Troubleshooting

This section covers common issues and solutions for the Script Control Center solution.

Connection Issues

Invalid API Token: Ensure the API_TOKEN in the Agent's .env file matches the Agent API Key in the MP Credentials.
Network Connectivity: Verify that the Cloud Proxy can reach the Agent VM on port 8080 (or your custom API_PORT). Use curl -v http://<AGENT_IP>:8080/health from the Cloud Proxy.
TLS Mismatch: If you enabled HTTPS on the Agent, ensure Skip TLS Validation is set to True in the MP settings if using a self-signed certificate.
Agent Not Running: Check the service status on the Agent VM: sudo systemctl status script-execution-agent.service.

Path Issue: Ensure the script exists in the /opt/script-execution-agent/scripts directory.
Relative Paths: In the MP configuration, the path should be relative to the scripts directory. If your script is at /opt/script-execution-agent/scripts/subdir/test.sh, enter subdir/test.sh.

Executable Bit: Ensure the script is executable: sudo chmod +x /opt/script-execution-agent/scripts/your_script.sh.
Ownership: The script must be readable by the script-agent user.

Output Format: Ensure your script output matches the selected Script Type (Nagios or Prometheus). Raw output without performance data (for Nagios) will not generate metrics.
Timeout: If the script takes too long to execute, it might be killed by the Agent. Check the Agent logs.

Logs for the adapter can be found on the Cloud Proxy at:

/storage/log/vcops/log/adapters/IndevopsScriptControlCenterAdapter/IndevopsScriptControlCenterAdapter_*.log

Systemd Logs: sudo journalctl -u script-execution-agent.service
Application Logs: /opt/script-execution-agent/logs/scheduler.log (as defined in .env).

Status	Meaning	Solution
`worker_error`	The Agent failed to start the worker process.	Check Agent logs for system resource issues or configuration errors.
`script_start_error`	The script failed to start (e.g., missing interpreter).	Verify the shebang line (e.g., `#!/bin/bash`) and that the interpreter is installed.
`script_failed`	The script executed but returned a non-zero exit code.	Check the `error` property on the Script Endpoint object for `stderr` output.
`to_cancel` / `cancelled`	The task was cancelled (often because a new collection cycle started before the previous one finished).	Increase the Interval in the MP configuration or optimize script performance.