Troubleshooting when the Local RADIUS server running on an Azure container instance becomes dormant

In this topic, you will find guidance to diagnose and resolve a situation where the Portnox™ Local RADIUS container on an Azure container instance (ACI) enters a dormant state. This problem is usually caused by DNS or networking issues within Azure or your network environment.

The Local RADIUS server can become dormant if the Azure instance stops running or if it cannot communicate with the Portnox Cloud back-end service. Example errors you may see in the Docker logs include:

curl == Info: Could not resolve host: radius.portnox.com
return code: '6', http code: '0'

In some cases, a timeout may occur:

return code: '28', http code: '0'

These issues usually originate from Azure’s infrastructure rather than the container or Portnox Cloud. To identify the root cause, examine the container logs carefully.

  1. Check the instance status in Azure.

    Verify whether the Azure container instance is running. If it is stopped, start or restart the instance and monitor whether the status in Portnox Cloud changes to Active.

  2. Collect logs for diagnosis.

    If the instance is running but still appears dormant in Portnox Cloud:

    1. Collect the complete logs from the container instance.
    2. Run the following debug command to capture detailed runtime behavior: 
      /local-radius-init.sh --debug
  3. Validate network and DNS behavior.

    In the Azure portal, open your container instance and go to Connect > Bash. Then enter the following commands:

    curl -v https://radius.portnox.com
curl -I https://radius.portnox.com
    Note:
    The Bash shell is sandboxed. If you see the error curl: command not found, install or reinstall curl using the following commands: 
    apk del curl
apk add curl
  4. Stop and start the affected container instance.

    Stop the container completely (do not just restart it) and wait a few minutes before starting it again. This action may move the container to a new Azure host node and reset its networking, which can resolve transient issues.

  5. Redeploy as a last resort.

    If the issue continues:

    1. Delete the affected container instance.
    2. Follow the official documentation to redeploy: Deploy the local RADIUS server container in Microsoft Azure.
  6. Contact Microsoft Support.

    For further root-cause investigation:

    1. Open a ticket with Microsoft Azure support.
    2. Provide logs showing DNS or connectivity errors.

Additional notes:

  • The Portnox container operates entirely at the application level.

  • Azure manages all DNS, routing, and networking.

  • The container does not modify or manage the underlying infrastructure.

  • Even when multiple container instances share the same subnet, isolated network faults can occur at the host node level.