Skip to main content

Fleet API

Prefix: /api/fleet | Tag: fleet

Unified fleet health scoring. Replaces the separate health/risk endpoints with a single data source from byoc_agent.unified_scorer. Results are cached for 60 seconds.

Endpoints​

MethodPathDescription
GET/api/fleet/scoresAll cluster scores (unified view)
GET/api/fleet/kpisFleet-wide KPI summary
GET/api/fleet/criticalWarning + Critical clusters only
GET/api/fleet/rulesUnified scoring rules
POST/api/fleet/refreshRecompute unified scores

GET /api/fleet/scores​

Returns all clusters scored by the unified scorer, sorted worst-first.

Response:

{
  "scores": [
    {
      "cluster_id": "uuid",
      "cluster_name": "prod-analytics",
      "customer_name": "Acme Corp",
      "account_id": "abc123",
      "customer_tier": "S",
      "overall_score": 42.5,
      "metrics_score": 38.0,
      "alerts_score": 55.0,
      "tier_adjusted_score": 40.0,
      "classification": "Critical",
      "top_risk": "Compaction: 35",
      "cluster_type": "ELASTIC",
      "region": "US East (N. Virginia) us-east-1",
      "dimension_scores": [
        {"name": "Compaction Score", "score": 35.0, "value": 8500.0, "weight": 0.2, "status": "critical"}
      ],
      "risk_reasons": ["High compaction score"],
      "suggested_actions": ["Investigate compaction backlog"],
      "firing_alert_count_7d": 12,
      "metrics": {
        "compaction_score": 8500.0,
        "disk_used_pct": 82.3,
        "query_errors_7d": 350,
        "query_error_pct": 1.2,
        "qps": 45.2,
        "node_count": 5
      }
    }
  ],
  "source": "unified",
  "generated_at": "..."
}

GET /api/fleet/kpis​

Response:

{
  "kpis": {
    "total_clusters": 172,
    "healthy_count": 150,
    "warning_count": 15,
    "critical_count": 7,
    "avg_score": 78.3,
    "avg_metrics_score": 75.0,
    "avg_alerts_score": 82.0,
    "tier_s_at_risk": 2
  },
  "source": "unified",
  "generated_at": "..."
}

GET /api/fleet/critical​

Returns only Warning and Critical clusters, sorted worst-first.

GET /api/fleet/rules​

Returns the unified scoring rules from byoc_agent/unified_rules.yaml.

Response:

{
  "lookback_days": 7,
  "weights": { "metrics": 0.7, "alerts": 0.3 },
  "tier_multipliers": { "S": 1.2, "A": 1.0, "B": 0.8 },
  "alert_bands": [...],
  "classification": { "healthy": 75, "warning": 40 },
  "ai_trigger": {...},
  "dimensions": [
    {
      "metric_key": "compaction_score",
      "display_name": "Max Compaction Score",
      "category": "storage",
      "weight": 0.2,
      "green": 200,
      "yellow": 1000,
      "red": 5000,
      "inverted": false,
      "action": "Investigate compaction backlog"
    }
  ]
}

POST /api/fleet/refresh​

Recomputes unified scores: loads rules, computes scores, persists, clears cache.

Response:

{
  "status": "ok",
  "clusters_scored": 172,
  "critical": 7,
  "warning": 15
}