Signs of cloud security hacks are there if you know what to look for

Monitoring of cloud activities should be incorporated into business-focused security KPIs

Cloud-based workloads may be configured and used for different scenarios than on-premises servers, but the process of picking out cybercriminal activity is not so far removed, according to a senior security executive who notes that telltale signs of penetration are there if you know where to look.

“The benefit of the cloud is that it's so easy to spin up new servers, but the drawback is that it's easy to forget them,” LightCyber senior director of product management David Thompson told CSO Australia as the company debuted versions of its Magna tools that extend visibility to workloads running on the Amazon Web Services (AWS) cloud platform.

“You leave them running,” he continued, “and they are attached to servers with production data but no one really know what's happening with them. The servers shouldn't be talking to each other, but often they are and attackers come and take advantage of that. This is happening all the time.”

Close monitoring of communications between AWS virtual environments can readily identify scenarios such as when an embedded attacker is probing the environment to find other running servers.

Such activity, Thompson said, would be glaringly obvious during flow-log analysis – even if attackers were using conventional and unremarkable security tools, as a recent LightCyber study found – due to a surge in connection anomalies created when attackers tried to connect to server addresses that didn't exist.

“A lot of the lateral movement steps can show up as connection anomalies,” he explained. “We can extract the vast majority of reconnaissance behaviour from flow logs. Mapping of the virtual environment has to happen if the attacker wants to figure out what other servers are running.”

Once the virtual workload monitors know what to look for, the increased visibility of such tools can be incorporated into high-level monitoring tools to inform cybersecurity monitoring processes. These processes will become increasingly important both as organisations shift an increasing proportion of their back-end workload into virtual cloud environments, and as their employees take up cloud-based productivity tools like Microsoft Office 365.

“There have been a lot of silos,” Thompson said, “and many are around the infrastructure that you operate. But those worlds will bridge and cohere, because the kinds of cybercriminal activities are often the same, and the attack surfaces are the same. The more you can get into the detection envelope, the better.”

Improving visibility into cloud workloads has been a steady focus for a number of security vendors, with cloud-monitoring firm Gigamon among those partnering with LightCyber to soon provide even higher-granularity visibility of cloud security compromises through integration of its own monitoring technology.

With the cloud visibility issue steadily being overcome, many caution that CISOs need to adjust their risk evaluation processes in a process informed by that enhanced visibility. Gartner analysts have highlighted the importance of establishing key risk indicators (KRIs) that measure risk exposure in business and technology-aligned terms and facilitate meaningful reporting to the executive.

Michael Shatter, partner for security and privacy services with consultancy RSM Australia, agrees. “Information security management is closely linked to an organisation’s risk management processes,” he recently said in a statement. “Therefore, security metrics reporting should be a key part of the risk assessment of mitigation strategies and actions that are either planned or already in place.”

Shatter outlined several key characteristics that organisations should incorporate into their security metrics, including meaningfulness, accuracy, genuineness, timeliness, predictivity, and independence.

“For some organisations, these activities and their associated costs become a material investment,” he said. “However, security spending is not and should not be excused from the normal business scrutiny of how funds are spent and the measurement of the return on these investments.”

“To really understand the value and success of the security measures and the respective investments, organisations should measure and report on agreed-upon metrics. These metrics should communicate clearly to the board and management whether the cyber and information system security controls and processes are effective and are delivering value.”

Tags CISOcloud securityhackingCSO OnlineAmazon Web Services (AWS)cybercriminalLightCybercommunication monitoring

Show Comments