Tag Archives: Zero-day Malware

Adobe Flash zero day exploited in the wild. Remote code execution vulnerabilities are hacker’s favorites!

8 December 2018

On December 5th, 2018 Adobe published security bulletin APSB18-41[1] for critical vulnerability CVE-2018-15928 in the widely used Flash Player. Gigamon Applied Threat Research (ATR) reported the vulnerability on November 29th, 2018 to Adobe. They detected the issue some days before while analyzing a malicious word document that was uploaded to VirusTotal from a Ukrainian IP address. For a detailed analysis of the attack and the vulnerability see [2][3].

Successful exploitation of CVE-2018-15928 could lead to Arbitrary Code Execution in the context of the current user. Due to RedHat the CVSS3 Base Metrics is CVSS:3.0/AV:N/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:H with a CVSS3 Base Score of 8.8.

Zero days are not a rare phenomenon. Between 2013 and 2017[4] about 60% of the exploits were disclosed before the related CVE was published.

For about 20% of vulnerabilities in the NVD exploits are published in the exploit database[5]. Only about 1% of the vulnerabilities are exploited in the wild. Thus CVE-2018-15928 is a really rare event.

Remote code/script execution (RxE) vulnerabilities like CVE-2018-15928 represent about 20% of all vulnerabilities. 43% of the exploits published between 1988 and 2018 are related to RxE vulnerabilities.

Remote Code Execution Vulnerabilities. Data: 1988-2018

RxE Vulnerabilities. Data: 1988-2018

Exploits for Remote Code Execution Vulnerabilities. Data: 1988-2018

Exploits for RxE Vulnerabilities. Data: 1988-2018

About 5% of the RxE vulnerabilities are exploited in the wild.

This means, that RxE vulnerabilities are 5 times more often exploited in the wild then Non-RxE vulnerabilities. They are hacker’s favorites!

What does the mean for our vulnerability management strategy?

  • The remediation process must be started directly upon publication of an RxE vulnerability in the NVD or the disclosure of an exploit for an RxE in the exploit database.
  • In scope for the first remediation wave must be at least all systems facing the internet, e.g. workstations, servers in the DMZ or in public clouds.
  • Gathering intelligence about new vulnerabilities from a plethora of publicly available sources (OSINT) is a time-consuming task. A threat intelligence service can speed-up information gathering and reduces the workload of your IT security staff.
  • In addition, since remediation takes some time, it makes sense to invest in means for enhancing the resilience of application systems.

Expect the worst and be prepared. Or, to echo Hamlet:

To be, or not to be, that is the question:
Whether ’tis nobler in the mind to suffer
The slings and arrows of outrageous fortune,
Or to take arms against a sea of troubles,
And by opposing, end them? To die: to sleep;

Have a good weekend.

  1. Adobe. Security updates available for Flash Player | APSB18-42 [Internet]. 2018 [cited 2018 Dec 8]. Available from: https://helpx.adobe.com/security/products/flash-player/apsb18-42.html

  2. Gigamon Threat Research Team. Adobe Flash Zero-Day Exploited In the Wild [Internet]. Gigamon ATR Blog. 2018 [cited 2018 Dec 8]. Available from: https://atr-blog.gigamon.com/2018/12/05/adobe-flash-zero-day-exploited-in-the-wild/

  3. Qihoo 360 Advanced Threat Response Team. Operation Poison Needles – APT Group Attacked the Polyclinic of the Presidential Administration of Russia, Exploiting a Zero-day [Internet]. 2018 [cited 2018 Dec 8]. Available from: http://blogs.360.cn/post/PoisonNeedles_CVE-2018-15982_EN.html

  4. Jochem K. About 60% of exploits are published before the CVE. What does this mean for your cyber security strategy? [Internet]. IT Security Matters. 2018 [cited 2018 Dec 8]. Available from: https://klausjochem.me/2018/11/04/about-60-of-exploits-are-published-before-the-cve-what-does-this-mean-for-your-cyber-security-strategy/

  5. Offensive Security. Offensive Security’s Exploit Database Archive [Internet]. Exploit Database. [cited 2018 Nov 4]. Available from: https://www.exploit-db.com/

Locky deployment methods just changed – Who cares?

30 July 2016

The Post ‘Locky Dropper Now Comes Embedded in the Loader’, published July 28, 2016 in the ReaQta Security Blog, clearly shows that the cyber criminals continuously develop and improve their products. In the past, Locky downloaded the encryption program from a command & control server. In the latest version the encryption program is embedded in the email attachment as strings. The moment the victim runs the loader, the encryption program is extracted from the strings to the User Space and executed from User Space.

This is no rocket science; simply the application of well-known obfuscation methods to the latest Locky variant.

And, with AppGuard installed on top of the security stack, this new Locky variant represents no real danger.

In my opinion, the next generation endpoint protection solutions available on the market will all deal effectively with this sort of zero-day malware. The example of AppGuard shows: It is simply install and forget.

With this, we will gain valuable time for the right and important things like the implementation of Two Factor Authentication or privileged accounts management, or the design of effective security procedures or user training. Unfortunately, the paradigm shift from prevention to detection prevents us from implementing and doing the right and important things. It’s time for a paradigm change…

Have a good weekend.

AppGuard is an important part of a comprehensive security stack

16 July 2016

In the past weeks I tried hard to get an idea of the capabilities of Blue Ridge Networks AppGuard. To be honest, I would not like to miss AppGuard anymore. AppGuard creates the really good feeling that, under certain conditions, many cyber-attacks are simply rendered ineffective.

AppGuard is a perfect means against all kind of Trojans and downloaders, in particular zero days. Characteristic for this kind of malware is that the malware directly drops a malicious program or downloads a malicious program from the attacker’s server and executes it afterwards. This includes e.g. most of the known Ransomware.

The User Space and MemoryGuard concept just blocks this kind of malware out-of-the-box, provided that the User Space concept is not undermined by a user who is working with high privileges permanently. In fact, if the user works with privileges which allow the Trojan program to store files outside the User Space, the concept will no longer work.

It is strongly recommended to work with the least possible privileges under normal conditions. For the case higher privileges are requested, set up an extra account with the required privileges and supply the credentials of this account if UAC requests higher privileges.

More advanced malware may try to use the Windows auto-elevation feature to acquire higher privileges and to compromise AppGuard. To protect from auto-elevation attacks just set UAC to ‘Always notify me’.

This works even in the case of a gaming computer, where e.g. WOW and TeamSpeak are heavily used. Why shouldn’t it work on a standard system?

In addition, it is strongly recommended to disable macro execution in all kind of office software, e.g. Microsoft Office, OpenOffice or LibreOffice.

Memory Guard protects against all kind of zero-day drive-by downloads, PuP (Potentially unwanted Programs) or file-less malware.

My comprehensive security stack

My comprehensive security stack. Click to enlarge.

 

AppGuard does not protect against any kind of password phishing attacks. Although popular internet browsers block many malicious URLs through URL reputation, e.g. SmartScreen Filtering in Internet Explorer or Firefox, this will not protect in the case of zero-days.

To reduce the likelihood of credential theft, turn on Two-Factor Authentication (TFA) for as many as possible internet services you use. If TFA cannot be enabled, choose a strong password and take care, means:

User awareness is the basic part of the entire security stack!

To put it succinctly: The proposed security stack will dramatically reduce the risk of cyber-attacks. Blue Ridge Networks AppGuard is an important component of this stack, in particular for the protection against all kind of zero-days.

Have a good weekend.

AppGuard successfully protects against PowerShell based zero-day malware

9 July 2016

To get a feel for the impact AppGuard has on daily operations I worked mainly on my test system in the past weeks. My test system is a 6 years old Dell Inspiron 1445 with 4 GB of RAM and a 240 GB SSD.  The latest version of Windows 10 is deployed and all out-of-the-box Windows security options like Windows Defender and SmartScreen are activated.

I work with standard user rights; UAC is set to ‘Always notify me’. Macro protection for the office suite is set to ‘Disable all macros with notification’. AppGuard is installed on top of this security stack to protect from all kind of zero days.

The impact on my daily work is hardly noticeable. Standard malware is blocked either by Defender or by SmartScreen. Even the download of e.g. JavaScript based malware from malwr.com for test purposes is a challenging task.

AppGuard does a really good job in blocking the execution all kind of zero-day malware from user space. But how well works AppGuard in the case of somewhat more advanced malware?

I searched for a new PowerShell based malware on malwr.com and found Invoice_201604469.doc.

A check on VirusTotal showed that only 3 of 56 anti-malware products identified malware:

Antivirus Result Update
Fortinet WM/Poseket.A!tr.dldr 20160706
Qihoo-360 heur.macro.powershell.a 20160706
Symantec W97M.Downloader 20160706

As always, the AutoOpen macro is password protected. But LibreOffice overrides the password protection and reveals a master piece of code:

AutoOpen Macro with Powershell code

AutoOpen Macro with PowerShell Code

I opened the document and followed the instructions to execute the AutoOpen macro.

Invoice_201604469.doc

Invoice_201604469.doc. Click to enlarge.

The effect was enormous. AppGuard’s MemoryGuard blocked the execution of the PowerShell script and prevented the download of the payload 18293.exe:

Blocked Program Message

Blocked Program Message 1

Blocked Program Message

Blocked Program Message 2

Thus the command shell wasn’t able to start the payload and Windows displayed the last error message:

Windows Error Message

Windows Error Message

MemoryGuard is a really charming concept, and out-of-the-box available after installation.

This concludes my tests. The experiments of the past weeks show that User Space and MemoryGuard are useful security features. They complete the Windows built-in security features, and provide additional protection, in particular in the case of zero-day malware.

Have a good weekend.