! The first OWASP Belgium Chapter was organised tonight. Here is my quick wrap-up.

This time the meeting started in the afternoon with a technical workshop organised by SPION. Due to agenda conflicts, I did not attend this one. I joined the meeting for the second part organised in a classic format: after a brief introduction with news about the Chapter and the OWASP foundation in general, two speakers came to present their researches.

The first one was Steven Van Acker who talked about remote JavaScript inclusions. They are plenty of publicly available JavaScript libraries on the Internet. It’s very easy for developers to do some shopping and use them without reinventing the wheel. Steven presented the results of a research about  the usage of those libraries in websites. Is it really safe to use them “as is“? Always keep in mind that browsers don’t care about what they execute. A crawler was developed to download websites content from the Internet (approximatively 3.3M URLS where visited) and included JavaScript content was extracted. Steven gave some statistics. The one which hit me was about the top-10 of JavaScript code used: 50% of this top-10 is related to Google services! (mainly Google Analytics) Once we saw the amount of JavaScript code included in websites, some questions arise:

  • Should websites trust remote providers?
  • Can we safely execute their code?
  • What’s the quality of their maintenance?

Then, again based on the finding, some weirdness:

  • Cross-user scripting (ex: http://localhost/script.js)
  • Cross-network scripting (ex: http://192.168.2.1/script.js)
  • Stale IP-based remote inclusions
  • State domain-based remote inclusions
  • Typo-squatting XSS
This last example was really weird. They found some sites trying to load JavaScript content from googlesyndicatio.com (with a missing “n”). What they did? They registered the domain and got hits! That’s an easy way to compromise websites. And what about the countermeasures? Steven gave two:
  • Executing the remote scripts in a sandbox (not always easy).
  • Download the script locally.
If the second one looks interesting, it could make difficult to implement. It’s do-able only if the files do not change often. A very nice presentation with clear explanations.
After a short break and pizzas, the second speaker for tonight was Dave van Stein. He talked about “modern information gathering” or how to grab interesting data from your targets event without sending any packet to them. The talk was a brief presentation of techniques and tools used by pentesters or auditors to collect information. Here is a short list of tools covered by Dave:
  • shodanhq.com
  • serversniff.net
  • robtex.com (with a good domain visualisation feature)
  • Google advanced searches (intent:, inurl:, filetype:, etc)
  • Google Hacking DB
  • Search engine optimisation tools (can crawl target websites for you)
  • FOCA
  • Maltego

Most of them are classic ones. But that was a good reminder or a good way to populate your bookmarks! That was a good meeting to start the new season!

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.