That's an internal failure that required the XBee to reset in order to recover. Ideally something you should never see under any circumstance. There is no to periodically transmit, etc to prevent watchdog resets.
If you have found a way to reliably reproduce such a failure you should reach out to the support team. They can try and reproduce based on your procedure and make sure it gets to the right team internally.