ICMP tunnelling is a technique that carries arbitrary data inside Internet Control Message Protocol packets. It uses fields in echo request and reply traffic, or similar ICMP messages, to move payloads past controls that mainly inspect TCP or UDP, which can hide command and control or data exfiltration.
What ICMP tunneling is used for
ICMP tunneling turns a control protocol into a covert transport layer. Because ICMP is commonly permitted for diagnostics and path testing, attackers can hide arbitrary payloads inside echo or similar messages to bypass filters that focus mainly on TCP and UDP. That makes it attractive for command and control, staged payload delivery, and quiet data transfer.
The technique is not about ICMP itself being malicious. The abuse comes from repurposing a protocol that many networks still allow, sometimes broadly, because blocking it outright can affect troubleshooting and monitoring. That is why ICMP tunneling is usually discussed alongside broader protocol tunnelling tradecraft.
How ICMP tunneling works
In a tunnel, one endpoint embeds data in ICMP packet fields and the other endpoint extracts it, reconstructing a bidirectional channel outside normal application flows. The payload may be encoded, fragmented, and paced to resemble ordinary diagnostic traffic. In practice, the tunnel can carry shell commands, proxy traffic, or exfiltrated fragments as long as both ends understand the encoding.
This is effective because security tooling often treats ICMP as low-value metadata rather than a carrier for business data. If inspection is limited to ports, sessions, or HTTP-style application protocols, the tunnel can slip past policy enforcement. For broader context on how defenders detect the surrounding command-and-control pattern, see the Command and Control (C2) Guide.
Why ICMP tunneling is difficult to spot
ICMP traffic is often low volume and intermittent, which helps a tunnel blend into background noise. It may also be masked by legitimate troubleshooting, latency checks, or path diagnostics, especially when the organisation lacks a baseline for normal ICMP use. A tunnel can therefore persist even when perimeter controls look healthy.
Detection usually depends on correlation rather than a single signature. Unusual request-reply symmetry, long-lived echo patterns, non-diagnostic payload sizes, and bursty traffic to uncommon destinations are all clues. For network telemetry and beacon-style analysis that complements this problem, the Beaconing and C2 Detection Guide is a useful reference.
How defenders reduce the attack surface
Defence is usually about narrowing where ICMP is allowed, not assuming it must be universally blocked. Egress filtering, strict ACLs, and controlled exception paths reduce the chance that arbitrary ICMP reaches the internet. Where diagnostics are needed, organisations should constrain ICMP to approved sources, destinations, and message types instead of permitting it network-wide.
Segmentation also matters because a tunnel is only useful if it can reach a host with internet access or an internal pivot point. Asset owners should know which systems genuinely need ICMP, which do not, and where monitoring must be strongest. That operational view fits naturally with the broader guidance in the Network Ports and Protocols Security Guide.
Risk and Threat Considerations
ICMP tunneling matters because it can bypass controls that are tuned to application ports rather than protocol abuse. It is a classic way to create a covert channel for command and control or exfiltration when outbound filtering is too permissive or when ICMP is trusted by default.
Failure mechanism: Security tools may allow ICMP for diagnostics, but fail to inspect payload content, correlate packet timing, or enforce policy on message types, letting a hidden channel survive normal network controls.
Impact: An attacker can maintain remote control, move data out of the environment, and keep a foothold even after obvious web or DNS channels are blocked.
Standards & Framework Alignment
This section maps relevant standards and security frameworks to the operational risks and controls described in this guidance.
MITRE ATT&CK addresses the attack and risk surface, while CIS Controls v8 and NIST CSF 2.0 set the governance and control requirements practitioners need to meet.
| Framework | Control / Reference | Relevance |
|---|---|---|
| MITRE ATT&CK | T1095 — Non-Application Layer Protocol | ICMP tunneling is a non-application-layer covert channel used for C2 and exfiltration. |
| T1071 — Application Layer Protocol | Tunneling hides C2 inside common protocols, making protocol-abuse tradecraft directly relevant. | |
| Recommendation — Map ICMP-only channels to T1095 and hunt for abnormal ICMP payloads and timing. Correlate suspicious ICMP flows with adjacent protocol-abuse telemetry and block unsolicited egress. | ||
| CIS Controls v8 | CIS-12 — Network Infrastructure Management | Network filtering and segmentation are central to constraining covert protocol abuse. |
| CIS-13 — Network Monitoring and Defense | Detection of covert tunnels depends on continuous network monitoring and anomaly review. | |
| Recommendation — Restrict ICMP at boundaries and segment systems that do not require it. Monitor ICMP baselines and alert on unusual volume, size, timing, or destinations. | ||
| NIST CSF 2.0 | PR.DS-01 — Data-at-rest is protected | Covert ICMP tunnels create a data-exposure path that protection controls must limit. |
| DE.CM-01 — Networks and network services are monitored to find potentially adverse events | ICMP tunneling is detected through network monitoring of unusual service behavior. | |
| Recommendation — Protect sensitive outbound data paths and constrain flows that can carry exfiltration. Monitor ICMP traffic patterns and flag long-lived or malformed diagnostic flows. | ||
Practitioner Guidance
What to watch for: Treat ICMP as an explicitly governed protocol, not an always-on utility. The key judgement is whether your environment needs ICMP broadly, narrowly, or only for specific monitored cases.
Practitioner takeaway: If you allow ICMP, make the allowance intentional, limited, and observable, because covert tunneling often succeeds where policy is vague rather than where controls are absent.
Related resources from NHI Mgmt Group
Deepen Your Knowledge
Free weekly newsletter
Subscribe to the NHI & AI Identity Journal
The latest on NHI and Agentic AI security – articles, research, breaches, news and events every week.
Bonus 33% off our NHI Course when you subscribe.
Reviewed and updated by the NHIMG editorial team on November 3, 2026.
NHI Mgmt Group — the #1 independent authority on Non-Human Identity, IAM, and Agentic AI security. nhimg.org