Added Agent for srs cg635m timing clock.

[mjrand]

Added an agent for the srs cg635m clock used for timing in the highbay office.

Description

This agent tracks the lock registers of the srs clock so we can monitor if the 10MHZ lock ever died on the srs clock.
The agent also tracks the frequency, CMOS output, and the running state as well.

Motivation and Context

Previously there was no tracking of the 10MHZ lock for the srs clock.
We need to be able to remotely track this lock and srs clock output.

This agent does not add any ability to actually adjust any clock outputs or output states!

Resolves #350.

How Has This Been Tested?

This agent was tested on daq-dev. All queries worked without error.
Grafana showed all lock statuses as well as the frequency, CMOS setting, and running state with no issues.

Types of changes

• Bug fix (non-breaking change which fixes an issue)
• New feature (non-breaking change which adds functionality)
• Breaking change (fix or feature that would cause existing functionality to not work as expected)

Checklist:

• My code follows the code style of this project.
• My change requires a change to the documentation.
• I have updated the documentation accordingly.

[msilvafe]

Thanks Michael, I think we should test unplugging and plugging back in the 10 MHz clock input and turning the output off and back on (with the run key) and double check that this agent catches the state changes. That can happen during a compute maintenance window otherwise we need to wait for all telescopes to have a break for data taking. Once tested I'm happy to merge, will give us more peace of mind knowing we have alarms on more failure points of the timing system.

[msilvafe]

Copied comment not relevant here.

[msilvafe]

I think you can do this as is but also could maybe do it a bit more compactly: register_list = [bool(msg & 1 << (8-i)) for i in range(1,9)] I tested this as follows:

In [64]: msg=int('01110000',2);print(msg)
112

In [65]: [bool(msg & 1 << (8-i)) for i in range(1,9)]
Out[65]: [False, True, True, True, False, False, False, False]

In [66]: msg=int('10000000',2);print(msg)
128

In [67]: [bool(msg & 1 << (8-i)) for i in range(1,9)]
Out[67]: [True, False, False, False, False, False, False, False]

In [68]: msg=int('00000001',2);print(msg)
1

In [69]: [bool(msg & 1 << (8-i)) for i in range(1,9)]
Out[69]: [False, False, False, False, False, False, False, True]

Then you can define registers = {rn: rl for r, rl in zip(reg_names, register_list)} where reg_names is a list of the register key names that you've defined in the dict in line 106.

[mjrand]

I knew there was probably a much cleaner way to do this that one of the friendly daq reviewers would surely suggest to me. :)

I can switch this decoding to this way for clarity sake. Though I'd like to keep them as ints so they're easily displayable on grafana, much like the leak detectors.

[BrianJKoopman]

This looks good to me, I don't have much to add beyond what @msilvafe pointed out, mostly documentation/style suggestions. I do agree that those state change tests that @msilvafe suggested should be done.

[BrianJKoopman]

What's the 'm' in 'SRSCG635m'? Looking at my photos from the site and the SRS website I don't see it as a part of a listed model number, just sometimes in filenames.

[BrianJKoopman]

By convention (and following PEP08) we use CapWords for classnames, so SRSCG635mInterface.

[BrianJKoopman]

I'd drop the "in the SO office", that's overly specific.

[BrianJKoopman]

Replace "send info to aggregator" with "publish to a feed." The current sentence doesn't really capture that it's also going to other subscribers, like the InfluxDB publisher agent, i.e. to Grafana.

[BrianJKoopman]

Needs a docstring.

Also, more generally, we need a docs page for this agent in socs/docs/agents/. Plenty of examples there, but also see https://ocs.readthedocs.io/en/main/developer/agent_references/documentation.html.

This is especially important for deciphering those returned integers.