FlowFuse Node-RED Instance States

The following list describes the possible states a hosted or remote Node-RED instance can find itself in.

Stable States

  • running: Instance is fully operational. Flows are deployed and executing normally.
  • suspended: Instance resources are freed and flows are paused. Can be resumed without redeploying.
  • stopped: Instance is not executing any flows and is intentionally shut down.
  • error: Instance cannot operate normally due to a critical configuration or runtime issue. Requires user intervention.
  • crashed: Instance terminated unexpectedly due to repeated runtime errors or failures.
  • rollback: A previous working version of the instance has been restored due to deployment failure or manual revert.
  • warning: Instance is running but has non-critical issues (e.g., node failures, resource limits close to threshold).
  • safe: Running in safe mode after multiple crashes. Editor works, flows are not started until a deploy action is triggered.
  • protected: Editor is disabled. Deployment can only occur via pipeline or controlled automation.
  • connected: Instance has established a successful connection to its runtime environment and is reachable.

Transitional States

  • loading: Instance UI or resources are being prepared (initial startup or page load).
  • installing: Required packages, dependencies, or container layers are being installed.
  • starting: Flows and runtime services are initializing.
  • stopping: Flows and runtime services are shutting down gracefully.
  • restarting: Instance is stopping and then immediately starting again, typically after a deployment or configuration update.
  • suspending: Instance flows are being paused and resources deallocated before entering suspended state.
  • importing: An external project or configuration is being applied to the instance.
  • pushing: Changes are being uploaded to the runtime or container registry.
  • pulling: Artifacts or project content are being fetched from a registry or pipeline source.