Description

Currently only the Relayer has Prometheus instrumentation. The API Server and Indexer expose /health but have no /metrics endpoint and no application-level metrics.

Suggested metrics for API Server:

• api_requests_total (counter, labels: method, endpoint, status)
• api_request_duration_seconds (histogram, labels: method, endpoint)
• api_active_connections (gauge)
• api_canton_rpc_duration_seconds (histogram — time spent calling Canton ledger)
• api_eth_rpc_requests_total (counter, labels: method, status)

Suggested metrics for Indexer:

• indexer_events_processed_total (counter, labels: event_type)
• indexer_sync_lag_seconds (gauge — how far behind the ledger)
• indexer_last_offset (gauge — last processed ledger offset)
• indexer_db_query_duration_seconds (histogram, labels: query)

Scope:

• Add Monitoring config section to API Server and Indexer configs
• Register promhttp.Handler() on each service
• Add middleware for HTTP request metrics (chi middleware or custom)
• Update Prometheus scrape config to include all three services
• Add metrics definitions in dedicated files or extend metrics

Acceptance Criteria:

• API Server(including Indexer) each expose /metrics
• Prometheus scrape config includes all three services
• Config files updated with monitoring section