The current state of ransomware — and its future
2020 did not mark the beginning of ransomware, nor its ascent, which has been happening gradually throughout the last decade or so. But 2020 may have been the year that saw the greatest concentration of ransomware attacks, magnifying the suffering of an already wary population.
Ransomware thrived in the chaos and fear sown by COVID-19 in the early days of the pandemic, and as the year wore on, the attacks became more sophisticated — targeting the embattled health and education sectors, even as hospitals became COVID-19 battlegrounds and schools struggled to invent an entirely new way to teach students.
As they navigated their way through vulnerable systems, and those systems attempted to protect themselves, attackers pioneered new ways to evade endpoint security products.
This behaviour is the standard modus operandi for ransomware attackers and has been for some time. But discoveries made by analysts at Sophos have unearthed a new development: ransomware code appeared to have been shared across ‘families’, and some of the ransomware groups seemed to work in collaboration more than in competition with one another.
This, according to Sophos, may well be the future of ransomware.
As potential victims shored up their defences, several ransomware families increased pressure on them to pay ransoms by holding machines hostage and stealing data on those machines to release if they didn’t pay the bounty.
These families seem to have settled on a common toolset to exfiltrate a victim’s network, which is much harder to detect by endpoint security products. The criminals use the toolset to copy sensitive internal information, compress it into an archive, and transfer it out of the network — and out of reach of the victim.
In a final act of destruction, ransomware attackers increasingly hunt for the local servers that contain backups of critical data; when found, they delete (or independently encrypt) these backups just before the network-wide encryption attack.
There are several everyday attacks that organisations like Sophos are exceedingly familiar with. But, while these attacks are well understood and easily contained, every attack carries with it the potential to get far worse if it isn’t dealt with effectively.
Here are some examples:Attacks targeting Windows & Linux servers
Attacks on servers increased considerably in 2020. They are attractive to attackers because they often run for extended periods while unmonitored, and they often occupy a privileged space on the network, having access to the most sensitive and valuable data in an organisation’s operation. Sophos projects that attacks on servers will continue to increase in 2021.
Cybersecurity teams can employ best practices by refraining from running conventional desktop apps, like email clients or web browsers, from the server. Server admins have not, historically, installed endpoint protection products on servers, but with the advent of these types of attacks,
that conventional wisdom has shifted.Delivery mechanisms
Sophos found Window’s Remote Desktop Protocol (RDP) brute force attempts to be a frequent root cause of ransomware attacks, whereby attacks gain initial access to a target’s network by through RDP, and then the use of those machines to gain a foothold within the network and take control of servers.
This has become an avenue for ransomware, and COVID-19 lockdowns have only exacerbated the problem. Many organisations rely on RDP to remain operational with workforces working from home.
The main risk here, according to Sophos, is that RDP was never meant to withstand the kind of onslaught it can receive from the public-facing internet. If the RDP password is weak, easily guessed or brute-forced through automated login attempts, the attacker gains a foothold inside the network they can exploit at their leisure.
Best practices suggested by Sophos direct security teams to place RDP behind a firewall instead of facing the public internet, with users requiring a VPN login to connect.
In research performed before the lockdown took effect, Sophos set up honeypots in 10 data centres worldwide to better understand just how bad the problem had become. Over 30 days, the honeypots received a median average of 467,000 RDP login attempts, or about 600 per hour at each location. The research revealed that each honeypot received a steadily increasing frequency and ferocity of login attempts.
There are many devices in our individual worlds that don’t look like computers or servers but can nonetheless be abused: routers, phones, smart TVs, as well as countless other bits of IoT hardware.
Malicious advertising (malvertising) is a significant source of threats to many devices, Sophos says. Throughout the research of current malvertising threats that fall outside the realm of malware attacks, Sophos discovered technical support scams using ‘browser locking’ web pages, and ads targeting mobile devices that are linked to fraudulent or ‘fleeceware’ apps.
Sophos classifies these as ‘fake alert’ attacks — malvertisements that attempt to scare their targets into taking an action that will enrich the scammers behind them.
These scams attempt to persuade targets to grant remote access to their devices — and convince them to purchase exorbitantly priced technical support software and services or obtain targets’ credit card data for fraudulent purposes.
Many attackers forego using malware altogether, and leverage tools already on operating systems; others may utilise tools used by two large segments of the information security industry: incident responders and penetration testers.
These are classified as LOL attacks, or living-off-the-land. Involving little-to-no malware, they instead harness existing components of the operating system or popular software packages. These attacks often use automation in the form of native scripting, collectively referred to as LOLscripts. The attackers use these LOLscripts to execute sequences of commands using living-off-the-land binaries (applications), colloquially called LOLbins.
Attackers can deploy security tools commonly used by network administrators and penetration testers, which are helpful to criminals as they are often used in a legitimate capacity and are also problematic for anti-virus to detect activity from such tools. As such, Sophos must rely more heavily on the study of the LOLscripted behaviour to identify potential malicious activity.
Despite Sophos tracking the use of 99 of these attack tools, the company says it’s ‘unlikely that we’ll see a reprieve from attackers continuing to take advantage of these well-written tools throughout 2021’.
To learn more about ransomware in Sophos’s 2021 Threat Report, click here.