Metrics

Operational metrics for monitoring ingestion progress, job queues, and graph growth.

Endpoints

Method	Path	Format
GET	/metrics	JSON (default)
GET	/metrics?format=prometheus	Prometheus text
GET	/metrics/prometheus	Prometheus text

# JSON snapshot
curl http://localhost:8090/metrics

# Prometheus scrape target
curl http://localhost:8090/metrics/prometheus

JSON response

{
  "collected_at": "2026-06-24T12:00:00Z",
  "campaigns": {
    "running": 1,
    "completed": 5,
    "total": 6
  },
  "jobs": {
    "by_status": {
      "queued": 2,
      "running": 1,
      "completed": 40,
      "failed": 1
    },
    "by_collector": {
      "dns": { "queued": 0, "running": 0, "completed": 10, "failed": 0, "total": 10 }
    },
    "total": 44
  },
  "intelligence": {
    "domains": 1200,
    "hosts": 8500,
    "certificates": 50000,
    "certificate_names": 120000,
    "graph_edges": 95000,
    "dns_records": 3200,
    "rdap_records": 800,
    "http_fingerprints": 400,
    "entities": 200,
    "campaign_edges": 1500,
    "observations": 600
  },
  "ct": {
    "logs_total": 25,
    "logs_active": 5,
    "logs_readonly": 20,
    "entries_fetched": 1500000,
    "entries_available": 12000000,
    "aggregate_progress_pct": 12.5,
    "certificates": 50000,
    "certificate_names": 120000,
    "domains": 1200
  },
  "infra": {
    "nats": "connected",
    "redis": "connected"
  }
}

Prometheus metrics

Metric	Labels	Description
atlas_campaigns	status	Campaign count by status
atlas_crawl_jobs	status	Crawl job count by status
atlas_crawl_jobs_by_collector	collector, status	Jobs per collector
atlas_intelligence_records	table	Row counts for graph tables
atlas_ct_logs	state	CT logs by state
atlas_ct_entries_fetched	—	Sum of last_fetched_index
atlas_ct_entries_available	—	Sum of last_tree_size
atlas_ct_aggregate_progress_pct	—	Fetched ÷ available × 100
atlas_infra_up	component	1 if NATS/Redis connected

Prometheus scrape config

scrape_configs:
  - job_name: atlas
    static_configs:
      - targets: ["control-api:8090"]
    metrics_path: /metrics/prometheus
    scrape_interval: 30s

Use cases

Question	Where to look
Is CT backfill progressing?	ct.aggregate_progress_pct or atlas_ct_aggregate_progress_pct
Are workers keeping up?	jobs.by_status.running vs queued
Which collector fails most?	jobs.by_collector.*.failed
Graph growth rate	intelligence.graph_edges, intelligence.domains over time
Stack health	infra.nats, infra.redis

Related docs

Guide	Description
Performance	Benchmarks and campaign throughput
Operations	Deployment and health checks
CT ingestor	Backfill configuration
API reference	All endpoints