If the average IT security threat is an opportunistic phishing e-mail or indiscriminate drive-by virus attack, then an advanced persistent threat (APT) attack is the exact opposite.
Tailored specifically to the victim, attacking from multiple directions at the same time and not giving up until they get their target, APTs are expensive, sophisticated tools used to obtain commercially valuable intellectual property or to bring large multinationals to a grinding halt. For the average company with average levels of security, an APT attack is the stuff of nightmares.
Some of the best known forms of APTs, such as Stuxnet, Flame and Gauss, have, amongst other things, targeted nuclear facilities in Iran, engaged in cyberespionage in the Middle East and, in the case of the Russian-based Red October group, have carried out highly targeted ‘spear phishing’ attacks on high-ranking government officials in the US. Some have lurked undetected for years before being brought to light.
While all of this is intriguing, and undoubtedly marks the beginning of a new era of sophistication in the kinds of threats out there, is an APT really something that Irish businesses need to worry about?
Challenge ahead
According to a number of industry experts, APTs are hotly tipped to be one of the top IT security challenges for this year. But does this really make sense? What does the average Irish company have that the kind of hacker that could direct an APT type attack would want? Is there a real threat here, or is a scaremongering security industry talking up a threat to make itself seem more indispensable?
"The reality is that there are really very few organisations in Ireland that are likely to be directly targeted commercially by an ATP attack. Speaking generally, rather than putting in security measures specifically to defend against that kind of threat, the average Irish business would be far better off tidying up their own internal processes and putting in reasonable and normal levels of control," said Michael Conway of Renaissance Contingency Services.
"True APT attacks are very tightly-focused and are something that, in the normal course of ordinary commercial operations, the average company shouldn’t be worried by. There are some organisations operating in Ireland that are typically the kind of targets that get hit by APT attacks-multinationals working with significant amounts of intellectual property for example-but they should all already have in place normal protection for that kind of thing."
While the average Irish enterprise may not be directly target by an APT, that does not mean that they can afford to totally ignore this developing element of the security landscape.
Obvious targets
"SMEs aren’t obvious targets for ATP actors, but in actual fact they are frequently targeted because they are a means to an end," said Don Smith, technology director for Dell Secureworks.
"You don’t have to be a large pharma or defence industry organisation to be targeted-you could just be a company that has dealings with one. This is something that is often missed, yet is a frequent tactic in security: attacking the soft underbelly of a small company’s systems, gaining a foothold, and using it as a stepping stone to a larger and more attractive target that the company has dealings with."
This is where the broadening focus on APTs is coming from-the idea that it is not just the main targets that become victims in APT attacks, but that it is often unsuspecting bystanders that become involved without ever being aware of it.
"While you might not be personally targeted, it’s entirely likely that you’ll be used as part of someone else’s APT campaign as a passive participant if your machines are taken over. This is most commonly seen when unsuspecting small companies with poor security are infected with malware that turns their machines into components of a botnet that is in turn used for an APT attack," said David Keating, security sales manager from Datasolutions.
"If that happens, it won’t take long until you can’t send an e-mail to anyone because you’ll be blacklisted by servers everywhere. If your machines are used as stepping stones in some scam to defraud a bank somewhere, you can probably talk your way out of the fallout by explaining that you’re also a victim. But do you need the hassle? You have to weigh that grief against the fact that you can put an awful lot of protection in place for not very much money."
Toolkits
A growing issue in this area is the availability of APT toolkits. As each new high level security threat is uncovered, inevitably toolkits based upon its core functions soon become widely available to hackers.
"Back in 2007 and 2008, Zeus was uncovered and very quickly became a downloadable toolkit. Likewise with Stuxnet-these things allow people to make their own APTs. As the more sophisticated ones break and are publicised, they become available and get more and more use," said Keating. "You no longer need to be a super sophisticated cyber-criminal in order to mount an APT attack."
When it comes to putting in place measures to counteract the threat posed by APTs and the toolkits associated with them, the most important single fact to remember, according to Don Smith of Dell Secureworks, is that it’s not a case of what but rather who is behind the attack.
"People talk about trying to identify and contain infections, malware or viruses, but with APTs the key thing to remember is that the threat is the person behind the attack, not the tools they’re using. You may eradicate the threat but the person behind it will probably try again using different and multiple attack vectors. That’s where the ‘persistent’ part of the acronym comes from," he said.
Deep running
What makes these threats so hard to deal with is that unlike traditional kinds of malware, which are designed to infect a system and cause damage or steal information and to be quite obvious in their effects, APTs are designed to do their work and remain undetected.
"APTs are designed with a primary objective of evading detection. If someone is trying to build a botnet, then that botnet is going to be used to do something nefarious, such as sending spam or performing a DDOS attack. The larger the network of bots, the more valuable the botnet is and the more rent the owner can charge for using it," said Shane Grennan, Irish country manager for Fortinet.
"However the botnet’s value is directly linked to its ability to evade detection. Once it’s detected, then that asset is lost. The persistent bit is that the technology is deployed to a particular end, and they’ll try multiple different attack vectors to reach their goal. If one attack doesn’t work they’ll try something else, and then something else, until they find a way that does work. This is quite different from automated malware that just randomly looks for unprotected machines on the Internet."
According to Grennan, the people using these technologies know exactly what they’re doing and the odds are good that there are large numbers of computers in Ireland that are currently deployed as part of botnets without their owners being aware of it.
Fair share
"If you picked the top Fortune 500 companies that have been the target of APTs and ask how many of them are in Ireland, the answer is pretty small. But if you look at the level of infection globally, we have our fair share of it," he said.
"We come across companies that have lots of infected PCs on their network, all part of a botnet. While it might not be a specific threat to the company that owns the machines, it’s still a small piece of an APT. If you become infected by a piece of malware and it successfully connects PCs on your network to a botnet, then it’s important to know that malware of that kind is usually multifunctional."
"From day one, it might make that machine one ten thousandth of a DDOS attack or it could be used to send spam or whatever, but it could also be gathering information as well. It is really only limited by the imagination of the person driving the software. Once it’s installed and evading detection then it can evolve, change and be sent instructions by the owner. You don’t know if it’s gathering keystroke information or otherwise seriously undermining your security."
The moral of the story then is that you do not have to be the direct target of an APT to be seriously affected by one. And once your security is compromised, there’s no real way of knowing how far the threat extends.
Fundamental change
However one thing the experts in this area agree upon is that successfully meeting the challenge that APTs pose involves fundamentally changing one aspect of how the average company views its security-the role of the perimeter firewall.
"Historically, people focused mostly on in-bound data and not outbound data, and the firewall was all-important," said Don Smith of Dell Secureworks. "That bias was identified very early on by the authors of malware-they knew that once they got installed on a desktop or laptop inside an organisation, then the outbound traffic initiated from within the organisation was not going to be subject to the same level of scrutiny."
"Because of the increasing importance of the internet in most people’s daily lives, web surfing policies have become increasingly lax, and that’s all outbound traffic. It’s become much more important than before to increase your focus on outbound traffic. It’s as important, if not more important, than your inbound data."
According to Smith, the bad guys aren’t interested in mounting a frontal assault on your perimeter firewall-instead they want to stealthily send spear phishing emails to members of staff looking to install malware that will allow their traffic to be concealed from within.
"There is no silver bullet to solve this threat, but at the very least your security needs to address weakness at every level, at the end point, at the application and at the network. That’s the only way you can start to address the persistent end of the threat, because if there are weaknesses, then attackers are quite successful at finding them," he said.
Firewall facing
Dermot Williams of Threatscape agrees.
"If all your firewall rules are designed to stop someone getting in, what happens if they get around those? Once they’re in, is there anything to stop them sending data out? Often there’s a presumption that if a user is behind the security perimeter then they must be one of the good guys rather than a bad guy," said Williams.
"In general though, dealing with the kind of threats posed by APTs is a matter of looking after basic security and also looking at the way in which the human beings use the technology. The easiest way to subvert the most sophisticated security is to compromise the user, to fool them into doing something silly. There are all sorts of interesting examples of how this has happened."
The simplest examples are also among the most effective, Williams says.
"Drop a USB stick printed with a company logo in the car park of the target company and then wait for someone to see it, pick it up, carry it inside the building and insert it into a machine to find out what’s on it-straight away malware starts to load up and it’s game over," he says. "That’s been successfully done more than once, and it makes sense. It’s a lot easier to get into a company’s car park than it is their IT network."





Ltd