Security researchers have many names for the hacking group that is one of the suspects for the cyber-attack on the US government's Office of Personnel Management: PinkPanther, KungFu Kittens, Group 72 and, most famously, Deep Panda. But to Jared Myers and colleagues at cyber-security company RSA, it is called Shell Crew, and Myers' team is one of the few who has watched it mid-assault - and eventually repulsed it.
Myers' account of a months-long battle with the group illustrates the challenges governments and companies face in defending against hackers that researchers believe are linked to the Chinese government - a charge Beijing denies.
"The Shell Crew is an extremely efficient and talented group," Myers said in an interview. Shell Crew, or Deep Panda, are one of several hacking groups that Western cyber-security companies have accused of hacking into US and other countries' networks and stealing government, defence and industrial documents.The attack on the OPM computers, revealed this month, compromised the data of 4 million current and former federal employees, raising US suspicions that Chinese hackers were building huge databases that could be used to recruit spies.
China has denied any connection with such attacks and little is known about the identities of those involved in them. But cyber-security experts are starting to learn more about their methods.
Researchers have connected the OPM breach to an earlier attack on US healthcare insurer Anthem Inc, which has been blamed on Deep Panda.
RSA's Myers says his team has no evidence that Shell Crew were behind the OPM attack, but believes Shell Crew and Deep Panda are the same group.
And they are no newcomers to cyberespionage. CrowdStrike, the cyber-security company which gave Deep Panda its name due to its perceived Chinese links, traces its activities to 2011, when it launched attacks on defence, energy and chemical industries in the United States and Japan. But few have caught them in the act.
Shell crew in action
In February 2014, a US firm that designs and makes technology products called in RSA, a division of technology company EMC, to fix an unrelated problem. RSA realised there was a much bigger one at hand: hackers were inside the company's network, stealing sensitive data. "In fact," Myers recalls telling the company, "you have a problem right now." Myers' team could see hackers had been there for more than six months. But the attack went back further than that. For months Shell Crew had probed the company's defences, using software code that makes use of known weaknesses in computer systems to try to unlock a door on its servers. Once Shell Crew found a way in, however, they moved quickly, aware this was the point when they were most likely to be spotted.
On July 10, 2013, they set up a fake user account at an engineering portal. A malware package was uploaded to a site, and then, 40 minutes later, the fake account sent emails to company employees, designed to fool one into clicking on a link which in turn would download the malware and open the door. "It was very well timed, very well laid out," recalls Myers. Once an employee fell for the email, the Shell Crew were in, and within hours were wandering the company's network. Two days later the company, aware employees had fallen for the emails - known as spearphish - reset their passwords. But it was too late: the Shell Crew had already shipped in software to create backdoors and other ways in and out of the system. For the next 50 days the group moved freely, mapping the network and sending their findings back to base. This, Myers said, was because the hackers would be working in tandem with someone else, someone who knew what to steal. "They take out these huge lists of what is there and hand it over to another unit, someone who knows about this, what is important," he said. Then in early September 2013, they returned, with specific targets. For weeks they mined the company's computers, copying gigabytes of data. They were still at it when the RSA team discovered them nearly five months later. Myers' team painstakingly retraced Shell Crew's movements, trying to catalogue where they had been in the networks and what they had stolen. They couldn't move against them until they were sure they could kick them out for good. It took two months before they closed the door, locking the Shell Crew out. But within days they were trying to get back in, launching hundreds of assaults through backdoors, malware and webshells. Myers says they are still trying to gain access today, though all attempts have been unsuccessful. "If they're still trying to get back in, that lets you know you're successful in keeping them out," he said.
© Thomson Reuters 2015