open-telemetry
/
opentelemetry-java-instrumentation
mirror of https://github.com/open-telemetry/opentelemetry-java-instrumentation.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity
opentelemetry-java-instrume.../dd-java-agent/README.md

9.4 KiB
Raw Blame History

Datadog Java Agent for APM

Minimal Java version required: 1.7

This is a Java Agent to instrument Java applications using the Datadog Tracer. Once attached to one of your JVM you should see traces in Datadog APM.

Tracing instrumentation can be done in 2 ways:

❗ Warning: This library is currently in Alpha. This means that even though we rigorously tested instrumentations you may experience strange behaviors depending on your running environment. Be sure to test thoroughly on a staging environment before releasing to production. For any help please contact tracehelp@datadoghq.com or reach out in the datadoghq slack channel: #apm-java.

Quick start

1. Install the Datadog Agent on your OS

The Java instrumentation library works in collaboration with a local agent that transmits the traces to Datadog. To install it with tracing please follow these steps:

[Main]
# Enable the trace agent.
apm_enabled: true

2. Instrument your application

To dynamically apply instrumentation you simply have to declare the provided jar file in your JVM arguments as a valid -javaagent:.

  • So first download the jar file from the main repository.

NOTE: While in beta, the latest version is best found on the Snapshot Repo.

# download the latest published version:
wget -O dd-java-agent.jar 'https://search.maven.org/remote_content?g=com.datadoghq&a=dd-java-agent&v=LATEST'
  • Then add the following JVM argument when launching your application (in your IDE, your maven or gradle application script, or your java -jar command):
-javaagent:/path/to/the/dd-java-agent.jar

That's it! If you did this properly the agent was executed at pre-main, had detected and instrumented the supported libraries and custom traces. You should then see traces on Datadog APM. For troubleshooting, look for the trace-agent.log along side the other agent logs.

Configuration

Configuration is done through a default dd-trace.yaml file as a resource in the classpath. You can also override it by adding the file path as a system property when launching the JVM: -Ddd.trace.configurationFile.

# Main service name for the app
defaultServiceName: java-app

# The writer to use.
# Could be: LoggingWritter or DDAgentWriter (default)
writer:
  # LoggingWriter: Spans are logged using the application configuration
  # DDAgentWriter: Spans are forwarding to a Datadog trace Agent
  #  - Param 'host': the hostname where the DD trace Agent is running (default: localhost)
  #  - Param 'port': the port to reach the DD trace Agent (default: 8126)
  type: DDAgentWriter
  host: localhost
  port: 8126

# The sampler to use.
# Could be: AllSampler (default) or RateSampler
sampler:
  # AllSampler: all spans are reported to the writer
  # RateSample: only a portion of spans are reported to the writer
  #  - Param 'rate': the portion of spans to keep
  type: AllSampler
  # Skip some traces if the root span tag values matches some regexp patterns
  # skipTagsPatterns: {"http.url": ".*/demo/add.*"}
  
# Enable custom tracing (Custom annotations for now)
# enableCustomAnnotationTracingOver: ["io","org","com"]

# Disable some instrumentations
# disabledInstrumentations: ["opentracing-apache-httpclient", "opentracing-mongo-driver", "opentracing-web-servlet-filter"]

Instrumented frameworks

When attached to an application the dd-java-agent dynamically applies the following opentracing-contrib instrumentation for the following set of frameworks & servers, though additional configuration may be required (JDBC):

Frameworks

Framework Versions Comments
OkHTTP 3.x HTTP client calls with cross-process headers
Apache HTTP Client 4.3 + HTTP client calls with cross-process headers
AWS SDK 1.11.119+ Trace all client calls to any AWS service
Web Servlet Filters Depending on server See Servers section
JMS 2 2.x Trace calls to message brokers, but distributed trace propagation not yet implemented

Servers

Server Versions Comments
Jetty 8.x, 9.x Trace all incoming HTTP calls with cross-process capabilities
Tomcat 8.0.x, 8.5.x & 9.x Trace all incoming HTTP calls with cross-process capabilities

Modern web application frameworks such as Dropwizard or Spring Boot are automatically instrumented thanks to these servers instrumentation. (See example projects)

Databases

DB Versions Comments
Spring JDBC 4.x Please check the following JDBC instrumentation section
Hibernate 5.x Please check the following JDBC instrumentation section
MongoDB 3.x Intercepts all the calls from the MongoDB client
Cassandra 3.2.x Intercepts all the calls from the Cassandra client

JDBC instrumentation

By enabling the JDBC instrumentation you'll intercept all the client calls to the following DBs: MySQL, PostgreSQL, H2, HSQLDB, IBM DB2, SQL Server, Oracle, MariaDB, etc...

But unfortunately this can not be done entirely automatically today. To enable tracing please follow the instructions provided on the java-jdbc opentracing contrib project.

We also provide an example project with Spring Boot & H2.

Disabling instrumentations

If for some reason you need to disable an instrumentation you should uncomment the disabledInstrumentations: attribute in the configuration and provide a list as illustrated below:

...

# Disable a few instrumentations
disabledInstrumentations: ["opentracing-apache-httpclient", "opentracing-mongo-driver", "opentracing-web-servlet-filter"]

The list of values that can be disabled are the top level keys found here

Custom instrumentation

The @Trace annotation

By adding the @Trace annotation to a method the dd-java-agent automatically measures the execution time.

@Trace
public void myMethod() throws InterruptedException{
		...
}

By default, the operation name attached to the spawn span will be the name of the method and no meta tags will be attached.

You can use the operationName customize your trace:

@Trace(operationName="Before DB")
public void myMethod() throws InterruptedException{
	....
}

Enabling custom tracing

  • Add the annotations jar as a dependency of your project
<dependency>
	<groupId>com.datadoghq</groupId>
	<artifactId>dd-trace-annotations</artifactId>
	<version>{version}</version>
</dependency>

or

compile group: 'com.datadoghq', name: 'dd-trace-annotations', version: {version}
  • Enable custom tracing in the dd-trace.yaml config file by setting the packages you would like to scan as follows enableCustomAnnotationTracingOver: ["io","org","com"].

If you want to see custom tracing in action please run the Dropwizard example.

Other useful resources

Before instrumenting your own project you might want to review the provided examples:

Other links that you might want to read: