Skip to content

Latest commit

 

History

History
276 lines (211 loc) · 7.62 KB

workflows.md

File metadata and controls

276 lines (211 loc) · 7.62 KB
Raw

Ham workflows

I wrote this document to capture how ham radio operator workflows can be broken down into different microservice flows with interactions. This also includes the notion of service health being integral to the system to prevent scenarios where something is inop and you don't find out until after the fact that information didn't get captured as you thought it would. If the service health is bad, for example, the logging flow is inop.

  1. Logging UI Loop (Fox or Hound)
    Calls Rig request, Log request, Call request, and Service Health

  2. Rig request
    Calls Publish Heartbeat

  3. Log request
    Calls Publish Heartbeat

  4. Call request
    Calls Publish Heartbeat

  5. Publish Heartbeat

  6. Service Health
    Consumes Publish Heartbeat messages.

  7. WSJT-X integration
    Consumes Publish Heartbeat messages.

Workflows

Logging UI Loop (Fox or Hound)

Description

Main UI loop for logging QSOs. If services are healthy, optional rig status, optional call lookups, and eventually QSO lookups are performed. Rig control is out of scope.

Diagram

flowchart LR;
    Start([Logging UI Loop]);
    Health0{Service<br>healthy?};
    Rig0{Rig status<br>integration?}
    Keyer0{Keyer<br>requested?};
    QSOUI0{Call<br>available?};
    QSOUI1{exchange?}
    QSOUI2{QSO<br>complete?}
    Lookup0{Call<br>lookup?};
    Lookup2{Log<br>lookup?};
    Clear([Clear]);
    Start --> Subscribed0{Subscribed?};
    Subscribed0 -->|No| Subscribed1[Subscribe];
    Subscribed0 -->|Yes| Health0;
    Subscribed1 --> Health0;
    Health0 -->|No| Sleep0[Sleep];
    Sleep0 -->|Retry| Subscribed0;
    Health0 -->|Yes| Rig0;
    Rig0 -->|Yes| Rig1([Request rig status]);
    Rig0 -->|No| Keyer0;
    Rig0 -->|Abandon| Clear;
    Rig1 --> Keyer0;
    Keyer0 -->|No| QSOUI0;
    Keyer0 -->|Yes| Keyer1([Run memory keyer]);
    Keyer1 --> QSOUI0;
    QSOUI0 -->|UI Loop| Subscribed0;
    QSOUI0 -->|Yes| Lookup0;
    QSOUI0 --> |Abandon| Clear;
    Lookup0 -->|Yes| Lookup1([Call lookup]);
    Lookup0 -->|No| Lookup2;
    Lookup1 --> Lookup2;
    Lookup2 -->|Yes| Lookup3([Log lookup]);
    Lookup2 -->|No| QSOUI1;
    Lookup3 --> QSOUI1;
    QSOUI1 -->|No| QSOUI2;
    QSOUI1 -->|Yes| Exchange0[capture exch];
    Exchange0 --> QSOUI2;
    QSOUI2 -->|UI Loop| QSOUI1;
    QSOUI2 -->|Yes| LogRequest0([Publish log QSO]);
    QSOUI2 -->|Abandon| Clear;
    LogRequest0 -->|Next QSO| Subscribed0;
    Clear --> Exit0{Exit?};
    Exit0 -->|No| Subscribed0;
    Exit0 -->|Yes| Exit1;
    Exit1([Exit]);
Loading

Rig request

Description

Perform a read or write to rigctld to interact with the rig. Publish a heartbeat. If an operation was performed, the heartbeat indicates success or failure.

Read or Write.

Diagram

flowchart LR;
    RigOp([Rig request]);
    Request0{Rig<br>request?}
    RigOp --> Subscribed0{Subscribed?};
    Subscribed0 -->|Yes| Request0;
    Subscribed0 -->|No| Subscribed1[Subscribe];
    Subscribed1 --> Sleep0[Sleep];
    Request0 -->|No| Heartbeat0(Heartbeat);
    Request0 -->|Yes| RigOp0[rigctld op];
    Sleep0 -->|Retry| Subscribed0;
    RigOp0 --> Publish0[Publish rig op status];
    Publish0 --> Heartbeat0
    Heartbeat0 -->|Loop| Subscribed0;
Loading

Log request

Description

Perform a log read or write against one or more defined databases. Publish a heartbeat. If an operation was performed, the heartbeat indicates success or failure.

Read or Write.

Diagram

flowchart LR;
    LogOp([Log request]);
    Request0{Log<br>request?};
    LogOp --> Subscribed0{Subscribed?};
    Subscribed0 -->|Yes| Request0;
    Subscribed0 -->|No| Subscribed1[Subscribe];
    Subscribed1 --> Sleep0[Sleep];
    Sleep0 -->|Retry| Subscribed0;
    Request0 -->|Yes| LogOp0[Log op];
    Request0 -->|No| Heartbeat0(Heartbeat);
    LogOp0 --> Publish0[Publish log op];
    Publish0 --> Heartbeat0;
    Heartbeat0 -->|Loop| Subscribed0;
Loading

Call request

Description

Perform a call database lookup one or more defined databases. Publish a heartbeat. If an operation was performed, the heartbeat indicates success or failure.

Reads only.

Diagram

flowchart LR;
    CallOp([Call request]);
    Request0{Log<br>request?};
    CallOp --> Subscribed0{Subscribed?};
    Subscribed0 -->|Yes| Request0;
    Subscribed0 -->|No| Subscribed1[Subscribe];
    Subscribed1 --> Sleep0[Sleep];
    Sleep0 -->|Retry| Subscribed0;
    Request0 -->|Yes| CallOp0[Call op];
    Request0 -->|No| Heartbeat0(Heartbeat);
    CallOp0 --> CallOp1{Success?};
    CallOp1 -->|No| Heartbeat0;
    CallOp1 -->|Yes| Publish0[Publish log op];
    Publish0 --> Heartbeat0;
    Heartbeat0 -->|Loop| Subscribed0;
Loading

Publish heartbeat

Description

Fragment which defines how services publish their heartbeat. Each heartbeat contains the name of the service reporting and once published the fragment is complete.

Writes only.

Diagram

flowchart LR;
    Heartbeat([Heartbeat]);
    Exit([Exit])
    Heartbeat --> Subscribed0{Subscribed?};
    Subscribed0 -->|No| Subscribed1[Subscribe];
    Subscribed1 --> Sleep0[Sleep];
    Sleep0 -->|Retry| Subscribed0;
    Subscribed0 -->|Yes| Publish0[Publish heartbeat for process];
    Publish0 --> Exit;
Loading

Service health

Description

Each workflow publishes a heartbeat when not performing work and after work is successfully completed. The heartbeat wait dead timer must be greater than the slowest operation to prevent the heartbeat from going stale. As long as all services check in at least once within a specified interval, the service health is alive. Otherwise it is dead.

Reads only.

Diagram

flowchart LR;
    Health([Start]);
    Health --> ZeroAll[Zero all service counters];
    ZeroAll --> Subscribed0{Subscribed?};
    Subscribed0 -->|No| Subscribed1[Subscribe];
    Subscribed1 --> Sleep0[Sleep];
    Sleep0 --> Subscribed0;
    Subscribed0 -->|Yes| Publish0[Publish overall and<br>per service health];
    Publish0 --> Heartbeat0{Heartbeat<br>available?};
    Heartbeat0 -->|Yes| Required0{Required?};
    Required0 -->|Yes| Zero0[Zero service timer];
    Zero0 -->|Check for more heartbeats| Heartbeat0;
    Heartbeat0 -->|No| Watermark0[Record highest service timer];
    Watermark0 --> Watermark1{High<br>watermark<br> reached?};
    Watermark1 --> |No| Alive0[Mark alive];
    Alive0 --> Sleep1[Sleep];
    Watermark1 --> |Yes| Dead0[Mark dead];
    Dead0 --> Sleep1;
    Sleep1 -->|Loop| Subscribed0;
Loading

WSJT-X Integration

Description

  1. Log QSO.
  2. Updated location.

Diagrams

(1)

flowchart LR;
    WSJTXApp([WSJT-X Multicast Receiver]);
    Subscribed0{Subscribed?};
    LogRequest0([Publish log QSO]);
    Health0{Service<br>healthy?};
    WSJTXApp -->|5: QSO Logged| Subscribed0;
    Subscribed0 -->|No| Subscribed1[Subscribe];
    Subscribed0 -->|Yes| Health0;
    Health0 -->|No| Sleep0;
    Subscribed1 --> Sleep0[Sleep];
    Sleep0 -->|Retry| Subscribed0;
    Health0 -->|Yes| LogRequest0;
Loading

(2)

flowchart LR;
    WSJTXApp([WSJT-X Sender]);
    Scheduler{Is it time?} -->|Yes| GPSPoll;
    Scheduler -->|No| Scheduler;
    GPSPoll[Poll GPS for Maidenhead grid locator] -->|11: Location| WSJTXApp;
Loading

Notes:

WSJT-X message format: https://sourceforge.net/p/wsjt/wsjtx/ci/master/tree/Network/NetworkMessage.hpp