STE WILLIAMS

Adobe has Patch Tuesdays, too – a reader reminds us!

Naked Security reader Haemish Edgerton just gave us a very polite but effective scolding (Linus Torvalds, take note!) for neglecting to mention the Adobe fixes that came out on Tuesday.

As Haemish pointed out:

I realize that the Apple iPhone 5S fingerprint sensor was automatically going to get a lot of attention in the context of security, but Adobe updates are important too :)

Sorry, Haemish.

For the record: there were three bulletins, four platforms affected (Windows, Linux, OS X and Android), five products updated, and fourteen vulnerabilities (CVEs) covered.

The bulletins are: APSB13-21, APSB13-22 and APSB13-23

All three bulletins list the vulnerabilities as potentially exploitable for Remote Code Execution (RCE), or, in Adobe’s own words, as bugs that could “allow an attacker to take control of the affected system.”

As is often the case with Adobe’s updates, there are lots of version numbers to take into account.

That’s especially true of Flash Player, where it seems that the product’s source code for the various platforms supported is currently at a wide range of different stations in Adobe’s railway network. (Platforms. Stations. Geddit?)

Here is what to look for if you want to see if you are vulnerable:

Confused?

Spare a thought for the guys in Adobe Quality Assurance!

Article source: http://feedproxy.google.com/~r/nakedsecurity/~3/V7SuBOWfWh8/

WordPress issues security fixes, advises “update your sites immediately”

Mega-popular blogging and content management system WordPress has just put out version 3.6.1.

Since it’s a maintenance release (an update from 3.6), it doesn’t have a huge raft of new features, but it does fix three security holes.

One of them is a Remote Code Execution vulnerability reported by a young Belgian web application security researcher named Tom Van Goethem.

Now that the fix is out, Van Goethem has published a very detailed description of the bug and the steps he went through to uncover it.

He also mentions that, by using a popular plugin, he was able to go from vulnerability (“there’s a hole, and it could be risky”) to exploit (“here’s how to use the vulnerability for unauthorised access”).

Fortunately, however, he hasn’t gone down the complete-and-total disclosure route, stopping short of giving you a working exploit and, saying:

Due to ethical considerations, I will not disclose a Proof of Concept of this exploit at this time, as there are too many vulnerable WordPress installations out there.

Van Goethem’s bug relates to PHP serialisation.

That’s where you take data, and perhaps even code, from a programming environment, and convert it into text string representation.

This means it can easily be saved, moved around on a network, and restored later.

It’s called serialisation because even data structures that have a complex layout in memory, such as arrays and tables, end up as a linear (i.e. serial) stream of bytes.

Going back from a serialised text string to live, run-time data inside a program is, unsurpisingly, known as unserialisation.

Any software that inadvertently passes unfiltered, remotely-supplied data into an unserialisation function is taking a bit of a chance, and that’s what was happening inside WordPress.

By the time you get round to validating that unserialised data, it already exists as a live data structure inside a live run-time environment: that’s a bit like dry-firing a handgun as a way of satisfying yourself that it isn’t loaded.

WordPress, which currently claims more than 7.5 million downloads, has unsurprisingly suggested that “you update your sites immediately.”

Note. Sophos Naked Security and the Sophos Corporate Blog are hosted on WordPress.com VIP servers. These servers were already updated by the time we received the advisory email from WordPress [2013-09-12T13:24Z].

Article source: http://feedproxy.google.com/~r/nakedsecurity/~3/XmEuPvznhtQ/

57% of college students think their Facebook postings aren’t vile at ALL!

Party guy. Image courtesy of Shutterstock57% of college students view their Facebook postings through rose-tinted glasses, blithely seeing nothing inappropriate.

That’s a dangerous perception mismatch, given that at least one previous industry survey found that 69% of recruiters have rejected candidates because of what their social media personas reveal.

The figures concerning students’ blissful ignorance come out of a new study from Persona, a social media utility dedicated to helping Facebook, Twitter, and Google+ users protect their professional reputations.

According to the earlier industry survey about recruiters’ use of social media to screen job applicants, which was done in 2011 by the online image management company TrustedID Reppler and posted on CNN, out of about 300 recruiters surveyed, the vast majority – 91% – said that they trawl the internet to screen job candidates.

In fact, recruiters are nearly unanimous: Persona cites a 2012 Jobvite survey that found that 92% of recruiters planned to mine social media for recruiting.

So, where do college students get the idea that their postings are pristine?

It’s not ignorance at work: 71% of students surveyed by Persona believe that Facebook profiles are “influential” or “very influential” components of hiring decisions.

In spite of that, students aren’t clearing away their tracks. Some of the survey’s findings:

  • 55% “never” delete or untag inappropriate photos and posts, or do so only “once a year.”
  • 80% would feel “comfortable” or “very comfortable” if a recruiter examined their profiles.
  • 57% rely on privacy settings rather than actively monitoring their profiles.

Could it be that students aren’t aware of what, exactly, a potential employer might find objectionable with, for example, pictures of candidates face-down in their oatmeal after one too many?

Students, I’m here to help. Read on.

In an April 2012 write-up of a CareerBuilder survey of some 2,300 hiring managers, AOL Jobs’ David Schepp reported that those responsible for hiring were turned off by these categories of social media missteps:

  • Candidate’s provocative/inappropriate photos/comments: 49%
  • Candidate drinking or using drugs: 45%
  • Candidate had “poor communication skills”: 35%
  • Candidate bad-mouthed a previous employer: 33%
  • Candidate made discriminatory comments related to race, gender or religion: 28
  • Candidate lied about qualifications: 22%

Sophos has recently been pumping readers full of tips on keeping safe on Facebook, plus some further tips and tricks for Facebook, like how to block someone, or remove certain posts.

Facebook. Image courtesy of ShutterstockIf college students need further help with keeping safe on Facebook, as in, keeping safe their chances of ever getting hired, it would behoove them to take a peek at one site that collates truly embarrassing Facebook postings.

If anything on that site raises a sense of déjà vu, you’ve got some cleaning up to do.

It’s not just students who need to be taken to task, here.

None of us should trust privacy settings to block all the inadvisable things our friends post.

We should all be actively monitoring our online presences, including deleting or untagging inappropriate photos and posts on a regular basis.

Have I done that recently? Well… Uhhh… Hmm….

Excuse me, I have a bit of work to do.

Image of party guy and Facebook screenshot courtesy of Shutterstock.

Article source: http://feedproxy.google.com/~r/nakedsecurity/~3/WqPvfSRo06E/

Apple ships OS X 10.8.5 security update – fixes “sudo” bug at last

I’m calling it a security update, though it’s officially a full-on point release of OS X Mountain Lion, taking the 10.8 variant of Apple’s OS X to version 10.8.5.

But with twice as many security fixes listed as regular bug fixes and improvements, I’m happy to call it a “security update,” if only in the hope you’ll feel a bit more urgency about deploying it.

There are 15 official security patches, one fix that Apple has appended to the list without explicitly admitting that it was a security issue, and one bonus patch that is mentioned on Apple’s website but not in its emailed security advisory.

I’ll start with the free bonus patch, because I’m delighted it’s happened and I think you should know about it.

The infamous sudo privilege escalation, documented and patched by sudo itself back in February and pointedly exposed on OS X by Metasploit last month, is no more.

Confusingly, if you run sudo -V to check the version number, you might get the impression it hasn’t been updated, since 1.7.4p6a has the same core version string as the version shipped with 10.8.4 (1.7.4p6).

Nevertheless, the sudo binary has been updated, and in my tests, the privilege escalation bug had vanished.

Until 10.8.4, doing a sudo -k (which is supposed to deauthenticate you, and thus does not require a password), followed by setting the time to just after midnight on 01 Jan 1970, would give you root access.

In 10.8.5, it does not.

Presumably, Apple yielded to public pressure to fix this long-running hole, but, instead of taking all the sudo changes from the past few months, just backported the sudo -k fix to version 1.7.4p6, a much less risky change.

Moving up the list, the not-a-security-fix I mentioned above is included, almost as an afterthought, as follows:

OS X Mountain Lion v10.8.5 also addresses an issue where certain Unicode strings could cause applications to unexpectedly terminate.

That’s the bug we decribed as “only six characters from a crash.”

Although it probably deserved to be called a denial of service rather than merely “an issue,” it was indeed more of an annoyance than a vehicle for cybercrooks.

At any rate, it’s good to see it patched quickly.

Other significant patches include potential remote code execution holes in:

  • JBIG2 decompression in PDF files by the CoreGraphics library.
  • JPEG2000 decompression in PDF files by the ImageIO component.
  • The web programming system PHP.
  • The handling of QuickTime movies by QuickTime.

If you’re an OS X user, you may have been unaware that PHP was installed at all, since it is usually considered a server-side component.

But it is present, and it was vulnerable, although it isn’t enabled by default, even if you turn on OS X’s built-in Apache web server.

PHP isn’t the only server-flavoured component to receive security attention in 10.8.5, with fixes also shipped for the following applications usually found on servers:

  • The Apache webserver. (Cross-site scripting.)
  • The name server Bind. (Denial of service.)
  • The database server PostgreSQL. (Privilege escalation.)

For users on the still-supported earlier versions of OS X, namely Snow Leopard (10.6) and Lion (10.7), the latest fixes come as Security Update 2013-004, rather than as a point release.

The list of fixes for 10.6 and 10.7 is similar to the list for 10.8.5, with the addition of a remote code execution flaw in ClamAV. (ClamAV is not part of the OS X 10.8 distribution.)

Also, the oldest supported OS X version, 10.6, gets a separate update for a remote code execution hole in Safari, which moves to version 5.1.10.

Neither Lion nor Mountain Lion need or receive this fix, as they are on Safari 6.

As usual, you can grab Apple’s updates by simply clicking on the Apple Menu and choosing Software Update… or by downloading them as DMG files from Apple’s download site.

Some useful pages on Apple’s site include:

To conclude, even though Macs don’t get malware (only kidding!), Apple has updated its plugin blocker following Adobe’s latest Patch Tuesday.

Safari will now refuse to use Flash plugins earlier than 11.8.800.94.

That doesn’t force you to be bang up to date with Flash – the September Patch Tuesday introduced 11.8.800.168 to fix remote code execution holes in the 11.8.800.94 – but ensuring you are at the latest-but-one is at least a start.

Happy patching!

(I did my 10.8.5 update early this morning: it may be only half a day, but so far, so good.)

Article source: http://feedproxy.google.com/~r/nakedsecurity/~3/LwapU0VaBSE/

Microsoft reissues September patches after user complaints

Supercharge your infrastructure

Problems with Microsoft’s last round of operating system and application patches have forced the company to reissue part of the update on Friday.

“Since the shipment of the September 2013 Security Bulletin Release, we have received reports of updates being offered for installation multiple times, or certain cases where updates were not offered via Windows Server Update Services (WSUS) or System Center Configuration Manager (SCCM),” said Redmond’s Office team in a blog post.


“We have investigated the issue, established the cause, and we have released new updates that will cease the unnecessary re-targeting of the updates or the correct offering of these updates.”

Register readers – and many other Microsoft users – started complaining about the patches shortly after their release on Tuesday. Some readers reported detection issues that left servers stuck in a loop of patching when the updates weren’t recognized, while others reported being unable to install flaw fixes.

Eight patches have now been reissued, covering security flaws in Excel, SharePoint Server, and Office suites going back to 2007. Two non-security patches for PowerPoint have also been reissued.

Unusually for Microsoft, not all the patches it promised for Patch Tuesday were in the final release, with one being pulled for quality-control issues. El Reg suspects there have been some harsh exchanges between management and the software testing teams at Redmond. ®

Supercharge your infrastructure

Article source: http://go.theregister.com/feed/www.theregister.co.uk/2013/09/13/microsoft_reissues_september_patches_after_user_complaints/

It’s about time: Java update includes tool for blocking drive-by exploits

Free ESG report : Seamless data management with Avere FXT

Oracle’s latest update to the Java SE Development Kit (JDK) version 7 adds new security features designed to help businesses avoid being stung by critical vulnerabilities in out-of-date versions of Java.

After a string of embarrassing Java security flaws was disclosed by independent researchers, Oracle has made addressing vulnerabilities its top priority for JDK 7, even going as far as to delay the release of JDK 8 so it could devote more resources to fixing bugs.


But many businesses still keep older versions of Java installed on client PCs because certain custom applications require them. That’s bad, because these out-of-date versions contain critical vulnerabilities that in some cases will never be fixed. Oracle discontinued support for JDK 6 in June.

JDK 7 Update 40, issued on Tuesday, implements a new feature called Deployment Rule Set that aims to address this problem. It allows businesses that centrally manage their Java desktop installations to establish a set of rules specifying which Java applets and Java Web Start applications – collectively termed Rich Internet Applications (RIAs) – are allowed to run on client PCs.

For example, an admin could create a rule blocking execution of all RIAs and then add additional rules to whitelist specific ones. Rules can be written to match any portion of an application’s URL, including the port number, and they can even specify the version of Java that should be used to run it. Full documentation on how this is done is available here.

By creating such rules, companies should be able to avoid many of the most serious Java exploits that have cropped up in recent months, most of which attack systems via the Java web plugin and do not affect server-side Java applications or desktop applications installed on the local machine.

Rules can additionally allow whitelisted RIAs to run without certain security prompts, such as warnings that the user is running an out-of-date version of Java.

The one caveat is that the Deployment Rule Set feature requires all client PCs to have the version of the Java web plugin that was distributed with Java SE 6 Update 10 and later. If a Deployment Rule Set is installed on a machine and the older version of the plugin is detected, all RIAs will be blocked.

Oracle also cautions companies to be careful not to let their rule sets fall into the wrong hands:

The Deployment Rule Set feature is optional and shall only be used internally in an organization with a controlled environment. If a JAR file that contains a rule set is distributed or made available publicly, then the certificate used to sign the rule set will be blacklisted and blocked in Java.

In addition to Deployment Rule Set, JDK 7 Update 40 brings several other new features and improvements, including Retina Display support for OS X, advanced monitoring and diagnostic tools for developers, new security warnings for unsigned and self-signed applications, and restrictions on use of certificates with keys less than 1024 bits in length. It also includes a number of bug fixes.

The update is available from the usual Java download website. ®

Free ESG report : Seamless data management with Avere FXT

Article source: http://go.theregister.com/feed/www.theregister.co.uk/2013/09/13/java_deployment_rule_set/

Apple Touch ID Fingerprint Scanner Unlocks Biometrics Debate

Giving the finger — so to speak — to Apple’s Touch ID feature may unlock the iPhone 5s and allow users to authorize purchases on iTunes, but whether the fingerprint scanning technology will push biometrics deep into the mainstream remains to be seen.

“Fingerprint readers, or biometrics, will not replace passwords in the near future for two reasons,” says Gene Meltser, technical director of security services firm Neohapsis. “First, fingerprints are not secret or even private — an average person leaves hundreds of fingerprints on various surfaces throughout the day. Second, count your fingers — if you’re like the 99.9 percent of the world, you have a total of 10 biometric passwords at your disposal for the rest of your life. A compromised fingerprint can never be effectively replaced like you can replace a password, and therefore cannot be relied on as future-proof authentication on its own.”

In Apple’s defense, the company has told the media that the iPhone does not store actual fingerprints — just “fingerprint data” in the iPhone’s processor. Also, users interested in using Touch ID must choose a password as a backup, and third-party applications are currently banned from using the scanner altogether.

On top of this, the most commonly mentioned attack, where someone steals a phone and lifts the fingerprints off of the device to unlock it, is not all that likely for the typical user, says Sebastien Taveau, founding board member of Fast IDentity Online Alliance (FIDO Alliance). Today’s sensors have multiple mechanisms built in to protect from fake fingers, template dissociation, and authentication replay, he says.

“It is misinformed to assume that the industry hasn’t developed and overcome the past vulnerabilities with extensive RD, making these fears a thing of the past,” he says. “That being said, even if someone were able to lift a fingerprint, odds are they have one-tenth of a chance to get the correct finger. If they were to be so lucky to capture an enrolled fingerprint, then they would have to have a ‘clean’ fingerprint to proceed at all.”

Meltser agrees, acknowledging the practicality of an attacker getting a clean fingerprint and duplicating that print on a polymer model of a human fingerprint with enough detail to account for skin pores and other features. Still, biometric technology should be viewed as only part of the solution to the challenge of authentication.

“It’s a third component to a well-known authentication adage: A good authentication solution must provide several of the following in addition to the username — something you have (a physical token), something you know (a password) and something you are (a retina scan, or a fingerprint). Biometrics is a great enhancement to a good authentication strategy, but counting on biometrics as a standalone authentication panacea is premature,” he says.

Apple is far from the only vendor to make recent announcements tied to biometrics. McAfee, for example, announced that its LiveSafe service would feature voice and face recognition. The same goes for the mobile space; two years ago, Motorola — now owned by Google — also released its Atrix phone with fingerprint-scanning technology of its own.

Even as weak passwords are often cited as a weak link in security, this has not, however, led to passwords falling off in use. Laptops have featured fingerprint scanners for years, but passwords are still the primary authenticator used by consumers, notes Neohapsis security consultant Joe Schumacher.

“The password is easier to accept by third parties without the worry of collaborating with other parties,” he says. “For example, if one website is using fingerprint identification, then it would need to store that information in a secure manner that is the same as website number two. Also, many fingerprint readers used for authentication still require a password.”

Still, Jamie Cowper, senior director at authentication solution provider Nok Nok Labs, says he expects there to be a significant growth in biometrics, particularly on mobile devices.

“This will be a combination of an improvement in sensor capability — voice, face, fingerprint, and others, plus the enhanced functions coming into the smartphone market — such as secure elements and operating systems that can be used to create and store biometric secrets,” he says. “We are also seeing a shift toward multifactor authentication by large-scale Web services, and biometrics represents a simple, user-friendly way of implementing multifactor. This growth may well occur in the consumer sector first, as enterprise IT will take their time to consider different security models.”

Have a comment on this story? Please click “Add Your Comment” below. If you’d like to contact Dark Reading’s editors directly, send us a message.

Article source: http://www.darkreading.com/mobile/apple-touch-id-fingerprint-scanner-unloc/240161303

12 arrested as UK cops foil Santander bank heist plot

Santander. Image courtesy of ShutterstockLondon Metropolitan Police are holding 12 men in connection with a scheme to access computers at Santander, one of the UK’s largest banks.

The plot involved planting remote-access hardware on a system in a local branch, from where the crooks hoped to navigate the bank’s networks from the inside.

Variously described as “audacious“, “sophisticated” and “significant“, the potential takings from the haul are estimated to have been in the millions, had it succeeded.

The plan centred around attaching KVM (Keyboard, Video, Mouse) kit to a machine at a Santander branch in Surrey Quays, southeast London. A man posing as an engineer from a telecoms firm fitted the device, but it was never operational.

If it had been activated, the crooks could have monitored all activity on the system or operated it themselves from a remote location. It’s not thought that any Santander employees were involved in the plot.

The Met police have had a busy day searching properties across London, and seizing equipment thought to be related to the case.

It seems the law had some advance warning of the scheme, as Santander claims to have been working with the police for several months prior to the attempt to fit the device. The arrests happened within hours of the hardware being put in place.

Santander UK, owned by global mega-bank Grupo Santander, was formed from the acquisition of several nationwide savings firms, and has over 1,300 branches and 25,000 employees in the UK.

This size, and the number of people involved in maintaining and running all those branches, must make it hard to keep an eye on all workstations for the addition of rogue hardware, which can be difficult if not impossible for security software to detect.

Hacker. Image courtesy of ShutterstockOf course, even with access to a workstation, there’s no telling how much further the crooks would have got.

If the systems were well controlled, secured and monitored, there should still have been plenty of obstacles to overcome before they could find their way into sensitive parts of the network, and move virtual cash out of the bank’s systems to somewhere they could turn it into bling and motors.

The partial success of the scheme seems to imply that despite centuries of evolution, the physical security of banks still needs more work, now that computers and cabling need to be monitored as closely as vaults, vans and drawers full of cash.

Let’s hope the digital protections are a little more solid.


Image of Santander and hacker courtesy of Shutterstock.

Article source: http://feedproxy.google.com/~r/nakedsecurity/~3/X9G6jZ67cwQ/

Apple ships OS X 10.8.5 security update

I’m calling it a security update, though it’s officially a full-on point release of OS X Mountain Lion, taking the 10.8 variant of Apple’s OS X to version 10.8.5.

But with twice as many security fixes listed as regular bug fixes and improvements, I’m happy to call it a “security update,” if only in the hope you’ll feel a bit more urgency about deploying it.

There are 15 official security patches, one fix that Apple has appended to the list without explicitly admitting that it was a security issue, and one bonus patch that is mentioned on Apple’s website but not in its emailed security advisory.

I’ll start with the free bonus patch, because I’m delighted it’s happened and I think you should know about it.

The infamous sudo privilege escalation, documented and patched by sudo itself back in February and pointedly exposed on OS X by Metasploit last month, is no more.

Confusingly, if you run sudo -V to check the version number, you’ll get the impression it hasn’t been updated, since 1.7.4p6a is the same as the version shipped with 10.8.4, and is officially listed by the sudo project as vulnerable.

Nevertheless, the sudo binary has been updated, and in my tests, the privilege escalation bug had vanished.

Until 10.8.4, doing a sudo -k (which is supposed to deauthenticate you, and thus does not require a password), followed by setting the time to just after midnight on 01 Jan 1970, would give you root access.

In 10.8.5, it does not.

Presumably, Apple yielded to public pressure to fix this long-running hole, but, instead of taking all the sudo changes from the past few months, just backported the sudo -k fix, a much less risky change.

Moving up the list, the not-a-security-fix I mentioned above is included, almost as an afterthought, as follows:

OS X Mountain Lion v10.8.5 also addresses an issue where certain Unicode strings could cause applications to unexpectedly terminate.

That’s the bug we decribed as “only six characters from a crash.”

Although it probably deserved to be called a denial of service rather than merely “an issue,” it was indeed more of an annoyance than a vehicle for cybercrooks.

At any rate, it’s good to see it patched quickly.

Other significant patches include potential remote code execution holes in:

  • JBIG2 decompression in PDF files by the CoreGraphics library.
  • JPEG2000 decompression in PDF files by the ImageIO component.
  • The web programming system PHP.
  • The handling of QuickTime movies by QuickTime.

If you’re an OS X user, you may have been unaware that PHP was installed at all, since it is usually considered a server-side component.

But it is present, and it was vulnerable, although it isn’t enabled by default, even if you turn on OS X’s built-in Apache web server.

PHP isn’t the only server-flavoured component to receive security attention in 10.8.5, with fixes also shipped for the following applications usually found on servers:

  • The Apache webserver. (Cross-site scripting.)
  • The name server Bind. (Denial of service.)
  • The database server PostgreSQL. (Privilege escalation.)

For users on the still-supported earlier versions of OS X, namely Snow Leopard (10.6) and Lion (10.7), the latest fixes come as Security Update 2013-004, rather than as a point release.

The list of fixes for 10.6 and 10.7 is similar to the list for 10.8.5, with the addition of a remote code execution flaw in ClamAV. (ClamAV is not part of the OS X 10.8 distribution.)

Also, the oldest supported OS X version, 10.6, gets a separate update for a remote code execution hole in Safari, which moves to version 5.1.10.

Neither Lion nor Mountain Lion need or receive this fix, as they are on Safari 6.

As usual, you can grab Apple’s updates by simply using Apple Menu | Software Update… or by downloading them as DMG files from Apple’s download site.

Some useful pages on Apple’s site include:

To conclude, even though Macs don’t get malware (only kidding!), Apple has updated its plugin blocker following Adobe’s latest Patch Tuesday.

Safari will now refuse to use Flash plugins earlier than 11.8.800.94.

That doesn’t force you to be bang up to date with Flash – the September Patch Tuesday introduced 11.8.800.168 to fix remote code execution holes in the 11.8.800.94 – but ensuring you are at the latest-but-one is at least a start.

Happy patching!

(I did my 10.8.5 update early this morning: it may be only half a day, but so far, so good.)

Article source: http://feedproxy.google.com/~r/nakedsecurity/~3/dakuwQk71JU/

Microsoft’s swipe’n’swirl pic passwords LESS secure than PINs, warn researchers

Free ESG report : Seamless data management with Avere FXT

Microsoft’s promotion of visual passwords, based on tapping pictures and making gestures instead of conventional text passwords, might be a boon for usability. Yet security experts warn the technology is less secure than even a simple 4-digit PIN.

The increased power of brute force attacks, password hash database leaks and the difficulty of getting users to choose secure text passwords in the first place means that attempts to create alternative login techniques are well worth exploring.


Windows 8 and Windows RT come with a feature called Picture Passwords. Users can choose any picture, and then “annotate” it with three finger movements: tapping a point, drawing a stroke, or sweeping a circle. This pattern becomes a users’ means to open or unlock a device as an alternative to a text password or PIN unlock code.

The picture helps you to remember where you made the gestures, so you can repeat them reliably enough to pass the test and unlock your device.

Tap, tap, pinto, stroke. Hack?

Four security researchers from Arizona State University and Delaware State University tried to measure the safety of picture passwords in a research paper, titled On the Security of Picture Gesture Authentication (PDF). The paper was presented at last month’s USENIX Security Symposium (summary and video here).

Microsoft’s own paper on the design, implementation and likely strength of picture passwords estimates that there are just over 1.155 billion possible picture passwords if three gestures are used.

The sounds like a lot, but is “only about four times as many as there are six-character passwords using the characters A to Z,” says security watcher Paul Ducklin. “No-one is seriously suggesting six-character, letters-only passwords these days,” he notes in a post on Sophos’s Naked Security blog.

“Furthermore, the equivalent of a dictionary attack is possible, too, if you can identify the most likely Points of Interest (PoIs) in the password picture. So a brute force attack is certainly possible, where you ignore the picture entirely and just try every possible tap-click-circle combination,” he adds.

All is not lost, however. Like credit cards that automatically block after three failed PIN entry attempts, making five mistakes in inputting a picture password obliges users to switch to using an old-fashioned text password. This, combined with the need for physical access to the device, limits the potential for potential misuse.

Can you see what it is yet?

The weakness, according to the researchers behind the USENIX paper, is that the point of interest in a picture users might tap on and the gesture they might make can be guessed. Microsoft’s own ad for picture passwords features a picture of someone’s two young daughters, heads close together and looking at some distant object. The password involved circling their heads and then drawing a line in the direction they were looking.

The chosen pattern is easily guessed. Using a test set of just over 10,000 passwords and 800 subjects, the Arizona State University and Delaware State University team reckon that automated point of recognition and other techniques can be used to guess visual gesture-based passwords correctly in 19 out of 1000 cases, given five attempts. The first guess alone would work in around nine in 1000 cases. Manual point of interest recognition offers even better results with a 26 in 1000 chance of hitting on the right gesture within five attempts.

So the security of picture passwords is a lot less than the three-in-10,000 chance of correctly guessing a randomly chosen four-digit SIM or credit card PIN before subsequent re-tries are blocked. In practice, however device unlock numbers are often not chosen randomly; something that limits their security.

Picture this

Picture gesture authentication has many of the same limitations of text passwords, as a blog post by Kaspersky Labs’ Threatpost news service notes. The computer scientists behind the USENIX research urge Microsoft and other suppliers to be more upfront to users about this point, as well as developing tools to provide an indication of the strength of visual passwords, similar to text password strength meters.

Those not deterred by these figures and still attracted by convenience of gesture-based visual passwords would be well advised to select hard-to-guess picture passwords.

“If you use Picture Passwords, don’t make it easy for the crooks: choose pictures with lots of PoIs, and don’t just ‘do the obvious’ when you choose the gestures you’re going to use,” Ducklin advises.

Microsoft itself offers tips on picking a secure picture password.

Per Thorsheim, an independent security consultant, who runs a set of conferences about password security, has a good overview of the multiple password options bundles with Windows 8 here.

Mapping function

Security researchers have shown how to extract passwords, hashes and password hints from Windows 8. Extracting PIN and picture password data might also be possible, and this wouldn’t need any attempt at guessing picture passwords. Both Ducklin and Thorsheim expressed interest in seeing more research into this area of offline attacks.

“How Picture Password data is stored, and how password attempts are tested against the database, is proprietary,” Ducklin writes. “With an effective key size of just 30 bits, it is vital to set a very high cost for testing each potential password against an offline copy of the password database. That requires a computationally expensive Key Derivation Function (KDF).”

Ducklin called on Microsoft should go public on how Picture Passwords work, from how they’re stored to how the KDF is calculated.

“You’d let outside experts assess the risk of offline attacks, which would be technically valuable. And you’d get great positive publicity for openness, considering the current brouhaha facing proprietary software vendors over the cryptographic influence of the world’s intelligence services,” he writes. ®

Free ESG report : Seamless data management with Avere FXT

Article source: http://go.theregister.com/feed/www.theregister.co.uk/2013/09/13/picture_passwords/