Core Features
Built for professional mining operations — secure, scalable, and zero-downtime
Full Traffic Encrypted Tunnel
Pain Point
Mining traffic transmitted in plaintext over complex networks is easily identified, throttled, and blocked by ISPs. Traffic sniffing and hashrate hijacking are real threats.
Solution
HashRelay inserts an encryption layer between miners and pools. The client encapsulates Stratum traffic in TLS 1.3 or QUIC tunnels before forwarding to the relay server. ISPs only see encrypted HTTPS/QUIC data.
Technical Highlights
- 1TLS 1.3 certificate mode (recommended for production)
- 2TCP + AES-256-GCM for no-certificate environments
- 3TCP + ChaCha20-Poly1305 for low-end CPUs without AES acceleration
- 4QUIC for cross-border weak networks (lowest latency)
Multi-Pool Multi-Coin Parallel Forwarding
Pain Point
Traditional mining farms running multiple coins must maintain separate proxies per coin — complex management, chaotic port assignments, and difficult troubleshooting.
Solution
Static port mappings allow one HashRelay instance to forward multiple coins to different pools simultaneously. Port 3333 → BTC pool, port 4444 → ETH pool. Miners don't change anything.
Technical Highlights
- 1Protocol-agnostic transparent forwarding for any Stratum-based coin
- 2Each mapping independently configured with pool address and aggregation settings
- 3Config changes pushed to clients in real time, no restart needed
- 4Per-pool Stratum parsing can be enabled or disabled independently
Request Aggregation
Pain Point
Large farms with thousands of miners each opening separate tunnel connections consume massive bandwidth and server resources — and risk triggering pool connection bans.
Solution
The built-in Aggregation Engine parses Stratum protocol client-side and merges thousands of miner requests into just a few upstream connections. 1,000 miners → 2 connections.
Technical Highlights
- 1Supports Stratum V1, V1-Ethash, and Stratum V2
- 2miners_per_conn controls ratio; default 500 miners/connection, auto-scaling
- 3extranonce2 space partitioning ensures each miner has a unique search space
- 4JSON-RPC ID remapping prevents concurrent request ID collisions
Unified Ops Dashboard & Alerts
Pain Point
Ops teams bounce between Bitmain, MicroBT, Canaan apps and more — fragmented views, slow response. Miners go offline for hours unnoticed; reject spikes mean logging into pool sites manually.
Solution
HashRelay rolls hashrate, uptime, reject rate, and hardware health into one ops dashboard. Mirror-side Stratum parsing runs async — zero impact on relay latency. Threshold breaches fire Email, Telegram, or Webhook alerts instantly.
Technical Highlights
- 1One screen for the whole farm — ditch the vendor app shuffle
- 2Miner offline: detected in seconds
- 3High reject rate: alert at ≥30% (min 10 shares)
- 4Hashrate drop: sliding-window detection, multi-channel push
Dynamic Config Hot Push
Pain Point
Traditional proxies require a service restart for every pool config change, causing brief miner disconnections — catastrophic at scale.
Solution
Admin changes mappings via dashboard or API; the server records the change in a database transaction and immediately pushes to the relevant client nodes via persistent connection. Miners feel nothing.
Technical Highlights
- 1Push event types: CREATED, UPDATED, ENABLED, DISABLED, DELETED
- 2Client hot-updates local listeners, no restart required
- 3Atomic database + push ensures consistency
- 4Changes take effect in under one second
TCP + QUIC Dual Protocol
Pain Point
High packet loss on cross-border links causes TCP throughput to collapse. But some networks block UDP, making QUIC unavailable. No single protocol suits all environments.
Solution
HashRelay simultaneously listens on TCP and QUIC. Choose your protocol in the config file based on actual network conditions. Both are production-ready and switchable on the fly.
Technical Highlights
- 1QUIC: best for cross-border/weak networks — low latency, packet-loss tolerant
- 2TCP + TLS cert: best compatibility for stable datacenter networks
- 3TCP + ChaCha20: best for no-cert internal networks with low-end CPUs
- 4Both protocols active simultaneously; clients choose per deployment
Smart Auto-Reconnect
Pain Point
Network fluctuations break connections and force miners to pause — losing hashrate revenue.
Solution
The client has built-in unlimited reconnect with a default 5-second interval. The local Stratum listener stays alive during reconnect. Miners experience zero interruption.
Technical Highlights
- 1Heartbeat: 60s Ping/Pong intervals
- 23 consecutive missed heartbeats (~180s) triggers reconnect
- 3Miners continue submitting shares to the local port during tunnel recovery
- 4Minimizes hashrate loss during network instability
Management API
Pain Point
Manual operation at scale is error-prone and slow. Teams need programmatic control over pool configs, mappings, and statistics.
Solution
A full RESTful JSON API at /api/v1/ enables scripting, CI/CD pipeline integration, and third-party system automation to manage all HashRelay resources.
Technical Highlights
- 1Pool management: CRUD operations
- 2Port mapping management with live status
- 3Client node management and token generation
- 4Statistics and alert event endpoints
Rust Performance & Stability
Pain Point
Mining relays run hot 24/7 at massive concurrency. Script languages and mixed stacks bring GC pauses, memory leaks, and jitter — unacceptable at million-miner scale.
Solution
HashRelay's core path is Rust from the ground up — memory safety, zero-cost abstractions, and zero-copy pipelines deliver rock-solid forwarding at minimal CPU cost.
Technical Highlights
- 1Full Rust stack: relay, stats, and admin share one runtime
- 2Zero-copy pipeline: bytes stay in-kernel where possible
- 3No GC pauses: low latency under sustained load
- 4Memory safety: entire bug classes eliminated at compile time
Hardware Telemetry & History
Pain Point
Thermal issues and power drift surface too late. Every vendor formats data differently — no cross-model trend view.
Solution
HashRelay periodically collects temperature, power draw, fan speed, and more. Unified storage with time-range lookback makes root-cause analysis straightforward.
Technical Highlights
- 1Hands-free periodic collection
- 2Core metrics: temp, power, fan RPM
- 3Full history retained for trend comparison
- 4Tied to the ops dashboard for linked alerts
Remote Miner Console Access
Pain Point
Miner web UIs live on LAN only. Remote staff can't troubleshoot — it's a site visit or a VPN per rig.
Solution
HashRelay tunnels through to each miner's web UI over encrypted channels. Open config pages, logs, and diagnostics from anywhere — like you're on-site.
Technical Highlights
- 1No public IP on miners needed
- 2Encrypted tunnel straight to the web UI
- 3Troubleshoot from anywhere, anytime
- 4Access controlled through the relay platform
Unified Ops Monitoring Wallboard
Pain Point
Mixed-vendor farms mean N monitoring apps — hashrate, hardware, and alerts scattered across screens. Ops efficiency tanks.
Solution
The HashRelay ops wallboard puts hashrate, uptime, reject rate, and hardware telemetry front and center. Anomalies highlight and notify automatically — one platform replaces N vendor tools.
Technical Highlights
- 1Whole-farm KPIs on one screen
- 2Auto-highlight and alert on anomalies
- 3Cross-vendor unified monitoring view
- 4Wallboard-ready for the control room