Details
-
Type:
Story
-
Status: Done
-
Resolution: Done
-
Fix Version/s: None
-
Component/s: ts_main_telescope
-
Labels:
-
Story Points:3
-
Epic Link:
-
Sprint:TSSW Sprint - Oct 26 - Nov 9, TSSW Sprint - Nov 9 - Nov 23
-
Team:Telescope and Site
-
Urgent?:No
Description
Update the M2 SAL interface based on the latest xml file. After the summit test at Feb. in this year, the M2 xml has been updated to support more use cases than before. This task is to update the software to support this.
The main focus is in this task is the state machine transition and figure out how to let the M2 to be controllable by SAL.
This task is in the phase 2.
Attachments
Issue Links
Activity
I got the system to be AvailableStateSubstate and then, back to the PublishOnlySubstate state. Maybe this is because of the TMA communication error:
2020-11-06T11:21:26.167-07:00,System state change to OfflineState |
2020-11-06T11:21:26.167-07:00,System substate change to AvailableStateSubstate |
2020-11-06T11:21:26.168-07:00,(WARNING) (E6086) Loss of TMA comm |
2020-11-06T11:28:38.462-07:00,System state change to OfflineState |
2020-11-06T11:28:38.462-07:00,System substate change to PublishOnlySubstate |
It looks like this OfflineState is just because I press the "Quit GUI" button. But I do see the value of "pass/block command" changes from False to True to False. Need to check this. In addition, may need to update the method to flush the SysLog message.
Based on T14900-0124 - LSST M2 Controller ICD_revB.pdf
, the LSST state machine should have been implemented.
I could enter the StandbyState (by SAL) by manually overwriting the enableCommand flag:
The start command to DisabledState will fail because the powers are off. I may need to test this on simulator or hardware on summit.
Finally, I could enter the StandbyState by the SAL without the manual override:
The system log message is:
2020-11-09T14:11:46.705-07:00,System state change to OfflineState |
2020-11-09T14:11:46.705-07:00,System substate change to AvailableStateSubstate |
2020-11-09T14:11:46.706-07:00,(WARNING) (E6086) Loss of TMA comm |
2020-11-09T14:12:07.416-07:00,System state change to StandbyState |
I would like to check the heartbeat event, acknowledge command, and logging of file tomorrow.
Used the log.json to replace the SysLog.xml and TelemetryLog.xml and updated the related code. Refactored the functions of Log.lvclass. Added the unit tests.
Please help to review the PR:
https://github.com/lsst-ts/ts_mtm2/pull/23
In this PR, I did the following:
- Remove the ConfigFiles/ in project.
- Fix the bugs in m2ms_incomingSAL.vi to allow the pass of SAL commands.
- Use the config/log.json to replace the SysLog.xml and TelemetryLog.xml.
- Refactor the functions in Log class and reflush the log message at each time of new entry.
Thanks!
Great work on unit tests!
Dave Mills
added a comment - Great work on unit tests!
Based on the m2ms_incomingSAL.vi, there should be a user event (SAL_Command) to decide the M2 control system can accept the SAL command or not. In the case structure inside, a enum of SAL_InterfaceCommandEnum.ctl is used.
The PublishOnlySubstate class has the systemReady.vi to send the SAL_Command event to allow the SAL command. This is triggered by the exec.vi of SystemReady class.
The createCommand.vi in CommandFactory class issues the SystemReady command. There is a SystemNotReady command as well.
The SystemMonitor class has the check_for_system_ready.vi to send the SystemReady command.