ts_hexrotcomm Overview

Introduction

Python code to communicate with the main telescope camera rotator and hexapod low level controllers (code running in PXI computers). Contents include:

• BaseCsc: base class for the main telescope and hexapod Commandable SAL Components (CSCs).

• CommandTelemetryClient: A TCP/IP client that communicates with the low level controllers to send commands and receive telemetry.

• BaseMockController: base class for mock controllers.

• Command, CommandStatus, and Header: C structures used for communication.

• SimpleMockController: a simple mock controller for testing BaseCsc and BaseMockController.

TCP/IP Communication Protocol

The CSC communicates with the low-level controller using TCP/IP. All data is sent as binary C data structures with no padding (the C structures are defined using __attribute__((__packed__))). All data received from the low-level controller begins with a standard header that includes a frame_id field, which identifies the kind of message: command status, configuration, or telemetry. This allows the reader to read the rest of the message (after the header) into the correct data structure.

The CSC connects to the low-level controller as part of the start command.

The CSC enables the low-level controller (including clearing errors) as part of the enable command.

The CSC disconnects from the low-level controller as part of the standby command. If the CSC loses its connection unexpectedly, it will go to the fault state.

If the low-level controller goes to fault state while the CSC is enabled, the CSC will also go to fault state, but remain connected. This gives users the ability to monitor recovery efforts using the EUI. Once you have resolved the underlying problem, you can recover the CSC using the usual command sequence:

• standby: the CSC disconnects (if connected).

• start: the CSC connects.

• enable: the CSC tries to clear the error, and, if successful, enables the low-level controller. If the CSC cannot clear the error state, it remains in disabled state. Once you have resolved the underlying problem, you can issue the enable command to try again (without having to go back to standby state).