Overview The Health Check API provides endpoints to monitor the health and readiness of your Conductor instance, check service dependencies, and retrieve system metrics.
GET /health Check basic system health.
Include detailed health information
Health check timeout in milliseconds
Response Health status: healthy, degraded, unhealthy
Unix timestamp of health check
System uptime in milliseconds
Service-specific health status (when details=true) Database health: up, down, degraded
services.database.latency
Database response time in ms
AI services health: up, down, degraded
Available AI provider statuses
Storage health: up, down, degraded
Available storage types (KV, R2, D1)
cURL
cURL (Detailed)
TypeScript
Python
curl https://your-worker.workers.dev/health
Basic Response
Detailed Response
Degraded Response
{ "status" : "healthy" , "timestamp" : 1704067200000 , "uptime" : 3600000 , "version" : "1.0.0" }
GET /health/live Liveness probe - check if service is running.
Response Always true if service responds
curl https://your-worker.workers.dev/health/live
{ "alive" : true , "timestamp" : 1704067200000 } GET /health/ready Readiness probe - check if service can handle requests.
Response : 200 OK or 503 Service Unavailabletrue if ready to handle requests
Detailed readiness checks Storage services availability
curl https://your-worker.workers.dev/health/ready
Ready (200 OK)
Not Ready (503)
{ "ready" : true , "timestamp" : 1704067200000 , "checks" : { "database" : true , "ai" : true , "storage" : true } }
GET /health/metrics Get system performance metrics.
Time period: 1m, 5m, 15m, 1h, 24h
Response Time period covered by metrics
Execution metrics Total executions in period
Average execution duration (ms)
Member usage statistics Executions by member type
Error statistics Errors grouped by error code
cURL
cURL (Last 24 hours)
TypeScript
curl https://your-worker.workers.dev/health/metrics
{ "period" : "1h" , "timestamp" : 1704067200000 , "executions" : { "total" : 1250 , "successful" : 1180 , "failed" : 70 , "avgDuration" : 234 }, "members" : { "total" : 5420 , "byType" : { "Think" : 2100 , "Function" : 1800 , "Data" : 980 , "API" : 540 } }, "errors" : { "total" : 70 , "byCode" : { "TIMEOUT" : 25 , "VALIDATION_ERROR" : 30 , "NETWORK_ERROR" : 15 } }, "resources" : { "cpu" : 125000 , "memory" : 52428800 } } Kubernetes Probes Use health endpoints in Kubernetes deployments:
apiVersion : apps/v1 kind : Deployment metadata : name : conductor spec : template : spec : containers : - name : conductor image : conductor:latest livenessProbe : httpGet : path : /health/live port : 8080 initialDelaySeconds : 10 periodSeconds : 30 readinessProbe : httpGet : path : /health/ready port : 8080 initialDelaySeconds : 5 periodSeconds : 10 Health Check Best Practices
Use liveness probes - Detect when service needs restartUse readiness probes - Prevent traffic to unready instancesMonitor metrics - Track performance over timeSet appropriate timeouts - Balance responsiveness vs accuracyCheck dependencies - Verify database, AI, storage healthUse detailed checks in staging - Get full diagnosticsUse basic checks in production - Minimize overheadAlert on degraded status - Catch issues earlyLog health events - Debug health issuesTest health endpoints - Ensure probes work correctly
Error Responses Service Unavailable { "status" : "unhealthy" , "timestamp" : 1704067200000 , "error" : "Database connection failed" , "services" : { "database" : { "status" : "down" , "error" : "Connection timeout" } } } Timeout { "error" : "Health check timeout" , "timeout" : 5000 , "timestamp" : 1704067200000 } 
