App Analytics

This page describes deprecated features with configuration information relevant to legacy App Analytics, useful for troubleshooting or modifying some old setups. To have full control over your traces, use ingestion controls and retention filters instead.

Migrate to the new configuration options

Navigate to the ingestion control page to see services with legacy configurations. These are flagged with a Legacy Setup status.

To migrate to the new configuration options, remove all legacy App Analytics configuration options from the services flagged with Legacy Setup. Then, implement the Datadog Agent and tracing libraries’ sampling mechanisms to send traces.

App Analytics setup

App Analytics configuration options are located in the Tracing Libraries and in the Datadog Agent. In the libraries, analytics spans from your services are generated either automatically or manually.

In Tracing Libraries

Automatic configuration

App Analytics is available starting in version 0.25.0 of the Java tracing client. It can be enabled globally for all web server integrations with one configuration parameter in the Tracing client:

  • System Property: -Ddd.trace.analytics.enabled=true
  • Environment Variable: DD_TRACE_ANALYTICS_ENABLED=true

App Analytics is available starting in version 0.19.0 of the Python tracing client. Enable App Analytics globally for all web integrations with one configuration parameter in the Tracing Client:

  • Tracer Configuration: ddtrace.config.analytics_enabled = True
  • Environment Variable: DD_TRACE_ANALYTICS_ENABLED=true

App Analytics is available starting in version 0.19.0 of the Ruby tracing client, and can be enabled for all web integrations with a global flag.

To do so, set either DD_TRACE_ANALYTICS_ENABLED=true in your environment, or configure with:

Datadog.configure { |c| c.tracing.analytics.enabled = true }
  • true enables analytics for all web frameworks.
  • false or nil disables analytics, except for integrations that explicitly enable it. (Default)

App Analytics is available starting in version 1.11.0 of the Go tracing client, and can be enabled globally for all web integrations using:

  • the WithAnalytics tracer start option, for example:

    tracer.Start(tracer.WithAnalytics(true))
    
  • starting in version 1.26.0 using environment variable: DD_TRACE_ANALYTICS_ENABLED=true

App Analytics is available starting in version 0.10.0 of the Node.js tracing client, and can be enabled globally for all web integrations with one configuration parameter in the tracing client:

tracer.init({
  analytics: true
})

You can also use the following configuration parameter:

  • Environment Variable: DD_TRACE_ANALYTICS_ENABLED=true

App Analytics is available starting in version 1.1.0 of the .NET tracing client, and can be enabled globally for all web integrations with one configuration parameter in the Tracing Client:

  • Environment Variable or AppSetting: DD_TRACE_ANALYTICS_ENABLED=true

This setting can also be set in code:

Tracer.Instance.Settings.AnalyticsEnabled = true;

App Analytics is available starting in version 0.17.0 of the PHP tracing client, and can be enabled globally for all web integrations with one configuration parameter in the Tracing Client:

  • Environment Variable: DD_TRACE_ANALYTICS_ENABLED=true

App Analytics is available starting in version 1.0.0 of the C++ tracing client, and can be enabled globally for all service entry spans by setting the environment variable: DD_TRACE_ANALYTICS_ENABLED to true. Note: This setting can also be set in the code directly:

datadog::opentracing::TracerOptions tracer_options;
  tracer_options.agent_host = "dd-agent";
  tracer_options.service = "<SERVICE_NAME>";
  tracer_options.analytics_rate = 1.0;
  auto tracer = datadog::opentracing::makeTracer(tracer_options);

To enable App Analytics for Nginx:

  1. Set the environment variable: DD_TRACE_ANALYTICS_ENABLED to true.

  2. Add env DD_TRACE_ANALYTICS_ENABLED; at the top of your nginx.conf file.

Configure additional services (optional)

Configure by integration

In addition to setting globally, you can enable or disable App Analytics for individual integrations using the following setting:

  • System Property: -Ddd.<integration>.analytics.enabled=true
  • Environment Variable: DD_<INTEGRATION>_ANALYTICS_ENABLED=true

Use this in addition to the global configuration for any integrations that submit custom services. For example, for JMS spans which comes in as a custom service, you can set the following to enable all JMS Tracing in App Analytics:

  • System Property: -Ddd.jms.analytics.enabled=true
  • Environment Variable: DD_JMS_ANALYTICS_ENABLED=true

Integration names can be found on the integrations table.

In addition to setting globally, you can enable or disable App Analytics for individual integrations using the following setting:

  • Tracer Configuration: ddtrace.config.<INTEGRATION>.analytics_enabled = True
  • Environment Variable: DD_<INTEGRATION>_ANALYTICS_ENABLED=true

Use this in addition to the global configuration for any integrations that submit custom services. For example, for Boto spans which comes in as a custom service, set the following to enable all Boto Tracing in App Analytics:

  • Tracer Configuration: ddtrace.config.boto.analytics_enabled = True
  • Environment Variable: DD_BOTO_ANALYTICS_ENABLED=true

Note: Several integrations require non-standard configuration due to the integration-specific implementation of the tracer. Consult the library documentation on App Analytics for details.

App Analytics can be enabled for specific integrations.

To do so, set either DD_<INTEGRATION>_ANALYTICS_ENABLED=true in your environment, or configure with:

Datadog.configure { |c| c.tracing.instrument :integration, analytics_enabled: true }

Where integration is the name of the integration. See the list of available integrations for options.

  • true enables analytics for this integration, regardless of the global setting.
  • false disables analytics for this integration, regardless of the global setting.
  • nil defers to global setting for analytics.

In addition to the global setting, you can enable or disable App Analytics individually for each integration. As an example, for configuring the standard library’s net/http package, you could do:

package main

import (
    httptrace "gopkg.in/DataDog/dd-trace-go.v1/contrib/net/http"
    "gopkg.in/DataDog/dd-trace-go.v1/ddtrace/tracer"
)

func main() {
    tracer.Start()
    defer tracer.Stop()

    mux := httptrace.NewServeMux(httptrace.WithAnalytics(true))
    // ...
}

In addition to setting globally, you can enable or disable App Analytics for individual integrations.

For example, to enable App Analytics for express:

tracer.use('express', {
  analytics: true
})

Integration names can be found on the integrations table.

In addition to setting globally, you can enable or disable App Analytics for individual integrations.

  • Environment Variable or AppSetting: DD_<INTEGRATION>_ANALYTICS_ENABLED=true

Or in code:

Tracer.Instance.Settings.Integrations["<INTEGRATION>"].AnalyticsEnabled = true;

For example, to enable App Analytics for ASP.NET MVC:

  • Environment Variable or AppSetting: DD_ASPNETMVC_ANALYTICS_ENABLED=true

Or in code:

Tracer.Instance.Settings.Integrations["AspNetMvc"].AnalyticsEnabled = true;

Integration names can be found on the integrations table. Note: On Linux, the names of environment variables are case-sensitive.

In addition to setting globally, you can enable or disable App Analytics for individual integrations using the following setting:

  • Environment Variable: DD_<INTEGRATION>_ANALYTICS_ENABLED=true

Use this in addition to the global configuration for any integrations that submit custom services. For example, for Symfony spans which comes in as a custom service, you can set the following to enable all Symfony Tracing in App Analytics:

  • Environment Variable: DD_SYMFONY_ANALYTICS_ENABLED=true

Integration names can be found on the integrations table.

Database services

Database tracing is not captured by App Analytics by default and you must enable collection manually for each integration. For example:

  • System Property: -Ddd.jdbc.analytics.enabled=true
  • Environment Variable: DD_JDBC_ANALYTICS_ENABLED=true

Database tracing is not captured by App Analytics by default and you must enable collection manually for each integration. For example:

  • Tracer Configuration: ddtrace.config.psycopg.analytics_enabled = True
  • Environment Variable: DD_PSYCOPG_ANALYTICS_ENABLED=true

Database tracing is not captured by App Analytics by default and you must enable collection manually for each integration. For example:

Datadog.configure { |c| c.tracing.instrument :mongo, analytics_enabled: true }

Database tracing is not captured by App Analytics by default. Enable collection manually for each integration, for example:

// Register the database driver with Analytics enabled.
sqltrace.Register("mysql", &mysql.MySQLDriver{}, sqltrace.WithAnalytics(true))

Database tracing is not captured by App Analytics by default and you must enable collection manually for each integration. For example:

tracer.use('mysql', {
  analytics: true
})

Database tracing is not captured by App Analytics by default and you must enable collection manually for each integration. For example, to enable App Analytics for ADO.NET:

  • Environment Variable or AppSetting: DD_AdoNet_ANALYTICS_ENABLED=true

Or in code:

Tracer.Instance.Settings.Integrations["AdoNet"].AnalyticsEnabled = true;

Integration names can be found on the integrations table. Note: On Linux, the names of environment variables are case-sensitive.

Database tracing is not captured by App Analytics by default. You can enable or disable App Analytics for individual integrations using the following setting:

  • Environment Variable: DD_<INTEGRATION>_ANALYTICS_ENABLED=true

Use this in addition to the global configuration for any integrations that submit custom services. For example, for mysqli:

  • Environment Variable: DD_MYSQLI_ANALYTICS_ENABLED=true

Integration names can be found on the integrations table.

Custom instrumentation

Applications with custom instrumentation can enable App Analytics by setting the ANALYTICS_SAMPLE_RATE tag on a span:

import datadog.trace.api.DDTags;
import datadog.trace.api.Trace;
import io.opentracing.Tracer;
import io.opentracing.util.GlobalTracer;

class MyClass {
  @Trace
  void myMethod() {
    final Span span = GlobalTracer.get().activeSpan();
    // Span provided by @Trace annotation.
    if (span != null) {
      span.setTag(DDTags.SERVICE, "<SERVICE_NAME>");
      span.setTag(DDTags.ANALYTICS_SAMPLE_RATE, 1.0);
    }
  }
}

Note: App analytics for dd.trace.methods or trace annotations spans can be enabled by setting -Ddd.trace-annotation.analytics.enabled=true.

Applications with custom instrumentation can enable App Analytics by setting the ddtrace.constants.ANALYTICS_SAMPLE_RATE_KEY tag on a span:

from ddtrace import tracer
from ddtrace.constants import ANALYTICS_SAMPLE_RATE_KEY

@tracer.wrap()
def my_method():
    span = tracer.current_span()
    span.set_tag(ANALYTICS_SAMPLE_RATE_KEY, True)

Applications with custom instrumentation can enable App Analytics by setting the Analytics::TAG_ENABLED tag on a span:

Datadog::Tracing.trace('my.task') do |span|
  # Set the analytics sample rate to 1.0
  span.set_tag(Datadog::Tracing::Metadata::Ext::Analytics::TAG_ENABLED, true)
end

For custom instrumentation, a special tag has been added to enable App Analytics on a span, as can be seen below:

span.SetTag(ext.AnalyticsEvent, true)

This marks the span as a App Analytics event.

Applications with custom instrumentation can enable App Analytics by setting the ANALYTICS tag on a span:

const { ANALYTICS } = require('dd-trace/ext/tags')

span.setTag(ANALYTICS, true)

Applications with custom instrumentation can enable App Analytics by setting the Tags.Analytics tag on a span:

using Datadog.Trace;

using(var scope = Tracer.Instance.StartActive("web.request"))
{
    // enable Analytics on this span
    scope.span.SetTag(Tags.Analytics, "true");
}

Applications with custom instrumentation can enable App Analytics by setting the ANALYTICS_KEY tag on a span:

<?php
  // ... your existing span that you want to enable for App Analytics
  $span->setTag(Tag::ANALYTICS_KEY, true);
?>

Applications with custom instrumentation can enable App Analytics by setting the analytics_event tag on a span:

...
#include <datadog/tags.h>
...
auto tracer = ...
auto span = tracer->StartSpan("operation_name");
// A boolean value of true enables App Analytics for the span,
// with a sample rate of 1.0.
span->SetTag(datadog::tags::analytics_event, true);
// A double value between 0.0 and 1.0 enables App Analytics
// and sets the sample rate to the provided value.
span->SetTag(datadog::tags::analytics_event, 0.5);

In the Datadog Agent

This section describes deprecated features with configuration information relevant to legacy App Analytics.

To configure a rate of spans to analyze by service, setup the following in the datadog.yaml file:

apm_config:
  analyzed_rate_by_service:
    service_A: 1
    service_B: 0.2
    service_C: 0.05

To configure a rate of spans to analyze by service and operation name, setup the following in the datadog.yaml file:

apm_config:
  analyzed_spans:
    service_A|operation_name_X: 1
    service_A|operation_name_Y: 0.25
    service_B|operation_name_Z: 0.01