Operator Toolbox

Every tool for monitoring, verifying, and operating your Livepeer orchestrator — Explorer, Prometheus metrics, Docker monitoring stack, community dashboards, and capability testing utilities.

Your node is running. These are the tools you reach for when you want to verify you are in the active set, track earnings, diagnose a routing problem, monitor GPU health, or understand what the network is doing.

Livepeer Explorer

URL: explorer.livepeer.org

The Explorer is the authoritative on-chain dashboard for the Livepeer protocol. It reads directly from the Arbitrum subgraph, so everything you see here reflects current on-chain state.

Your orchestrator profile lives at:

Explorer orchestrator profile URL

https://explorer.livepeer.org/accounts/<your-address>/orchestrating

Active Set and Stake

See your rank by total stake (own + delegated). The top 100 active orchestrators receive transcoding work. Check here to confirm you are in the set and track your rank trend.

Reward Call History

Round-by-round history of reward calls. Missed rounds mean lost inflation — for you and your delegators. Target is 100% call rate. Any gaps here warrant investigation.

Earnings

Cumulative ETH transcoding fees and LPT inflation rewards. Use this to verify your node is earning and to track long-term trend.

Governance

Vote on Livepeer Improvement Proposals and treasury proposals directly from the Explorer interface.

Also on the Explorer:

• Your published reward cut and fee share — verify these match what you configured on-chain
• Your delegator list, with individual stake amounts and unbonding status
• Rounds active (your seniority on the network)

The Explorer shows on-chain data only. Worker-level data for pool contributors lives outside Explorer, and off-chain pool payouts do not appear there. For AI model warm status, use tools.livepeer.cloud/ai/network-capabilities.

Node Metrics (Prometheus)

go-livepeer exposes a standard Prometheus metrics endpoint. Scrape it with any monitoring stack.

Enable metrics with the -monitor flag:

Enable Prometheus metrics

livepeer \
  -orchestrator \
  -transcoder \
  -monitor \
  -metricsPerStream \
  # ... other flags

Flags:

Endpoint: http://localhost:7935/metrics

Key metric categories exposed:

• Session count and capacity utilisation
• Segment success and failure rates by stream
• Winning ticket counts (probabilistic approximation)
• Job processing latency
• GPU utilisation

For the full metric reference, see the Prometheus Metrics Reference.

Docker monitoring stack (fastest setup)

Livepeer maintains a Docker image that bundles Prometheus and Grafana with starter dashboard templates:

Docker monitoring stack for a single node

# Single node
docker run --net=host \
  --env LP_MODE=standalone \
  --env LP_NODES=localhost:7935 \
  livepeer/monitoring:latest

Docker monitoring stack for multiple nodes

# Multi-node
docker run --net=host \
  --env LP_MODE=standalone \
  --env LP_NODES=node1.example.com:7935,node2.example.com:7935 \
  livepeer/monitoring:latest

Environment variable options:

For Kubernetes, pods with the prometheus.io/scrape label are discovered automatically.

livepeer/livepeer-monitoring

Source, Dockerfile, and prometheus.yml for the monitoring supercontainer.

Custom Prometheus setup

Add this to your prometheus.yml to scrape go-livepeer directly:

Prometheus scrape config

scrape_configs:
  - job_name: 'livepeer'
    static_configs:
      - targets: ['localhost:7935']
    scrape_interval: 15s

Cloud SPE Tools

URL: tools.livepeer.cloud

A community-built suite of orchestrator and network monitoring tools, maintained by the Cloud SPE.

Key sections:

• AI network capabilities (/ai/network-capabilities) — lists all AI-capable orchestrators by pipeline, showing which models are warm. Use this to verify your AI pipeline registrations are visible to the network after configuring aiModels.json.
• Payout reports — daily, weekly, and monthly earnings breakdowns per orchestrator address
• Orchestrator view — per-orchestrator statistics including pricing and job activity

This is community infrastructure maintained outside the Livepeer Foundation. It has experienced intermittent availability, so keep Explorer and Prometheus in the loop as primary sources.

Network Dashboards (Dune Analytics)

For protocol-level macro data — inflation, staking participation, fee volumes, token distribution:

Use these dashboards for network trends. Use Explorer and Prometheus for per-node operations.

Capability Testing

Tools for verifying your node is correctly serving jobs end-to-end. Use them to confirm behaviour from the outside instead of inferring everything from logs.

stream-tester

Submits transcoding jobs to an orchestrator and measures response times and success rates. Used to validate your transcoding capability from the outside.

livepeer-ai-job-tester

Community-built tool (Cloud SPE) that submits AI inference jobs and records performance metrics. Used to verify AI pipeline availability before relying on network routing.

Tool selection guide

Livepeer Explorer Guide

Full walkthrough of every Explorer metric and what to do when things look wrong.

Metrics and Monitoring

Setting up Prometheus, Grafana, and the Docker monitoring stack.