sing-box Fork Expands from Desktop VPN Launcher to Android and Router Deployments
The sing-box-lx project, an actively maintained fork of sing-box, has completed its transition from a desktop VPN engine to a unified platform running on Android devices and OpenWrt routers. The developer behind L×Box and Sing-box Launcher reports more than 1,700 commits, 24 stable 1.14 releases, and roughly seventy specialized builds since June.
Two major protocol additions stand out. A native MASQUE outbound implements CONNECT-IP as defined in RFC 9484, tunneling complete IP packets rather than UDP datagrams and reusing the same userspace network stack employed by WireGuard endpoints. The implementation treats Cloudflare WARP as a configuration preset while keeping device registration outside the core. AmneziaWG 2.0 received exhaustive verification against amneziawg-tools, the upstream kernel module netlink interface, and live IPC round-trips, confirming parity on all sixteen obfuscation parameters including Jc/Jmin/Jmax junk packets, S1–S4 padding, H1–H4 magic headers, and I1–I5 signature packets.
Android required extensive runtime optimizations. A full state machine now suspends idle WireGuard and AmneziaWG endpoints, sometimes destroying userspace devices entirely, and restores them lazily on first use. Health checks skip cycles when recent traffic has succeeded. The resulting savings reach 31 percent lower resident set size and 80 percent lower kernel CPU usage on production configurations. Additional fixes eliminated 127-second TCP dial timeouts caused by unbounded gVisor SYN backoff and ensured prompt recovery after Android Doze cycles.
Several long-standing silent failures were resolved during the summer. A three-byte change restored REALITY compatibility after Xray 26.7.11 began enforcing minimum client versions. XHTTP path handling now correctly appends trailing slashes, and mutual uplink/downlink deadlock in packet-up scenarios was eliminated. XMUX connection reuse and automatic TLS record fragmentation for detoured outbounds were also added.
The newest component, sing-box lxd, is a headless daemon that hosts the core and exposes a persistent gRPC management channel. Designed for routers, it survives configuration reloads, validates candidates before apply, performs automatic rollback on failure, and delivers the same structured observability previously available only to Android clients. An additional REST endpoint aggregates DHCP leases, ARP tables, bridge FDB entries, hostapd data, and operator labels to map IP addresses to human-readable device names.
Releases continue to follow three self-imposed rules: drift from upstream is measured by merge-base rather than commit count, every submodule is integrity-checked before merge, and no stable tag is cut without execution on real hardware. Static musl binaries for router architectures, including big-endian MIPS, are published alongside detailed OpenWrt recipes for isolated VPN SSIDs.
Related articles
OpenAI Releases GPT-6.1 Terra: Cost-Effective Model Focused on Computer Control, Image Analysis and Practical Automation
OpenAI has launched GPT-6.1 Terra, a new model positioned between its high-end GPT-6 Astra and more affordable options in the GPT-6 family. The model emphasizes real-world computer interaction, visual understanding of documents and interfaces, and reduced error rates rather than raw reasoning benchmarks. It supports a 1.05 million token context window, image inputs, and tool use through the Responses API while cutting input token prices by half compared with its predecessor. Published benchmarks show substantial gains on Terminal-Bench 4.0, OSWorld 2.1 and Agents' Last Exam, with error rates dropping sharply at both low and maximum reasoning levels. OpenAI stresses that the model still requires human oversight for high-stakes actions and does not directly process audio or video. The release targets developers and enterprises seeking reliable automation of repetitive multi-step tasks at lower cost than flagship models.
Yandex Black October: How Architectural Flaws and Capacity Mismanagement Turned Data Center Losses Into a 11-Day Fintech Coma
In October 2026 three Yandex data centers in Central Russia were physically destroyed within days, triggering the collapse of Yandex Cloud and multiple core services. Between the first and third site failures engineers had 74 hours to restore quorum and rebalance workloads yet failed to do so because Control Plane components remained tied to the lost zone and spare capacity had been eliminated to cut costs. Core banking and BNPL systems entered Read-Only mode after losing quorum, forcing a complete stop-the-world reconciliation that lasted until 22 October. Backups stored inside the same regional S3 buckets proved unusable, resulting in a seven-day RPO and primary-key collisions after restoration. The incident exposed that Metro-Cluster designs had been incorrectly treated as full Disaster Recovery and that aggressive AP microservice patterns created logical split-brain conditions outside the database layer.
Astra Group Releases Low-Code Astra AI Platform for Building Corporate AI Agents in Closed Environments
Astra Group has launched the Astra AI Platform, a low-code constructor designed for creating corporate AI applications that operate entirely within closed circuits without reliance on foreign clouds. The platform targets small and medium businesses as well as large corporations, enabling rapid deployment of AI secretaries, technical support assistants, and other agents in days rather than months. It forms part of the broader Astra AI ecosystem and includes two main constructors: one for machine learning model training and deployment, and another for LLM-based language assistants, document search, and process automation. Users can assemble scenarios from visual blocks, debug them in real time, or extend functionality with custom Python code when needed. The solution supports RAG-based knowledge bases, integration with external tools such as CRM and ERP systems, and the coordination of multiple agents for distributed task handling. One demonstrated use case involves processing ten pages of engineering drawings and specifications in five minutes to prepare data for downstream systems. The product is already listed in the Russian domestic software registry and available to customers.
Multi-Tenancy Isolation in VictoriaMetrics, VictoriaLogs and Grafana: Building Client Separation in a Shared Stack
Voltir runs a single shared monitoring stack consisting of VictoriaMetrics, VictoriaLogs, Grafana and vmalert for all customers instead of deploying separate instances per tenant. Isolation relies on numeric account IDs enforced at every layer: nginx validates tokens against the platform API before proxying writes, Grafana data sources hard-code tenant accounts, and vmalert rules receive automatic account filters. The team discovered that a single malformed rule or template could break alerting for every tenant because vmalert loads the entire rules directory at once. They also found that Grafana automatically created personal organizations granting Org Admin rights, accumulating hundreds of orphaned organizations and users that consumed disk space and resources. A dedicated cleanup module plus hourly reconciliation now removes rules, organizations, series and users when a tenant is deleted, while isolation tests verify both data and API objects across tenants.