Google attacks: A wake-up call or curtain call for agencies?
It may already be too late for agencies to ward off the kinds of cyberattacks that targeted Google
- By John Zyskowski
- Feb 04, 2010
Until now, it might have been easy for government agencies that are not in the defense or intelligence business to think that state-sponsored cyberattacks were something they didn't have to worry about. But if there is any lesson from the news that Google and dozens of other nondefense companies were recent targets of sophisticated hacks run through China, it’s that civilian agencies are not as safe from such threats as some of them might think.
“These were regular old businesses being attacked,” said Alan Paller, director of research at the SANS Institute, which provides cybersecurity training programs. “This means that regular old federal agencies are being attacked the same way, and they are, but their managers don’t know it.”
What’s worse, agencies have been required to take an approach to cybersecurity that makes it extremely difficult to protect themselves from these kinds of assaults, Paller and other security experts say.
The Google attacks are also significant because, besides the wider range of targets, of the methods and intent.
Hacking used to be mostly about mischief and larceny. In the early days of the Internet, hackers would unleash software viruses and worms designed to slip through firewalls and cripple or disable infected computers and networks. Then along came criminals using so-called phishing techniques to try to trick computer users into divulging account passwords and other personal information in response to phony e-mails that appear to be from the random recipient’s bank or other trusted source.
However, the Google attacks fall into a different class of skulduggery, called advanced persistent threat (APT), which combines more sophisticated versions of those other hacking methods with greater resources and patience and the pursuit of different objectives.
“This was not a smash-and-grab,” said Scott Crawford, research director of security and risk management at Enterprise Management Associates. “This comes from an adversary with a long-term view and motivation and resources pressing toward a specific goal.”
APT perpetrators are highly selective about their targets, whether it's getting access to trade secrets and business plans, government records and programs, or anything else that might provide them with political, economic or strategic advantage.
And unlike the grandstanding hacker of old who might deface an organization’s public Web site to show the world how clever he or she is, an APT strike aims to avoid detection.
Its first goal is to gain entry to an organization’s computer network, through targeted phishing ploys and the exploitation of system vulnerabilities, such as the flaw in Microsoft’s Internet Explorer used by those behind the Google job. Once inside, the APT attacker leaves software programs that can steal or modify information undetected but then will lay low, so that the dirty tricks can be used again and again.
DOD officials and others from some of the more sensitive civilian departments, such as Commerce, Treasury and State, have known about and have been targeted by APT for years. But security experts contend the APT threat applies to a much broader range of interests and agencies, many of them still unaware of the risks. What would happen to public trust in the government if APT hackers could modify records in entitlement programs or national economic reports?
Outside defense and intelligence agencies, only State — and to a lesser extent, the Transportation Department — have the types of cybersecurity resources needed to even start to fight back against APT, Paller and others said. Those assets are not just tools and products, though those things do play an important role, but teams of experts, such as those at State, who monitor and analyze network traffic and system logs, conduct forensics, and reverse-engineer past attacks to better anticipate future ones.
“If you understand how burglars get into houses really well, you don’t have to protect the whole house, you only have to protect the spots where they go,” Paller said.
Unfortunately, most agencies don’t have those resources. Instead, they’ve devoted most of their cybersecurity spending and efforts to compiling security system inventory data, detailing contingency and response plans, and writing it all up into the annual audit reports required by the 2002 Federal Information Security Management Act.
The problem with FISMA is that it focuses on compliance effectiveness, which doesn’t correlate with how well security resources are performing to protect government systems and data, said James Lewis, a senior fellow and cybersecurity expert at the Center for Strategic and International Studies.
Support is growing to refocus FISMA on more meaningful metrics, and there are other initiatives heading in the right direction. But it’s still hard to gauge the level of resolve behind them.
A proper and serious deployment of cybersecurity resources “was never a priority, and it is still not a priority,” Lewis said.
It remains to be seen whether Google’s recent run-in with APT will help to change that.
John Zyskowski is a senior editor of Federal Computer Week. Follow him on Twitter: @ZyskowskiWriter.