A Jenkins plugin for automatically forwarding metrics, events, and service checks to a Datadog account.

Note: The Jenkins CI plugin page for this plugin references this documentation.

Setup

Installation

This plugin requires Jenkins 2.361.4 and Java 11.

For older versions of Jenkins (i.e 1.632+), you can find the 1.2.0 version of the plugin here.

This plugin can be installed from the Update Center (found at Manage Jenkins -> Manage Plugins) in your Jenkins installation:

1. Select the Available tab, search for Datadog, and select the checkbox next to Datadog Plugin.
2. Install the plugin by using one of the two install buttons at the bottom of the screen.
3. To verify the plugin is installed, search for Datadog Plugin on the Installed tab.

Continue below for configuration.

Note: If you see an unexpected version of the Datadog Plugin, run Check Now from the Manage Jenkins -> Manage Plugins screen.

Configuration

There are two ways to configure your plugin to submit data to Datadog:

• Using a Datadog Agent that acts as a forwarder between Jenkins and Datadog (recommended).
  • When using a DogStatsD server instead of a full Datadog Agent, only metrics and events are supported.
  • For data submitted from an external host, the Datadog Agent requires the following configuration: dogstatsd_non_local_traffic: true and apm_non_local_traffic: true. This can be configured using the datadog.yaml configuration file.
• Sending data directly to Datadog through HTTP.
  • The HTTP client implementation used is blocking with a timeout duration of 1 minute. If there is a connection problem with Datadog, it may slow your Jenkins instance down.

The configuration can be done from the plugin user interface with a Groovy script, or through environment variables.

Plugin user interface

To configure your Datadog Plugin, navigate to the Manage Jenkins -> Configure System page on your Jenkins installation. Once there, scroll down to find the Datadog Plugin section:

HTTP forwarding

1. Select the radio button next to Use Datadog API URL and Key to report to Datadog (selected by default).
2. Paste your Datadog API key in the API Key textbox on the Jenkins configuration screen. If you would like to store your API key with the Credentails Manager, create a Credential for the API key and select that credential in the Datadog API Key (Select from Credentials) dropdown.
3. Test your Datadog API key by using the Test Key button on the Jenkins configuration screen directly below the API key textbox.
4. (optional) Enter the hostname of the Jenkins server in the Advanced tab to include it with the events.
5. (optional) Enter your Datadog Log Intake URL and select “Enable Log Collection” in the Advanced tab.
6. (optional) Select “Enable CI Visibility”, optionally configuring your CI Instance name.
7. Save your configuration.

Datadog Agent forwarding

1. Select the radio button next to Use the Datadog Agent to report to Datadog.
2. Specify your Datadog Agent hostname and port.
3. (optional) Enter the hostname of the Jenkins server in the Advanced tab to include it with the events.
4. (optional) Enter your Log Collection Port, configure log collection in the Datadog Agent, and select “Enable Log Collection”.
5. (optional) Enter your Trace Collection Port and select “Enable CI Visibility”, optionally configuring your CI Instance name.
6. Save your configuration.

Groovy script

Configure your Datadog plugin to forward data through HTTP or DogStatsD using the Groovy scripts below. Configuring the plugin this way might be useful if you’re running your Jenkins Master in a Docker container using the official Jenkins Docker image or any derivative that supports plugins.txt and Groovy init scripts.

HTTP forwarding using Groovy

import jenkins.model.*
import org.datadog.jenkins.plugins.datadog.DatadogGlobalConfiguration

def j = Jenkins.getInstance()
def d = j.getDescriptor("org.datadog.jenkins.plugins.datadog.DatadogGlobalConfiguration")

// If you want to use Datadog API URL and Key to report to Datadog
d.setReportWith('HTTP')
d.setTargetApiURL('https://api.datadoghq.com/api/')
d.setTargetApiKey('<DATADOG_API_KEY>')

// Customization, see dedicated section below
d.setExcluded('job1,job2')

// If you want to collect logs
d.setLogIntakeUrl('https://http-intake.logs.datadoghq.com/v1/input/')

// Save config
d.save()

Datadog Agent forwarding using Groovy

import jenkins.model.*
import org.datadog.jenkins.plugins.datadog.DatadogGlobalConfiguration

def j = Jenkins.getInstance()
def d = j.getDescriptor("org.datadog.jenkins.plugins.datadog.DatadogGlobalConfiguration")

d.setReportWith('DSD')
d.setTargetHost('localhost')
d.setTargetPort(8125)

// If you want to collect logs
d.setTargetLogCollectionPort(10518)
d.setCollectBuildLogs(true)

// If you want to enable CI Visibility
d.setTargetTraceCollectionPort(8126)
d.setEnableCiVisibility(true)
d.setCiInstanceName("jenkins")

// Customization, see dedicated section below
d.setExcluded('job1,job2')

// Save config
d.save()

Environment variables

Configure your Datadog plugin using environment variables with the DATADOG_JENKINS_PLUGIN_REPORT_WITH variable, which specifies the report mechanism to use.

HTTP forwarding using environment variables

1. Set the DATADOG_JENKINS_PLUGIN_REPORT_WITH variable to HTTP.
2. Set the DATADOG_JENKINS_PLUGIN_TARGET_API_URL variable, which specifies the Datadog API endpoint (defaults to https://api.datadoghq.com/api/).
3. Set the DATADOG_JENKINS_PLUGIN_TARGET_API_KEY variable, which specifies your Datadog API key.
4. (optional) Log Collection:

• Set the DATADOG_JENKINS_PLUGIN_COLLECT_BUILD_LOGS variable to true in order to enable log collection (disabled by default).
• Set the DATADOG_JENKINS_PLUGIN_TARGET_LOG_INTAKE_URL variable, which specifies the Datadog Log Intake URL (defaults to https://http-intake.logs.datadoghq.com/v1/input/).

5. (optional) CI Visibility (trace collection):

• Set the DATADOG_JENKINS_PLUGIN_ENABLE_CI_VISIBILITY variable to true in order to enable CI Visibility (disabled by default).
• Set the DATADOG_JENKINS_TARGET_WEBHOOK_INTAKE_URL variable, which specifies the Datadog Webhook Intake URL (defaults to https://webhook-intake.datadoghq.com/api/v2/webhook/).
• Set the DATADOG_JENKINS_PLUGIN_CI_VISIBILITY_CI_INSTANCE_NAME variable, which specifies the name of the Jenkins instance for CI Visibility (defaults to jenkins).

Datadog Agent forwarding using environment variables

1. Set the DATADOG_JENKINS_PLUGIN_REPORT_WITH variable to DSD.
2. Set the DATADOG_JENKINS_PLUGIN_TARGET_HOST variable, which specifies the DogStatsD server host (defaults to localhost).
3. Set the DATADOG_JENKINS_PLUGIN_TARGET_PORT variable, which specifies the DogStatsD server port (defaults to 8125).
4. (optional) Log Collection:
   • Enable log collection in the Datadog Agent.
   • Set the DATADOG_JENKINS_PLUGIN_COLLECT_BUILD_LOGS variable to true in order to enable log collection (disabled by default).
   • Set the DATADOG_JENKINS_PLUGIN_TARGET_LOG_COLLECTION_PORT variable, which specifies the Datadog Agent log collection port.
5. (optional) CI Visibility (trace collection):
   • Set the DATADOG_JENKINS_PLUGIN_ENABLE_CI_VISIBILITY variable to true in order to enable CI Visibility (disabled by default).
   • Set the DATADOG_JENKINS_PLUGIN_TARGET_TRACE_COLLECTION_PORT variable, which specifies the Datadog Agent trace collection port (defaults to 8126).
   • Set the DATADOG_JENKINS_PLUGIN_CI_VISIBILITY_CI_INSTANCE_NAME variable, which specifies the name of the Jenkins instance for CI Visibility (defaults to jenkins).

Additionally, you can use the standard Datadog environment variables:

• Set the DD_AGENT_HOST variable, which specifies the Datadog Agent host.
• Set the DD_AGENT_PORT variable, which specifies the DogStatsD server port.
• Set the DD_TRACE_AGENT_PORT variable, which specifies the Datadog Agent trace collection port.
• Set the DD_TRACE_AGENT_URL variable, which specifies the Datadog Agent URL to send traces. When set this takes precedence over DD_AGENT_HOST and DD_TRACE_AGENT_PORT.

The environment variables with the DATADOG_JENKINS_PLUGIN namespace take precedence over the standard Datadog environment variables.

Logging

Logging is done by utilizing the java.util.Logger, which follows the best logging practices for Jenkins.

The plugin automatically registers a custom logger named “Datadog Plugin Logs” that writes the plugin’s logs with level INFO or higher. The custom logger registration can be disabled by setting the DD_JENKINS_PLUGIN_LOG_RECORDER_ENABLED environment variable to false. If you want to see the plugin logs with maximum detail, manually change the level of the custom logger to ALL.

Customization

Pipeline customization

The Datadog plugin adds a datadog step that provides some configuration option for your pipeline-based jobs.

In declarative pipelines, add the step to a top-level options block like so:

pipeline {
    agent any
    options {
        datadog(collectLogs: true, tags: ["foo:bar", "bar:baz"])
    }
    stages {
        stage('Example') {
            steps {
                echo "Hello world."
            }
        }
    }
}

In scripted pipeline, wrap the relevant section with the datadog step like so:

datadog(collectLogs: true, tags: ["foo:bar", "bar:baz"]) {
  node {
    stage('Example') {
      echo "Hello world."
    }
  }
}

Note: Pipeline customizations are only registered after a job has started. Tags specified in pipeline customization will not be associated with jenkins.job.started.

Global customization

To customize your global configuration, in Jenkins navigate to Manage Jenkins -> Configure System then click the Advanced button. The following options are available:

Job customization

From a job specific configuration page:

Test Optimization Configuration

The plugin can automatically configure Datadog Test Optimization for a job or a pipeline (see the Test Optimization documentation for your language to make sure that the testing framework that you use is supported; also note that automatic configuration is not supported for tests that are executed inside containers - follow manual instrumentation steps to enable Test Optimization for containerized test runs).

Before enabling Test Optimization, be sure to properly configure the plugin to submit data to Datadog.

There are two options to enable automatic Test Optimization configuration:

1. Using Jenkins UI (available in the plugin v5.6.0 or newer): go to the Configure page of the job or pipeline whose tests need to be traced, tick the Enable Datadog Test Optimization checkbox in the General section, and save your changes. This option is unavailable if you are using Multibranch Pipelines, Organization Folders, or other types of pipelines that are configured entirely with Jenkinsfile.
2. Using datadog pipeline step (available in the plugin v5.6.2 or newer):

In declarative pipelines, add the step to a top-level options block like so:

pipeline {
    agent any
    options {
        datadog(testOptimization: [ 
            enabled: true, 
            serviceName: "my-service", // the name of service or library being tested
            languages: ["JAVA"], // languages that should be instrumented (available options are "JAVA", "JAVASCRIPT", "PYTHON", "DOTNET", "RUBY")
            additionalVariables: ["my-var": "value"]  // additional tracer configuration settings (optional)
        ])
    }
    stages {
        stage('Example') {
            steps {
                echo "Hello world."
            }
        }
    }
}

In scripted pipelines, wrap the relevant section with the datadog step like so:

datadog(testOptimization: [ enabled: true, serviceName: "my-service", languages: ["JAVA"], additionalVariables: [:] ]) {
  node {
    stage('Example') {
      echo "Hello world."
    }
  }
}

The other datadog settings, such as collectLogs or tags can be added alongside the testOptimization block.

Please bear in mind that Test Optimization is a separate Datadog product that is billed separately.

Data collected

This plugin is collecting the following events, metrics, and service checks:

Events

Default events type

NOTE: event_type is always set to default for above events and metrics.

Systems events type

NOTE: event_type is always set to system for above events and metrics.

Security events type

NOTE: event_type is always set to security for above events and metrics.

Filtering events

This plugin allows you to filter events by the event type as well as the specific event names listed above. To include/exclude all events of the system or security type:

• In the UI: Uncheck the checkboxes for these events.
• In a groovy script: Fetch the Datadog global descriptor and call either d.setEmitSystemEvents() or d.setEmitSecurityEvents().
• In the environment variables section: Set the environment variables for the emitting security or system events.

To get more specific control over what events are sent, three configuration options are provided to allow a comma-separated include/exclude list of strings of event names. The include/exclude list has precedence over filtering by event type. For example, security events can be toggled off, but including UserAuthenticated takes precedence, so only UserAuthenticated events will be sent from the security type. In the UI, text boxes are provided for both the included and excluded lists. In a groovy script, the methods d.setIncludeEvents() and d.setExcludeEvents() accept a comma-separated list of event names as input which is another valid configuration method. Lastly, there are provided environment variables for manually setting included/excluded lists.

NOTE: As mentioned in the job customization section, there are job-specific toggles to send SCMCheckout events. If the SCMCheckout event is excluded globally, this toggle will have no effect.

Metrics

Log Collection for Agents

Note: This configuration only applies to those using the Datadog Agent configuration.

1. Collecting logs is disabled by default in the Datadog Agent, enable it in your datadog.yaml file:

   logs_enabled: true
   

2. To collect Jenkins logs, create a custom log source file for your Agent by creating a conf.yaml inside conf.d/jenkins.d with the following:

   logs:
     - type: tcp
       port: <PORT>
       service: <SERVICE>
       source: jenkins
   

3. In Jenkins, submit the port you specified above as the Log Collection Port. You can set this using environment variables, a Groovy script, or the Jenkins UI.

4. Restart the Agent.

Service checks

Build status jenkins.job.status with the default tags: : jenkins_url, job, node, user_id

Troubleshooting

Generating a diagnostic flare.

Plugin diagnostic flare contains data that can be used to diagnose problems with the plugin. At the time of this writing the flare includes the following:

• plugin configuration in XML format
• plugin connectivity checks results
• runtime data (current versions of JVM, Jenkins Core, plugin)
• recent exceptions that happened inside the plugin code
• plugin logs with level INFO and above, and recent Jenkins controller logs
• current stacks of the threads of the Jenkins controller process
• environment variables starting with DD_ or DATADOG_ (except API key and/or APP key)

To generate a flare go to the Manage Jenkins page, find the Troubleshooting section, and select Datadog. Click on Download Diagnostic Flare (requires “MANAGE” permissions) to generate the flare.

Issue tracking

GitHub’s built-in issue tracking system is used to track all issues relating to this plugin: jenkinsci/datadog-plugin/issues. However, given how Jenkins plugins are hosted, there may be issues that are posted to JIRA as well. You can check this jenkins issue for those issue postings.

Note: Unresolved issues on JIRA mentioning Datadog.

Changes

See the CHANGELOG.md.

How to contribute code

First of all and most importantly, thank you for sharing.

Checkout the contributing guidelines before you submit an issue or a pull request. Checkout the development document for tips on spinning up a quick development environment locally.