Stealing Data By 'Living Off The Land'

Organizations should be on the alert for adversaries attempting to breach their computer systems by using little or no malware in their attacks, warns Dell SecureWorks’ Counter Threat Unit (CTU) senior researcher Phil Burdette. 

According Burdette, cyber criminals are using the target company’s own system credentials and legitimate software administration tools to move freely throughout their network, infecting and collecting valuable data. Burdette, who is part of the CTU operations team, says this has been the method to gain access to networks in nearly all of the intrusions responded to by the Incident Response Team over the past year.

The CTU has coined this tactic: “living off the land.” 

The tactic of stealing system credentials and administration tools has been an ongoing challenge for network defenders for some time.  What makes the task even more challenging is that many standard security technologies to uncover cyber threats -- intrusion detection and prevent systems and firewalls -- do not pick up these threats.  So these types of attacks are often missed by many companies until it is too late, according to incident response teams.  

“Threat actors will often follow the path of least resistance to achieve their objectives,” says Burdette.  “If there are legitimate tools and solutions available to them, they will attempt to use those before they need to deploy and concentrate craft to exfiltrate data from a system environment.”

Often the method deployed is as simple as a message purporting to be from the information technology department, telling a company’s employees to go to a certain web site to update their credentials, Burdette says.  For instance, one incident investigated by the SecureWorks team involved a threat group that targeted a pharmaceutical manufacturer.  The group got their initial foothold into the company via a convincing spearphishing campaign where the attackers, posing as the organization’s IT staff, sent an email to several company employees.

They used the ploy that the IT department was testing a new webmail solution and asked all of the employees to click a link and provide their domain username and password.  One employee responded and within hours the cyber thieves were using the compromised credentials, accessing the network via a virtual private network tunnel.  Using stolen system administrator credentials, the group moved laterally and connected to other systems using the Remote Desktop Protocol. RDP is used by legitimate sys admins and helpdesk employees to maintain systems.  Using the File Transfer Protocol, the group accessed and stole sensitive Intellectual Property information from many files, Burdette notes.

Another incident involved the stealing of hundreds of credit and debit cards from an organization’s Point-of-Sale (POS) Terminals. Burdette’s incident response team discovered that the adversaries got their initial foothold into their target’s IT environment by getting an employee’s network credentials for the company’s Citrix server. The server did not require employees to use two-factor authentication to log in remotely. Once they got into the network, the cyber thieves captured the domain administrator’s credentials. This allowed them to conduct a thorough reconnaissance of the target’s IT environment, finding a “trusted” system to breach more systems. 

They gained access to the company’s Centralized Security Management Server, which was used to deploy and manage the organization’s anti-virus software for all of their endpoints, including their POS terminals.  With the domain administrator credentials, the thieves had easy access to this trusted system. They then pushed malware down to the POS Terminals that captured all of the credit and debit card data entered into each terminal.  The company’s anti-virus software did detect the financial data-stealing malware, however, the cyber thieves cleverly instructed the Security Management Server to whitelist the malware, allowing for its continued use.

Often it is difficult to discover how an adversary initially penetrated a network because the attacker has been in the network for such a long time the data is no longer available.  Or, in some cases, the victim organization did not have the proper system implementation in place for the incident response team to determine how credentials might have been compromised, Burdette says.

However, knowing something about what is the normal behavior of a system administrator versus an adversary is one of the first steps in combatting these types of attacks that use sys admin credentials and tools, Burdette says.  This can be difficult because large organizations usually have hundreds of system administrators to support business operations.  Still, identifying and detecting best practice standards for what is legitimate activity and how adversaries would act differently to achieve their objectives are necessary steps.

Organizations should also implement endpoint security systems that focus on threat behavior and can detect if malicious activity is going on in the network along with their IDS/IPS and firewall security.  Ultimately, the defense has to be a coordinated effort between incident response teams that know how adversaries operate plus the internal IT staff who know how their networks operate, Burdette says.