How to automate mining rig reboots? (Smart Plug Setup)

Understanding Smart Plug Integration

1. Smart plugs serve as intermediary hardware between the mining rig’s power supply and the wall outlet, enabling remote power cycling via Wi-Fi or Bluetooth protocols.

2. Compatibility must be verified with local voltage standards—110V in North America versus 220–240V in Europe and Asia—alongside maximum load ratings exceeding the rig’s peak draw, typically 800W–1600W for multi-GPU setups.

3. Firmware support for MQTT or HTTP API access is essential for integration into automation frameworks like Home Assistant, Node-RED, or custom Python scripts.

4. Physical placement requires secure mounting near the rig’s PSU to avoid accidental disconnection and ensure stable signal reception during high electromagnetic interference from GPUs and PSUs.

5. Some models include built-in energy monitoring, allowing real-time wattage tracking that correlates with hash rate stability and thermal throttling events.

Network Configuration and Authentication

1. The smart plug must join the same subnet as the control device—either through 2.4GHz Wi-Fi (required for most low-cost plugs) or Ethernet-to-Wi-Fi bridging where signal strength is weak.

2. Default credentials are changed immediately after onboarding to prevent unauthorized access, especially critical when rigs operate on public-facing subnets or shared ISP gateways.

3. Static IP assignment via DHCP reservation ensures consistent endpoint resolution, avoiding failed reboot attempts due to dynamic address changes after router reboots.

4. TLS/SSL termination is disabled on most consumer-grade smart plugs; therefore, all API calls must occur within trusted LAN boundaries, never exposed to WAN interfaces without reverse proxy hardening.

5. Time synchronization using NTP servers guarantees accurate scheduling of automated reboots, particularly when triggering based on uptime thresholds or cron-style intervals.

Scripting Power-Cycle Triggers

1. A Python script using the TuyaSmartHome or KasaPy library can issue ON/OFF commands by querying device state before initiating a cycle.

2. System-level checks such as nvidia-smi --query-gpu=temperature.gpu --format=csv,noheader,nounits feed temperature data into decision logic—if GPU temp exceeds 92°C for over 60 seconds, trigger plug OFF → wait 8 seconds → ON.

3. Log rotation captures every command execution with timestamps, device ID, and response status codes, stored locally and optionally forwarded to centralized logging services like Graylog.

4. Fail-safe timeouts prevent indefinite hanging: if no ACK arrives from the plug within 5 seconds, the script logs an error and retries once before escalating to manual alert via Telegram webhook.

5. Dual-plug redundancy is implemented for critical rigs—one controls main PSU, another handles riser-powered peripherals—allowing granular restart sequences without full system blackouts.

Monitoring Rig Health Post-Reboot

1. Within 90 seconds of power restoration, the rig must report its ethminer --version output and active pool connection status to confirm software initialization.

2. Hash rate stabilization is validated by comparing current 5-minute average against baseline values stored in SQLite; deviations beyond ±18% trigger secondary diagnostics.

3. Memory allocation errors logged by dmesg | grep -i 'out of memory' indicate insufficient RAM or driver corruption requiring deeper investigation beyond simple power cycles.

4. Fan speed telemetry from ipmitool sdr type fan or GPU BIOS registers confirms thermal management subsystem readiness before resuming full-load hashing.

5. Each successful post-reboot validation adds an entry to a rolling health ledger, used later for predictive failure modeling based on reboot frequency and recovery latency.

Frequently Asked Questions

Q: Can I use a standard Belkin Wemo plug for GPU mining rigs?A: Yes, but only if its rated load exceeds the rig’s surge current—many Wemo Mini units cap at 15A/1800W, which may fail under transient loads from 4x RTX 4090s during boot.

Q: Does turning off the smart plug cut power to the motherboard RTC battery?A: No—the CMOS battery remains powered independently; however, BIOS clock drift may occur if the plug cycles during OS suspend states due to inconsistent ACPI signaling.

Q: Will repeated smart plug cycling damage the PSU capacitors?A: Not significantly—modern ATX PSUs are designed for thousands of cold starts; capacitor stress occurs primarily during sustained overload, not controlled AC interruption.

Q: How do I test the automation without risking downtime?A: Use a dummy load resistor bank connected to the same plug, simulate rig behavior with a timed LED blink pattern, and validate script logic before attaching actual mining hardware.