Auteur Sujet: [FireEye]GAME OVER: Detecting and Stopping an APT41 Operation  (Lu 2493 fois)

0 Membres et 1 Invité sur ce sujet

Hors ligne igor51

  • Admin
  • Mega Power Members
  • *****
  • Messages: 10419
GAME OVER: Detecting and Stopping an APT41 Operation

In August 2019, FireEye     href="/content/fireeye-www/en_US/blog/threat-research/2019/08/apt41-dual-espionage-and-cyber-crime-operation.html">released
    the “Double Dragon” report on our newest graduated threat group,
  APT41. A China-nexus dual espionage and financially-focused group,
  APT41 targets industries such as gaming, healthcare, high-tech, higher
  education, telecommunications, and travel services. APT41 is known to
  adapt quickly to changes and detections within victim environments,
  often recompiling malware within hours of incident responder activity.
  In multiple situations, we also identified APT41 utilizing
  recently-disclosed vulnerabilities, often weaponzing and exploiting
  within a matter of days.


 

Our knowledge of this group’s targets and activities are rooted in
  our Incident Response and Managed Defense services, where we encounter
  actors like APT41 on a regular basis. At each encounter, FireEye works
  to reverse malware, collect intelligence and hone our detection
  capabilities. This ultimately feeds back into our Managed Defense and
  Incident Response teams detecting and stopping threat actors earlier
  in their campaigns.


 

In this blog post, we’re going to examine a recent instance where     href="https://www.fireeye.com/solutions/managed-defense.html">FireEye
    Managed Defense came toe-to-toe with APT41. Our goal is to
  display not only how dynamic this group can be, but also how the
  various teams within FireEye worked to thwart attacks within hours of
  detection – protecting our clients’ networks and limiting the threat
  actor’s ability to gain a foothold and/or prevent data exposure.


 

GET TO DA CHOPPA!


 

In April 2019, FireEye’s Managed Defense team identified suspicious
  activity on a publicly-accessible web server at a U.S.-based research
  university. This activity, a snippet of which is provided in Figure 1,
  indicated that the attackers were exploiting   href="https://nvd.nist.gov/vuln/detail/CVE-2019-3396">CVE-2019-3396,
  a vulnerability in Atlassian Confluence Server that allowed for path
  traversal and remote code execution.


 


 
 
 Figure 1: Snippet of PCAP showing
    attacker attempting CVE-2019-3396 vulnerability


 

This vulnerability relies on the following actions by the attacker:


 
  • Customizing the     class="code">_template field to utilize a template that
        allowed for command execution.
  • Inserting a     class="code">cmd field that provided the command to be
      executed.

 

Through custom JSON POST requests, the attackers were able to run
  commands and force the vulnerable system to download an additional
  file. Figure 2 provides a list of the JSON data sent by the attacker.


 


 
 
 Figure 2: Snippet of HTTP POST requests
    exploiting CVE-2019-3396


 

As shown in Figure 2, the attacker utilized a template located at
      class="code">hxxps[:]//github[.]com/Yt1g3r/CVE-2019-3396_EXP/blob/master/cmd.vm.
  This publicly-available template provided a vehicle for the attacker
  to issue arbitrary commands against the vulnerable system. Figure 3
  provides the code of the file cmd.vm.


 


 
 
 Figure 3: Code of cmd.vm, used by the
    attackers to execute code on a vulnerable Confluence system


 

The HTTP POST requests in Figure 2, which originated from the IP
  address 67.229.97[.]229, performed system
  reconnaissance and utilized Windows certutil.exe to download a file
  located at   class="code">hxxp[:]//67.229.97[.]229/pass_sqzr.jsp and save it
  as test.jsp (MD5:
  84d6e4ba1f4268e50810dacc7bbc3935
). The file   class="code">test.jsp was ultimately identified to be a variant
  of a       href="https://www.fireeye.com/blog/threat-research/2013/08/breaking-down-the-china-chopper-web-shell-part-i.html">China
    Chopper
webshell.


 

A Passive Aggressive Operation


 

Shortly after placing test.jsp on the
  vulnerable system, the attackers downloaded two additional files onto
  the system:


 

  •     64.dat (MD5:
      51e06382a88eb09639e1bc3565b444a6
    )

  •     Ins64.exe (MD5:
      e42555b218248d1a2ba92c1532ef6786
    )

 

Both files were hosted at the same IP address utilized by the
  attacker, 67[.]229[.]97[.]229. The file
    Ins64.exe was used to deploy the HIGHNOON
  backdoor on the system. HIGHNOON is a backdoor that consists of
  multiple components, including a loader, dynamic-link library (DLL),
  and a rootkit. When loaded, the DLL may deploy one of two embedded
  drivers to conceal network traffic and communicate with its command
  and control server to download and launch memory-resident DLL plugins.
  This particular variant of HIGHNOON is tracked as HIGHNOON.PASSIVE by
  FireEye. (An exploration of passive backdoors and more analysis of the
  HIGHNOON malware family can be found in our full     href="https://content.fireeye.com/apt-41/rpt-apt41">APT41 report).


 

Within the next 35 minutes, the attackers utilized both the   class="code">test.jsp web shell and the HIGHNOON backdoor to
  issue commands to the system. As China Chopper relies on HTTP
  requests, attacker traffic to and from this web shell was easily
  observed via network monitoring. The attacker utilized China Chopper
  to perform the following:


 
  • Movement of     class="code">64.dat and Ins64.exe
        to C:\Program
      Files\Atlassian\Confluence
  • Performing a directory
        listing of C:\Program
      Files\Atlassian\Confluence
  • Performing a directory
        listing of C:\Users

 

Additionally, FireEye’s FLARE team reverse engineered the custom
  protocol utilized by the HIGHNOON backdoor, allowing us to decode the
  attacker’s traffic. Figure 4 provides a list of the various commands
  issued by the attacker utilizing HIGHNOON.


 


 
 
 Figure 4: Decoded HIGHNOON commands
    issued by the attacker


 

Playing Their ACEHASH Card


 

As shown in Figure 4, the attacker utilized the HIGHNOON backdoor to
  execute a PowerShell command that downloaded a script from   href="https://github.com/PowerShellMafia/PowerSploit">PowerSploit,
  a well-known PowerShell Post-Exploitation Framework. At the time of
  this blog post, the script was no longer available for downloading.
  The commands provided to the script – “    class="code">privilege::debug sekurlsa::logonpasswords exit
  exit” – indicate that the unrecovered script was likely a copy
  of   href="https://raw.githubusercontent.com/mattifestation/PowerSploit/master/Exfiltration/Invoke-Mimikatz.ps1">Invoke-Mimikatz,
  reflectively loading Mimikatz 2.0 in-memory. Per the observed HIGHNOON
  output, this command failed.


 

After performing some additional reconnaissance, the attacker
  utilized HIGHNOON to download two additional files into the     class="code">C:\Program Files\Atlassian\Confluence directory:


 

  •     c64.exe (MD5:
      846cdb921841ac671c86350d494abf9c
    )

  •     F64.data (MD5:
      a919b4454679ef60b39c82bd686ed141
    )

 

These two files are the dropper and encrypted/compressed payload
  components, respectively, of a malware family known as ACEHASH.
  ACEHASH is a credential theft and password dumping utility that
  combines the functionality of multiple tools such as Mimikatz,
  hashdump, and Windows Credential Editor (WCE).


 

Upon placing c64.exe and   class="code">F64.data on the system, the attacker ran the command


 


  c64.exe f64.data "9839D7F1A0 -m”


 

This specific command provided a password of   class="code">“9839D7F1A0” to decrypt the contents of   class="code">F64.data, and a switch of   class="code">“-m”, indicating the attacker wanted to replicate
  the functionality of Mimikatz. With the correct password provided,
    c64.exe loaded the decrypted and
  decompressed shellcode into memory and harvested credentials.


 

Ultimately, the attacker was able to exploit a vulnerability,
  execute code, and download custom malware on the vulnerable Confluence
  system. While Mimikatz failed, via ACEHASH they were able to harvest a
  single credential from the system. However, as Managed Defense
  detected this activity rapidly via network signatures, this operation
  was neutralized before the attackers progressed any further.


 

Key Takeaways From This Incident


 
  • APT41 utilized multiple
        malware families to maintain access into this environment; impactful
        remediation requires full scoping of an incident.
         
  • For effective Managed Detection &
        Response services, having coverage of both Endpoint and Network is
        critical for detecting and responding to targeted attacks.
         
  • Attackers may weaponize vulnerabilities
        quickly after their release, especially if they are present within a
        targeted environment. Patching of critical vulnerabilities ASAP is
        crucial to deter active attackers.

 

Detecting the Techniques


 

FireEye detects this activity across our platform, including
  detection for certutil usage, HIGHNOON, and China Chopper.


 
   
     
   
     
   
     
   
     
   
     
   
     
   
     
   
     
   
     


          Detection


          Signature Name


          China Chopper

        valign="top">

FE_Webshell_JSP_CHOPPER_1


           

        valign="top">

FE_Webshell_Java_CHOPPER_1


           

        valign="top">

FE_Webshell_MSIL_CHOPPER_1


          HIGHNOON.PASSIVE

        valign="top">

FE_APT_Backdoor_Raw64_HIGHNOON_2


           

        valign="top">

FE_APT_Backdoor_Win64_HIGHNOON_2


          Certutil Downloader

          valign="top">

CERTUTIL.EXE DOWNLOADER (UTILITY)


           

          valign="top">

CERTUTIL.EXE DOWNLOADER A (UTILITY)


          ACEHASH

        valign="top">

FE_Trojan_AceHash


 

Indicators


 
   
     
   
     
   
     
   
     
   
     
   
     
   
              width="28%" valign="top">

67.229.97[.]229

        width="55%" valign="top">

N/A


          Type


          Indicator


          MD5 Hash (if applicable)

File

        valign="top">

test.jsp

        valign="top">

84d6e4ba1f4268e50810dacc7bbc3935

File

        valign="top">

64.dat

        valign="top">

51e06382a88eb09639e1bc3565b444a6

File

        valign="top">

Ins64.exe

        valign="top">

e42555b218248d1a2ba92c1532ef6786

File

        valign="top">

c64.exe

        valign="top">

846cdb921841ac671c86350d494abf9c

File

        valign="top">

F64.data

        valign="top">

a919b4454679ef60b39c82bd686ed141

IP Address


 

Looking for more?     href="https://www.brighttalk.com/webcast/7451/366611/double-dragon-apt41-a-dual-espionage-and-cyber-crime-operation">Join
    us for a webcast on August 29, 2019 where we detail more of
  APT41’s activities. You can also find a direct link to the public
  APT41 report here.


 

Acknowledgements


 

Thanks to Raymond Leong of Advanced Practices and Willi Ballenthin
  of FLARE for identification and reversing of the HIGHNOON.PASSIVE malware.


Source: GAME OVER: Detecting and Stopping an APT41 Operation

Tags: