Showing posts with label Security. Show all posts
Showing posts with label Security. Show all posts

Tuesday, March 5, 2013

Kaspersky PURE 3.0 Total Security adds online shopping protection


On Monday, Kaspersky released version 3.0 of PURE Total Security--the company's most-feature-packed PC antivirus and security product--priced at $90 for a 3-user, 1-year license. There are no major visual changes, but this new version has some new and improved features in addition to the existing file encryption, file backup, password manager, parental controls, and remote access features, among others.
The new Safe Money feature helps you better secure your online banking and shopping. Before you log onto a banking or payment site, it automatically verifies the site's security certificate and reputation. If the site is found genuine, Safe Monet opens it in a secure browser with secure keyboard input to prevent your login or payment information from being captured by any keyloggers or malware you might have on your PC.
Kaspersky also made improvements to the suite's existing backup utility, which now allows you to backup your important files online with the integrated Dropbox storage in addition to local drives. And if lose your files or your PC crashes, you can restore them via the program or access them from the Dropbox website.
The existing Online Password Manager that can store your passwords in a protected vault now offers a synchronization feature. Login details, notes, and identity information you save can automatically be encrypted, stored online, and synced with your other PURE-protected machines.
Kaspersky has also made several other minor improvements in version 3.0, including better protection against exploits and phishing sites, and improvements to the content filtering for the Paternal Controls.
Want to give Kaspersky a try? Though the company doesn't provide a full free edition, it does offer a free scan-only download and 30-day free trials for all its products.

Evernote hack shows that passwords aren't good enough


Evernote revealed over the weekend that it was the victim of a data breach, emailing users and posting a notice on its Web site that attackers had gained access to usernames, email addresses, and encrypted passwords associated with Evernote accounts. As a precaution, Evernote forced all 50 million users to reset their passwords. That's a good step, but it's not really not good enough--so Evernote is accelerating its plan to roll out two-factor authentication.
Evernote wasn't originally designed as a business service, at least until the December release ofEvernote for Business. Evernote is primarily a note-taking and organizational tool similar to Microsoft's OneNote. Evernote provides a range of services--including Evernote Food, Evernote Peek, Skitch, Penultimate and more--as Web-based tools or apps across a range of operating systems and mobile platforms. Its capability to access and sync data across a broad range of devices makes it appealing as a business tool.
By its nature, Evernote is a prime example of a service where you stash both personal and professional data. Like any cloud-based service, it comes with some inherent risk. Any time you place business data in the cloud--particularly sensitive information such as customer names or addresses, banking or financial details, or proprietary company research--you are trusting the vendor to protect it. The big caveat, though, is that you are still ultimately responsible for what happens to your data.
One password to rule them all?
Evernote claims that the password data captured by the attackers was encrypted, but it still made all users select new passwords, just in case. As respected security authority Brian Krebs notes in his blog post on the Evernote breach, the standard hashing and salting algorithms used by vendors to encrypt password data offers trivial protection that can be cracked with relative ease.
One solution would be to use stronger passwords or passphrases, and to ensure that you don't use the same password for more than one service. When you do, a data breach at one vendor can expose your password, which could then allow the attacker to access all of your accounts instead of limiting the damage to the one that was breached.
Of course, remembering tens or hundreds of passwords is a bit of a Herculean task--especially if you're using strong, complex passwords. My PCWorld peer John Mello suggests a few options for simplifying password management, such as OneID, KeePass, and RoboForm.
The real lesson of the Evernote hack, though, is that passwords don't offer very good protection for your data. Unique passwords that are complex offer better protection than using your dog's name or no password at all, but ultimately all passwords can be cracked or guessed, given enough time and effort.
Moving to multi-factor authentication
With that in mind, Evernote is joining Facebook, Dropbox, Microsoft SkyDrive, PayPal, Gmail, and a growing list of online service providers by adopting two-factor authentication.
Multi-factor authentication provides an extra layer of protection to safeguard your data. Phone-based authentication, for instance, can dramatically boost security. You've probably encountered a prompt for phone-based authentication when you try to log on to a bank's website from a device you don't normally use.
With phone-based authentication, a random or one-time code is sent to a mobile phone, and must be entered in addition to the standard username and password. Some solutions use a mobile app to generate a one-time PIN. Either way, in order for an attacker to access the account they'd have to both crack your password and be in possession of your mobile phone.
There are many other options aside from phone-based authentication, such as access tokens, smartcards and email verification. The exact method varies widely. No matter the implementation, two-factor authentication provides an extra layer of protection, and Evernote should be commended for offering it.

Tuesday, January 29, 2013

New bug makes moot Java's latest anti-exploit defenses, claims researcher


Java's new security settings, designed to block "drive-by" browser attacks, can be bypassed by hackers, a researcher announced Sunday.
The news came in the aftermath of several embarrassing "zero-day" vulnerabilities, and a recent commitment by the head of Java security that his team would fix bugs in the software.
The Java security provisions that can be circumvented were introduced last December with Java 7 Update 10, and let users decide which Java applets are allowed to run within their browsers. The most stringent of the four settings is supposed to block any applet not signed with a valid digital certificate. Other settings freely allow most unsigned applets, execute unsigned applets only if Java itself is up to date, or display a warning before unsigned applets are allowed to run.
But according to Adam Gowdiak, CEO of Security Explorations, none of the settings can stymie an attacker.
"What we found ... is that unsigned Java code can be successfully executed on a target Windows system regardless of the four Java Control Panel settings," Gowdiak wrote in a message posted Sunday to the Bugtraq mailing list.
In an email reply to questions Sunday, Gowdiak said there was a single vulnerability that makes the bypass possible. "It could be used to successfully launch unsigned Java code on a target system regardless of the security level set by the user in Java Control Panel. [The] 'High' or 'Very High' security [setting] does not matter here, the code will still run," he said.
After discovering the vulnerability and creating a proof-of-concept exploit that worked on Java 7 Update 11 -- the version released two weeks ago -- running on Windows 7, Gowdiak reported the bug to Oracle.
His discovery makes moot -- in theory at least -- Oracle's latest security change. When it shipped an emergency update on Jan. 13 to quash two critical Java browser plug-in vulnerabilities, including one that was actively being exploited by cyber criminals, Oracle also automatically reset Java to the "High" security level. At that setting, Java notifies users before they can run unsigned applets.
Although there's no evidence of hackers exploiting the newest vulnerability, Gowdiak hinted that it wouldn't be difficult for them to do so. "It should be considered in terms of a big miss by Oracle," Gowdiak said. "We were truly surprised to find out how trivial it is to bypass these new security settings."
Hackers have stepped up their attacks against Java and its browser plug-in, with some security firms estimating that they account for more than half of all attempted exploits. Most often, Java exploits are used to conduct "drive-by" attacks, or ones that install malware on PCs and Macs after their owners simply browse to compromised or malicious websites.
Gowdiak published his claim just days after Oracle released a recording of a conference call between Milton Smith, the senior principal product manager who oversees Java security, and Java user group leaders, to discuss the recent vulnerabilities and steps Oracle was taking.
During the call, Smith touted the security enhancements to Java 7, including the introduction of the settings in Update 10, and the change of the default from "Medium" to "High" in Update 11. "[They] effectively make it so that unsigned applets won't run without a warning," Smith said of the security settings. "Some of the things we were seeing were silent exploits, where people would click on a link in an email and unwittingly compromise a machine. But now those features really prevent that. Even if Java did have an exploit, it would be very hard to do it silently."
According to Gowdiak, that's exactly what the newest vulnerability could let attackers do. "Recently made security improvements to Java 7 don't prevent silent exploits at all," Gowdiak wrote on Bugtraq.
When asked how users who must run Java in their browser should protect themselves against possible exploits, Gowdiak repeated his earlier suggestion that people turn to a browser with "click-to-play," a feature that forces users to explicitly authorize a plug-in's execution. Both Chrome and Firefox include click-to-play.
"That may help prevent automatic and silent exploitation of known and not-yet-addressed Java plug-in vulnerabilities," Gowdiak said.
Gregg Keizer covers Microsoft, security issues, Apple, Web browsers and general technology breaking news for Computerworld. Follow Gregg on Twitter at @gkeizer, on Google+ or subscribe to Gregg's RSS feed. His email address is gkeizer@computerworld.com.
Read more about malware and vulnerabilities in Computerworld's Malware and Vulnerabilities Topic Center.

FTC Online Privacy Protection Campaign Kicks Into High Gear


As the Federal Trade Commission continues its work in evaluating the privacy practices of businesses in the Internet age, agency staffers are focusing not only on what personal information companies are collecting and how they're using it, but also on the security measures in place to keep that data out of the hands of would-be identity thieves and other bad actors.
Speaking here at an event to mark Data Privacy Day, an annual initiative led by the nonprofit National Cyber Security Alliance, Commissioner Maureen Ohlhausen stressed that the FTC's privacy work is closely coupled with its consideration of industry security practices.
When businesses fail to implement or enforce strong security practices, they run the risk of suffering a major data breach that can expose sensitive information about their customers, severely damaging the firm's brand and inviting an enforcement action from federal authorities, Ohlhausen warns.
"Data is an increasingly vital asset and companies need to protect their ... customers' personal information from theft and unauthorized access that can hurt customers and harm the business's reputation. That's where data security comes in. Data security is part of the broader topic of data privacy," she says. "Regardless of how one feels about the use of consumer data for marketing or targeting purposes, I believe we can all agree that failure to take reasonable precautions to secure data identity thieves and other malicious parties hurts consumers and legitimate businesses alike."
The timing of Ohlhausen's keynote address was apt. Earlier today, the FTC announced that it had reached a settlement with Cbr Systems, the operator of a cord blood bank, concerning allegations of a data breach that may have exposed sensitive information of nearly 300,000 consumers.
The FTC's complaint against Cbr Systems, which stores umbilical cord blood and tissue, dates to December 2010, when unencrypted backup tapes, a laptop and other equipment were stolen from an employee's car, according to the commission. As a result, sensitive health information, credit card and Social Security numbers and other data were compromised, and the laptop and a hard drive that were stolen included passwords and protocols that could have provided access to Cbr Systems' internal network.
The FTC based its complaint on its authority under the section of its charter statute concerning unfair or deceptive practices, maintaining that the company violated its own privacy policy by failing to have in place reasonable policies and procedures for safeguarding its customers' information, and that it courted further risk by carelessly transporting portable storage devices.
Under the settlement agreement, Cbr Systems submitted to 20 years of independent audits of its data-management practices.
The FTC has brought more than three dozen complaints against companies concerning data breaches, Ohlhausen said. Many of those cases had little to do with the technical protections in place to safeguard data, but instead were the product of soft policies, uneven implementation or a weak chain of custody.
"This really seems very simple, but many of the data security cases that the commission has brought involve companies who engaged in careless practices, such as dumping sensitive medical or financial information into open trash bins, and not even shredded," Ohlhausen says.
Over the coming year, the FTC intends to ramp up its scrutiny of data brokers, a sector that the agency has ide ntified as an area of concern for consumer privacy. In December, the FTC sent letters to nine leading brokers asking for detailed information about their data-collection practices, with responses expected next month. At that point, Ohlhausen says, the FTC's in-house economists and other agency staffers will review the information with an eye toward recommendations for reforms within the industry, and potentially legislation authorizing new regulations.
In the meantime, lawmakers could move to pass a bill to establish a nationwide requirement for notifying customers whose information might have been compromised in a data breach. National data-breach notification legislation, long supported by many in the tech sector, would preempt the patchwork of requirements across the 46 states with data- breach laws on the book.
"Although some of the laws are similar, they are not identical. And this means that companies need to comply with separate state notice requirements, and consumer may get notifications that are different and are triggered by different kinds of breaches," Ohlhausen says, adding that she believes there is a good chance that Congress will pass a bill this year. "I believe a single standard would let companies know what to do and let consumers know what to expect."
Ohlhausen also advises business to take steps to limit their risk of a data breach with common-sense measures like incorporating security and privacy protections in the design phase of their products and systems, securing storage, and promoting privacy through education and training programs across business units.
Then, too, they must ensure that they are living up to the security and privacy assurances they make to their customers.
"It's also really critical that businesses honor the promises they make to protect consumer privacy, and this is really at the heart of the commission's law enforcement against deceptive practices," Ohlhausen says. "But because breaches may still occur even in the most security-conscious company, it's also critical to have a plan for responding to data breaches before they happen. So putting together a response plan now may help reduce the impact of a data breach on a business and its customers later."
Kenneth Corbin is a Washington, D.C.-based writer who covers government and regulatory issues for CIO.com.


Friday, January 18, 2013

Pwn2Own hacking contest puts record $560K on the line


HP TippingPoint, the long-time organizer of the annual Pwn2Own hacking contest, has revamped the challenge for the second year running and will offer cash awards exceeding half a million dollars, more than five times the amount paid out last year, the company said yesterday.
The 2013 edition of the contest will offer $560,000 in potential prize money to hackers who demonstrate exploits of previously-unknown vulnerabilities in Chrome, Firefox, Internet Explorer (IE) or Safari, or the Adobe Reader, Adobe Flash or Oracle Java browser plug-ins.
Prizes will be awarded on a sliding schedule, with $100,000 for the first to hack Chrome on Windows 7 or IE10 on Windows 8. From there, payments will fall to $75,000 for IE9 and slide through a number of targets before ending at $20,000 for Java. Prizes will also be given for exploiting Adobe Flash and Adobe Reader ($70,000 each), Safari ($65,000) and Firefox ($60,000).
About the Java award, Kostya Kortchinsky, a researcher who now works for Microsoft, quickly tweeted, "ZDI giving out $20k for free," referring to the Oracle software's recent vulnerabilities.
Pwn2Own will run March 6-8 at the CanSecWest security conference in Vancouver, British Columbia.
According to Brian Gorenc, a researcher with TippingPoint's DVLabs, HP will sponsor this year's Pwn2Own in conjunction with Google. Last year, Google was initially a co-sponsor, but withdrew over disagreements with TippingPoint about that year's rules.
Google then ran its own hacking contest, dubbed Pwnium, at CanSecWest, where it handed out $120,000 to two researchers for exploiting Chrome.
This year's contest is another revamp of the process and rules, the second in two years. The 2012 challenge used a complicated point system that awarded prizes to the researcher or team of researchers who exploited the most targets during a three-day stretch. It also challenged hackers to devise exploits on the spot.
With 2013's Pwn2Own, TippingPoint has essentially dumped last year's model and returned to earlier contest rules: Researchers will draw their order of appearance before the contest begins, each will have 30 minutes to try his or her luck, and the first to exploit a given target wins the prize.
Another change from last year is that researchers must provide TippingPoint with a fully-functional exploit and all the details of the vulnerability used in the attack. That's different from last year, whenGoogle backed out because Pwn2Own did not require hackers to divulge full exploits, or all of the bugs used, so that vendors, including Google, could then fix the flaws.
The rule changes and the large infusion of cash hint that Google returned to Pwn2Own sponsorship only after it convinced TippingPoint to revise the exploit disclosure policy. Yesterday, Google declined to comment on whether it would again run a Pwnium contest at CanSecWest, but did confirm it will host its Chrome-specific challenge at some point in 2013.
But it was the cash that caught researchers' attention.
The $100,000 prize for an exploit of Chrome or IE10, for example, was 67% more than Google paid last year in its inaugural Pwnium contest, and over six times the maximum paid at Pwn2Own in 2011 for hacking a desktop browser.
The always-quotable Charlie Miller, who won prizes at Pwn2Own four years in a row -- the only "four-peat" in the contest's history -- bemoaned the high awards.
"I have to say the Pwn2Own prize money is serious," Miller said on Twitter yesterday. "I feel like a 1950's pro athlete wondering why current athletes are paid so much."
Miller, who won at Pwn2Own while a security consultant, now works for Twitter.
Others took up Miller's line of thought, with Larry Seltzer, a long-time security reporter and now the editorial director of Byte, chiming in with, "They're all using exploit-enhancing drugs these days."
TippingPoint has published the 2103 Pwn2Own rules on its website, and will provide updates during the contest via a dedicated Twitter account.
Gregg Keizer covers Microsoft, security issues, Apple, Web browsers and general technology breaking news for Computerworld. Follow Gregg on Twitter at @gkeizer, on Google+ or subscribe to Gregg's RSS feed. His email address is gkeizer@computerworld.com.
Read more about cybercrime and hacking in Computerworld's Cybercrime and Hacking Topic Center.

Saturday, December 29, 2012

Security in 2013: The rise of mobile malware and fall of hacktivism


It's always fun this time of year to look back on the year we're about to leave behind, and to make some predictions about what we might expect the year ahead to look like. We already have a comprehensive look back at the biggest security stories of 2012, and now here's a look at some trends to watch for 2013.
These prognostications come from McAfee--the respected information security and antimalware company, not its international man of mystery founder. McAfee is one of the companies with a global presence and the skills and knowledge required to distill evolving trends and determine where things are headed.
For starters, McAfee believes that mobile malware will both evolve in technique and grow in volumeby leaps and bounds in 2013. The explosion of mobile devices--iPhones, iPads, Android smartphones and tablets, etc.--have driven malware developers to expand their horizons and focus on new, fertile ground.
The attackers seem to have an advantage here. Malware developers can apply the lessons learned from years of evolving exploits against PCs to target the weaknesses of mobile devices, but users seem to be oblivious to the threat and feel like mobile devices are somehow just inherently secure.
One thing to watch for in the year ahead is an increase in ransomware attacks. These attacks basically lock you out of your smartphone or tablet unless you agree to pay the ransom. A McAfee spokesperson explained, "The harsh reality of these schemes is that users have no way of knowing if their device will be unlocked even if they do meet the perpetrator's demands," adding, "Since attackers hijack the users' ability to access data, victims will be faced with either losing their data or paying a ransom in the hope of regaining access."
McAfee predicts that the concept of a Trojan app will evolve to a more pernicious threat when attackers merge it with a mobile worm so it can spread itself without relying on users to install the app. McAfee also warns that malware developers will create mobile worms capable of spreading through NFC-enabled devices to spread wirelessly and steal money from "digital wallets".
On the other side of the coin, McAfee suggests that Anonymous will fade away in the next year. McAfee doesn't expect hacktivism itself--or even Anonymous--to disappear completely, but some of the more malicious attacks attributed to hacktivist groups have fallen short of the "Robin Hood" aura that initially made Anonymous popular. Potential victims have also learned more about hacktivist tactics and are more prepared to defend against such attacks.
These aren't the only trends in security for 2013, though. McAfee says the trend of cyber attacks as a political or military tool will become more sophisticated, and suspicions of state-sponsored cyber attacks will continue to grow, and warns that crimeware and hacking-as-a-service business models will become more mature.
For more about what to expect in 2013 according to McAfee, check out the 2013 Threat Predictions report from McAfee Labs.

Saturday, December 1, 2012

Cisco's internal security team fights to corral BYOD, malware and Wild West environment


Many organizations have a computer security incident response team (CSIRT) that swoops into action to battle malware outbreaks, other types of cyberattacks and possible insider threats, and at networking giant Cisco, that CSIRT team is made up of about 60 people trying to protect a business with about 75,000 employees.
"We're tasked with monitoring for and investigating policy violations against Cisco," says Matthew Valites, Cisco's CSIRT manager for information security investigations. That means protecting corporate IT assets used directly by employees or the business for processing purposes so that sensitive information isn't compromised. However, since Cisco has embraced a "bring your own device" (BYOD) strategy, policy enforcement matters for Cisco's CSIRT have become more complicated.
"With user-owned devices, enforcement has become an issue," acknowledges Valites, in the course of discussing some of Cisco's security incident response practices. "BYOD is a real problem." In what's regarded as a cost-saving move, Cisco typically doesn't supply smartphones to any employee anymore, expecting them to use their own, unless their job falls under government regulatory restrictions where it's plainly spelled out an employee must be using a corporate-issued device. "This is a really big problem for my team," acknowledges Valites.
Above and beyond the BYOD conundrum, the Cisco CSIRT group each day faces the prospect of stopping desktop malware outbreaks, monitoring for unauthorized traffic on the network and guarding against stealthy online attacks from attackers going after key assets. There's also the inevitable spate of things like faulty log-ins but CSIRT's hard job is trying to ascertain unauthorized access.
This all has to be done within the framework for regulatory compliance. "We have a healthcare center in San Jose on premises with nurses and doctors," points out Valites, saying making healthcare professionals available on site is seen as a benefit for employees. And this means that security and privacy policies related to any data associated with it must adhere to federal HIPAA rules, he notes.
Valites says high-level executives at Cisco, not surprisingly, get special attention in terms of whatever computer or network they use since these executives are recognized as being valuable targets for cyber-espionage and the like. In comparison to other employees, "we pay more attention to their assets," says Valites.
And then there are whole groups at Cisco, such as an entire lab, that are known to all too frequently be getting into trouble, breaking with usage policies and their computers erupting with malware. "The labs are a little like the Wild West," acknowledges Valites. With repeat offenders there, Cisco CSIRT has no choice but to clamp down with additional controls, such a blackholing an entire lab on the network so they can't get online or shutting off network segments so they're restricted to an internal LAN.
But the main day-to-day challenge is in getting visibility into security events of any type and quickly deciding when and how to escalate the response. Cisco designed its own incident-response tracking system, where trouble of any type is recorded and pushed toward closure.
When an incident arises, the first task is to associate the computer device in question with its specific owner, says Valites. "We need the asset owners to provide that information to us," and in a large organization of global scope, that can be a challenge. Although lots of technical tools for antivirus, VPN, Web application control, intrusion detection and the like are in use, in the end much often rides on communication between people to share information accurately and quickly.
The CSIRT division also has to be mindful that there's the potential for an insider threat as there would be in any organization. That's the rogue employee or contractor with access to the network willing to steal data or do other damage. It's a prickly situation where escalation would mean reaching out to human resources and legal.
"We have good partnerships there," says Valites, noting that at Cisco, the legal counsel has made it clear about their role in incident response investigation and they want to be involved in the potential investigations into things such as leaks of sensitive information. Investigations of all sorts could require computer forensics, and Cisco's CSIRT is equipped to do that.
As Cisco is a global company, there is the need to coordinate the CSIRT across time zones and continents spanning North America to the Asia-Pacific region. "It's a follow-the-sun model," says Valites, adding that Cisco would benefit from physical security operations centers (SOCs). He says Cisco is now undertaking to construct two such SOCs -- one in San Jose, Calif., and the other in India -- that will make use of technologies of many types, including Cisco's own dedicated TelePresence systems for collaboration.
Ellen Messmer is senior editor at Network World, an IDG publication and website, where she covers news and technology trends related to information security. Twitter: @MessmerE. Email: emessmer@nww.com.
Read more about wide area network in Network World's Wide Area Network section.

Wednesday, October 3, 2012

Sandia builds massive Android network to study security, more


Government scientists have built a network of hundreds of thousands of simulated Android mobile devices that could be used for building better security on the most popular mobile devices.
By early spring 2013, the Sandia National Laboratories in California plans to make software tools available to private and government organizations that want to build their own environment for studying the behaviors of smartphone networks.
Sandia scientists have built a network of as many as 300,000 virtual handheld computing devices, but say the technology can scale up to run on supercomputer-class machines, or scale down to a workstation.
What the researchers have done is link together instances of generic Android, each running on a separate virtual machine. The network, which runs on racks of off-the-shelf, x86 desktops, can be built up into a realistic computing environment that includes a full domain name service (DNS), an Internet relay chat (IRC) server, a web server and multiple subnets.
A key component of MegaDroid is an imitation Global Positioning System (GPS) that includes simulated data of a smartphone user in an urban environment. Since Wi-Fi and Bluetooth capabilities depend on GPS data, the feature is important for studying how the two communication features could be used by cybercriminals to steal data.
Researchers also could run malware on any of the simulated devices to see how it would behave within the network.
"If you have something you're capable of running on an Android device, be it malware, an application or whatever, this platform could test it for you," Keith Vanderveen, manager of Sandia's Scalable and Secure Systems Research department, said.
Android is the favorite mobile platform of cybercriminals. Reasons include the platform's large user base and the fact that any organization can set up an app market. In August, Android accounted for almost 53% of the smartphone market, comScore said.
Besides malware, Megadroid has a much broader use. Because it can scale to the size of real-life cellular networks, it is expected to be valuable in finding ways to limit damage from network disruptions due to glitches in software or protocols, natural disasters or acts of terrorism.
In addition, the platform would be useful in studying methods for preventing unauthorized data from leaving a device, a major concern for corporations and the departments of Defense and Homeland Security.
MegaDroid will be released as an open-source project, so other researchers can modify the technology to fit their needs. While Android was chosen for the initial platform, the technology could be used in testing Apple's iOS devices.
"The platform is really designed to be flexible," David Fritz, a Sandia researcher, said.
MegaDroid is an offshoot of simulation platforms built for studying large-scale networks of Windows and Linux computers. Over the last three years, Sandia has spent a total of $3.5 million on the various projects.
The laboratory is open to working with academia and private industry on the MegaDroid project. In the 1990s, Sandia helped advise the President's Commission of Critical Infrastructure Protection, which led to its current focus on network security.Ã'Â
Read more about wireless/mobile security in CSOonline's Wireless/Mobile Security section.

Friday, September 28, 2012

You might be surprised at which Web browser aced this security test


Microsoft IE9 blew away Google Chrome, Mozilla Firefox and Apple Safari in new tests by NSS Labs to measure the ability of web browsers to block malware and catch click fraud.
NSS subjected Apple Safari 5, Google Chrome 15-19, Microsoft Internet Explorer 9 and Mozilla Firefox 7-13 to over 3 million test runs against over 84,000 URLs determined to be active and malicious out of a unique sample set of 227,841. Out of 750,000 test cases per browser, NSS labs found in its 75-day reviewthat IE9's malware block rate was 95%, whle Firefox and Safari trailed far behind at 6% apiece and Chrome was somewhere in the middle, with its rate varying from 13% to 74%.
(IE and Chrome are neck and neck, by the way, for the global lead in Web browser share, followed by Firefox and Safari.)
The NSS "Is your Browser Putting You at Risk?" report conclude that users should "evaluate browser security as part of their layered security strategy."
NSS Labs says browser protection entails an "'in-the-cloud' reputation-based system that scours the Internet for malicious websites and categorizes content accordingly, either by adding it to a black or white list, or assigning a score (depending on the vendor's approach)." When a browser detects a site as "bad," it will re-direct the user to a warning message or page informing them that the URL is malicious. Sometimes the browser will instruct the user that content is malicious or should be cancelled, the report says.
NSS Labs also says its testing determined that the SafeBrowsing API 2.0, which provides reputation services for executable files (otherwise called "malicious downloads"), has been integrated into Chrome but not Firefox or Safari.
NSS Labs also tested the browsers for protection against click fraud, a crime that abuses pay-per-click advertising through use of malware infection. Click fraud "causes minimal direct harm to the typical end user, as the ultimate target is the ad buyer," the NSS Labs report points out.
According to the testing, the click-fraud catch rates are as follows: IE9 at 96.6%, followed by Chrome at 1.6%, Firefox at 0.8% and Safari at 0.7%. The lab notes the average lifespan of a click fraud URL was 32 hours with over 50% expiring within 54 hours.
NSS Labs recommends that ad buyers "put pressure on Google to increase the click fraud protection capabilities of Chrome and the SafeBrowsing API."
NSS Labs is making its browser-security test reports available for free here and here.
Ellen Messmer is senior editor at Network World, an IDG publication and website, where she covers news and technology trends related to information security. Twitter: MessmerE. E-mail: emessmer@nww.com.
Read more about wide area network in Network World's Wide Area Network section.

Hacktivists strike U.S. Bank with volunteer-powered DDoS


U.S. Bank's website was disrupted on Wednesday in a people-powered distributed denial of service (DDoS) attack, launched by a group of Islamic hacktivists who have claimed responsibility for similar cyberattacks against four other banks in the U.S.
The attack involved hundreds of thousands of computers sending an overwhelming number of requests that downed the site for roughly an hour, starting at around 3:30 Pacific, said Atif Mushtaq, a security researcher at FireEye who has been monitoring the attack.
The disruption of U.S. Bank's website comes one day after a similar attack against Wells Fargo & Co. The group has taken credit for other attacks that occurred last week, against Bank of America, JPMorgan Chase and Citigroup.
A representative of U.S. Bancorp, which operates as U.S. Bank, confirmed it was under attack. "We apologize that some customers experienced intermittent delays today on our website. We have been working hard to restore full connectivity," the spokeswoman said.
"We are asking customers who are experiencing issues with our online or mobile sites and have an urgent banking need to please call us at 1-800-US-BANKS, or stop by one of our branches," she said.
She said the issues were "related to unusual and coordinated high traffic volume designed to slow down the system -- similar to what other banks have experienced in the past week."
"We are working closely with federal law enforcement officials to address the issue. In the meantime, we can assure customers that their data and funds are secure," the spokeswoman added.
A new twist on DDoS
Rather than launch the attack from a network of compromised machines, called a botnet, the attackers are apparently using volunteers, Mushtaq said. Participants go to either one of two file-sharing sites and download a program written in a scripting language. Once the program is running, a person only has to click on a "start attack" button to send continuous requests to the target's website.
This method makes it more difficult for authorities to stop the attack, because there are no control servers. "They know [servers] can be blocked very easily," Mushtaq said.
The group calling itself "Mrt. Izz ad-Din al-Qassam Cyber Fighters" had said on a Pastebin post that it would attack Wells Fargo on Tuesday, U.S. Bank on Wednesday and PNC Financial Services Group on Thursday.
DDoS attacks, which experts say banks can only hope for best with, are considered crude because they do not require any sophisticated technology, just a large enough network of computers to overwhelm a site. Banks the size of the ones under attack would have to be hit by a network of hundreds of thousands of computers in order to disrupt their sites, Mushtaq said.
The use of volunteers launching attacks from their own computers makes it difficult for banks to separate traffic and redirect the DoS requests, Mushtaq said. "There's no way you can distinguish between the benign traffic and this DDoS traffic," Mushtaq said. "It's simply mixed up."
The group claiming to be behind the attacks indicated in the Pastebin post that it was in retaliation for thevideo trailer denigrating the Prophet Muhammad. The amateurish YouTube video made in the U.S. has sparked violent protests in the Middle East and other regions.

Thursday, September 27, 2012

Sophos rolls out mobile apps for anti-malware and encryption


Sophos today announced a free anti-virus defense app for Google Android devices which works via the Sophos cloud-based service. Separately Sophos also made available Android and Apple iOS apps for securing files held in the cloud file-storage service Dropbox.
The Sophos anti-virus app for Android devices will be available for free through November, at which time it will be integrated into The Sophos mobile-device management software Sophos Mobile Control. However, a standalone free version is still expected to be made available after November. "It acts as a scanner and looks for malicious apps and malware," says Payal Mehrotra, Sophos mobile product manager.
With regard to the encryption for mobile, the Sophos Mobile Encryption apps for Android or iOS devices are intended to work in conjunction with another product from Sophos called SafeGuard Enterprise, which can be used to ensure files stored in cloud-based services are encrypted.
The first version of the Sophos Mobile Encryption apps allows user access to encrypted files stored in the Dropbox cloud-based file storage. Future versions will be available for other services, including Egnyte and Google Drive, says Mehrotra.
There are free and paid versions of Sophos Mobile Encryption, with the paid versions (ranging from $2.99 for Apple iOS to $9.99 for the Android version) being more feature-rich to allow offline access to encrypted files when connectivity isn't available, says Mehrotra.
The Google apps are available here from Google Play, and the Sophos Mobile Encryption app is availablehere for Apple iOS devices.
Ellen Messmer is senior editor at Network World, an IDG publication and website, where she covers news and technology trends related to information security. Twitter: MessmerE. E-mail: emessmer@nww.com.
Read more about wide area network in Network World's Wide Area Network section.

USSD attack not limited to Samsung Android devices, can also kill SIM cards


The story, "USSD attack not limited to Samsung Android devices, can also kill SIM cards," posted on the wire on Thursday, wrongly stated the dates at paragraphs 12 and 16.
The story has been corrected on the wire and the paragraphs now read:
Paragraph 12
This is possible because of a MMI code that allows changing a SIM card's PIN (Personal Identity Number) number using the PUK (Personal Unblocking Key), Collin Mulliner, a mobile security researcher who works in the SECLAB at Northeastern University in Boston, said Wednesday via email.(
Paragraph 16
Samsung has already fixed the USSD/MMI code execution issue for Galaxy S III devices. "We would like to assure customers that the recent security issue affecting the Galaxy S III has already been addressed in a software update," Samsung said Wednesday in a statement via email. "We believe this issue was isolated to early production devices, and devices currently available are not affected by this issue. To ensure customers are fully protected, Samsung advises checking for software updates through the 'Settings: About device: Software update' menu. We are in the process of evaluating other Galaxy models."