Showing posts with label exploit. Show all posts
Showing posts with label exploit. Show all posts

April 01, 2022

Linux Networking Bug

 

A graduate student at the University of California has found a vulnerability (CVE-2022-27666) in IPSec’s esp6 crypto module which can be potentially abused for local privilege escalation. The problem lies with heap overflow – a receiving buffer of a user message in esp6 is 8 pages, but you can send a message larger than that causing a buffer overflow.

Red Hat states this can allow an attacker to overwrite kernel heap objects and escalate their local privileges. Both NIST and Red Hat have given this vulnerability a CVSS score of 7.8 – which is very high. If IPSec is already in use on a Linux system (more than likely) and has IPSec Security Associations configured (which is essential to the network security protocol), there are no additional privileges needed to exploit this bug, meaning many people are open to this attack including users of the popular Linux distros such as Ubuntu, Debian, and Fedora.

Linux Brand Tux Penguin

The student was able to get around the Kernel Address-space Layout Randomisation (KASLR) which typically makes it much harder to exploit memory because processes are placed at random instead of at fixed memory addresses. After hanging this process, an attacker could use Filesystem in User Space (FUSE) to create their own filesystem and map memory to it. All read/write functions will then be handled by the attacker’s filesystem which makes it trivial to then gain root access to a system.

Please see the updates for your Linux distro in order to patch this vulnerability!

March 30, 2022

Cybersecurity Lessons from 2021

 

SecureWorks, an incident response service provider, covered over 450 incidents last year and recently published its feedback. 85% of the incidents they responded to were financially motivated and a further 5% were seemingly government-sponsored attacks. The remaining attacks were accidental or deliberate actions of employees.

43% of the initial access gained was through threat actors exploiting vulnerabilities in internet-connected devices and credentials theft represented another 18% of initial system access. These credentials can be obtained through the dark web, brokers, credential stealing, brute-force attacks, or password spraying. In previous years, credentials theft was the number one approach to compromising a target, so the focus for security professionals needs to shift to patching vulnerabilities.

The rise in multi-factor authentication may mean that attackers are focusing on exploiting vulnerabilities that do not require authentication. Alternatively, it can be easy for an attacker to exploit proof-of-concept code that is published shortly after a vulnerability is publicly disclosed. This can lead to wide scale exploitation of any vulnerable devices in multiple targets simultaneously.

Despite ransomware attackers being increasingly imprisoned for their actions and the US government prioritising ransomware the same with it does terrorism, SecureWorks has not seen a reduction in ransomware attacks in 2021.

Many of the attacks that relied on credential theft and abuse occurred because the target organisation failed to implement multi-factor authentication mechanisms at all or properly. However, attackers have been able to bypass MFA by exploiting legacy authentication protocols (e.g. IMAP and SMTP) which are either still in use or haven’t been disabled. These protocols cannot enforce MFA and pose a significant security risk to businesses.

Security Vulnerability Exploit

Even when MFA is implemented correctly, users may still eventually decide to approve an MFA request if attackers continuously send them due to “notification fatigue”. To mitigate this issue, consider implementing MFA that request a code from the user rather than a one-click solution.

If an enterprise is using cloud solutions, carefully investigate all of the security components and controls offered by the cloud provider to ensure logging and controlled access is offered by the cloud service. It may be attractive for businesses to implement these cloud solutions, but there are security considerations that must be accounted for before moving resources online.

To prevent cyberattacks going forward, SecureWorks recommends that IT and security professional regularly perform vulnerability scans, control access carefully and make use of IP lists, monitor newly registered domains that spoof or impersonate your company, improve your backup strategies and procedures to mitigate ransomware attacks, implement MFA properly, and implement DKIM and SPF authentication for email clients to avoid fake emails being sent by attackers.

Of course, you should also ensure your systems and software and kept up to date, use the principle of least privilege for account access, and ensure you implement an endpoint detection and response solution.

March 27, 2022

Google Stops Hackers

 

Google has announced that it has prevented 2 North Korean hacking groups from exploiting a zero-day bug in its Chrome browser. The bug (CVE-2022-0609) was patched in February, but it was being exploited a whole month earlier than this, with reports it was being used as early as January 4th 2022. The US agency, CISA, mandated that federal agencies must patch the bug in February. The North Korean hacking groups suspected at using the exploit are linked to Lazarus – another North Korean group accused of hacking Sony Pictures and attacking the SWIFT international bank-messaging system.

It is suspected that all of these groups are working for the same entity (potentially the North Korean government) and have a shared supply chain, thus they are all able to use the same exploits. However, each group seems to use different techniques and have different missions when deploying attacks. The hackers linked their exploit kits inside hidden iframes embedded on both websites they owned and websites they had compromised. This kit contained multiple different stages and components.

Anonymous Hacker Group

The groups have primarily targeted US tech, cryptocurrency, news media, and fintech organisations, but similar companies in other countries may also have been targeted according to Google. The groups have also targeted web-host providers, software vendors, and domain registrars with fake job offered in emails that impersonate recruiters from the likes of Oracle, Disney, and Google. These emails contained links to spoofed versions of job board websites that are popular in the US for hiring tech talent.

Once the group was discovered, Google added all known websites and domains to their Safe Browsing service which prevented users from further exploits. Google also sent alerts to all users that were targeted through Gmail and Workspace. A recent Google acquisition, Mandiant, has identified the hacking groups as Lab 110, TEMP.Hermit, APT38, Andariel, and Bureau 325 – all operating under North Korea’s foreign intelligence agency which deals with technology, operations, and reconnaissance. Mandiant claims that North Korea is borrowing China’s strategy of cajoling hacker groups to work with the government.

Each one of these groups is designated to target separate industries and gather intelligence for the purposes of geopolitical strategy or raise revenue for North Korea’s ballistic missile programs through cryptocurrency theft. Information collected through these attacks may also be used to develop vaccines, bypass sanctions, fund other weapon programs, and produce other strategies and internal items relevant to the North Korean regime.