STE WILLIAMS

Live chat One month ago a rookie IT mistake crippled the banking network of the RBS Group.

Sixteen million customers – individuals and companies – of RBS, NatWest and the Bank of Ulster were locked out of accounts, unable to withdraw cash or pay into accounts for days.

The cause? An “inexperienced operative” pressing the wrong button on a rather routine CA-7 batch processing job.

And although RBS denied it, The Reg uncovered that the person responsible worked offshore – in one of many positions sent abroad as part of wide-ranging outsourcing to cut costs and so-called efficiencies.

Join Reg reporter Anna Leach, who nailed the story, city IT recruitment expert Dominic Connor and fellow Reg readers for a one-hour, post-work, interactive Live Chat on the IT meltdown and for and update one month on. We tackle:

• What has RBS done to fix the problem?
• Could this hit other banks?
• What was the role of Computer Associates and who is really to blame?
• Will/should heads roll?
• What does this mean for your job?
• Does this spell the end of cheapo outsourcing deals?

Join the discussion below at 19.00 BST (18.00 GMT) on 25 July for QA and convo. You can register before the Live Chat for free and receive an email reminder before we go live.

®

Article source: http://go.theregister.com/feed/www.theregister.co.uk/2012/07/20/rbs_meltdown_live_chat/

Analysis Iranian state media has angrily rejected suggestions that the Madi cyber espionage campaign is anything to do with the Islamic Republic.

Madi had claimed more than 800 victims located in Iran, Israel, Afghanistan and elsewhere, according to the results of an eight month investigation into the cyber-espionage tool by Kaspersky Lab and Seculert, first publicised on Tuesday.

Where previous high profile attacks such as Flame, Duqu, and Stuxnet utilised zero-day vulnerabilities or forged digital certificates, Madi relies on basic well-established engineering techniques to infect targeted computers. The attackers send spear-phishing emails to the targeted entities with different attachments (PowerPoint documents, fake Word documents, fake images, etc) which have the malware embedded.

Attacks came under the guise of a video of a missile test, PowerPoint files featuring serene wilderness images and a copy of a Daily Beast article discussing electronic warfare between Israel and Iran, among others.

Researchers at Kaspersky Labs described the techniques used by the malware as amateurish and rudimentary, an assessment shared by Aviv Raff, CTO at Seculert.

“The malware itself is written is either an amateurish, or written in a rush in order to start the campaign as early as possible,” Raff told El Reg. “It looks like native Persian speakers” created the malware, he added.

Iranian state media outlet reacted with indignation to perceived suggestions that either its spies or its ordinary citizens were only capable of producing lame malware, such as Madi.

“If this was a product of Iran it would be professional and at least as advanced as Stuxnet and Flame,” an English language editorial carried by the semi-official FARS news agency said.

The story bristled at the perceived suggestion that Madi was developed in response to the Flame and Stuxnet malware attacks against Iran’s controversial nuclear weapon’s programme.

Madi (AKA Mahdi), named after files used in the malware, references the moniker of the Muslim messiah expected to cleanse the world of wrongdoing and bestow peace and justice before Judgment Day.

Whatever the significance or otherwise of this name it’s generally agreed that the malware is primarily designed to steal information from compromised machines. Almost three quarters (72 per cent) of Madi’s cyber espionage targets were based in Israel, according to an analysis by Symantec.

Targets included oil companies, US-based think tanks, a foreign consulate, as well as various governmental agencies some of which were in the energy sector. Isolated infections have occurred in the US and New Zealand, but most have been clustered in the Middle East according to Symantec.

Kaspersky and Seculert, by contrast, argue that the highest concentration of Madi’s confirmed 800 victims is in Iran. The malware has claimed 387 victims in Iran and 54 in Israel, the duo report.

The trojan has been communicating with command-and-control servers hosted in Canada, Iran and, more recently, Azerbaijan. Madi allows remote hackers to steal sensitive files from infected Windows computers, monitor email and instant messages exchanges, record audio, log keystrokes, and take screenshots of victims’ activities. Gigabytes of data were stolen over the last eight months.

Madi has emerged as the likely pathogen behind the malware-based attacks against Bank Hapoalim in Israel in Feb 2012.

Security experts are split over the likelihood or otherwise that a nation-state might be behind the attack.

“Targets like Iran, Israel, and Saudi Arabia might suggest involvement of a nation state, however our research has not found evidence that this is the case. Instead, the current research indicates these attacks are being conducted by an unknown Farsi-speaking hacker with a broad agenda,” Symantec concludes.

However Seculert’s Raff is more open to the state-backed possibility, while dismissing any possible suggestion that Madi might be the work of regular cybercrooks.

“It’s unclear if it is state-sponsored or not. But it does look like a campaign which requires large investment or financial backing,” Raff told El Reg. “This malware also records sound and steals 27 different file types. So, I find it unlikely that the motive is cybercrime.”

“It is definitely suited for surveillance against the targeted entities,” he concluded. ®

Article source: http://go.theregister.com/feed/www.theregister.co.uk/2012/07/20/madi_cyberspy_analysis/

Cambridgeshire police are reviewing a potential contract with troubled Olympic security provider G4S. The deal on the table would see G4S take over back-office functions inclusing ICT for three forces – Cambridgeshire, Hertfordshire and Bedfordshire – and would result in 1,191 staff being rehired by G4S, according to the Cambridge News.

The police chiefs want to cut £73 million by outsourcing and have been talking to G4S since June – but the Olympic screw-up has forced a rethink:

The Chief Constables of Bedfordshire, Cambridgeshire and Hertfordshire police forces have met and discussed at considerable length the current situation regarding the work to develop a full business case for the outsourcing of organisational support services through the Lincolnshire Police contract with G4S.

The ten year contract G4S already have with the Lincolnshire police service, which started in April this year, would form the basis for the new deal: under that scheme everyone except actual police officers is employed by G4S rather than the police service itself. That includes staff working behind counters, in custody centres and suspect identification, in HR, IT and in the criminal justice unit.

Since the Lincolnshire contract took effect in April, there have been several controversies: including one when G4S put its own logo next to the Police one on new uniforms for staff and also over threatened redundancies.

The Surrey police have already pulled out of a deal with the company.

Commenting on the use of G4S in police forces earlier this month, public sector union Unison said:

The problems that G4S has had delivering on the Olympics contract exposes one of the many pitfalls of hiving off services to the private sector. The government may claim that risk is transferred, but in reality it is not. When things go wrong, the public sector has to come to the rescue and the taxpayer ends up paying twice.

We understand that the part of the G4S that runs police services – G4S Care and Justice Services – is separate to the events wing involved with the Olympics – G4S Secure Solutions.

G4S have not replied to our request for comment. ®

Article source: http://go.theregister.com/feed/www.theregister.co.uk/2012/07/20/police_g4s_contract/

Android Jelly Bean 4.1 promises to be more secure than previous versions of the Google’s mobile OS.

The big news is that the software now properly implements Address Space Layout Randomization (ASLR), a technique designed to make malware-based attacks more difficult. The latest Jelly Bean iteration was released to select devices last week but is not due to come bundled with mainstream Android smartphones and tablets much before the end of the year.

What is ASLR?

ASLR shuffles the pack and position of libraries, heap, and stack, in a process’s address space. That means even if a hacker finds a vulnerability in a device, they will have a far greater problem finding the location of infected shellcode in a stack, for example, making it much harder to craft a working exploit. The technology is often combined with other memory protection techniques, such as non-executable memory protection.

ASLR has been a mainstay of security defences in desktop machines for years, appearing in Windows Vista and Mac OS X since 2007, for example. More recently it has featured in iOS 4.3, released in March 2011, marking its migration onto smartphones, and more recently in Android 4.0.

The initial debut of the tech on Android – in Ice Cream Sandwich – was somewhat half-cocked and the Android 4.1 Jelly Bean implements the technology properly for the first time, according to smartphone security experts at Duo Security.

Why it didn’t quite work in Ice Cream Sandwich

Jon Oberheide, the security firm’s CTO, praised the implementation of the tech in Jelly Bean as fixing the deficiencies of ASLR’s implementation in Ice Cream Sandwich.

“The ASLR support in Android 4.0 did not live up to expectations and is largely ineffective for mitigating real-world attacks, due to the lack of randomization of the executable and linker memory regions,” he wrote in a blog post at the time the technology was introduced in February 2012.

ASLR is only effective if everything is randomised. If portions of memory are consistently left alone by the process then hackers can still gain a foothold by sticking to these areas. In a blog post on Monday, Oberheide said Google has now applied randomisation (properly) across the board.

But Oberheide qualified his praise – noting there was plenty of room for further improvement. For example, Google could push ASLR down to the Android kernel, allowing it to catch up with Apple.

While Android is still playing a bit of catch-up, other mobile platforms are moving ahead with more innovative exploit mitigation techniques, such as the in-kernel ASLR present in Apple’s iOS 6. One could claim that iOS is being proactive with such techniques, but in reality, they’re simply being reactive to the type of exploits that typically target the iOS platform.

An overview from Google of how exploit mitigation has evolved on the Android platform can be found here. ®

Article source: http://go.theregister.com/feed/www.theregister.co.uk/2012/07/20/android_jelly_bean_security_revamp/

The average Brit maintains 26 online accounts but only uses five different passwords to keep them secure.

A poll of 2,000 by Experian found that one in four people uses a single password for the majority of profiles, and one in 25 stick with the same one for ALL their accounts. In addition to chronic password reuse, failing to close unused accounts poses an addition privacy risk.

Two-thirds of those surveyed (66 per cent) copped to having defunct profiles which hold valuable personal and financial information – including social network profiles (26 per cent), email addresses (18 per cent) and shopping accounts (21 per cent).

Leaked personal information opens the doors to identity fraud, a crime suffered by an estimated two million UK citizens. About 90 per cent of the 12 million pieces of personal information illegally traded online globally between January and April this year involved password and login combinations, according to figures from Experian’s identity web monitoring service.

Victims of identity fraud can have their credit histories thrashed while suffering the consequences of false loans run up in their name. Many victims remain ignorant that their ID has been stolen until they are refused loans, cards or mobe contracts, according to Experian.

Peter Turner, MD at Experian Consumer Services in the UK and Ireland, commented: “If password information is stolen from a website, all accounts using the same details will be compromised, and this information can spread among fraudsters rapidly.”

Leaky logins in the news

Leaks of password data by one or another high-profile website have become everyday news over recent weeks. Last week alone news that Yahoo! Voices had spilled more than 450,000 login credentials was followed quickly by warnings about the loss of secured IDs from Nvidia’s developer and user forums after password hashes were swiped. The Yahoo! Voices hack was the only one of the trio that directly exposed users’ passwords, but even the less-serious Nvidia hack prompted the graphics chip firm to suspend its forums.

Other recent victims of consumer password security breaches have included Last.fm, eHarmony and LinkedIn.

Around 20 per cent of the passwords found on lists of compromised login credentials match those from Microsoft Accounts because of password reuse, according to a blog post by Microsoft on the issue.

Eric Doer, group programme manager at Microsoft, said: “These attacks… highlight the longstanding security advice to use unique passwords, as criminals have become increasingly sophisticated about taking a list of usernames and passwords from one service and then ‘replaying’ that list against other major account systems.”

Experian argues that better password choices combined with the use of monitoring services (such as the one it supplies, of course), can address the problems passed by account hijacking. However other security experts argue that the survey illustrates the growing problems with using passwords as a security defence.

Carl Leonard, senior manager at Websense Security Labs, commented: “Passwords are simply not enough to protect vital data – they’re as strong as a simple lock against professional thieves. Passwords can be guessed, cracked or stolen through social engineering.

“Worse still businesses can be attacked and stories of breached password databases make for uneasy reading. Businesses need to think carefully how they secure password information for which they are responsible – encrypting password records and securing the database makes good sense.” ®

Article source: http://go.theregister.com/feed/www.theregister.co.uk/2012/07/20/password_reuse_survey/

What if you could use a password with 38 bits of entropy without memorizing it? Stanford University researchers think they’ve found a way to deliver.

Their argument is that attackers can steal passwords from ill-defended servers, install keyloggers in drive-by attacks, or force people to hand over their security tokens. The aim is to defeat the “rubber hose” attack (ie, beating the information out of some unfortunate insider) – since you can’t divulge what you don’t know.

In a paper to be presented at the August Usenix conference in Washington, Stanford University scientists led by Hristo Bojinov designed a game to deliver a password that the user can remember without knowing it.

The researchers apply a neuroscience concept called implicit learning to the question of cryptography. Activities such as bike-riding or guitar-playing are learned by repetition: people know they’re learning, but they’re not conscious of the processes by which they learn.

The same applies, the researchers say, to repetitive game-playing. While complex games pose explicitly intellectual challenges, there are plenty of games in which the only way to get better is to keep trying.

To turn this into a crypto technique, the researchers created a game into which they planted patterns – what the paper refers to as Serial Interception Sequence Learning. The game itself is simple: you use keystrokes to intercept falling blobs before they reach a sink at the bottom of the screen (see image).

Beat the game to get your password. Source: Hristo Bojinov, bojinov.org

However, unlike a “real” game – in which the column in which the blobs appear would be randomized – the researchers embedded a pattern into the game which represented passwords, but couldn’t be memorized. “All of the sequences presented to the user are designed to prevent conspicuous, easy to remember patterns from emerging,” the paper states. The sequences “are designed to contain every ordered pair of characters exactly once, with no character appearing twice in a row”.

What they found was that users could be trained to “beat” the game (and thus capture the password) in between 30 and 45 minutes without ever knowing the password they used – and that even after an interval of two weeks, people would remember how they played.

Of course, the system is still vulnerable to a hack: if attackers gained access to the authentication system itself, all bets are off. And a 30-minute learn time to achieve a password is impractical in the real world. However, the researchers told New Scientist that “If the time required for training and authentication can be reduced, then some of the benefits of biometrics, namely effortlessness and minimal risk of loss, can be coupled with a feature that biometrics lack: the ability to replace a biometric that has been compromised.” ®

Article source: http://go.theregister.com/feed/www.theregister.co.uk/2012/07/19/neuroscience_as_password_protector/

Mozilla has plugged a privacy-related security hole in Firefox 13 and released a fixed version of its web browser. The flaw allowed the software’s speed-dial-alike “new tab” feature to take snapshots of supposedly secure HTTPS sessions.

Punters sounded the alarm over the feature that, for example, revealed online bank account details or private messages in webmail sessions to the next user of a shared PC. Mozilla quickly acknowledged the behaviour was undesirable and issued a workaround and privacy advice in early June.

The browser maker bundled in a more comprehensive fix with Firefox 14, which stops the caching of content from sensitive websites, as a statement by Mozilla explained:

The new tab feature that displays thumbnails of your favorite and recently visited sites in Firefox now omits privacy-sensitive websites like banking or webmail sites. The new tab thumbnails are based on your browsing history and you can easily control the experience by moving or deleting the thumbnails.

Firefox 14, released on 17 July, automatically encrypts web searches through Google, while leaving the back porch accessible to advertisers, as explained in our earlier story here.

Firefox 14 also changes the way the globe icon (to the left of the URL in the address bar) works. The icon remains a globe when the browser accesses a site that is unencrypted, but becomes a a grey padlock icon if a site uses SSL encryption. If a site is secure with the added benefit of an EV (extended validation) certificate, the browser flashes up a green padlock icon and includes the name of the site’s owner. These changes are also designed to make spoofing harder, as Mozilla explains here.

Firefox 14 also introduces other features and performance tweaks (listed here) as well as fixing various security bugs, five of which Mozilla lists as critical.

The “new tab” page introduced alongside Firefox 13 is comparable to the Speed Dial feature already present in other browsers displaying cached copies of a user’s most visited websites. ®

Article source: http://go.theregister.com/feed/www.theregister.co.uk/2012/07/19/firefox_14_new_tab_fix/

A seventh computer-hacking suspect was arrested this morning by detectives who are also probing separate allegations of voicemail interception at Rupert Murdoch’s British newspapers.

Scotland Yard is yet to put out an official statement, but officers confirmed to The Register that an unnamed man, whose age and alleged offences were also kept under wraps by the Met, was cuffed at his home in north London today.

“For operational reasons we are not providing further details of this arrest at this time but will release more information in due course,” a spokesman at the Yard added.

The arrest is the seventh of its kind under Operation Tuleta – which is probing allegations of criminal breaches of privacy.

El Reg asked Murdoch’s News International to comment on this story, but at time of writing it hadn’t got back to us. ®

Article source: http://go.theregister.com/feed/www.theregister.co.uk/2012/07/19/op_tuleta_arrest/

Security researchers have dealt a knockout blow to Grum, one of the most prolific spam-distribution botnets.

Command-and-control servers in the Netherlands were taken out on Monday, but that still left zombie control nodes in Russia and Panama up and running. According to security researchers, pressure was applied on a Panamanian ISP hosting a botnet-linked server to clean up its act or risk losing upstream connectivity. The tactic worked by Tuesday, apparently, but that still left the Russian motherlode as well as secondary command servers hosted in Ukraine, a country that’s been something of a safe haven for cybercrime in the past.

According to researchers, after some lobbying, the plug was pulled on the Ukrainian hosted servers. Meanwhile, action by an upstream provider null-routed the Russian-hosted node, despite a reported unwillingness to heed complaints by GazInvestProekt, the local ISP.

“All the known command and control (CnC) servers are dead, leaving their zombies orphaned,” Atif Mushtaq, a researcher at network security and malware intelligence firm FireEye, announced on Wednesday. FireEye worked with other security researchers at Spamhaus, the Russian Computer Security Incident Response Team and elsewhere on the takedown operation.

Grum was the world’s third-biggest botnet and responsible for 18 per cent of global junk mail around the time of its takedown, or 18 billion spam messages a day. The zombie network has been around for around five years and most often associated with rogue pharmacy and fake Rolex spam. Estimates vary but the number of infected drones on its network may number 800,000 or more. The stream of crud is rapidly drying up, according to FireEye.

“According to data coming from Spamhaus, on average, they used to see around 120,000 Grum IP addresses sending spam each day, but after the takedown, this number has reduced to 21,505,” Mushtaq writes. “I hope that once the spam templates expire, the rest of the spam will fade away as well.”

The Grum takedown operation follows similar exercises against other junkmail distribution networks such as Srizbi, Rustock, Ozdok and Cutwail. The latest case is noteworthy because it showed that even ISPs within Russia and the Ukraine can be pressured to end their cooperation with bot herders. “There are no longer any safe havens,” Mushtaq concluded. ®

Article source: http://go.theregister.com/feed/www.theregister.co.uk/2012/07/19/grum_botnet_takedown/

The Chinese government has released sweeping new information security guidelines designed to enable public and private bodies to protect themselves more effectively against new cyber threats.

The State Council’s long list of recommendations spans just about every conceivable aspect of information security, painting a picture of a nation under siege from attackers and increasingly vulnerable thanks to its reliance on the internet.

It points to the need to better secure “energy, transport, finance and other fields of the national economy” as well as government departments.

On the government side, the guidelines include more auditing, security reporting and monitoring and a pledge to “reduce the number of internet connection points” – presumably to isolate highly classified data on specific machines.

The government also acknowledged the risk to industrial control systems, pledging to “strengthen the protection” of nuclear facilities, aerospace, advanced manufacturing, oil and gas pipelines, power systems, and more.

China also wants to “improve the information security certification and accreditation system”, step-up password protection in e-commerce and e-government, promote the use of “e-signatures” in banking and e-commerce and use strong encryption to protect classified information systems.

In addition, the plans include working towards better information sharing and exchange on cyber security matters, improving emergency response teams, and strengthening and promoting the ranks of information security professionals in the country.

Although short on any detail of exactly how all of this is going to be achieved, as a statement of intent it’s pretty comprehensive and with significant financial and human resources to hand, you can be pretty sure China will meet its goals.

However, throwing more technology at the problem may not be the best way for China to go, according to Kenny Lee, a principal consultant with Verizon Business Asia Pacific.

“Companies simply adopting more layers of technology may lead to false sense of security,” he told The Reg.

“Many of today’s malware are undetectable due to increased customisation which renders anti-virus tools less effective. For example, on a case Verizon worked on, we identified a backdoor which was only recognised by one out of 40 AV vendors.”

The government’s proposals are nothing new, but if anything can be seen as a recognition of the importance of evolving information security strategies in key industries to protect national security and economic advantage.

If nothing else, a more secure China should at least reduce the number of unprotected machines which can be co-opted by cyber criminals.

There’s no doubt China is an increasing target of attack for other states and cyber criminals.

Stats revealed in March claimed that attacks from outside the country had infected 8.9 million machines in 2011, up from five million a year earlier.®

Article source: http://go.theregister.com/feed/www.theregister.co.uk/2012/07/19/china_government_cyber_security_guidelines/