I wonder how security manager is handled in ros2 environment with dds-security

[kwonL57]

Hello! I am a student studying ROS2 with DDS-Security applied.

1. In FastDDS, on what basis is SecurityManager::init() called? For example, is it based on the GUID?
2. Is this function called only when a connection between Pub/Sub is first established? Or can it be called multiple times even for nodes with the same GUID?
3. How is the SecurityManager handled or deleted in the connection flow?