A study has revealed that the Flame malware, developed to target Middle East-based companies, may be just one of four pieces of malware created by the same unknown development team.
Flame became public in May after Iran discovered the malware had been used to infect computers in its oil ministry. It was quickly linked with Stuxnet, which was used to disrupt Iran’s uranium refinement equipment, and was suspected to have been developed by other countries, possibly the U.S. and Israel, due to its sophistication and targeting.
The latest research is the result of a combined effort by Symantec, Kaspersky Lab, the International Telecommunication Union’s (ITU) IMPACT cyber security team, Germany’s Computer Emergency Response Team for Bundesbehörden and its Federal Office for Information Security (BSI).
The study found that the command-and-control mechanisms for Flame may have been developed as far back as December 2006, making the malware much older than previously thought, according to a write-up by Kaspersky Lab. It had been circulating in the wild since around 2010, but its creation date had been unknown until now.
Among the other significant findings is that Flame’s command-and-control system handled three types of malware that have not been examined by researchers and whose purpose is unknown. Other clues showed those malware types may still be under development or not yet activated.
Flame is the Swiss Army knife of spying tools: It can collect data entered into forms, collect passwords, record audio and capture screenshots. It may have played a reconnaissance role to scout out systems for later infection by Stuxnet, which disrupted industrial control systems made by Siemens and used by Iran for refining uranium.
According to the research, Flame’s command-and-control system was designed to look more like a content management system (CMS). Simple commands were used such as “upload,” “client,” “news,” “blog” and “ads” that mimic a CMS interface rather than looking like other botnet interfaces.
“We believe this was deliberately done to deceive hosting company sys-admins who might run unexpected checks,” Kaspersky wrote.
Surprisingly, those who modified Flame’s code left nicknames embedded in the code along with timestamps. The researchers identified four people who modified the code, though the nicknames were censored in Kaspersky’s write-up. One of the coders appeared to be the team’s leader, while another specialized in advanced cryptography.
Kaspersky said it had received many questions about the amount of data Flame might have taken. Although it appeared that stolen information was transferred to other servers every 30 minutes, Flame’s operators for some reason became locked out of one of the drop servers used to collect their stolen data. Flame may have collected as much as 5.5GB of compressed information in just a week.
Flame’s controllers made another mistake: forgetting to delete log files that showed how many victim computers were sending back information. During a one-week period between the end of March and early April, some 3,702 unique IP addresses from Iran connected to that drop server, with another 1,280 from Sudan.
“Our previous statistics did not show a large number of infections in Sudan, so this must have been a dedicated campaign targeting systems in Iran and Sudan,” Kaspersky wrote.
