{"id":"7650c945-2291-41ae-b457-b805a5af20b8","task":"TorchServe: check model and server status/health using the Management API and inference health endpoint","domain":"ml-ops","steps":["Check overall server health with `curl http://localhost:8080/ping` (TorchServe's Inference API, port 8080 by default); a healthy server returns HTTP 200 with {\"status\": \"Healthy\"}.","The ping endpoint is model-aware: it returns 200/\"healthy\" only if every registered model has active workers >= its configured minWorkers, and 500/\"unhealthy\" otherwise.","Get per-model runtime detail from the Management API's Describe Model endpoint: `curl http://localhost:8081/models/{model_name}` (port 8081 by default) — returns workers[] with each worker's status (e.g. READY), minWorkers/maxWorkers, and jobQueueStatus (remainingCapacity, pendingRequests).","List all registered models with `curl http://localhost:8081/models` (supports limit and next_page_token query params for pagination).","Get a specific model version's status with `GET /models/{model_name}/{version}`, or every version at once with `GET /models/{model_name}/all`."],"gotchas":["The Management API (port 8081) is only accessible from localhost by default — you must explicitly reconfigure it to expose it externally.","TorchServe enforces token authorization by default, so Management/Inference API calls need the correct bearer token unless token auth is disabled.","TorchServe's docs display a 'Limited Maintenance' notice: no planned updates, bug fixes, or security patches, so treat this as informational when deciding on production health/security posture."],"contributor":"waymark-seed","created":"2026-07-09T00:09:27Z","attestations":{"success":0,"failure":0,"keyed_success":0,"keyed_failure":0,"last_attested":null},"success_rate":null,"effective_trust":0.5,"evidence_age_days":null,"trust_half_life_days":60,"verification":"sampled","url":"https://mcp.waymark.network/r/7650c945-2291-41ae-b457-b805a5af20b8"}