Java agent class transformer settings

When enabled, enhanced Spring transaction naming will name any transaction that originates from a Spring controller after the defined route and HTTP method. For example: /customer/v1/edit (POST). This includes controllers that implement or extend interfaces/classes with WebMVC related annotations (@RestController, @Controller, @RequestMapping, etc).

By default, this is configured to false, which will name transactions for those types of controllers based on the controller class name and method. For example; CustomerController/edit. This is the naming logic carried over from previous agent versions. "Standard" controllers, with all relevant annotations present on the actual class, will still get named based on route and HTTP method.

This config exists in Java agent version 8.9.0+.

When enabled, all Spring Controller transactions will be named using the controller class name and method name (e.g., /CustomerController/edit) instead of using request mappings (e.g., /api/v1/customer (POST)).

This setting takes precedence over enhanced_spring_transaction_naming and helps prevent transaction name cardinality issues with complex URI patterns.

Applies to Spring 4.3.x+ (spring-4.3.0 module) and Spring 6.x+ (spring-6.0.0 module).

This config exists in Java agent version 8.25.0+.

By default, built-in actuator endpoints and custom actuator endpoints (using the @endpoint annotation and its subclasses) will all be named as OperationHandler/handle in New Relic. Enabling this setting will result in the transaction name reflecting the actual base actuator endpoint URI. For example, invoking /actuator/loggers or actuator/loggers/com.newrelic will result in the transaction name actuator/loggers (GET). This is to prevent MGI.

Supported when using Spring Boot v3 and later.

This config exists in Java agent version 8.15.0+.

String containing the full class name of the annotation class the agent uses to determine which user-specified methods to instrument. For more information about custom annotations, see Java custom metric collection.

Enable this option to capture the userPrincipal name. This name is included as a transaction trace attribute (with attribute names user and enduser.id), and can be queried. This also allows you to see in errors inbox how many users are impacted by an error group.

By default, @TraceLambda annotations are disabled. To use the annotation functionality for tracing lambda methods you must explicitly enable the feature as follows:

You can configure this via YAML:

instrumentation:
  trace_lambda:
    enabled: true

You can also use the system property newrelic.config.instrumentation.trace_lambda.enabled=true or the environment variable NEW_RELIC_INSTRUMENTATION_TRACE_LAMBDA_ENABLED.