STE WILLIAMS

Hack reveals passwords from locked iPhones and iPads

Researchers have devised a method for stealing passwords stored on locked iPhones and iPads that doesn’t require cracking of the device’s passcode.

The technique, disclosed on Thursday by members of the Fraunhofer Institute for Secure Information Technology, requires physical access to the targeted iPhone or iPad, so remote attacks aren’t possible. But it takes less than six minutes and carry out, and the after effects are easy to conceal, making it ideal to carry out on devices that are lost, stolen or temporarily unattended.

The hack exploits cryptography in the iOS password management system – known as keychain – that uses a secret key that is completely independent of the device’s passcode. That saves attackers who manage to access the file system the hassle of deducing a key that’s based on a passphrase set up by the user.

“After using a jailbreaking tool, to get access to a command shell, we run a small script to access and decrypt the passwords found in the keychain,” the researchers wrote in a paper (PDF). “The decryption is done with the help of functions provided by the operating system itself.”

The script also reveals always-encrypted account settings for things like user names and server addresses for all stored accounts, as well as the account clear-text secrets. The hack worked on a locked iPhone 4 running iOS 4.2.1, which was the most current firmware version at time of writing. A demo of the attack is available on YouTube – you can view it below.

“The accessibility of keychain secrets without requiring the passcode is considered a result of a trade-off between system security and usage convenience,” the researchers wrote. “The passwords for network related services should be available directly from device startup, without having to enter the passcode first.”

The technique doesn’t retrieve passwords stored in parts of the device that remain off limits until the passcode is entered.

Still, the hack can reveal a wealth of sensitive codes, including those used for virtual private networks, Wi-Fi networks, LDAP accounts, voicemail systems and Microsoft Exchange accounts. And that’s likely to spook large business customers with employees that use the devices to connect to sensitive company systems. ®

Source

Join in the Wikileaks DDoS war from your iPhone or iPad

The online “infowar” precipitated by the media circus surrounding Wikileaks and Julian Assange continues, with DDoS attacks occurring against a bewildering variety of websites assessed as having either aided or failed to aid the leak-publisher – or often merely for commenting on the brouhaha.

Meanwhile, interest has focused on the methods used to mount the DDoS attacks. It appears that in general most of the muscle is coming from botnets of the usual sort: ones made up of zombie machines infected with malware using the same methods as ordinary online criminals and spammers (and just as illegal).

However, some of the battling communities – for instance the loosely organised hacktivist collective Anonymous, aligned in support of Assange and Wikileaks – also use collaborative tools where supporters can voluntarily attach their machines to a botnet in order to assist with a DDoS attack. The preferred tools are usually some version of the Low Orbit Ion Cannon (LOIC) software. Machines running LOIC can then be controlled via IRC or some other channel (again the campaigners are aping criminals by using Twitter of late).

Downloading and installing LOIC (the code is freely available at such places as Sourceforge) is simple enough, but evidently off-putting enough that not many people are doing it. The LOIC hivemind net run by Anonymous has generally had only a few hundred machines in it, far too few to mount a serious DDoS, and most of the grunt has been delivered by larger malware-based botnets controlled by individual Anonymous members (just one reportedly containing more than 30 times as many machines as the anonops.net hivemind).

But in the last day or two, a new wrinkle has begun to gain prominence. It is now possible to visit a webpage which will convert your browser into a pocket LOIC instance, delivering DDoS packets from whatever device you are using to browse – not necessarily even a computer.

As Panda Labs analyst Sean-Paul Correll notes:

Only a browser is needed, so you can even launch the attack from your fone, I just tested it with my iPhone … Of course I tested that it was real and worked, but I didn’t send any attack out.

Such a webpage will typically give you the option of adjusting how many requests per second to send to the target website (handy in the case of a phone or perhaps a fondle-slablet device with a limited data package and/or bandwidth) and allow you to attach an insulting message of your own devising.

This would appear to be rather less sophisticated than a proper IRC or Twitter-controlled LOIC install, but has the merit of being simpler. Whether this tremendously simple way of joining in botnets will finally mobilise large numbers of pro- or anti-Wikileaks vigilantes remains to be seen. For now, it appears that the effective DDoS attacks – and other more sophisticated meddling going on – are emanating from relatively small numbers of people.

It would seem that in general most people are aware how relatively unimportant and easily replaceable a part Julian Assange and Wikileaks have played in the release of the classified US files, which continue to mildly interest the outside world. ®

Bootnote
1) Reader be warned: Participating willingly in a DDoS attack is a crime in many countries. Even if this doesn’t bother you, you download software and visit webpages of this sort at your own significant risk: campaigners on both sides have shown little in the way of scruples, and ordinary criminal scammers are now exploiting the situation too.