Gather telemetry

Main > Using Liquid for building your application > Gather telemetry

In software, telemetry is used to gather data on the use of applications and application components, e.g. how often certain features are used, measurements of start-up time and processing time, hardware, application crashes, and general usage statistics.

Take a look at related key concepts of Liquid
Business Logic Separation
Leveling up Platform Providers

##From back-end

In the project StartUp file it is necessary to define that Liquid will use the telemertry, in this case, Azure AppInsights.

WorkBench.UseTelemetry<AppInsights>();

##appsettings.json It is also necessary to set in the appsettings, the instrumentation key of AppInsights to Liquid load the configuration.

  "AppInsights": {
    "InstrumentationKey": "3de84743-7611-4768-95c7-250c54546257"
  }

See how this is done in Liquid Hotel360 sample application.

##Generate custom events and metrics

Now it's possible to use Telemetry anywhere in the code

public async Task<DomainResponse> AddAsync(BasketModel basket)
{
  Telemetry.TrackEvent("New Basket");

  //Persist data on database
  var data = await Repository.AddOrUpdateAsync(basket);
  
  //Then it sends to AppInsights a metric of how much time was spent for inserting data.
  Telmetry.TrackMetric("time-save-database", Convert.ToDouble(stopWatch.Elapsed));
}

See how this is done in Liquid Hotel360 sample application.

##Generate computed (aggregated) metrics

  public async Task<DomainResponse> ItemProcess()
  {
        //Declaration of dictionary key that will be used for begin count the timestamp.
        Telemetry.BeginMetricComputation("item-process-basket");

        //Create values for input aggregate metrics on AppInsights
        var fibNumbers = new List<int> { 0, 1, 1, 2, 3, 5, 8, 13, 21, 34, 55 };

        foreach (int item in fibNumbers)
        {
            //Just for check a final timestamp
            Thread.Sleep(10000);

            //Track values for generate aggregation.
            Telemetry.ComputeMetric("item-process-basket", item);
         }

         //Finishing aggregating metrics and submitting to AppInsights 
         Telemetry.EndMetricComputation("item-process-basket");

         return new DomainResponse() { };

  }

See how this is done in Liquid Hotel360 sample application.

##Management of Telemetry Hierarchies

  public IActionResult Post([FromBody] BasketViewModel basket)
  {
          Telemetry.EnqueueContext("Basket", basket.BasketId, "Add");

          //When call another method, all telemetry events could be visualized on the same tree.
          var data = Factory<BasketService>().AddAsync(basket);
       
         //After executing all calls we cause the telemetry client to lose the context.
         Telemetry.DequeueContext();
  }

See how this is done in Liquid Hotel360 sample application.

##Telemetry on Kubernates

public override void Initialize()
        {
            AppInsightsConfiguration appInsightsConfiguration = LightConfigurator.Config<AppInsightsConfiguration>("AppInsights");
                      
            TelemetryConfiguration aiConfig = new TelemetryConfiguration(appInsightsConfiguration.InstrumentationKey);

            if(appInsightsConfiguration.EnableKubernetes)
            {
                aiConfig.EnableKubernetes();
            }

            TelemetryClient = TelemetryClient ?? new TelemetryClient(aiConfig) { InstrumentationKey = aiConfig.InstrumentationKey } ;
        }

See how this is done in Liquid Hotel360 sample application.

#From front-end

For the configuration in FrontEnd it is necessary to set the instrumentation key in the environment.ts file.

  appInsights: {
    instrumentationKey: '5af50bad-3bb6-4504-b353-45cae74a416a'
  }

In the project module, in this case, app.module.ts is necessary to tell which class will be injected at runtime.

providers: [
     WorkBench,
     AuthenticationService,
    { provide: 'ILightTelemetry', useClass: AppInsightsTelemetry }
  ]

See how this is done in Liquid Hotel360 sample application.

##Generate custom events and metrics

export class CreateComponent implements OnInit {

  basketId = 2;
  constructor(private api: APIService, @Inject('ILightTelemetry') private telemetryService: ILightTelemetry,
              private auth: AuthenticationService) { }

  ngOnInit() {
    this.telemetryService.TrackEvent('New view in Basket page');
    this.telemetryService.TrackMetric('time-load-page', OnInit.pageRender.Elapsed);  
  }

See how this is done in Liquid Hotel360 sample application.