Auto-Resume

Auto-resume Stateful

How auto-resume works

The connector tracks its last known configuration and runtime state in a small JSON file under the user’s home directory (for example ~/.edge-connector/state.json).

Stored fields include:

• whether the connector was running
• the last process ID and start time
• the Tool RPC URL and custom implementations path
• deployment credentials (if applicable)

Using —auto-resume

Start the CLI with:

edge-connector --auto-resume

If a previous run marked the connector as running and stored configuration, the CLI:

• skips interactive setup,
• reuses the stored Tool RPC URL and deployment credentials,
• launches the runtime with the same effective configuration.

If no previous state exists, the CLI falls back to the interactive menu or normal argument parsing.

When not to use auto-resume

Avoid --auto-resume immediately after changing environments (for example moving from staging to production). Clear state or deploy a fresh download instead to prevent accidental cross-environment reuse.