Skip to navigation

Threat reportIntrusionTL-2026-1269

SQL Injection Leads to MSSQL Compromise, BadIIS Backdoor, and XMRig Deployment

highACTIVE

Threat actors exploited a SQL injection flaw in an unvalidated web application input to compromise a technology-sector organization's Microsoft SQL Server, then pivoted to create a rogue admin account, enable RDP, disable Windows Defender, install malicious BadIIS IIS modules, and deploy the XMRig cryptominer via nssm.exe.

Key facts

IdentifiersTL-2026-1269
Also known asSQLi-to-BadIIS-XMRig intrusion, Home Field Advantage intrusion
SeverityHIGH
Affected productsMicrosoft SQL Server; Microsoft Internet Information Services (IIS)
Weaknesses (CWE)CWE-89, CWE-284, CWE-269, CWE-506
AttributionUAT-8099 (low confidence)
Nation-state nexusChina
MalwareIISpy, xmrig
ToolsCnCrypt Protect, nssm.exe
MITRE ATT&CK techniquesT1005 Data from Local System, T1007 System Service Discovery, T1021 Remote Services, T1027 Obfuscated Files or Information, T1036 Masquerading, T1041 Exfiltration Over C2 Channel, T1057 Process Discovery, T1059 Command and Scripting Interpreter, T1069 Permission Groups Discovery, T1071 Application Layer Protocol, T1078 Valid Accounts, T1090 Proxy and 14 more in the attack flow
MotivationFINANCIAL
Target sectorstechnology
Target regionsGlobal
Detection rules9 (1 shown free)
Sources12 cited sources

Detection coverage

As of 2026-07-13, Threadlinqs Intelligence publishes 9 detection rule(s) for TL-2026-1269 across Splunk SPL, Microsoft KQL and Sigma, covering 21 indicator(s) of compromise. The whole corpus is readable without an account. Every threat comes with one complete detection in Splunk SPL, Microsoft KQL and Sigma, free for everyone; a free account adds daily TL Creds to unlock the rest. Paid plans add a larger daily allowance, correlation and the MCP server. Threadlinqs MCP server · View plans.

Free detection: Rogue Local Admin Account Creation Followed by RDP Enablement and Defender Disable

Detects creation of a service-account-styled local administrator account (e.g. trailing-$ naming like adminweb2$) followed closely by Terminal Services/RDP enablement and Windows Defender real-time protection being disabled — the persistence and defense-evasion chain observed after the SQLi foothold.

  • Languages: Splunk SPL, Microsoft KQL, Sigma
  • ATT&CK: T1136, T1078, T1133, T1021
  • Data sources: windows_security_event_log, windows_system_event_log, edr_registry_telemetry
  • Severity: critical
  • Confidence: high

Splunk SPL

index=wineventlog sourcetype="WinEventLog:Security" (EventCode=4720 OR (EventCode=4732 AND Group_Name="Administrators"))
| eval t=_time
| join type=inner ComputerName [ search index=wineventlog sourcetype="WinEventLog:Security" OR sourcetype="WinEventLog:System" (EventCode=7040 CommandLine="*TermService*") OR (EventCode=5001 Product="*Defender*")
    | eval t2=_time | table ComputerName, t2, EventCode ]
| where t2>=t AND t2<=t+1800
| table t, ComputerName, TargetUserName, Group_Name, t2, EventCode

Microsoft KQL

DeviceEvents
| where Timestamp > ago(24h)
| where ActionType in ("UserAccountCreated", "UserAccountAddedToLocalGroup") and AdditionalFields has "Administrators"
| project AcctTime = Timestamp, DeviceName, AccountName, ActionType
| join kind=inner (
    DeviceRegistryEvents
    | where Timestamp > ago(24h)
    | where RegistryKey has "Terminal Server" and RegistryValueName == "fDenyTSConnections"
    | project RdpTime = Timestamp, DeviceName2 = DeviceName
) on $left.DeviceName == $right.DeviceName2
| where RdpTime between (AcctTime .. AcctTime + 30m)
| project AcctTime, RdpTime, DeviceName, AccountName

Sigma

title: New Local Admin Account Followed by RDP Enablement Within 30 Minutes
id: 8e6a1f6a-1269-4a01-9c1a-badiis0000009
status: experimental
description: Detects a new local administrator account creation followed within 30 minutes by Terminal Services/RDP being enabled, matching the adminweb2$ persistence pattern observed in this intrusion.
references:
    - https://www.huntress.com/blog/sql-injection-attacker-persistence
author: Threadlinqs Team
date: 2026/07/13
tags:
    - attack.persistence
    - attack.privilege-escalation
    - attack.lateral-movement
    - attack.t1136
    - attack.t1078
    - attack.t1133
    - attack.t1021.001
logsource:
    category: process_creation
    product: windows
detection:
    selection:
        Image|endswith: '\net.exe'
        CommandLine|contains|all:
            - 'user'
            - '/add'
    selection2:
        Image|endswith: '\net.exe'
        CommandLine|contains|all:
            - 'localgroup'
            - 'administrators'
    condition: 1 of selection*
falsepositives:
    - Legitimate IT onboarding scripts creating and elevating new local admin accounts, especially service-account provisioning
level: high

Known false positives:

  • Legitimate IT provisioning of a new local admin/service account combined with an unrelated coincidental RDP configuration change

Tuning notes: Flag account names matching service-account conventions (trailing $, generic prefixes like adminweb/svc/admin) with elevated weight; require change-management ticket correlation to suppress authorized provisioning.

8 more detection rules (Splunk SPL, Microsoft KQL, Sigma) · unlock with TL Creds. Sign up free to unlock the detection rules or query them with the Threadlinqs MCP server (Purple)
  • Recon Command Followed by OAST/Collaborator-Style HTTP Exfil · unlock with TL Creds
  • Malicious IIS Module Installed via appcmd.exe (BadIIS Pattern) · unlock with TL Creds
  • File Hidden via attrib.exe Under Masquerading Vendor Update Path · unlock with TL Creds
  • Outbound HTTP/DNS to oastify.com OAST Infrastructure from Server Tier · unlock with TL Creds
  • Defender Real-Time Protection Disabled Shortly After New Local Admin Account Creation · unlock with TL Creds
  • PowerShell Hidden Non-Interactive Execution With Bypass Policy · unlock with TL Creds
  • SQL Server Process Spawning Command Shell (SQLi-to-RCE) · unlock with TL Creds
  • Hidden PowerShell Download From Public Object-Storage Bucket Followed by nssm Service Registration · unlock with TL Creds

Indicators of compromise

As of 2026-07-13, Threadlinqs Intelligence tracks 21 indicators of compromise for TL-2026-1269.

Sample indicators (2 of 21)

  • Domain: 334thribetlhkyo977gqrcht1k7bvdj2[.]oastify[.]com
  • Domain: pub-c4c8e8c336c3429d97195076bf3bb6eb[.]r2[.]dev

Defanged, TLP:CLEAR. Full set of 21 indicators: unlock with TL Creds.

21 indicators of compromise · unlock with TL Creds. Compare plans

Technical analysis

Huntress DFIR investigators observed a confirmed intrusion beginning with SQL injection against an IIS-hosted web application where user-supplied input was not validated before reaching the backend Microsoft SQL Server. The injection allowed the attacker to reach OS command execution, observed as sqlservr.exe spawning cmd.exe — consistent with abuse of MSSQL command-execution stored procedures (e.g., xp_cmdshell) reachable from an injectable query. Once on the host, the actor ran reconnaissance (tasklist /svc) and exfiltrated the output via an HTTP POST to an out-of-band interaction domain (334thribetlhkyo977gqrcht1k7bvdj2.oastify.com), a Burp Suite Collaborator/OAST-style subdomain typically used for blind SQLi/SSRF confirmation, indicating the intrusion may itself have originated from automated or semi-automated SQLi tooling reuse, or that the actor repurposed an OAST domain for DNS/HTTP beaconing and exfiltration. The actor created a new local administrator account (adminweb2$), enabled Remote Desktop Services/Terminal Services for interactive follow-on access, and disabled Windows Defender to prepare the host for further tooling. For defense evasion the actor used attrib.exe to mark dropped cryptominer files and directories as system, hidden, archive, and read-only, staging them inside a masquerading path under C:\Program Files (x86)\Microsoft\EdgeUpdate\ to blend in with a legitimate Microsoft update service. The actor installed two malicious IIS modules — HttpFastCgiModule.dll and HttpCgiModule.dll — via appcmd.exe, consistent with the BadIIS malware family (tracked publicly by Cisco Talos as used by the China-nexus group UAT-8099/REF4033 for SEO fraud, illicit traffic redirection, reverse-proxying, and content hijacking on compromised IIS servers worldwide). BadIIS integrates into the IIS request pipeline as a native module, conditionally serving keyword-stuffed content to search crawlers while redirecting real visitors to attacker-controlled destinations (gambling, adult content, crypto-phishing) and can also facilitate credential theft — giving the actor a durable, web-server-native backdoor independent of the initial SQLi foothold. For financial impact, the actor downloaded PowerShell scripts (qdcjoke1.2.ps1, c_joke1.2.ps1) and a batch launcher (qd_tjoke.bat) from a Cloudflare R2 public bucket (pub-c4c8e8c336c3429d97195076bf3bb6eb.r2.dev), executed with powershell.exe -NoLogo -NoProfile -NonInteractive -ExecutionPolicy Bypass -WindowStyle Hidden to suppress visible windows and logging prompts. These scripts staged the XMRig Monero cryptominer (delivered as xmr-1.zip) and registered it as a Windows service using nssm.exe (Non-Sucking Service Manager), a legitimate open-source service-wrapper tool abused here to auto-restart the miner process and blend in among normal Windows services. The actor additionally installed a tool referred to as "CnCrypt Protect" alongside the miner, likely intended to hinder analysis/detection of the miner binary or its configuration. No CVE was assigned by the source reporting since the vulnerability is a generic unvalidated-input SQL injection rather than a specific product flaw; no formal CVSS score was published. The concrete, actionable artifacts — the rogue account name, IIS module filenames, staging paths, script names, and the OAST/R2 delivery infrastructure — make this intrusion highly suited to detection-engineering coverage despite the lack of a CVE/CVSS anchor.

MITRE ATT&CK techniques by tactic:

Collection

T1005 Data from Local System

Discovery

T1007 System Service Discovery; T1057 Process Discovery; T1069 Permission Groups Discovery

Lateral Movement

T1021 Remote Services

Defense Evasion

T1027 Obfuscated Files or Information; T1036 Masquerading; T1218 System Binary Proxy Execution; T1564 Hide Artifacts

Exfiltration

T1041 Exfiltration Over C2 Channel; T1567 Exfiltration Over Web Service

Execution

T1059 Command and Scripting Interpreter; T1569 System Services

Command and Control

T1071 Application Layer Protocol; T1090 Proxy; T1102 Web Service; T1105 Ingress Tool Transfer

Privilege Escalation

T1078 Valid Accounts

Persistence

T1133 External Remote Services; T1136 Create Account; T1505 Server Software Component; T1543 Create or Modify System Process

Initial Access

T1190 Exploit Public-Facing Application

Impact

T1496 Resource Hijacking

Credential Access

T1557 Adversary-in-the-Middle

defense-impairment

T1685 Disable or Modify Tools

Remediation

Patches

  • No vendor patch applicable — remediate the application-layer input validation defect, not a specific CVE

Immediate actions

  • Validate and parameterize all SQL queries; remove any dynamic string-concatenated SQL built from unvalidated web input
  • Disable or tightly restrict xp_cmdshell and other MSSQL command-execution stored procedures at the database engine level
  • Audit local Administrators group membership and remove unauthorized accounts such as adminweb2$ or similarly named service-like accounts
  • Disable RDP/Terminal Services on internet-facing or database-tier servers unless explicitly required and MFA-protected
  • Re-enable and verify Windows Defender / EDR real-time protection and tamper protection on affected hosts
  • Inspect IIS applicationHost.config / web.config for unauthorized <globalModules> or <modules> entries referencing HttpFastCgiModule.dll, HttpCgiModule.dll, or unrecognized native/managed modules; remove via appcmd.exe
  • Hunt for hidden/system-attributed files under C:\Program Files (x86)\Microsoft\EdgeUpdate\ and other legitimate-looking update directories
  • Identify and remove unauthorized nssm.exe-registered Windows services, especially those launching unsigned or renamed binaries
  • Block outbound traffic to *.oastify.com and the identified R2 bucket subdomain at the network egress layer pending investigation

Workarounds

  • Restrict application database accounts to parameterized stored procedures only; deny ad hoc SQL execution rights
  • Apply Attack Surface Reduction (ASR) rules to block LOLBin abuse patterns (e.g., attrib.exe hiding files, appcmd.exe module manipulation from unexpected parent processes)

Longer-term hardening

  • Deploy a web application firewall (WAF) in front of internet-facing applications with SQLi rule sets tuned to the application
  • Implement least-privilege SQL service accounts and disable unnecessary MSSQL surface area (Surface Area Configuration / sp_configure)
  • Deploy EDR with behavioral detection for LOLBin abuse (attrib.exe, appcmd.exe, nssm.exe) and PowerShell script-block logging
  • Establish asset inventory and a formal patch/config management process for internet-facing SQL and IIS infrastructure
  • Implement network segmentation so database-tier hosts cannot reach arbitrary outbound HTTP/DNS destinations
  • Enforce MFA on all remote access and administrative authentication paths

Affected and fixed versions

  • Microsoft — SQL Server
    Vulnerable versions: unspecified/any version reachable via unvalidated application input
    Fixed in: N/A — application-layer input validation fix required
  • Microsoft — Internet Information Services (IIS)
    Vulnerable versions: unspecified version hosting the vulnerable application and IIS module pipeline
    Fixed in: N/A — remove unauthorized modules; not a version-specific vulnerability

Timeline

  • Huntress publishes DFIR report 'Home Field Advantage: How Attackers Reshape Victim Environments' documenting the confirmed intrusion at a technology-sector organization.
  • Attacker registers the XMRig miner as a resilient Windows service using nssm.exe (Non-Sucking Service Manager) to auto-restart the miner process.
  • Attacker deploys XMRig cryptominer (locked-ioc-9) and installs the CnCrypt Protect anti-analysis/protection tool alongside it.
  • Attacker uses attrib.exe to set cryptominer component files/directories to system, hidden, archive, read-only under a masquerading path in locked-ioc-1.
  • Attacker downloads PowerShell scripts locked-ioc-6 and locked-ioc-7 plus batch file locked-ioc-8 from Cloudflare R2 bucket locked-ioc-3, executed hidden via powershell.exe with -ExecutionPolicy Bypass -WindowStyle Hidden.
  • Attacker installs malicious IIS modules locked-ioc-4 and locked-ioc-5 via appcmd.exe, consistent with the BadIIS malware family used for SEO fraud/traffic hijacking.
  • Attacker disables Windows Defender to prevent detection of subsequent tooling.
  • Attacker enables Terminal Services/Remote Desktop Services to allow interactive follow-on access using the new admin account.
  • Attacker creates rogue local administrator account adminweb2$ and adds it to the local Administrators group.
  • Attacker runs tasklist /svc for service enumeration and exfiltrates output via HTTP POST to locked-ioc-2.
  • Attacker exploits SQL injection in an unvalidated web application input field, gaining code execution on the backend Microsoft SQL Server (sqlservr.exe spawns cmd.exe).

Sources cited

More from UAT-8099

Drafted by the Threadlinqs research pipeline from 12 sources; published ; last updated —.

Threadlinqs Intelligence — Real-Time Threat Detection Platform

// threat_feed
$ sort --newest
Showing all threats

Live intelligence console

Threat level

Latest Threats