British intelligence agency GCHQ used port scanning as part of the "Hacienda" programme to find vulnerable systems it and other agencies could compromise across at least 27 countries, German news site Heise Online has revealed.
The use of so-called port scanning has long been a trusty tool used by hackers to find systems they can potentially access. In top-secret documents published by Heise on August 15, it is revealed that in 2009, GCHQ started using the technology against entire nations.
One of the documents states that full scans of network ports of 27 countries and partial scans of another five countries had been carried out. Targets included ports using protocols such as SSH (Secure Shell) and SNMP (Simple Network Management Protocol), which are used for remote access and network administration.
The results were then shared with other spy agencies in the US, Canada, the UK, Australia and New Zealand. "Mailorder" is described in the documents as a secure way for them to exchange collected data.
Gathering the information is only the first step, according to Heise Online.
The documents also reveal "Landmark," a programme started by the Canadian spy agency CSEC to find what it calls ORBs (Operational Relay Boxes), which are used to hide the location of the attacker when it launches exploits against targets or steals data, Heise said. For example, during an exercise in February 2010, eight groups of three "network exploitation analysts" were able to find 3,000 potential ORBs, which could then potentially be used by CSEC.
"It isn't surprising the intelligence organisations were technically able to do this. That they attack people they have no reason to attack and then install malware on their systems to attack even more systems is really shocking and sickening to see. On that I think we can all agree," said Christian Grothoff, one of the co-authors of the Heise article, in an interview with IDG News Service.
At the Technische Universität München, he has led the development of TCP Stealth, which can help prevent Hacienda and similar tools from identifying systems. The development of TCP Stealth was started during a course on peer-to-peer systems and security that Grothoff taught last year.
TCP Stealth works by adding a passphrase on the user's device and on the system that needs to be protected.
"For example, if you have remote administration of routers or servers you don't want that access to be public. You typically have a small group of administrators that are authorized, so between them you share a passphrase and also add it where they want to connect," Grothoff said.
If the passphrase is incorrect when the connection is started, the system simply doesn't answer, and the service appears to be dead.
For this to work, operating systems and applications have to be upgraded to be able to use TCP Stealth. Linux has already been upgraded and there is a library application developers can use to add TCP Stealth to their software without having to recompile. Windows, Chrome OS and Mac OS haven't been ported to TCP Stealth.
The hope is now that the technology will be standardised by the IETF (Internet Engineering Task Force). A first draft has already been filed with the organisation. It was co-authored by Jacob Appelbaum with the Tor project and edited by Holger Kenn from Microsoft in Germany.
"I think there is a chance we can convince people this is necessary," Grothoff said.