Operate Goauld

Daily operator workflows for Goauld

Use this runbook for routine work after the server and agents are deployed.

1. Check fleet health

Start with the agent list and TUI:

tealc list
tealc tui

Confirm that the expected agents are online, their last update is recent, and their transport matches the deployment. Use the server’s metrics to check reconnects, heartbeat latency, failures, channels, and traffic over time.

2. Open an operator session

Select an agent in the TUI or connect directly:

tealc ssh AGENT_NAME

Use a persistent session when work must survive a network interruption. Enable ControlMaster when several client commands will target the same agent.

3. Move data or expose a service

Bind local listeners to loopback unless another host explicitly needs access.

4. Record and review activity

Use shell logging when the engagement requires a session recording. Retrieve the agent’s own buffered logs when diagnosing transport or feature failures. Redact credentials, addresses, and hostnames before sharing logs.

5. Diagnose a failure

Determine whether the failure is in the control path, SSH data path, or one specific feature. Follow Diagnose an agent before changing several transport settings at once.

6. Update or retire an agent

Use agent update to replace a running binary. Verify that the agent reconnects and that its expected features still work after the update.

Use tealc kill AGENT_NAME to stop an agent. Use --delete only when the binary should also be removed, and use tealc delete AGENT_NAME to remove the server-side agent record when it is no longer needed.

Before ending an operation

  • Close temporary forwards and local proxy listeners.
  • Confirm whether persistent sessions should remain running.
  • Secure or remove transferred files and session recordings.
  • Record unexpected reconnects or transport failures.
  • Confirm the status of agents that should have been stopped or retired.

The client guide contains the complete command list.