• United States



Shellshock Bash hackers found gearing up for broader attacks

Sep 25, 20144 mins
AppleData and Information SecurityLinux

AlienVault used a honeypot to trap malware uploaded by attackers exploiting the Shellshock Bash vulnerability

Attackers exploiting the Shellshock vulnerability that was discovered in a widely used component of Mac OS X, Linux and Unix had infected by Thursday hundreds of systems with exploratory malware.

Security vendor AlienVault used a honeypot to trap two pieces of malware used to infect systems compromised through the critical flaw in GNU Bash, the default command shell for the affected operating systems. Bash stands for Bourne Again SHell.

Proof-of-concept exploits have shown that the zero-day vulnerability published Wednesday can be used to commandeer systems and access others on a network. The flaw poses the biggest immediate threat to Web servers running Linux. However, Bash is also found in routers and in Linux-based devices found on the Internet of Things.

Researchers believe the number of potentially vulnerable devices could be at least 500 million, making the flaw as widespread as the infamous Heartbleed OpenSSL bug.

Bash is able to run an application-sent command that sets environment variables, which are dynamic, named values that affect the way processes are run by a computer. The flaw enables an attacker to tack on malicious code to an environment variable and send it for execution on a vulnerable server via the Common Gateway Interface (CGI), which is a widely used system for generating dynamic Web content.

While there are other paths to exploit Bash, CGI is the most likely route of attack on Web servers, researchers say.

The attackers seen by AlienVault sent packets carrying the exploit to Web servers, loading malware on those that were compromised.

The command-and-control server for one piece of malware was down, so little was known about infected systems. However, based on the C&C server of the second malware, more than 700 systems were likely compromised, Jaime Blasco, director of AlienVault Labs, said.

The malware was capable of checking files, looking for the user name on a server and checking the version of the operating system, indicators that the attackers were on a reconnaissance mission.

“It’s like a proof-of-concept,” Blasco said. “They are trying to find out how big this (exploit) can be and how they can use it.”

The malware appeared to be a repurposed Internet Relay Chat (IRC) bot that received instructions from an IRC server acting as the C&C system.

Such experimentation on the part of hackers will ultimately determine how successful attackers are in compromising systems through Shellshock. The vulnerability is potentially more serious than Heartbleed, because Shellshock can be used to take over systems and infect others, while Heartbleed was used mostly to steal data, Kevin Haley, director of product management for Symantec Security Response, said.

“It leaves systems much more vulnerable than Heartbleed did,” Haley said.

On Wednesday, AusCERT, the Computer Emergency Response Team of Australia, was among the first to report that Shellshock exploits were in the wild.

The discovery sent shockwaves across the Internet. Linux distributors, including Debian, Unbunt, Red Hat, CentOS and Novell/SUSE, scrambled to push out temporary patches while vendors released signatures to block exploits traveling through intrusion prevention systems.

Web servers running CGI are the most vulnerable to immediate attack, while the risk to Mac OS X computers is less.

That’s because the most likely path to compromise Macs would be through SSH, a secure communications protocol, Symantec said. To take that route, the attacker would need valid SSH credentials, which means he would already have to be logged into an SSH session.

Older Internet of Things and embedded devices running Bash would also be vulnerable, while new devices would not, Symantec said. The latter typically runs a set of tools called BusyBox, which does not have the vulnerability.