Search
Contactez-nous Suivez-nous sur Twitter En francais English Language
 

De la Théorie à la pratique











Freely subscribe to our NEWSLETTER

Newsletter FR

Newsletter EN

Vulnérabilités

Unsubscribe

Lieberman Software says e-voting flaw could have been prevented

March 2012 by Lieberman Software

As Super Tuesday got under way in the US elections last week, reports started coming in about an electronic voting platform whose security was massively compromised after the election board invited external researchers to test its systems.

Lieberman Software says that it took coding experts with the University of Michigan just a few hours to compromise the system’s security and ‘elect’ a cartoon character from ’Futurama’ to the Washington DC election board.

Philip Lieberman, president and Chief Executive of the security software specialist, says that the multi-stage crack could almost certainly have been prevented had the developers of the electronic voting system built privileged identity management into the system.

“The methodology used by the researchers was precise. First they identified that the platform was susceptible to a shell injection attack, and then they started writing output to the images directory. After that they encrypted their IP traffic to stop the Intrusion Detection System (IDS) from triggering,” he said.

“The final piece in the cracking jigsaw was when they guessed the login details of the server - so granting them high level access to the e-voting platform. The right privileged identity management controls would have eliminated the obviously weak admin credentials, making it all but impossible for outsiders to access sensitive accounts,” he added.

The Lieberman Software president went on to say that the encryption of the cracking IP traffic was an interesting twist, as it prevented the IDS from spotting the rogue traffic.

But, he says, this loophole could also have been closed through the use of deep level packet inspection and/or heuristic packet analysis, which are security processes increasingly being adopted by large corporations to defend their critical IT systems.

Lieberman goes on to explain that the real safety net on any Internet-facing system – and especially with public platforms such as electronic voting servers – is the use of privileged identity management, since this protects organisations against unauthorised users and malicious programs from gaining unrestricted, anonymous access to sensitive data on the network.

The problem that most large organisations face, he says, is that nearly all have powerful, unsecured privileged accounts present in their cross-platform enterprise.

Reducing the level of access to their various functions on all but the most critical of admin accounts – and then limiting these high-privilege admin accounts to specific terminals – dramatically lowers the security risk profile of the organization concerned.

“This isn’t rocket science,” said Lieberman. “It’s the application of multi-level security controls in a cost-effective manner.”

“I think the most interesting takeaway from this high profile incident is that the voting board opened up its platform to a public security test and the security was found wanting. There are almost certainly a great many other public-facing government systems that have similar weaknesses – or worse,” he said.

“At the end of the day, the Washington DC election board should be applauded for its openness which – though unconventional – helped them highlight the fact that necessary controls were missing from their environment,” he added.




See previous articles

    

See next articles