WARPscan: Cloudflare WARP Abused to Hijack Cloud Services
Cado Security (now a part of Darktrace) found attackers are abusing Cloudflare's WARP service, a free VPN, to launch attacks. WARP traffic often bypasses firewalls due to Cloudflare's trusted status, making it harder to detect. Campaigns like "SSWW" cryptojacking and SSH brute-forcing exploit this trust, highlighting a significant security risk for organizations.
Darktrace cyber analysts are world-class experts in threat intelligence, threat hunting and incident response, and provide 24/7 SOC support to thousands of Darktrace customers around the globe. Inside the SOC is exclusively authored by these experts, providing analysis of cyber incidents and threat trends, based on real-world experience in the field.
Written by
Nate Bill
Threat Researcher
Share
17
Jul 2024
Researchers from Cado Security Labs (now part of Darktrace) have observed several recent campaigns making use of Cloudflare’s WARP[1] service in order to attack vulnerable internet-facing services. In this blog we will explain what Cloudflare WARP is, the implications for its use in opportunistic attacks, and provide a few case studies on real-world attacks taking advantage of WARP.
What is Cloudflare WARP?
Cloudflare WARP is effectively a Virtual Private Network (VPN) that uses Cloudflare’s international backbone to “optimize” user’s traffic. This is a free service, meaning anyone can download and use it for their own purposes. In practice, WARP just tunnels traffic to the nearest Cloudflare data center over a custom implementation of WireGuard, which they claim will speed up your connection.
Cloudflare WARP is designed to present the IP of the end user to Cloudflare CDN customers. However, attacks observed by Cado researchers exclusively connect directly to IP addresses rather than Cloudflare’s CDN, with the attacker in control of the transport and application layers. As such, it is not possible to determine the IP of the attackers.
Implications of attacks originating from WARP
Network administrators are far more likely to inherently trust or overlook traffic originating from Cloudflare’s ASN as it is not a common attack origin, and is often used in many organizations as a part of regular business operations. As a result of this, the IP ranges used by WARP may even be allowed in firewalls, and might be missed during triage of alerts by Security Operations Center (SOC) teams.
Cado Security has observed several threads on sysadmin forums, where network operators are advised to “allowlist” all of Cloudflare’s IP ranges instead of just those specific to a given service, which is a serious security risk that makes their infrastructure directly vulnerable to attackers using WARP to launch their attacks.
These factors make attacks using WARP potentially more dangerous unless an organization takes preventive action, such as educating security teams and ensuring WARP IP ranges are not included in Cloudflare related firewall rules.
Case study - SSWW mining campaign
The SSWW campaign is a novel cryptojacking campaign targeting exposed Docker which utilizes Cloudflare WARP for initial access. Based on the TLS certificate used by the C2 server, it would appear that the C2 was created on September 5, 2023. However, the first attack detected against Cado’s honeypot infrastructure was on February 21, 2024, which lines up with the dropped payload’s Last-Modified header of February 20, the day before. This is likely when the current campaign began.
The attack began with a container being created with elevated permissions, and access to the host. The image used is simply selected from images that are already available on the host, so the attacker does not have to download any new images.
The attacker then creates a Docker VND stream in order to run commands within the created container:
This downloads the main SSWW script from the attacker’s command and control (C2) infrastructure and sets it running. The SSWW script is fairly straightforward and does the following set up tasks:
Attempts to stop “systemd” services that belong to competing miners.
Exits if the system is already infected by the SSWW campaign.
Disables “SELinux”.
Sets up huge pages and enables drop_caches, common XMRig optimizations
Downloads https://94[.]131.107.38:58282/sst, an XMRig miner with embedded config, and saves it as /var/spool/.system
Attempts to download and compile https://94[.]131.107.38:58282/phsd2.c, which is a simple off-the-shelf process hider designed to hide the .system process. If this fails, it will download https://94[.]131.107.38:58282/li instead. The resultant binary of either of these processes is saved to /usr/lib/libsystemd-shared-165.so
Adds the above to /etc/ld.so.preload such that it acts as a usermode rootkit.
Saves https://94[.]131.107.38:58282/aa82822, a SystemD unit file for running /var/spool/.system, to /lib/systemd/system/cdngdn.service, and then enables it.
The configuration file can be extracted out of the miner, and observe that it is using the wallet address: 44EP4MrMADSYSxmN7r2EERgqYBeB5EuJ3FBEzBrczBRZZFZ7cKotTR5airkvCm2uJ82nZHu8U3YXbDXnBviLj3er7XDnMhP on the monero ocean gulf mining pool. We can then use the mining pool’s wallet lookup feature to determine the attacker has made a total of 9.57 XMR (~£1269 at time of writing).
While using Cloudflare WARP affords the attacker a layer of anonymity, we can see the IPs the attacks originate from are consistently deriving from the Cloudflare data center in Zagreb, Croatia. As Cloudflare WARP will use the nearest data center, this suggests that the attacker’s scan server is located in Croatia. The C2 IPs on the other hand are hosted using a Netherlands-based VPS provider.
The main benefit to the attacker of using Cloudflare WARP is likely the relative anonymity afforded by WARP, as well as the reduced suspicion around traffic related to Cloudflare. It is possible that some improperly configured systems that allow all Cloudflare traffic have been compromised as a result of this, however, it is not possible to say with certainty without having access to all compromised hosts infected by the malware.
Case study - opportunistic SSH attacks
Since 2022, Cado Security has been tracking SSH attacks originating from WARP addresses. Initially these were fairly limited, however around the end of 2023 they surged to a few thousand per month. These frequently rise and fall with quite a high velocity, suggesting that the surges are the result of individual campaigns rather than a more general trend.
Image 1: SSH attacks originating from WARP addresses since the end of 2023
Interestingly, a number of SSH campaigns we have seen previously originating from commonly abused VPS providers now appear to have migrated to using Cloudflare WARP. As these VPS providers are soft on abuse, it is unlikely that the purpose of this was for anonymity. Instead, the attackers are likely trying to take advantage of Cloudflare’s “clean” IP ranges (many “dirty” ranges belonging to bulletproof hosting are blocklisted, e.g. by spamhaus [2]), as well as the higher likelihood of the Cloudflare ranges being overlooked or blindly allowed in the victim’s firewall.
All of the attacks seen so far from Cloudflare WARP appear to be simple SSH brute forcing attacks, however it is alleged that the recent CVE-2024-6387 is now being exploited in the wild [3]. An attacker could perform this exploit via Cloudflare WARP in order to take advantage of overly trusting firewalls to attack organizations that may not otherwise have the vulnerable SSH server exposed.
Conclusion
The main threat posed by attackers using Cloudflare’s WARP service is the inherent trust administrators may have in traffic originating from Cloudflare, and the dangerous advice to “allow all Cloudflare IPs” being circulated online. Ensure your organization has not granted permission for 104[.]28.0.0/16 in your firewall. Follow a defense in-depth approach and additionally ensure services such as SSH have strong authentication (via SSH keys instead of passwords) and are up-to-date. Do not expose Docker to the internet, even if it is behind a firewall.
Darktrace cyber analysts are world-class experts in threat intelligence, threat hunting and incident response, and provide 24/7 SOC support to thousands of Darktrace customers around the globe. Inside the SOC is exclusively authored by these experts, providing analysis of cyber incidents and threat trends, based on real-world experience in the field.
How Organizations are Addressing Cloud Investigation and Response
The importance of cloud investigation and incident response are compounded by an expanded attack surface in the cloud, lack of advanced tooling to upskill teams, and increasing regulatory pressure from compliance regulations. This blog dives into these challenges and explores potential solutions for security teams attempting to secure their cloud environment
Minimizing Permissions for Cloud Forensics: A Practical Guide to Tightening Access in the Cloud
Most cloud environments struggle to strike the right balance between security and accessibility. This blog breaks down why traditional approaches to cloud forensics often fail and outlines practical, security-first strategies to solve the access dilemma. You’ll learn how to enable effective investigations without over-permissioning your environment.
How CDR & Automated Forensics Transform Cloud Incident Response
This blog walks through an example of how Darktrace’s CDR and automated cloud forensics capabilities automate cloud detection, and deep forensic investigation in a way that’s fast, scalable, and deeply insightful.
Rethinking Signature-Based Detection for Power Utility Cybersecurity
Lessons learned from OT cyber attacks
Over the past decade, some of the most disruptive attacks on power utilities have shown the limits of signature-based detection and reshaped how defenders think about OT security. Each incident reinforced that signatures are too narrow and reactive to serve as the foundation of defense.
2015: BlackEnergy 3 in Ukraine
According to CISA, on December 23, 2015, Ukrainian power companies experienced unscheduled power outages affecting a large number of customers — public reports indicate that the BlackEnergy malware was discovered on the companies’ computer networks.
2016: Industroyer/CrashOverride
CISA describes CrashOverride malwareas an “extensible platform” reported to have been used against critical infrastructure in Ukraine in 2016. It was capable of targeting industrial control systems using protocols such as IEC‑101, IEC‑104, and IEC‑61850, and fundamentally abused legitimate control system functionality to deliver destructive effects. CISA emphasizes that “traditional methods of detection may not be sufficient to detect infections prior to the malware execution” and recommends behavioral analysis techniques to identify precursor activity to CrashOverride.
2017: TRITON Malware
The U.S. Department of the Treasury reports that the Triton malware, also known as TRISIS or HatMan, was “designed specifically to target and manipulate industrial safety systems” in a petrochemical facility in the Middle East. The malware was engineered to control Safety Instrumented System (SIS) controllers responsible for emergency shutdown procedures. During the attack, several SIS controllers entered a failed‑safe state, which prevented the malware from fully executing.
The broader lessons
These events revealed three enduring truths:
Signatures have diminishing returns: BlackEnergy showed that while signatures can eventually identify adapted IT malware, they arrive too late to prevent OT disruption.
Behavioral monitoring is essential: CrashOverride demonstrated that adversaries abuse legitimate industrial protocols, making behavioral and anomaly detection more effective than traditional signature methods.
Critical safety systems are now targets: TRITON revealed that attackers are willing to compromise safety instrumented systems, elevating risks from operational disruption to potential physical harm.
The natural progression for utilities is clear. Static, file-based defenses are too fragile for the realities of OT.
These incidents showed that behavioral analytics and anomaly detection are far more effective at identifying suspicious activity across industrial systems, regardless of whether the malicious code has ever been seen before.
Strategic risks of overreliance on signatures
False sense of security: Believing signatures will block advanced threats can delay investment in more effective detection methods.
Resource drain: Constantly updating, tuning, and maintaining signature libraries consumes valuable staff resources without proportional benefit.
Adversary advantage: Nation-state and advanced actors understand the reactive nature of signature defenses and design attacks to circumvent them from the start.
Recommended Alternatives (with real-world OT examples)
Figure 1: Alternative strategies for detecting cyber attacks in OT
Behavioral and anomaly detection
Rather than relying on signatures, focusing on behavior enables detection of threats that have never been seen before—even trusted-looking devices.
Real-world insight:
In one OT setting, a vendor inadvertently left a Raspberry Pi on a customer’s ICS network. After deployment, Darktrace’s system flagged elastic anomalies in its HTTPS and DNS communication despite the absence of any known indicators of compromise. The alerting included sustained SSL increases, agent‑beacon activity, and DNS connections to unusual endpoints, revealing a possible supply‑chain or insider risk invisible to static tools.
Darktrace’s AI-driven threat detection aligns with the zero-trust principle of assuming the risk of a breach. By leveraging AI that learns an organization’s specific patterns of life, Darktrace provides a tailored security approach ideal for organizations with complex supply chains.
Threat intelligence sharing & building toward zero-trust philosophy
Frameworks such as MITRE ATT&CK for ICS provide a common language to map activity against known adversary tactics, helping teams prioritize detections and response strategies. Similarly, information-sharing communities like E-ISAC and regional ISACs give utilities visibility into the latest tactics, techniques, and procedures (TTPs) observed across the sector. This level of intel can help shift the focus away from chasing individual signatures and toward building resilience against how adversaries actually operate.
Real-world insight:
Darktrace’s AI embodies zero‑trust by assuming breach potential and continually evaluating all device behavior, even those deemed trusted. This approach allowed the detection of an anomalous SharePoint phishing attempt coming from a trusted supplier, intercepted by spotting subtle patterns rather than predefined rules. If a cloud account is compromised, unauthorized access to sensitive information could lead to extortion and lateral movement into mission-critical systems for more damaging attacks on critical-national infrastructure.
This reinforces the need to monitor behavioral deviations across the supply chain, not just known bad artifacts.
Defense-in-Depth with OT context & unified visibility
OT environments demand visibility that spans IT, OT, and IoT layers, supported by risk-based prioritization.
Moreover, by integrating contextual risk scoring, considering real-world exploitability, device criticality, firewall misconfiguration, and legacy hardware exposure, utilities can focus on the vulnerabilities that genuinely threaten uptime and safety, rather than being overwhelmed by CVE noise.
Regulatory alignment and positive direction
Industry regulations are beginning to reflect this evolution in strategy. NERC CIP-015 requires internal network monitoring that detects anomalies, and the standard references anomalies 15 times. In contrast, signature-based detection is not mentioned once.
This regulatory direction shows that compliance bodies understand the limitations of static defenses and are encouraging utilities to invest in anomaly-based monitoring and analytics. Utilities that adopt these approaches will not only be strengthening their resilience but also positioning themselves for regulatory compliance and operational success.
Conclusion
Signature-based detection retains utility for common IT malware, but it cannot serve as the backbone of security for power utilities. History has shown that major OT attacks are rarely stopped by signatures, since each campaign targets specific systems with customized tools. The most dangerous adversaries, from insiders to nation-states, actively design their operations to avoid detection by signature-based tools.
A more effective strategy prioritizes behavioral analytics, anomaly detection, and community-driven intelligence sharing. These approaches not only catch known threats, but also uncover the subtle anomalies and novel attack techniques that characterize tomorrow’s incidents.
From PowerShell to Payload: Darktrace’s Detection of a Novel Cryptomining Malware
What is Cryptojacking?
Cryptojacking remains one of the most persistent cyber threats in the digital age, showing no signs of slowing down. It involves the unauthorized use of a computer or device’s processing power to mine cryptocurrencies, often without the owner’s consent or knowledge, using cryptojacking scripts or cryptocurrency mining (cryptomining) malware [1].
Unlike other widespread attacks such as ransomware, which disrupt operations and block access to data, cryptomining malware steals and drains computing and energy resources for mining to reduce attacker’s personal costs and increase “profits” earned from mining [1]. The impact on targeted organizations can be significant, ranging from data privacy concerns and reduced productivity to higher energy bills.
As cryptocurrency continues to grow in popularity, as seen with the ongoing high valuation of the global cryptocurrency market capitalization (almost USD 4 trillion at time of writing), threat actors will continue to view cryptomining as a profitable venture [2]. As a result, illicit cryptominers are being used to steal processing power via supply chain attacks or browser injections, as seen in a recent cryptojacking campaign using JavaScript [3][4].
Therefore, security teams should maintain awareness of this ongoing threat, as what is often dismissed as a "compliance issue" can escalate into more severe compromises and lead to prolonged exposure of critical resources.
This blog will discuss Darktrace’s successful detection of the malicious activity, the role of Autonomous Response in halting the cryptojacking attack, include novel insights from Darktrace’s threat researchers on the cryptominer payload, showing how the attack chain was initiated through the execution of a PowerShell-based payload.
Darktrace’s Coverage of Cryptojacking via PowerShell
In July 2025, Darktrace detected and contained an attempted cryptojacking incident on the network of a customer in the retail and e-commerce industry.
The threat was detected when a threat actor attempted to use a PowerShell script to download and run NBMiner directly in memory.
The initial compromise was detected on July 22, when Darktrace / NETWORK observed the use of a new PowerShell user agent during a connection to an external endpoint, indicating an attempt at remote code execution.
Specifically, the targeted desktop device established a connection to the rare endpoint, 45.141.87[.]195, over destination port 8000 using HTTP as the application-layer protocol. Within this connection, Darktrace observed the presence of a PowerShell script in the URI, specifically ‘/infect.ps1’.
Darktrace’s analysis of this endpoint (45.141.87[.]195[:]8000/infect.ps1) and the payload it downloaded indicated it was a dropper used to deliver an obfuscated AutoIt loader. This attribution was further supported by open-source intelligence (OSINT) reporting [5]. The loader likely then injected NBMiner into a legitimate process on the customer’s environment – the first documented case of NBMiner being dropped in this way.
Figure 1: Darktrace’s detection of a device making an HTTP connection with new PowerShell user agent, indicating PowerShell abuse for command-and-control (C2) communications.
Script files are often used by malicious actors for malware distribution. In cryptojacking attacks specifically, scripts are used to download and install cryptomining software, which then attempts to connect to cryptomining pools to begin mining operations [6].
Inside the payload: Technical analysis of the malicious script and cryptomining loader
To confidently establish that the malicious script file dropped an AutoIt loader used to deliver the NBMiner cryptominer, Darktrace’s threat researchers reverse engineered the payload. Analysis of the file ‘infect.ps1’ revealed further insights, ultimately linking it to the execution of a cryptominer loader.
Figure 2: Screenshot of the ‘infect.ps1’ PowerShell script observed in the attack.
The ‘infect.ps1’ script is a heavily obfuscated PowerShell script that contains multiple variables of Base64 and XOR encoded data. The first data blob is XOR’d with a value of 97, after decoding, the data is a binary and stored in APPDATA/local/knzbsrgw.exe. The binary is AutoIT.exe, the legitimate executable of the AutoIt programming language. The script also performs a check for the existence of the registry key HKCU:\\Software\LordNet.
The second data blob ($cylcejlrqbgejqryxpck) is written to APPDATA\rauuq, where it will later be read and XOR decoded. The third data blob ($tlswqbblxmmr)decodes to an obfuscated AutoIt script, which is written to %LOCALAPPDATA%\qmsxehehhnnwioojlyegmdssiswak. To ensure persistence, a shortcut file named xxyntxsmitwgruxuwqzypomkhxhml.lnk is created to run at startup.
Figure 3: Screenshot of second stage AutoIt script.
The observed AutoIt script is a process injection loader. It reads an encrypted binary from /rauuq in APPDATA, then XOR-decodes every byte with the key 47 to reconstruct the payload in memory. Next, it silently launches the legitimate Windows app ‘charmap.exe’ (Character Map) and obtains a handle with full access. It allocates executable and writable memory inside that process, writes the decrypted payload into the allocated region, and starts a new thread at that address. Finally, it closes the thread and process handles.
The binary that is injected into charmap.exe is 64-bit Windows binary. On launch, it takes a snapshot of running processes and specifically checks whether Task Manager is open. If Task Manager is detected, the binary kills sigverif.exe; otherwise, it proceeds. Once the condition is met, NBMiner is retrieved from a Chimera URL (https://api[.]chimera-hosting[.]zip/frfnhis/zdpaGgLMav/nbminer[.]exe) and establishes persistence, ensuring that the process automatically restarts if terminated. When mining begins, it spawns a process with the arguments ‘-a kawpow -o asia.ravenminer.com:3838 -u R9KVhfjiqSuSVcpYw5G8VDayPkjSipbiMb.worker -i 60’ and hides the process window to evade detection.
Figure 4: Observed NBMiner arguments.
The program includes several evasion measures. It performs anti-sandboxing by sleeping to delay analysis and terminates sigverif.exe (File Signature Verification). It checks for installed antivirus products and continues only when Windows Defender is the sole protection. It also verifies whether the current user has administrative rights. If not, it attempts a User Account Control (UAC) bypass via Fodhelper to silently elevate and execute its payload without prompting the user. The binary creates a folder under %APPDATA%, drops rtworkq.dll extracted from its own embedded data, and copies ‘mfpmp.exe’ from System32 into that directory to side-load ‘rtworkq.dll’. It also looks for the registry key HKCU\Software\kap, creating it if it does not exist, and reads or sets a registry value it expects there.
Zooming Out: Darktrace Coverage of NBMiner
Darktrace’s analysis of the malicious PowerShell script provides clear evidence that the payload downloaded and executed the NBMiner cryptominer. Once executed, the infected device is expected to attempt connections to cryptomining endpoints (mining pools). Darktrace initially observed this on the targeted device once it started making DNS requests for a cryptominer endpoint, “gulf[.]moneroocean[.]stream” [7], one minute after the connection involving the malicious script.
Figure 5: Darktrace Advanced Search logs showcasing the affected device making a DNS request for a Monero mining endpoint.
Though DNS requests do not necessarily mean the device connected to a cryptominer-associated endpoint, Darktrace detected connections to the endpoint specified in the DNS Answer field: monerooceans[.]stream, 152.53.121[.]6. The attempted connections to this endpoint over port 10001 triggered several high-fidelity model alerts in Darktrace related to possible cryptomining mining activity. The IP address and destination port combination (152.53.121[.]6:10001) has also been linked to cryptomining activity by several OSINT security vendors [8][9].
Figure 6: Darktrace’s detection of a device establishing connections with the Monero Mining-associated endpoint, monerooceans[.]stream over port 10001.
Darktrace / NETWORK grouped together the observed indicators of compromise (IoCs) on the targeted device and triggered an additional Enhanced Monitoring model designed to identify activity indicative of the early stages of an attack. These high-fidelity models are continuously monitored and triaged by Darktrace’s SOC team as part of the Managed Threat Detection service, ensuring that subscribed customers are promptly notified of malicious activity as soon as it emerges.
Figure 7: Darktrace’s correlation of the initial PowerShell-related activity with the cryptomining endpoint, showcasing a pattern indicative of an initial attack chain.
Darktrace’s Cyber AI Analyst launched an autonomous investigation into the ongoing activity and was able to link the individual events of the attack, encompassing the initial connections involving the PowerShell script to the ultimate connections to the cryptomining endpoint, likely representing cryptomining activity. Rather than viewing these seemingly separate events in isolation, Cyber AI Analyst was able to see the bigger picture, providing comprehensive visibility over the attack.
Figure 8: Darktrace’s Cyber AI Analyst view illustrating the extent of the cryptojacking attack mapped against the Cyber Kill Chain.
Darktrace’s Autonomous Response
Fortunately, as this customer had Darktrace configured in Autonomous Response mode, Darktrace was able to take immediate action by preventing the device from making outbound connections and blocking specific connections to suspicious endpoints, thereby containing the attack.
Figure 9: Darktrace’s Autonomous Response actions automatically triggered based on the anomalous connections observed to suspicious endpoints.
Specifically, these Autonomous Response actions prevented the outgoing communication within seconds of the device attempting to connect to the rare endpoints.
Figure 10: Darktrace’s Autonomous Response blocked connections to the mining-related endpoint within a second of the initial connection.
Additionally, the Darktrace SOC team was able to validate the effectiveness of the Autonomous Response actions by analyzing connections to 152.53.121[.]6 using the Advanced Search feature. Across more than 130 connection attempts, Darktrace’s SOC confirmed that all were aborted, meaning no connections were successfully established.
Figure 11: Advanced Search logs showing all attempted connections that were successfully prevented by Darktrace’s Autonomous Response capability.
Conclusion
Cryptojacking attacks will remain prevalent, as threat actors can scale their attacks to infect multiple devices and networks. What’s more, cryptomining incidents can often be difficult to detect and are even overlooked as low-severity compliance events, potentially leading to data privacy issues and significant energy bills caused by misused processing power.
Darktrace’s anomaly-based approach to threat detection identifies early indicators of targeted attacks without relying on prior knowledge or IoCs. By continuously learning each device’s unique pattern of life, Darktrace can detect subtle deviations that may signal a compromise.
In this case, the cryptojacking attack was quickly identified and mitigated during the early stages of malware and cryptomining activity. Darktrace's Autonomous Response was able to swiftly contain the threat before it could advance further along the attack lifecycle, minimizing disruption and preventing the attack from potentially escalating into a more severe compromise.
Credit to Keanna Grelicha (Cyber Analyst) and Tara Gould (Threat Research Lead)
The content provided in this blog is published by Darktrace for general informational purposes only and reflects our understanding of cybersecurity topics, trends, incidents, and developments at the time of publication. While we strive to ensure accuracy and relevance, the information is provided “as is” without any representations or warranties, express or implied. Darktrace makes no guarantees regarding the completeness, accuracy, reliability, or timeliness of any information presented and expressly disclaims all warranties.
Nothing in this blog constitutes legal, technical, or professional advice, and readers should consult qualified professionals before acting on any information contained herein. Any references to third-party organizations, technologies, threat actors, or incidents are for informational purposes only and do not imply affiliation, endorsement, or recommendation.
Darktrace, its affiliates, employees, or agents shall not be held liable for any loss, damage, or harm arising from the use of or reliance on the information in this blog.
The cybersecurity landscape evolves rapidly, and blog content may become outdated or superseded. We reserve the right to update, modify, or remove any content without notice.