System status

One status center for global AI compute operations.

Monitor the health of the control plane, Supabase data layer, telemetry ingestion, inference gateway, and distributed regional network before opening UmamiEdge to pilot customers.

Open status API Readiness gates Observability

Overall postureincident

computed from operational components

Regions5

7 MW modeled capacity

Demo nodes5

1 active incident components

Ready areas3

launch-readiness controls completed

Components

Operational health by layer.

Control plane

operational

Next.js app, global navigation, documentation, auth recovery, and operator console are available.

Evidence: GET /api/health and GET /api/status should return 200 responses.

Supabase data layer

degraded

Schema, RLS, and seed files are included, but production health depends on migration order and project credentials.

Evidence: Run compatibility fixes, schema, RLS, then seed. Generate Supabase types after migration.

Node telemetry

maintenance

Secure telemetry endpoints and node-agent starter exist; production needs signed agent release and queue retry handling.

Evidence: Install agent/node-agent.ts on a test VM and validate POST /api/telemetry/ingest.

Inference gateway

degraded

OpenAI-compatible endpoints are scaffolded and route through the selection engine; real runtime proxy is next.

Evidence: GET /api/v1/models works. Connect vLLM/Ollama/TGI before customer traffic.

Global network

incident

Demo topology includes APAC, Africa, EU, MENA, and North America corridors with regional failover concepts.

Evidence: Review /network and /api/network/topology for corridor and node health.