Skip to content
This repository has been archived by the owner on Dec 19, 2018. It is now read-only.

Latest commit

 

History

History
19 lines (16 loc) · 990 Bytes

index.rst

File metadata and controls

19 lines (16 loc) · 990 Bytes

Authenticating and Authorizing Connections

The agent is a process that runs with root access inside of a server. It connects to DCM and takes instructions from DCM that run system changing commands. These commands do things like add additional users that also have root level access. Because of this it is crucial that this connection is safely formed and that both sides of the connection are assured they are in communication with the trusted and expected part.

When the agent connects to DCM it need to be assured that there is no bad actor spoofing the expected DCM server and that there is no man in the middle. How this is achieved is discussed here agent_trust.

DCM must also decide if the agent that connected is safe in the following ways:
  • It is running on the server it claims to be. This is discussed here: handshake.
  • It is talking to a trusted user on that server. This is discussed here token.