Habr•August 4, 2026•🇷🇺Translated from Russian

Mapping Logical Air Gap Techniques for Secure Network Segmentation

A comprehensive technical guide examines practical methods for constructing logical air gaps that isolate sensitive network segments while still permitting controlled data exchange. The material builds on the principle that security is an architectural property rather than a simple perimeter fence, positioning logical air gaps between the extremes of a complete physical disconnect and an ordinary reverse proxy.

Why Firewalls Alone Are Insufficient

Traditional perimeter firewalls maintain an always-present network path that can be exploited through rule errors, unknown vulnerabilities, or configuration changes. For segments such as industrial control systems, backup repositories, hardware security modules, and machine-learning models, the cost of compromise justifies stronger isolation than packet filtering can provide.

Definition and Evaluation Criteria

A logical air gap exists when no end-to-end network path remains between an untrusted and a trusted segment, yet meaningful data exchange continues through an intermediary that terminates one session and originates another. Three primary axes are used for comparison: persistence of a direct route, which party initiates connections, and support for synchronous replies. Cost in money, latency, and operational effort forms an implicit fourth dimension.

Level 0 – Physical Isolation

At the lowest level, isolation is enforced by the absence or physical one-way nature of the transmission medium. Classic air gap and sneakernet rely on removable media with manual inspection. Hardware optical diodes provide unidirectional fiber links whose directionality is proven by circuitry rather than configuration. Stations with manual or robotic confirmation add human oversight to each transfer.

Level 1 – Network-Layer Separation

Separate VRF or VLAN instances without route leakage, dual-homed hosts with forwarding disabled, and bump-in-the-wire Layer-2 bridges all prevent packets from traversing the boundary. One-way UDP firewall rules offer a low-cost pilot before hardware diode deployment, though they remain dependent on correct configuration.

Level 2 – Transport-Layer Mediation

Proxies that fully terminate TCP sessions, pull-based models where the trusted side always initiates outbound connections, and mutual outbound connections to a meeting point in a DMZ are presented. Two common misconceptions are highlighted: reverse tunnels that re-establish an end-to-end stream and single-packet authorization schemes that ultimately permit direct connectivity once the port is opened.

Level 3 – Storage-Based Exchange

Message brokers such as Redis, RabbitMQ, Kafka, and NATS, file spools, object storage used as mailboxes, database tables acting as transfer queues, and one-way replication or change-data-capture mechanisms allow the two sides to interact only through a shared intermediary. Shared memory between virtual machines on the same hypervisor provides microsecond latency at the cost of placing both segments on a single host.

Level 4 – Semantic Validation

Above the transport layer, narrow RPC contracts using gRPC and protobuf, request reconstruction by proxies, signed envelopes, JSON Schema validation, two-phase ticket-and-callback flows, and human-in-the-loop approval further reduce the attack surface by operating on meaning rather than raw traffic.

Level 5 – Exotic Channels

Screen-to-camera QR-code chains, printed barcodes, and acoustic modems illustrate that isolation is a property of the overall construction rather than any single commercial product.

Related articles

Habr•Other

Bot Traffic Overtakes Human Traffic in 2024 as AI Agents and Scrapers Surge

Analysis of internet traffic from 2013 to 2026 shows automated bots steadily eroding human dominance online. Imperva data reveals human traffic fell to 47 percent by 2025 while malicious bots reached 40 percent. Good bots such as search crawlers remain stable, but gray AI agents and scrapers now drive much of the growth. Companies face rising infrastructure costs from bot traffic that generates no revenue, described as an invisible tax. Cloudflare and Akamai reports confirm high volumes of automated requests, with many classified as harmful scraping. The trend raises concerns about a synthetic internet shaped more by AI recommendations than human activity.

Habr•Other

Understanding Security Alerts Only Becomes Clear When You Start Writing Detection Rules Yourself

A cybersecurity professional who spent years as a SOC analyst reviewing alerts before moving into detection engineering shares key insights on why alerts behave differently depending on perspective. The article explains the inherent trade-offs in tuning detection rules, the architectural reasons alerts often arrive without full context, and why default priorities rarely match any single organization's environment. It highlights how false positives are not simply developer errors but calculated risks between coverage and noise. The piece also stresses the value of feedback loops between SOC teams and detection developers to improve overall security effectiveness.

安全客•Other

NVIDIA Accelerates Physical AI Push with $500 Billion Infrastructure Plan at 2026 World Robot Conference

The 2026 World Robot Conference opened with 373 companies showcasing over 3,000 exhibits and more than 300 new products focused on embodied intelligence. NVIDIA is deepening its commitment to physical AI by partnering with Apollo, Blackstone, KKR and other major asset managers to create an independent financing platform targeting over $500 billion in third-party capital for AI infrastructure. Madison Huang, NVIDIA’s Senior Director of Physical AI and Robotics, visited the event to review advances in human data, simulation, and real-world deployment. The company is reframing compute resources as revenue-generating AI factories that produce tokens, simulation data, and action policies rather than treating them as cost centers. Humanoid robot development faces a critical bottleneck because high-fidelity physics simulation for millions of virtual agents demands enormous cloud compute, while sim-to-real transfer gaps and edge inference constraints remain unsolved engineering challenges.

Habr•Other

Anthropic Releases Eight Claude Code Updates in August Focused on Multi-Agent Workflows

Between August 13 and 21, Anthropic shipped eight consecutive Claude Code releases from version 2.1.232 to 2.1.239. The updates center on enabling multiple long-running agents that can share context, communicate across sessions, and continue work automatically after hitting usage limits. Key additions include subagent forking that inherits prompt cache and conversation history, cross-session messaging via @mentions, and an automatic mode that uses a classifier model to approve actions. Additional improvements cover GitLab merge request integration, memory management fixes for extended sessions, and support for native add-ons in musl-based Alpine environments. The changes significantly expand the scale of tasks that can be delegated to Claude Code without constant human oversight.