OTEL-LGTM-Boilerplate

Production-ready observability boilerplate using OpenTelemetry and the LGTM stack (Loki, Grafana, Tempo, Mimir). This repo ships a Docker Compose setup, a preconfigured OpenTelemetry Collector, and Grafana provisioning with a host metrics dashboard.

• Grafana for visualization
• Loki for logs
• Tempo for traces
• Mimir for metrics
• OpenTelemetry Collector for ingest and routing

1. Start the stack:

docker compose --env-file .env -f <your_composer_1> -f...  -f docker-compose.monitoring.yml up -d

2. Open Grafana:

• URL: http://localhost:3000
• User: admin
• Password: admin

You can override credentials with environment variables:

export GF_SECURITY_ADMIN_USER=myuser
export GF_SECURITY_ADMIN_PASSWORD=mypassword

• grafana is exposed on port 3000
• loki, tempo, mimir, and otel-collector run on the internal Docker network

If you need to send telemetry from the host machine, add explicit port mappings to otel-collector and update the collector config.

Key config files:

• monitoring/otel/collector-config.yaml - OpenTelemetry Collector pipelines, receivers, exporters, and processors.
• monitoring/loki/loki-config.yaml - Loki storage, retention, and ingestion settings.
• monitoring/tempo/tempo-config.yaml - Tempo trace storage and receiver settings.
• monitoring/mimir/mimir-config.yaml - Mimir metrics storage and limits.
• monitoring/grafana/provisioning/datasources/datasources.yaml - Grafana data source definitions for Loki, Tempo, and Mimir.
• monitoring/grafana/provisioning/dashboards/dashboards.yaml - Grafana dashboard provisioning rules.
• monitoring/grafana/dashboards/metrics.json - Prebuilt host metrics dashboard.