STE WILLIAMS

Worldwide, the rollout of DNSSEC can comfortably be described as “glacial”, but Google valiantly continues to try to give it profile. Having launched its own DNSSEC service three years ago, Mountain View has now added DNSSEC validation to its public DNS resolvers.

Announced in this blog post, Google says the move means “we can better protect people from DNS-based attacks and make DNS more secure overall by identifying and rejecting invalid responses from DNSSEC-protected domains.”

Until now, Google notes, it accepted and forwarded DNSSEC-formatted messages without validating them. This meant that while its own DNSSEC responses were trustworthy, those it forwarded may not have been.

As noted by DSL Reports in January of this year, DNSSEC implementations remain rare.

Google agrees, stating that currently only seven per cent of the 130 billion DNS requests it resolves on an average day are DNSSEC-enabled, and only one per cent of server responses are signed. That means most of the Internet remains vulnerable to the so-called “Kaminsky bug” – fake DNS records that redirect users to malicious Web sites, injected by DNS cache poisoning.

What’s it all about?

As readers will know, DNS – the Internet’s lookup system that matches a readable URL (theregister.co.uk) to an IP address (92.52.96.89) – was originally based on trust, and it assumed that trust long after bad actors learned how to exploit the DNS to direct users to malicous sites.

To get some context for the announcement, The Register spoke to APNIC chief scientist Geoff Huston, who says it’s a good move, partly because his research suggests Google is gaining importance as a DNS resolver.

In research that identified 2.5 million DNS clients, Huston said Google DNS was in use by as many as 15 percent of them, making it a big enough resolver to influence the rest of the industry.

Prior to this announcement, he explained, an address might resolve even though Google had not tested the DNS signature key against the site’s certificate. With validation in place, Google is committing itself to signature-checking all responses. If there’s a mismatch – which Huston noted could just as easily be because of an error rather than malice – Google will return a SERVFAIL (the impact of which is that the end user will get a “site not found” error).

“I’ve seen some very big sites that only worked by accidental glue”, he remarked. “Anyone doing proper DNSSEC on them will get a SERVFAIL from Google.

“You can’t do DNS at 4.59 on a Friday afternoon! You have to follow the instructions to the letter, no shortcuts, no crap – because otherwise you will have a badly signed domain and a large amount of the internet won’t see you any more. DNSSEC validation means anything wrong with a domain certificate, or the parent, all the way back to the root, makes a SERVFAIL.”

That, he said, will ultimately be a good thing because it will encourage sites to lift their game.

Huston believes any site that’s entrusted with customer data should already be using DNSSEC. “I would expect all banks, all government sites, to use signed domains … if you haven’t locked the DNS, it’s close to negligence”.

There is, however, a hole in the local industry which may well be common worldwide. It’s still very uncommon for companies selling domain name services to offer usable tools for site owners to create their own site signatures. That needs to change to become a basic part of the service, “because the beneficiary of DNSSEC signing is everybody”. ®

Article source: http://go.theregister.com/feed/www.theregister.co.uk/2013/03/20/google_adds_dnssec_validation/

A web analytics firm has sniffed out a botnet that was raking in $6m a month from online advertisers.

The so-called Chameleon botnet mimicked human visitors on select websites, causing billions of display ad impressions to be served to compromised machines. As many as 120,000 infected drones have been discovered so far. Almost all of the over 202 websites targeted in the scam are located in the US. In some cases, two-thirds of the websites’ traffic was generated from zombie machines.

All the bot browsers report themselves as being Internet Explorer 9.0 running on Windows 7.

The advertisers cough up a few pennies every time an ad is viewed, and the ad network, ad exchanges and the publisher all take their share.

The malign traffic was difficult to identify because the malware used a hundreds of thousands of different ad-exchange cookies. These characteristics earned the malware behind the scam the Chameleon moniker.

Click fraud as a revenue generation model for zombie networks is far from unprecedented. For example, the Bamital botnet taken down by Microsoft and Symantec last month also made money through advertising fraud. However the Chameleon malware is reckoned to be the most sophisticated botnet software of its type to appear to date.

Chameleon affects display ad advertisers and not just text link advertisers. Individual bots within the Chameleon botnet run on host machines with Microsoft Windows as the operating system. Bots access the web through a flash-enabled Trident-based browser that executes JavaScript.

The bots subject host machines to heavy load, and the bots appear to crash and restart regularly. These crashes and idiosyncratic site-traversal patterns are just two of the many bot features that provide for a distinctive bot signature that eventually allowed London-based web analytics firm spider.io to track fraudulent behaviour associated with the malware and draw up a blacklist of compromised IP addresses in association with its ad exchange partners.

“Spider.io has been tracking anomalous behaviour associated with Chameleon botnet since December, 2012, and in February of this year the extent of the Chameleon botnet’s principal web-browsing activity was established,” an advisory by spider.io explains.

“This was achieved as part of spider.io’s broader work with leading display ad exchanges and demand-side platforms to identify deviant consumption of display advertising media. In particular, DataXu and media6degrees have been proactive partners.”

Spider.io was born out of Imperial College London. ®

Article source: http://go.theregister.com/feed/www.theregister.co.uk/2013/03/19/chameleon_botnet/

Andrew Auernheimer, a member of the grey-hat hacking collective Goatse Security, has been sent down for three years and five months in the slammer after he helped leak users’ private email addresses via a flaw in ATT’s servers.

Auernheimer, known online as Weev, received his sentence wearing shackles after he tried to bring a mobile phone into the courtroom. After completing his term he will have to pay over $72,000 in restitution to ATT and undergo three years of supervised release.

“I didn’t come here today to ask for forgiveness,” Auernheimer told US District Judge Susan Wigenton, Bloomberg reports. “The Internet is bigger than any law can contain. Many, many governments that have attempted to restrict the freedoms of the Internet have ended up toppled.”

In 2010, Auernheimer found a flaw in a public-facing ATT server that could be used, via the iPad’s integrated circuit card identifier (ICC-ID), to uncover the names and email addresses of 114,067 early adopters of Apple’s 3G-equipped fondleslab. His colleague Daniel Spitler wrote a PHP script called “iPad 3G Account Slurper” to harvest the data, and then handed it over to online magazine Gawker.

The data caused huge embarrassment to ATT and Apple, since it included the personal emails of then-White House Chief of Staff Rahm Emanuel, New York Mayor Michael Bloomberg, film mogul Harvey Weinstein, and several high-ranking US Army officials. ATT fixed the flaw, and there’s no evidence Auernheimer did anything more than highlight the sloppy coding.

His defense lawyers argued that he was accessing information on a public web server and that if this was a crime then most internet users are guilty too. This cut little ice with the presiding judge.

“While you consider yourself to be a hero of sorts, without question the evidence that came out at trial reflected criminal conduct,” Judge Wigenton said in imposing the sentence. “You’ve shown absolutely no remorse. You’ve taken no responsibility for these criminal acts whatsoever. You’ve shown no contrition whatsoever.”

Auernheimer’s colleague Spitler now looks likely to face a similar sentence after pleading guilty, andsome in the security field are warning that the verdict will have a deadening effect of flaw exposure. Former National Security Agency (NSA) programmer and now Apple-cracker and security consultant Charlie Miller said the decision was highly troublesome.

We could all go to jail for security research at any moment, and a jury would happily convict us.

— Charlie Miller (@0xcharlie) March 18, 2013

In this hack’s opinion, Auernheimer’s sentence is far too severe. You could argue that he should have submitted the flaw to ATT, waited for the problem to be fixed, and then reaped the publicity. He could also have profited from selling the flaw on the grey or black markets, but chose not to go for the money, but to get embarrassment value instead.

“My regret is being nice enough to give ATT a chance to patch before dropping the dataset to Gawker. I won’t nearly be as nice next time,” he said in a Reddit forum.

With no evidence of harm done, sending someone down for over three years, near-bankrupting them with fines, and setting such a long probation victim looks less like justice and more like judicial spite. ®

Article source: http://go.theregister.com/feed/www.theregister.co.uk/2013/03/18/auernheimer_ipad_hack_prison/

Security researchers have warned that the controversial FinFisher spyware has been updated to evade detection and has now been discovered in 25 countries across the globe, many of them in APAC.

FinFisher, also known as FinSpy, is produced by Anglo/German firm Gamma International and marketed as a “lawful interception” suite designed for law enforcers to monitor suspected criminals.

However, it’s alleged to have been used by repressive regimes in countries like Bahrain and Egypt to target human rights activists and other enemies of the state.

The British government has already apparently told Gamma it needs a license to export the software outside the EU, while human rights groups last month requested that the OECD investigate whether the firm violated its guidelines by failing to carry out due diligence on how the software would be used in Bahrain.

Now, researchers at Toronto University’s Munk School of Global Affairs, who have been tracking the use of this surveillance-ware for over a year, say they’ve found 36 new command and control servers, 30 of which are new, in 19 countries.

This brings the total number of countries where the spyware has been found to 25, they said.

The researchers also claimed that as of October last year “the behavior of FinSpy servers began to change”, in a bid to evade detection.

There’s a heavy APAC bias to the countries where servers have been found, with Australia, Bahrain, Bangladesh, India, Malaysia, Singapore, and Vietnam all on the list.

The report has a handy map showing the breakdown of countries and a list of the newly found servers and related ISPs.

The researchers added the following caveat:

Importantly, we believe that our list of servers is incomplete due to the large diversity of ports used by FinSpy servers, as well as other efforts at concealment. Moreover, discovery of a FinSpy command and control server in a given country is not a sufficient indicator to conclude the use of FinFisher by that country’s law enforcement or intelligence agencies. In some cases, servers were found running on facilities provided by commercial hosting providers that could have been purchased by actors from any country.

In the case of Vietnam, the report details the discovery of a mobile version of FinSpy featuring GPS tracking and the ability to snoop on conversations close to the handset, as well as pilfering text messages from the device.

In Indonesia, meanwhile, the three ISPs involved are under investigation by the authorities on suspicion of spying on their customers, according to TechInAsia.

Kenny Lee, principal for Verizon APAC’s Investigative Response unit, cautioned that “just because a piece of software is found within a particular country, doesn’t necessarily mean it originates from that country”.

“As we have seen time and time again, anyone anywhere could have set up the server,” he told The Reg, adding that corporates should always be alert to the risks posed by spyware.

He recommended IT admins monitor event logs for suspicious activity, eliminate any unnecessary data from the organisation, and “ensure fundamental and common sense security countermeasures are in place and functioning correctly”.

In light of the findings, the report calls for a “policy debate about surveillance software and the commercialisation of offensive cyber capabilities”.

It should be noted that, as mentioned in the report, Gamma International has repeatedly denied any links to the spyware and servers revealed by Munk School researchers.

However, it was named on a recent report from Reporters Without Borders as one of five corporate “enemies of the internet” which has produced technology “which has repeatedly been discovered in countries who mistreat journalists”. ®

Article source: http://go.theregister.com/feed/www.theregister.co.uk/2013/03/19/finfisher_spyware_apac_countries/

Researchers at the University of Glasgow have found that cloud storage apps that say they send files to the cloud also leave retrievable versions of files on the devices.

The files aren’t there for all to see, but do represent a resource useful to forensic investigators or those willing to look hard – for whatever purpose.

The extent to which data remains on phones is detailed in a paper titled Using Smartphones as a Proxy for Forensic Evidence contained in Cloud Storage Services, delivered at the 46th Hawaii International Conference on System Sciences and authored by George Grispos, William Bradley Glisson and Tim Storer.

The paper explains the authors performed a hard reset on an iPhone 3G running iOS 3 and an HTC Desire running Android 2.1. Those phones were then equipped with Dropbox (iOS version 1.4.7, Android version 2.1.3), Box (iOS version 2.7.1, Android version 1.6.7) and SugarSync (iOS version 3.0, Android version 3.6). 20 files – JPGs, .DOCS, .PDFs, .MP3s and .MP4s – were created on each device. Some of the files were opened or altered and some left alone. The phones were then “manipulated in one of the following ways”:

• Active power state – the smartphone was not powered down and the application’s cache was not cleared;
• Cache cleared – the applications cache was cleared;
• Powered off – the smartphone was powered down; and
• Cache cleared and powered off

Next, the phones were “processed to create a forensic dump of its internal memory.” Which is when the researchers found lots of files in lots of places.

The exact results are detailed here (PDF), but the long and short of the study is that the HTC Desire’s SD card and the iPhone’s main storage both yielded files users would reasonably expect to have been vaporised and condensed in the cloud services mentioned above, rather than remaining on the handsets.

Both phones even yielded the unique file ID number for items uploaded to Box, an authentication token for that service and a URL. Together, “This information can be merged to reconstruct a URL, which will result in the file associated with the ZBOXID being downloaded.”

The paper offers the following conclusion:

“The results from this research have shown that smartphone devices which access cloud storage services can potentially contain a proxy view of the data stored in a cloud storage service. The recovery of data from these devices can in some scenarios provide access to further data stored in a cloud storage account. From the client perspective, it can potentially provide a partial view of the data without access to the data provider. The recovery of this evidence is dependent on two factors. First, the cloud storage application has been used to view the files in the cloud. Second, the user has not attempted to clear the cache of recently viewed files.”

The paper also suggests more research is needed, because the whole point of cloud storage is access from multiple devices and security of those devices is therefore very important.

More research also seems prudent because the authors used superseded versions of the apps. Dropbox on iOS, for example, is now at version 2.1.3, many updates beyond the version 1.4.7 used for this analysis.

The authors’ long-term plan is to get that additional research done and to eventually “propose a set of security measures for both cloud providers and smartphone users to mitigate the potential risk of data leakage.” ®

Article source: http://go.theregister.com/feed/www.theregister.co.uk/2013/03/19/cloud_storage_smartphone_research/

Japanese boffins at Fujitsu have announced a new imaging technology that can take a user’s pulse simply by monitoring the changing brightness of their face, and in so doing potentially stop shifty souls at the airport.

The ingenious video-based system captures the subject’s face, calculating the average value of red, green and blue colours in a pre-determined part of the face for each frame, before working out the brightness waveform from the green component and discarding other data.

This is important because haemoglobin absorbs green light, so a pulse can be calculated from the peaks in that brightness waveform, created as blood flows through the face, Fujitsu said.

Impressively, all of this can be done in as little as five seconds, lending it potential security applications by flagging people acting “suspiciously” at airports or public events, according to the Japanese IT giant.

Presumably, if it were used in these scenarios some work would have to be done to make sure aviophobics are not mistaken for nervy terrorists.

Fujitsu is also touting the facial imaging tech as a consumer application using PC webcams or smartphone cameras to help users monitor their health and understand how it changes over time.

The firm’s vision for a “Human Centric Intelligent Society” imagines customers will upload and manage this kind of data in the cloud through a dedicated health monitoring service. The firm has already unveiled plans for a similar kind of service for pet dogs.

It also announced the Hada Memori smartphone, a device designed to let owner’s monitor their skin tone.

This is one of a series of planned devices to monitor stress levels, exercise habits and even sleeping patterns, with the anonymised data sent up to the cloud where Fujitsu hopes to sell it on to beauty product manufacturers.

Being peered at by cameras is not alien to Asian airports, as some have used heat sensors to detect passengers with elevated temperatures as a way of detecting possible carriers of SARS or other infectious diseases. ®

Article source: http://go.theregister.com/feed/www.theregister.co.uk/2013/03/19/fujitsu_takes_pulse_with_camera/

A flaw in EA’s Origin game store puts its 40 million or so users at risk of remote execution vulnerabilities

The vulnerability was described by security researchers Luigi Auriemma and Donato Ferranta of ReVuln, in a paper released on Saturday.

Origin is the distribution platform behind just-launched SimCity, along with other popular EA games such as Crysis 3. It lets EA roll out updates to its games, sell titles, and also provides DRM capabilities by authenticating players’ games.

But the way the software authorizes players can also be used to hijack computers and install malicious software, the researchers found.

“The Origin platform allows malicious users to exploit local vulnerabilities or features, by abusing the Origin URI handling mechanism,” they write. “In other words, an attacker can craft a malicious internet link to execute malicious code remotely on victim’s system, which has Origin installed.

Origin works by using uniform resource identifiers (URIs) to authenticate and initiate games on players’ machines. The attack works by spoofing the URI via an URL on a third-party website, so that when a person clicks it, Origin silently opens and loads a file onto the users’ machine.

In a demonstration at the Black Hat Security Conference in Amsterdam on Friday, the researchers showed that the exploit could be used to load a Windows dynamic link library file onto the machines. However, because Origin functions on multiple platforms, the exploit works on other systems as well, they say.

A possible fix for the flaw is to disable URLs prefaced with “origin://” via software such as the urlprotocolview, but this will also render useless any desktop shortcuts for Origin games.

At the time of writing, EA had not responded to our requests for further information. This news comes alongside the abrupt departure of EA chief executive John Riccitiello. ®

Article source: http://go.theregister.com/feed/www.theregister.co.uk/2013/03/19/ea_origin_bug_allows_remote_exploits/

Stuxnet gave the world a graphic demonstration just how high the stakes can be when malware hits machinery.

This week, the world is starting to plan a response to an even scarier incident, in which an online attack is aimed at a working nuclear or radiological facility.

Leading the fight is the International Atomic Energy Agency, which this week hosts an event titled “Consultancy Meeting Incident Response Planning for Computer Security Events at Nuclear/Radiological Facilities” in its Vienna home.

The agency isn’t saying much about the week-long talk-fest, but the European Network and Information Security Agency (ENISA), which will attend to lend its expertise, says the objective is to “… provide guidance on the key elements required to develop and implent [sic] a comprehensive security response plan associated with an information and computer/computer systems attacks at nuclear material, or other associated activities.”

ENISA says it also expects to discuss the following:

• Information and computer security incident scenarios
  
  -Categorisation of information and computer security incidents
  
  -Organisational roles and structures
  
  -Outline of response and recovery procedures for incidents oon industrial computer systems (ICS, including computer based systems used for physical prtection, nuclear safety and nuclear material accountancy and control
• National reporting and response structures , including Computer Emergency Response Teams.
• Restoration and recovery activities.

ENISA won’t be alone at the event: The Reg has seen letters to various nations’ nuclear regulators asking them to send representatives to the event.

Those letters say the hoped-for outcome of the meeting is “the initial development of a new Nuclear Security Series document: ‘Incident Response Planning for Computer Security Events at Nuclear/Radiological Facilities’.” That document is expected to serve as a useful guide for others. Let’s hope that: a) They do a good job; and b) It never gets used. reg

Article source: http://go.theregister.com/feed/www.theregister.co.uk/2013/03/19/iaea_nuke_hack_defence_meeting/

Andrew Auernheimer, a member of the grey-hat hacking collective Goatse Security, has been sent down for three years and five months in the slammer after he helped splay open iPad user’s private email addresses via a flaw in ATT’s servers.

Auernheimer, known online as Weev, received his sentence wearing shackles after he tried to bring a mobile phone into the courtroom. After completing his term he will have to pay over $72,000 in restitution to ATT and undergo three years of supervised release.

“I didn’t come here today to ask for forgiveness,” Auernheimer told US District Judge Susan Wigenton, Bloomberg reports. “The Internet is bigger than any law can contain. Many, many governments that have attempted to restrict the freedoms of the Internet have ended up toppled.”

In 2010, Auernheimer found a flaw in a public-facing ATT server that could be used, via the iPad’s integrated circuit card identifier (ICC-ID), to uncover the names and email addresses of 114,067 early adopters of Apple’s 3G-equipped fondleslab. His colleague Daniel Spitler wrote a PHP script called “iPad 3G Account Slurper” to harvest the data, and then handed it over to online magazine Gawker.

The data caused huge embarrassment to ATT and Apple, since it included the personal emails of then-White House Chief of Staff Rahm Emanuel, New York Mayor Michael Bloomberg, film mogul Harvey Weinstein, and several high-ranking US Army officials. ATT fixed the flaw, and there’s no evidence Auernheimer did anything more than highlight the sloppy coding.

His defense lawyers argued that he was accessing information on a public web server and that if this was a crime then most internet users are guilty too. This cut little ice with the presiding judge.

“While you consider yourself to be a hero of sorts, without question the evidence that came out at trial reflected criminal conduct,” Judge Wigenton said in imposing the sentence. “You’ve shown absolutely no remorse. You’ve taken no responsibility for these criminal acts whatsoever. You’ve shown no contrition whatsoever.”

Auernheimer’s colleague Spitler now looks likely to face a similar sentence after pleading guilty, andsome in the security field are warning that the verdict will have a deadening effect of flaw exposure. Former NSA programmer and now Apple-cracker and security consultant Charlie Miller said the decision was highly troublesome.

We could all go to jail for security research at any moment, and a jury would happily convict us.

— Charlie Miller (@0xcharlie) March 18, 2013

In this hack’s opinion, Auernheimer’s sentence is far too severe. You could argue that he should have submitted the flaw to ATT, waited for the problem to be fixed, and then reaped the publicity. He could also have profited from selling the flaw on the grey or black markets, but chose not to go for the money, but to get embarrassment value instead.

“My regret is being nice enough to give ATT a chance to patch before dropping the dataset to Gawker. I won’t nearly be as nice next time,” he said in a Reddit forum.

With no evidence of harm done, sending someone down for over three years, near-bankrupting them with fines, and setting such a long probation victim looks less like justice and more like judicial spite. ®

Article source: http://go.theregister.com/feed/www.theregister.co.uk/2013/03/18/auernheimer_ipad_hack_prison/

Huawei has been accused of poor security practice by Russian researcher Nikita Tarakanov, who told Black Hat Europe last week that the vendor’s 3G and 4G devices are vulnerable and its update server is a massive attack vector.

The update server in the Netherlands that Tarakanov tested probably isn’t the only one used by Huawei, but he found it was running on the hoary code of Windows IIS 6.0 – a relic from Windows Server 2003. If the server were compromised, he said, an attacker could distribute a malicious update to millions of dongle users.

While the executable the USB keys run is signed, he says the modems’ plaintext configuration files are another matter: easy to modify in ways that would point the modems to malicious software. Two examples given in Network World are that an attacker could change the DNS the modems use, pointing users to malicious Websites; and a built-in anti-virus installer parameter, which could be modified to install malicious software.

There are also privilege escalation vulnerabilities under both Windows and OS X. According to The H Security, the latter vulnerability was a last-second addition to the presentation after iOS researcher Stefan Esser discovered and tweeted it from the conference: the Huawei OS X update app (ouc.app) has unrestricted access to /usr/local.

“Can anyone verify that the Telekom LTE Stick from Huawei makes /usr/local world writable on OSX? WTF?”, Esser posted.

While the research was conducted in Russia, Tarakanov believes the vulnerabilities aren’t specific to that country. ®

Article source: http://go.theregister.com/feed/www.theregister.co.uk/2013/03/17/huawei_3g_4g_vulnerability/