Author Archives: wunderbarb

Twitter’s selection of bad password

Posted on by
Reply

When signing up to Twitter and selecting your password, Twitter bans some passwords although they have the minimal size. These passwords are hard coded in the code, and thus can be consulted (which is not an issue at all).

If you don’t want to reverse engineer code, the full list of 370 bad passwords is published by the Wundercounter.

It is interesting to see them. They are totally logic. Interestingly, this password ban is only for English native speakers. Foreigners can still use stupid passwords. For instance, as expected QWERTY, but also ASDFGH are banned. In france, among the worst passwords, we have AZERTY and QSDFGH due to our specific keyboards.

It is really fun to see the selection. My preferred one is THX1178. How many geeks would use this one? It is an ultimate recognition to be listed with STARWARS and STARTREK.

In any case, it is a good initiative. But internationalisation would be better. Do you know any repository of bad passwords for many languages?

Ubisoft’s DRM torpedoed!

Posted on by
Reply

Ubisoft recently launched its new game “Silent Hunter 5”, a simulation of submarine. The game was protected with a new generation of DRM that required constant online connection to servers. Of course, if you are online, it is easier to fight piracy. As usually, in the game arena, this new DRM generated a huge fury (remember Spore).

Unfortunately, 24 hours after the launch, a cracked version appeared on the P2P networks (see TorrentFreak). The cracked version does not require online connection!

Quickly, Ubisoft denied that the game was cracked.

“You have probably seen rumors on the web that Assassin’s Creed II and Silent Hunter 5 have been cracked. Please know that this rumor is false and while a pirated version may seem to be complete at start up, any gamer who downloads and plays a cracked version will find that their version is not complete,”

Unfortunately, when scouting the forums, I never find any person complaining that the game was not working. Nevertheless, the crack requires to avoid any connection to Ubisoft servers.

Ubisoft was expecting to deploy the same DRM for Assassin Creed II. Some delay may be foreseen.

As a citizen, I see the need of DRM. As a security expert, I “build” DRM. Piracy is bad. As a gamer, I hate DRM that requires a permanent connection for a game that does not need interaction with other entities. I often play games in train or plane. Such a DRM requirement would be deterrent for me. One of the most important requirements for DRM is that DRM should be as transparent as possible for honest user.

Software protection is one of the most complex tasks.

ReFormat: Automatic Reverse Engineering of Encrypted Messages

Posted on by
Reply

Five researchers, Z. WANG, X. JANG, W. CUI, W. WANG and M. GRACE presented, according to me, a nice piece of work at Esorics 2009.

The objective was to automatically reverse engineer encrypted messages without breaking the algorithms. The basic idea is simple. When a piece of software receives an encrypted message, it performs two steps (regardless of the used cryptographic algorithms and protocols). First, it decrypts the message and then it processes the clear message. This means that the message is during a while in the clear in the memory. if you identify the location of this buffer, and when it is used, then game over.

To succeeed, they used two tricks. The first was to distinguish between decryption routines and normal processing routines. Cryptographic functions use far more bit wise and arithmetic operations than normal software. They measured (on OpenSSL) that more than 80% of the operations were bit wise and arithmetic for cryptographic functions. The rate dropped beneath 25% for normal processing. This heuristic allows to detect the encryption/decryption phases.

The second step is to locate the buffer containing the clear text. They identify all the buffers that are written while in decryption phase. Then, they identify all the buffers that are read during the processing phase. The expected buffer should be in the intersection between the two sets.

Obviously, there are many ways to deter this attack. For instance code obfuscation may change the rate. Dynamic code encryption is of course a must. Nevertheless, I found the approach extremely clever.

Once more, it proves that writing secure implementations is extremely difficult. And it requires clearly a twisted mindset. :Happy:

If you are interested in tamper resistance, you have to read this paper. It is available here.

MyDRMSpace

Posted on by
Reply

Do you want to add a digital watermark to your personal pictures or audio records? The MyDRMspace site offers such a free service free (although in the terms of use, there is a possibility to charge).

Polish developers and entrepreneurs opened this site in 2009. Once registered, you can upload a photo or a piece of audio to the server. Then, you enter information related to the file and mydrmspace will add a watermark to it (signature in their vocabulary). Of course, you can upload a file to check if any watermark is present. If the watermark is present, then mydrmspace will point to the corresponding information.

The developers had the bright idea to offer the infrastructure as a platform for watermark designers. Any watermark designer can upload his/her embedders and detectors (so called coder and decoder on the site :Sad: ) Then the public may use this technology. In other words, when uploading to mark, you can choose the embedder (coder). I have no clue on which criteria, normal users would use one technology rather than another one. The most probable would be that normal user would choose either the first one presented or the most popular one. When checking a file, you have to select the right detector (decoder).

Unfortunately, they seem not to have tremendeous success. There is only one audio watermark technology available and two image watermark technologies (from the same author). The place holder for video watermark is empty.

I have not tested if the watermarks are robust.

Do you know any similar site that offers free watermarking service?

Thank you to TF for the pointer :Happy:

Augmented Identity

Posted on by
Reply

Technologies were around. Social networks become prevalent. People are not aware of risks (or at least they are less paranoid than me 🙂 ). Somebody had to do it.

The Astonishing Tribe (TAT) did shake all the elements together and presented recognizr. Using a mobile phone to capture a picture of the person, recognizr extracts 3D parameters of the face and sends it to a server. The server uses a face recognition software and looks in social networks to identify the person and provide data.

TAT was privacy aware. Thus, the system works on strict opt in. In other words, you must register to the system to be recognized.

TAT is privacy aware, but what prevents another company/agency/group to design a similar software and scout the social networks without the consent of the people? Law enforcement? Well, it uses only public data. The data that you/we posted on the Internet.

This application was inevitable. Everything was available. Nevertheless, it scares me. Once more, be very cautious on what personal data you post on the Internet.

Digital Personal Property

Posted on by
Reply

Paul Sweazey believes he has found the solution that mitigates the problem of DRM. He wanted to emulate a property of physical goods: rivalry (If you want more information about rivalry, please have a look on Bomsel’s works). In a nutshell, rivalry is the fact that when consuming a good you reduce the access for others. For instance, when you play your DVD, someone else cannot play it on another player. This is not true for electronic files. By definition, electronic goods are non-rival. One of the purposes of DRM is to add a pinch of rivalry.

To do so, Sweazey created the concept of Digital Personal property. How does it work? Content has two elements: an encrypted folder containing the essence and a playkey that you preciously keep in a vault. Sounds familiar, isn’t it? In DRM vocabulary, his playkey is called a license. You may freely distribute the encrypted folder but will give your playkey only to trusted people who would not steal your license. The license must be UNIQUE in the sense that there is one unique instance at any time. Thus, if the person you gave your playkey does not return it, you lost its ownsership.

The technical trick will be to be able to create a rival license that should not be linked to a device (else you end up with the typical problem of interoperability).

He just moved the problem of DRM towards the license. He will still have to find a method to generate a license (playkey) that can exist only as one unique instance in the world and that could be played everywhere. This is the Holy Grail of DRM that we have been all looking for years. TCreating rivalry is difficult without introducing physical constraints.

It reminds me one of the concepts we built in an old system called SmartRight. The objective was to control the size of an authorized domain for a familly but without any central online authority. We used an electronic token that was passed to the newly joining device. Of course, you could add a device from your neighbour, but then your neighbour “owned” the electronic token. Would the neighbour leave or not collaborate anymore, you could not anymore add devices to your domain. It was based on the use of secure processors and on the fear of loosing the token.

Will DPP work? If Sweazy finds a robust and user friendly way to create this unicity of instance, it would work. This would also offer a lot more applications. But is it feasible? Bruce Schneier would probably say no. (Wait our next security newsletter with his interview.) And many brains are researching this topic.

For more information, read Goodbye, DRM; hello “stealable” Digital Personal Property at Ars technica.

Attacking the BitLocker Boot Process

Posted on by
Reply

TPM and BitLocker are interesting targets for security experts. Tarnovsky has recently reverse engineered a Trusted Platform Module (TPM) chip from Infineon. Five researchers from German Fraunhofer Institute have explored attacks on BitLocker when using TPM to seal the data.

The paper is interesting even if you are not familiar with TPM. The team targets the boot loader and especially the recovery strategy. If you illegaly modify the environment of the machine, the TPM will detect it but the sealing data for BitLocker will not be accurate anymore. Thus, Bitlocker uses a recovery mechanism independent from the TPM. The idea is to trick the user in this mode. They suggest five attacks: create a false plausible recovery situation, spoof the recovery message, Spoof then hide, replace the computer by a “‘phishing” computer, and preemptive modification (i.e. modify the computer before activating BitLocker. The two last attacks are less plausible. All attacks require physical access to the target.

Lesson: The attacks target the operating mode and process and not the technology itself. Therefore, they are clever.
Recovery systems are always BACKDOORS in a system!!

The paper is available at here.