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

Use Spotless for formatting (DataDog/dd-trace-java#1619)

This commit is contained in:
Tyler Benson 2020-06-19 15:17:38 -04:00 committed by Trask Stalnaker
parent b558c50229
commit 4a943c8411
26 changed files with 356 additions and 201 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

6
.editorconfig
View File

@ -663,10 +663,8 @@ ij_groovy_while_brace_force = never
ij_groovy_while_on_new_line = false
ij_groovy_wrap_long_lines = false
[{*.gradle.kts,*.kt,*.kts,*.main.kts}]
indent_size = 4
tab_width = 4
ij_continuation_indent_size = 8
[{*.kt,*.kts}]
ij_continuation_indent_size = 2
ij_kotlin_align_in_columns_case_branch = false
ij_kotlin_align_multiline_binary_operation = false
ij_kotlin_align_multiline_extends_list = false

14
.githooks/pre-commit Executable file
View File

@ -0,0 +1,14 @@
#!/bin/bash
# http://redsymbol.net/articles/unofficial-bash-strict-mode/
set -euo pipefail
IFS=$'\n\t'
if ! ./gradlew spotlessCheck; then
./gradlew spotlessApply
echo ""
echo ""
echo -e "\033[0;33mCode has been formatted; please git diff/add and recommit."
echo ""
echo ""
exit 1
fi

37
CONTRIBUTING.md
View File

@ -83,8 +83,8 @@ Gradle plugin.
#### Snapshot builds
For developers testing code changes before a release is complete, there are
snapshot builds of the `master` branch. They are available from
[JFrog OSS repository](https://oss.jfrog.org/artifactory/oss-snapshot-local/io/opentelemetry/auto/)
snapshot builds of the `master` branch. They are available from
[JFrog OSS repository](https://oss.jfrog.org/artifactory/oss-snapshot-local/io/opentelemetry/auto/)
#### Building from source
@ -145,31 +145,34 @@ We follow the [Google Java Style
Guide](https://google.github.io/styleguide/javaguide.html). Our build will
fail if source code is not formatted according to that style.
To verify code style manually run the following command, which uses
[google-java-format](https://github.com/google/google-java-format) library:
The main goal is to avoid extensive reformatting caused by different IDEs having different opinion
about how things should be formatted by establishing.
`./gradlew verifyGoogleJavaFormat`
Running
or on Windows
```bash
./gradlew spotlessApply
```
`gradlew.bat verifyGoogleJavaFormat`
reformats all the files that need reformatting.
Instead of fixing style inconsistencies by hand, you can run gradle task
`googleJavaFormat` to automatically fix all found issues:
Running
`./gradlew googleJavaFormat`
```bash
./gradlew spotlessCheck
```
or on Windows
`gradlew.bat googleJavaFormat`
runs formatting verify task only.
#### Pre-commit hook
To completely delegate code style formatting to the machine, you can add [git
pre-commit hook](https://git-scm.com/docs/githooks). We provide an example
script in `buildscripts/pre-commit` file. Just copy or symlink it into
`.git/hooks` folder.
To completely delegate code style formatting to the machine,
there is a pre-commit hook setup to verify formatting before committing.
It can be activated with this command:
```bash
git config core.hooksPath .githooks
```
#### Editorconfig

16
README.md
View File

@ -10,8 +10,8 @@
This project provides a Java agent JAR that can be attached to any Java 7+
application and dynamically injects bytecode to capture telemetry from a
number of popular libraries and frameworks.
The telemetry data can be exported in a variety of formats.
number of popular libraries and frameworks.
The telemetry data can be exported in a variety of formats.
In addition, the agent and exporter can be configured via command line arguments
or environment variables. The net result is the ability to gather telemetry
data from a Java application without code changes.
@ -29,11 +29,11 @@ The instrumentation agent is enabled using the `-javaagent` flag to the JVM.
java -javaagent:path/to/opentelemetry-auto-all.jar \
-jar myapp.jar
```
By default OpenTelemetry Java agent uses
By default OpenTelemetry Java agent uses
[OTLP exporter](https://github.com/open-telemetry/opentelemetry-java/tree/master/exporters/otlp)
configured to send data to
configured to send data to
[OpenTelemetry collector](https://github.com/open-telemetry/opentelemetry-collector/blob/master/receiver/otlpreceiver/README.md)
at `localhost:55680`.
at `localhost:55680`.
Configuration parameters are passed as Java system properties (`-D` flags) or
as environment variables (see below for full list). For example:
@ -67,7 +67,7 @@ A simple wrapper for the Zipkin exporter of opentelemetry-java. It POSTs json in
|----------------------------------|----------------------------------|----------------------------------------------------------------------|
| ota.exporter=zipkin | OTA_EXPORTER=zipkin | To select Zipkin exporter |
| ota.exporter.zipkin.endpoint | OTA_EXPORTER_ZIPKIN_ENDPOINT | The Zipkin endpoint to connect to. Currently only HTTP is supported. |
| ota.exporter.zipkin.service.name | OTA_EXPORTER_ZIPKIN_SERVICE_NAME | The service name of this JVM instance
| ota.exporter.zipkin.service.name | OTA_EXPORTER_ZIPKIN_SERVICE_NAME | The service name of this JVM instance
#### OTLP exporter
@ -94,7 +94,7 @@ attributes to stdout. It is used mainly for testing and debugging.
*This is highly advanced behavior and still in the prototyping phase. It may change drastically or be removed completely. Use
with caution*
The OpenTelemetry API exposes SPI [hooks](https://github.com/open-telemetry/opentelemetry-java/blob/master/api/src/main/java/io/opentelemetry/trace/spi/TracerProviderFactory.java)
The OpenTelemetry API exposes SPI [hooks](https://github.com/open-telemetry/opentelemetry-java/blob/master/api/src/main/java/io/opentelemetry/trace/spi/TracerProviderFactory.java)
for customizing its behavior, such as the `Resource` attached to spans or the `Sampler`.
Because the auto instrumentation runs in a separate classpath than the instrumented application, it is not possible for customization in the application to take advantage of this customization. In order to provide such customization, you can
@ -163,7 +163,7 @@ For this reason the following instrumentations are disabled by default:
- `jdbc-datasource` which creates spans whenever `java.sql.DataSource#getConnection` method is called.
- `servlet-filter` which creates spans around Servlet Filter methods.
- `servlet-service` which creates spans around Servlet methods.
To enable them, add `ota.integration.<name>.enabled` system property:
`-Dota.integration.jdbc-datasource.enabled=true`

8
RELEASING.md
View File

@ -9,7 +9,7 @@ to calculate the current version based on git tags. This plugin looks for the la
## Snapshot builds
Every successful CI build of the master branch automatically executes `./gradlew snapshot` as the last task.
This signals Nebula plugin to build and publish to
This signals Nebula plugin to build and publish to
[JFrog OSS repository](https://oss.jfrog.org/artifactory/oss-snapshot-local/io/opentelemetry/auto/)next _minor_ release version.
This means version `vX.(Y+1).0-SNAPSHOT`.
@ -23,12 +23,12 @@ Do the following:
On new tag creation a CI will start a new release build.
It will do the following:
- Checkout requested tag.
- Run `./gradlew -Prelease.useLastTag=true final`.
This signals Nebula plugin to build `X.Y.0` version and to publish it to
- Run `./gradlew -Prelease.useLastTag=true final`.
This signals Nebula plugin to build `X.Y.0` version and to publish it to
[Bintray repository](https://bintray.com/open-telemetry/maven/opentelemetry-java-instrumentation).
## Patch releases
Whenever a fix is needed to any older branch, a PR should be made into the corresponding maintenance branch.
When that PR is merge, CI will notice the new commit into maintenance branch and will initiate a new build for this.
That build, after usual building and checking, will run `./gradlew -Prelease.scope=patch final`.
This will signal Nebula plugin to build a new version `vX.Y.(Z+1)` and publish it to Bintray repo.
This will signal Nebula plugin to build a new version `vX.Y.(Z+1)` and publish it to Bintray repo.

21
build.gradle
View File

@ -8,14 +8,11 @@ plugins {
id 'org.unbroken-dome.test-sets' version '2.2.1'
id 'com.github.ben-manes.versions' version '0.27.0'
// Not applying google java format by default because it gets confused by stray java build
// files in 'workspace' build directory in CI
id 'com.github.sherter.google-java-format' version '0.8' apply false
id 'com.dorongold.task-tree' version '1.5'
id "com.github.johnrengelman.shadow" version "5.2.0"
id "com.diffplug.gradle.spotless" version "3.28.1"
id "com.diffplug.gradle.spotless" version "4.3.0"
id "com.github.spotbugs" version "4.0.1"
}
@ -67,3 +64,19 @@ task writeMuzzleTasksToFile {
.join('\n')
}
}
apply plugin: 'com.diffplug.gradle.spotless'
spotless {
// this formatting is applied at the root level, as some of these files are not in a submodules
// and would be missed otherwise
format 'misc', {
target '**/.gitignore', '**/*.md', '**/*.sh'
indentWithSpaces()
trimTrailingWhitespace()
endWithNewline()
}
}
task formatCode(dependsOn: ['spotlessApply'])
check.dependsOn 'spotlessCheck'

1
buildSrc/build.gradle.kts
View File

@ -1,6 +1,7 @@
plugins {
groovy
`java-gradle-plugin`
id("com.diffplug.gradle.spotless") version "4.3.0"
}
gradlePlugin {

3
buildscripts/pre-commit
View File

@ -1,3 +0,0 @@
#!/usr/bin/env sh
./gradlew googleJavaFormat

12
gradle/checkstyle.gradle
View File

@ -34,15 +34,3 @@ check.dependsOn checkstyleTasks
tasks.withType(Test).configureEach {
mustRunAfter checkstyleTasks
}
// Verification seems broken on Java 9.
apply plugin: 'com.github.sherter.google-java-format'
googleJavaFormat {
source = sourceSets*.allJava
exclude '**/build/**/*.java'
}
tasks.withType(Checkstyle).configureEach {
mustRunAfter verifyGoogleJavaFormat
}

12
gradle/enforcement/spotless-groovy.properties Normal file
View File

@ -0,0 +1,12 @@
# Disable formatting errors
ignoreFormatterProblems=true
org.eclipse.jdt.core.formatter.tabulation.char=space
org.eclipse.jdt.core.formatter.tabulation.size=2
org.eclipse.jdt.core.formatter.indentation.size=1
org.eclipse.jdt.core.formatter.indentation.text_block_indentation=indent by one
org.eclipse.jdt.core.formatter.indent_empty_lines=false
org.eclipse.jdt.core.formatter.continuation_indentation=1
org.eclipse.jdt.core.formatter.continuation_indentation_for_array_initializer=1
groovy.formatter.longListLength=50
groovy.formatter.multiline.indentation=1
groovy.formatter.remove.unnecessary.semicolons=true

7
gradle/spotless.gradle
View File

@ -2,6 +2,7 @@ apply plugin: 'com.diffplug.gradle.spotless'
spotless {
java {
googleJavaFormat()
licenseHeaderFile rootProject.file('gradle/enforcement/spotless.license.java'), '(package|import|public)'
target 'src/**/*.java'
}
@ -9,9 +10,15 @@ spotless {
licenseHeaderFile rootProject.file('gradle/enforcement/spotless.license.java'), '(package|import|class)'
}
scala {
scalafmt()
licenseHeaderFile rootProject.file('gradle/enforcement/spotless.license.java'), '(package|import|public)'
}
kotlin {
// ktfmt() // only supports 4 space indentation
ktlint().userData(['indent_size': '2', 'continuation_indent_size': '2'])
licenseHeaderFile rootProject.file('gradle/enforcement/spotless.license.java'), '(package|import|public)'
}
}
task formatCode(dependsOn: ['spotlessApply'])
check.dependsOn 'spotlessApply'

76
instrumentation-core/spring/README.md
View File

@ -2,11 +2,11 @@
<!-- ReadMe is in progress -->
<!-- TO DO: Add sections for starter guide -->
This package streamlines the manual instrumentation process of OpenTelemetry for [Spring](https://spring.io/projects/spring-framework) and [Spring Boot](https://spring.io/projects/spring-boot) applications. It will enable you to add traces to requests and database calls with minimal changes to application code. This package will not fully automate your OpenTelemetry instrumentation, instead, it will provide you with better tools to instrument your own code.
This package streamlines the manual instrumentation process of OpenTelemetry for [Spring](https://spring.io/projects/spring-framework) and [Spring Boot](https://spring.io/projects/spring-boot) applications. It will enable you to add traces to requests and database calls with minimal changes to application code. This package will not fully automate your OpenTelemetry instrumentation, instead, it will provide you with better tools to instrument your own code.
The [first section](#manual-instrumentation-with-java-sdk) will walk you through span creation and propagation using the OpenTelemetry Java API and [Spring's RestTemplate Http Web Client](https://spring.io/guides/gs/consuming-rest/). This approach will use the "vanilla" OpenTelemetry API to make explicit tracing calls within an application's controller.
The [first section](#manual-instrumentation-with-java-sdk) will walk you through span creation and propagation using the OpenTelemetry Java API and [Spring's RestTemplate Http Web Client](https://spring.io/guides/gs/consuming-rest/). This approach will use the "vanilla" OpenTelemetry API to make explicit tracing calls within an application's controller.
The [second section](#manual-instrumentation-using-handlers-and-filters) will build on the first. It will walk you through implementing spring-web handler and filter interfaces to create traces with minimal changes to existing application code. Using the OpenTelemetry API, this approach involves copy and pasting files and a significant amount of manual configurations.
The [second section](#manual-instrumentation-using-handlers-and-filters) will build on the first. It will walk you through implementing spring-web handler and filter interfaces to create traces with minimal changes to existing application code. Using the OpenTelemetry API, this approach involves copy and pasting files and a significant amount of manual configurations.
The third section will walk you through the annotations and configurations defined in the opentelemetry-instrumentation-spring package. This section will equip you with new tools to streamline the setup and instrumentation of OpenTelemetry on Spring and Spring Boot applications. With these tools you will be able to setup distributed tracing with little to no changes to existing configurations and easily customize traces with minor additions to application code.
@ -16,14 +16,14 @@ In this guide we will be using a running example. In section one and two, we wil
## Create two Spring Projects
Using the [spring project initializer](https://start.spring.io/), we will create two spring projects. Name one project `MainService` and the other `TimeService`. In this example `MainService` will be a client of `TimeService` and they will be dealing with time. Make sure to select maven, Spring Boot 2.3, Java, and add the spring-web dependency. After downloading the two projects include the OpenTelemetry dependencies and configuration listed below.
Using the [spring project initializer](https://start.spring.io/), we will create two spring projects. Name one project `MainService` and the other `TimeService`. In this example `MainService` will be a client of `TimeService` and they will be dealing with time. Make sure to select maven, Spring Boot 2.3, Java, and add the spring-web dependency. After downloading the two projects include the OpenTelemetry dependencies and configuration listed below.
## Setup for Manual Instrumentation
Add the dependencies below to enable OpenTelemetry in `MainService` and `TimeService`. The Jaeger and LoggingExporter packages are recommended for exporting traces but are not required. As of May 2020, Jaeger, Zipkin, OTLP, and Logging exporters are supported by opentelemetry-java. Feel free to use whatever exporter you are most comfortable with.
Add the dependencies below to enable OpenTelemetry in `MainService` and `TimeService`. The Jaeger and LoggingExporter packages are recommended for exporting traces but are not required. As of May 2020, Jaeger, Zipkin, OTLP, and Logging exporters are supported by opentelemetry-java. Feel free to use whatever exporter you are most comfortable with.
### Maven
#### OpenTelemetry
```xml
<dependency>
@ -35,7 +35,7 @@ Add the dependencies below to enable OpenTelemetry in `MainService` and `TimeSer
<groupId>io.opentelemetry</groupId>
<artifactId>opentelemetry-sdk</artifactId>
<version>0.5.0</version>
</dependency>
</dependency>
<dependency>
<groupId>io.grpc</groupId>
<artifactId>grpc-context</artifactId>
@ -73,7 +73,7 @@ Add the dependencies below to enable OpenTelemetry in `MainService` and `TimeSer
```
### Gradle
#### OpenTelemetry
```gradle
compile "io.opentelemetry:opentelemetry-api:0.5.0"
@ -97,7 +97,7 @@ compile "io.grpc:grpc-netty:1.27.2"
To enable tracing in your OpenTelemetry project configure a Tracer Bean. This bean will be auto wired to controllers to create and propagate spans. This can be seen in the `Tracer otelTracer()` method below. If you plan to use a trace exporter remember to also include it in this configuration class. In section 3 we will use an annotation to set up this configuration.
A sample OpenTelemetry configuration using LoggingExporter is shown below:
A sample OpenTelemetry configuration using LoggingExporter is shown below:
```java
import org.springframework.context.annotation.Bean;
@ -114,21 +114,21 @@ import io.opentelemetry.exporters.logging.*;
@Configuration
public class OtelConfig {
private static final tracerName = "fooTracer";
private static final tracerName = "fooTracer";
@Bean
public Tracer otelTracer() throws Exception {
final Tracer tracer = OpenTelemetry.getTracer(tracerName);
SpanProcessor logProcessor = SimpleSpanProcessor.newBuilder(new LoggingSpanExporter()).build();
OpenTelemetrySdk.getTracerProvider().addSpanProcessor(logProcessor);
return tracer;
}
}
```
The file above configures an OpenTelemetry tracer and a span processor. The span processor builds a log exporter which will output spans to the console. Similarly, one could add another exporter, such as the `JaegerExporter`, to visualize traces on a different back-end. Similar to how the `LoggingExporter` is configured, a Jaeger configuration can be added to the `OtelConfig` class above.
The file above configures an OpenTelemetry tracer and a span processor. The span processor builds a log exporter which will output spans to the console. Similarly, one could add another exporter, such as the `JaegerExporter`, to visualize traces on a different back-end. Similar to how the `LoggingExporter` is configured, a Jaeger configuration can be added to the `OtelConfig` class above.
Sample configuration for a Jaeger Exporter:
@ -141,11 +141,11 @@ SpanProcessor jaegerProcessor = SimpleSpanProcessor
.build();
OpenTelemetrySdk.getTracerProvider().addSpanProcessor(jaegerProcessor);
```
### Project Background
Here we will create REST controllers for `MainService` and `TimeService`.
`MainService` will send a GET request to `TimeService` to retrieve the current time. After this request is resolved, `MainService` then will append a message to time and return a string to the client.
`MainService` will send a GET request to `TimeService` to retrieve the current time. After this request is resolved, `MainService` then will append a message to time and return a string to the client.
## Manual Instrumentation with Java SDK
@ -190,7 +190,7 @@ import HttpUtils;
public class MainServiceController {
private static int requestCount = 1;
private static final String TIME_SERVICE_URL = "http://localhost:8081/time";
@Autowired
private Tracer tracer;
@ -246,7 +246,7 @@ public class HttpUtils {
headers.set(key, value);
}
};
@Autowired
private Tracer tracer;
@ -314,7 +314,7 @@ public class TimeServiceController {
@GetMapping
public String time() {
Span span = tracer.spanBuilder("time").startSpan();
try (Scope scope = tracer.withSpan(span)) {
span.addEvent("TimeServiceController Entered");
span.setAttribute("what.am.i", "Tu es une legume");
@ -328,17 +328,17 @@ public class TimeServiceController {
### Run MainService and TimeService
***To view your distributed traces ensure either LogExporter or Jaeger is configured in the OtelConfig.java file***
***To view your distributed traces ensure either LogExporter or Jaeger is configured in the OtelConfig.java file***
To view traces on the Jaeger UI, deploy a Jaeger Exporter on localhost by running the command in terminal:
`docker run --rm -it --network=host jaegertracing/all-in-one`
`docker run --rm -it --network=host jaegertracing/all-in-one`
After running Jaeger locally, navigate to the url below. Make sure to refresh the UI to view the exported traces from the two web services:
`http://localhost:16686`
Run MainService and TimeService from command line or using an IDE. The end point of interest for MainService is `http://localhost:8080/message` and `http://localhost:8081/time` for TimeService. Entering `localhost:8080/message` in a browser should call MainService and then TimeService, creating a trace.
Run MainService and TimeService from command line or using an IDE. The end point of interest for MainService is `http://localhost:8080/message` and `http://localhost:8081/time` for TimeService. Entering `localhost:8080/message` in a browser should call MainService and then TimeService, creating a trace.
***Note: The default port for the Apache Tomcat is 8080. On localhost both MainService and TimeService services will attempt to run on this port raising an error. To avoid this add `server.port=8081` to the resources/application.properties file. Ensure the port specified corresponds to port referenced by MainServiceController.TIME_SERVICE_URL. ***
@ -346,13 +346,13 @@ Congrats, we just created a distributed service with OpenTelemetry!
## Manual Instrumentation using Handlers and Filters
In this section, we will implement the javax Servlet Filter interface to wrap all requests to MainService and TimeService controllers in a span.
In this section, we will implement the javax Servlet Filter interface to wrap all requests to MainService and TimeService controllers in a span.
We will also use the RestTemplate HTTP client to send requests from MainService to TimeService. To propagate the trace in this request we will also implement the ClientHttpRequestInterceptor interface. This implementation is only required for projects that send outbound requests. In this example it is only required for MainService.
We will also use the RestTemplate HTTP client to send requests from MainService to TimeService. To propagate the trace in this request we will also implement the ClientHttpRequestInterceptor interface. This implementation is only required for projects that send outbound requests. In this example it is only required for MainService.
### Set up MainService and TimeService
Using the earlier instructions [create two spring projects](#create-two-spring-projects) and add the required [dependencies and configurations](#setup-for-manual-instrumentation).
Using the earlier instructions [create two spring projects](#create-two-spring-projects) and add the required [dependencies and configurations](#setup-for-manual-instrumentation).
### Instrumentation of TimeService
@ -396,16 +396,16 @@ public class TimeServiceController {
#### Create Controller Filter
Add the class below to wrap all requests to the TimeServiceController in a span. This class will call the preHandle method before the REST controller is entered and the postHandle method after a response is created.
Add the class below to wrap all requests to the TimeServiceController in a span. This class will call the preHandle method before the REST controller is entered and the postHandle method after a response is created.
The preHandle method starts a span for each request. This implementation is shown below:
The preHandle method starts a span for each request. This implementation is shown below:
```java
@Component
public class ControllerFilter implements Filter {
private static final Logger LOG = Logger.getLogger(ControllerFilter.class.getName());
@Autowired
Tracer tracer;
@ -420,7 +420,7 @@ public class ControllerFilter implements Filter {
public void doFilter(ServletRequest request, ServletResponse response, FilterChain chain)
throws IOException, ServletException {
LOG.info("start doFilter");
HttpServletRequest req = (HttpServletRequest) request;
Span currentSpan;
try (Scope scope = tracer.withSpan(currentSpan)) {
@ -432,10 +432,10 @@ public class ControllerFilter implements Filter {
} finally {
currentSpan.end();
}
LOG.info("end doFilter");
LOG.info("end doFilter");
}
private Span createSpanWithParent(HttpServletRequest request, Context context) {
return tracer.spanBuilder(request.getRequestURI()).setSpanKind(Span.Kind.SERVER).startSpan();
}
@ -478,7 +478,7 @@ public class MainServiceController {
@Autowired
private Tracer tracer;
@Autowired
private RestTemplate restTemplate;
@ -504,7 +504,7 @@ As seen in the setup of TimeService, implement the javax servlet filter interfac
Next, we will configure the ClientHttpRequestInterceptor to intercept all client HTTP requests made using RestTemplate.
To propagate the span context from MainService to TimeService we must inject the trace parent and trace state into the outgoing request header. In section 1 this was done using the helper class HttpUtils. In this section, we will implement the ClientHttpRequestInterceptor interface and register this interceptor in our application.
To propagate the span context from MainService to TimeService we must inject the trace parent and trace state into the outgoing request header. In section 1 this was done using the helper class HttpUtils. In this section, we will implement the ClientHttpRequestInterceptor interface and register this interceptor in our application.
Include the two classes below to your MainService project to add this functionality:
@ -544,10 +544,10 @@ public class RestTemplateInterceptor implements ClientHttpRequestInterceptor {
@Override
public ClientHttpResponse intercept(HttpRequest request, byte[] body,
ClientHttpRequestExecution execution) throws IOException {
String spanName = request.getMethodValue() + " " + request.getURI().toString();
Span currentSpan = tracer.spanBuilder(spanName).setSpanKind(Span.Kind.CLIENT).startSpan();
try (Scope scope = tracer.withSpan(currentSpan)) {
OpenTelemetry.getPropagators().getHttpTextFormat().inject(Context.current(), request, setter);
ClientHttpResponse response = execution.execute(request, body);
@ -589,12 +589,12 @@ public class RestClientConfig {
}
```
### Create a distributed trace
### Create a distributed trace
By default Spring Boot runs a Tomcat server on port 8080. This tutorial assumes MainService runs on the default port (8080) and TimeService runs on port 8081. This is because we hard coded the TimeService end point in MainServiceController.TIME_SERVICE_URL. To run TimeServiceApplication on port 8081 include `server.port=8081` in the resources/application.properties file.
By default Spring Boot runs a Tomcat server on port 8080. This tutorial assumes MainService runs on the default port (8080) and TimeService runs on port 8081. This is because we hard coded the TimeService end point in MainServiceController.TIME_SERVICE_URL. To run TimeServiceApplication on port 8081 include `server.port=8081` in the resources/application.properties file.
Run both the MainService and TimeService projects in terminal or using an IDE (ex. Eclipse). The end point for MainService should be `http://localhost:8080/message` and `http://localhost:8081/time` for TimeService. Type both urls in a browser and ensure you receive a 200 response.
Run both the MainService and TimeService projects in terminal or using an IDE (ex. Eclipse). The end point for MainService should be `http://localhost:8080/message` and `http://localhost:8081/time` for TimeService. Type both urls in a browser and ensure you receive a 200 response.
To visualize this trace add a trace exporter to one or both of your applications. Instructions on how to setup LogExporter and Jaeger can be seen [above](#tracer-configuration).
To visualize this trace add a trace exporter to one or both of your applications. Instructions on how to setup LogExporter and Jaeger can be seen [above](#tracer-configuration).
To create a sample trace enter `localhost:8080/message` in a browser. This trace should include a span for MainService and a span for TimeService.

36
instrumentation/akka-http-10.0/src/test/scala/AkkaHttpTestAsyncWebServer.scala
View File

@ -34,20 +34,27 @@ object AkkaHttpTestAsyncWebServer {
val asyncHandler: HttpRequest => Future[HttpResponse] = {
case HttpRequest(GET, uri: Uri, _, _, _) =>
Future {
val endpoint = HttpServerTest.ServerEndpoint.forPath(uri.path.toString())
HttpServerTest.controller(endpoint, new Closure[HttpResponse](()) {
def doCall(): HttpResponse = {
val resp = HttpResponse(status = endpoint.getStatus) //.withHeaders(headers.Type)resp.contentType = "text/plain"
endpoint match {
case SUCCESS => resp.withEntity(endpoint.getBody)
case QUERY_PARAM => resp.withEntity(uri.queryString().orNull)
case REDIRECT => resp.withHeaders(headers.Location(endpoint.getBody))
case ERROR => resp.withEntity(endpoint.getBody)
case EXCEPTION => throw new Exception(endpoint.getBody)
case _ => HttpResponse(status = NOT_FOUND.getStatus).withEntity(NOT_FOUND.getBody)
val endpoint =
HttpServerTest.ServerEndpoint.forPath(uri.path.toString())
HttpServerTest.controller(
endpoint,
new Closure[HttpResponse](()) {
def doCall(): HttpResponse = {
val resp = HttpResponse(status = endpoint.getStatus) //.withHeaders(headers.Type)resp.contentType = "text/plain"
endpoint match {
case SUCCESS => resp.withEntity(endpoint.getBody)
case QUERY_PARAM => resp.withEntity(uri.queryString().orNull)
case REDIRECT =>
resp.withHeaders(headers.Location(endpoint.getBody))
case ERROR => resp.withEntity(endpoint.getBody)
case EXCEPTION => throw new Exception(endpoint.getBody)
case _ =>
HttpResponse(status = NOT_FOUND.getStatus)
.withEntity(NOT_FOUND.getBody)
}
}
}
})
)
}
}
@ -56,7 +63,10 @@ object AkkaHttpTestAsyncWebServer {
def start(port: Int): Unit = synchronized {
if (null == binding) {
import scala.concurrent.duration._
binding = Await.result(Http().bindAndHandleAsync(asyncHandler, "localhost", port), 10 seconds)
binding = Await.result(
Http().bindAndHandleAsync(asyncHandler, "localhost", port),
10 seconds
)
}
}

33
instrumentation/akka-http-10.0/src/test/scala/AkkaHttpTestSyncWebServer.scala
View File

@ -34,19 +34,25 @@ object AkkaHttpTestSyncWebServer {
val syncHandler: HttpRequest => HttpResponse = {
case HttpRequest(GET, uri: Uri, _, _, _) => {
val endpoint = HttpServerTest.ServerEndpoint.forPath(uri.path.toString())
HttpServerTest.controller(endpoint, new Closure[HttpResponse](()) {
def doCall(): HttpResponse = {
val resp = HttpResponse(status = endpoint.getStatus)
endpoint match {
case SUCCESS => resp.withEntity(endpoint.getBody)
case QUERY_PARAM => resp.withEntity(uri.queryString().orNull)
case REDIRECT => resp.withHeaders(headers.Location(endpoint.getBody))
case ERROR => resp.withEntity(endpoint.getBody)
case EXCEPTION => throw new Exception(endpoint.getBody)
case _ => HttpResponse(status = NOT_FOUND.getStatus).withEntity(NOT_FOUND.getBody)
HttpServerTest.controller(
endpoint,
new Closure[HttpResponse](()) {
def doCall(): HttpResponse = {
val resp = HttpResponse(status = endpoint.getStatus)
endpoint match {
case SUCCESS => resp.withEntity(endpoint.getBody)
case QUERY_PARAM => resp.withEntity(uri.queryString().orNull)
case REDIRECT =>
resp.withHeaders(headers.Location(endpoint.getBody))
case ERROR => resp.withEntity(endpoint.getBody)
case EXCEPTION => throw new Exception(endpoint.getBody)
case _ =>
HttpResponse(status = NOT_FOUND.getStatus)
.withEntity(NOT_FOUND.getBody)
}
}
}
})
)
}
}
@ -55,7 +61,10 @@ object AkkaHttpTestSyncWebServer {
def start(port: Int): Unit = synchronized {
if (null == binding) {
import scala.concurrent.duration._
binding = Await.result(Http().bindAndHandleSync(syncHandler, "localhost", port), 10 seconds)
binding = Await.result(
Http().bindAndHandleSync(syncHandler, "localhost", port),
10 seconds
)
}
}

16
instrumentation/akka-http-10.0/src/test/scala/AkkaHttpTestWebServer.scala
View File

@ -33,14 +33,21 @@ object AkkaHttpTestWebServer {
implicit val executionContext = system.dispatcher
val exceptionHandler = ExceptionHandler {
case ex: Exception => complete(HttpResponse(status = EXCEPTION.getStatus).withEntity(ex.getMessage))
case ex: Exception =>
complete(
HttpResponse(status = EXCEPTION.getStatus).withEntity(ex.getMessage)
)
}
val route = { //handleExceptions(exceptionHandler) {
path(SUCCESS.rawPath) {
complete(HttpResponse(status = SUCCESS.getStatus).withEntity(SUCCESS.getBody))
complete(
HttpResponse(status = SUCCESS.getStatus).withEntity(SUCCESS.getBody)
)
} ~ path(QUERY_PARAM.rawPath) {
complete(HttpResponse(status = QUERY_PARAM.getStatus).withEntity(SUCCESS.getBody))
complete(
HttpResponse(status = QUERY_PARAM.getStatus).withEntity(SUCCESS.getBody)
)
} ~ path(REDIRECT.rawPath) {
redirect(Uri(REDIRECT.getBody), StatusCodes.Found)
} ~ path(ERROR.rawPath) {
@ -55,7 +62,8 @@ object AkkaHttpTestWebServer {
def start(port: Int): Unit = synchronized {
if (null == binding) {
import scala.concurrent.duration._
binding = Await.result(Http().bindAndHandle(route, "localhost", port), 10 seconds)
binding =
Await.result(Http().bindAndHandle(route, "localhost", port), 10 seconds)
}
}

2
instrumentation/finatra-2.9/src/test/scala/ResponseSettingExceptionMapper.scala
View File

@ -21,7 +21,7 @@ import javax.inject.{Inject, Singleton}
@Singleton
class ResponseSettingExceptionMapper @Inject()(response: ResponseBuilder)
extends ExceptionMapper[Exception] {
extends ExceptionMapper[Exception] {
override def toResponse(request: Request, exception: Exception): Response = {
response.internalServerError(exception.getMessage)

15
instrumentation/java-concurrent/akka-testing/src/test/scala/AkkaActors.scala
View File

@ -24,7 +24,8 @@ import scala.concurrent.duration._
// ! == send-message
object AkkaActors {
val TRACER: Tracer = OpenTelemetry.getTracerProvider.get("io.opentelemetry.auto")
val TRACER: Tracer =
OpenTelemetry.getTracerProvider.get("io.opentelemetry.auto")
val system: ActorSystem = ActorSystem("helloAkka")
@ -33,8 +34,10 @@ object AkkaActors {
val howdyGreeter: ActorRef =
system.actorOf(Greeter.props("Howdy", printer), "howdyGreeter")
val forwarder: ActorRef = system.actorOf(Forwarder.props(printer), "forwarderActor")
val helloGreeter: ActorRef = system.actorOf(Greeter.props("Hello", forwarder), "helloGreeter")
val forwarder: ActorRef =
system.actorOf(Forwarder.props(printer), "forwarderActor")
val helloGreeter: ActorRef =
system.actorOf(Greeter.props("Hello", forwarder), "helloGreeter")
def tracedChild(opName: String): Unit = {
TRACER.spanBuilder(opName).startSpan().end()
@ -86,7 +89,8 @@ class AkkaActors {
}
object Greeter {
def props(message: String, receiverActor: ActorRef): Props = Props(new Greeter(message, receiverActor))
def props(message: String, receiverActor: ActorRef): Props =
Props(new Greeter(message, receiverActor))
final case class WhoToGreet(who: String)
@ -129,7 +133,8 @@ class Receiver extends Actor with ActorLogging {
}
object Forwarder {
def props(receiverActor: ActorRef): Props = Props(new Forwarder(receiverActor))
def props(receiverActor: ActorRef): Props =
Props(new Forwarder(receiverActor))
}
class Forwarder(receiverActor: ActorRef) extends Actor with ActorLogging {

14
instrumentation/java-concurrent/kotlin-testing/src/test/kotlin/KotlinCoroutineTests.kt
View File

@ -17,13 +17,22 @@
import io.opentelemetry.OpenTelemetry
import io.opentelemetry.trace.Tracer
import io.opentelemetry.trace.TracingContextUtils.currentContextWith
import kotlinx.coroutines.*
import java.util.concurrent.TimeUnit
import kotlinx.coroutines.CompletableDeferred
import kotlinx.coroutines.CoroutineDispatcher
import kotlinx.coroutines.CoroutineScope
import kotlinx.coroutines.CoroutineStart
import kotlinx.coroutines.async
import kotlinx.coroutines.awaitAll
import kotlinx.coroutines.channels.actor
import kotlinx.coroutines.channels.consumeEach
import kotlinx.coroutines.channels.produce
import kotlinx.coroutines.channels.toChannel
import kotlinx.coroutines.launch
import kotlinx.coroutines.runBlocking
import kotlinx.coroutines.selects.select
import java.util.concurrent.TimeUnit
import kotlinx.coroutines.withTimeout
import kotlinx.coroutines.yield
class KotlinCoroutineTests(private val dispatcher: CoroutineDispatcher) {
val tracer: Tracer = OpenTelemetry.getTracerProvider().get("io.opentelemetry.auto")
@ -144,4 +153,3 @@ class KotlinCoroutineTests(private val dispatcher: CoroutineDispatcher) {
}
}
}

58
instrumentation/java-concurrent/scala-testing/src/test/scala/ScalaConcurrentTests.scala
View File

@ -24,11 +24,12 @@ import scala.concurrent.duration._
import scala.concurrent.{Await, Future, Promise}
class ScalaConcurrentTests {
val TRACER: Tracer = OpenTelemetry.getTracerProvider.get("io.opentelemetry.auto")
val TRACER: Tracer =
OpenTelemetry.getTracerProvider.get("io.opentelemetry.auto")
/**
* @return Number of expected spans in the trace
*/
* @return Number of expected spans in the trace
*/
def traceWithFutureAndCallbacks() {
val parentSpan = TRACER.spanBuilder("parent").startSpan()
val parentScope = TRACER.withSpan(parentSpan)
@ -71,14 +72,15 @@ class ScalaConcurrentTests {
1
}
goodFuture onSuccess {
case _ => Future {
2
} onSuccess {
case _ => {
tracedChild("callback")
latch.countDown()
case _ =>
Future {
2
} onSuccess {
case _ => {
tracedChild("callback")
latch.countDown()
}
}
}
}
latch.await()
@ -89,8 +91,8 @@ class ScalaConcurrentTests {
}
/**
* @return Number of expected spans in the trace
*/
* @return Number of expected spans in the trace
*/
def traceWithPromises() {
val parentSpan = TRACER.spanBuilder("parent").startSpan()
val parentScope = TRACER.withSpan(parentSpan)
@ -137,26 +139,22 @@ class ScalaConcurrentTests {
}
/**
* @return Number of expected spans in the trace
*/
* @return Number of expected spans in the trace
*/
def tracedWithFutureFirstCompletions() {
val parentSpan = TRACER.spanBuilder("parent").startSpan()
val parentScope = TRACER.withSpan(parentSpan)
try {
val completedVal = Future.firstCompletedOf(
List(
Future {
tracedChild("timeout1")
false
},
Future {
tracedChild("timeout2")
false
},
Future {
tracedChild("timeout3")
true
}))
val completedVal = Future.firstCompletedOf(List(Future {
tracedChild("timeout1")
false
}, Future {
tracedChild("timeout2")
false
}, Future {
tracedChild("timeout3")
true
}))
Await.result(completedVal, 30 seconds)
} finally {
parentSpan.end()
@ -165,8 +163,8 @@ class ScalaConcurrentTests {
}
/**
* @return Number of expected spans in the trace
*/
* @return Number of expected spans in the trace
*/
def tracedTimeout(): Integer = {
val parentSpan = TRACER.spanBuilder("parent").startSpan()
val parentScope = TRACER.withSpan(parentSpan)

13
instrumentation/java-concurrent/src/slickTest/scala/SlickUtils.scala
View File

@ -22,11 +22,13 @@ import scala.concurrent.duration.Duration
import scala.concurrent.{Await, Future}
class SlickUtils {
val TRACER: Tracer = OpenTelemetry.getTracerProvider.get("io.opentelemetry.auto")
val TRACER: Tracer =
OpenTelemetry.getTracerProvider.get("io.opentelemetry.auto")
import SlickUtils._
val database = Database.forURL(Url,
val database = Database.forURL(
Url,
user = Username,
driver = "org.h2.Driver",
keepAliveConnection = true,
@ -34,7 +36,12 @@ class SlickUtils {
// wrapped runnables.
executor = AsyncExecutor("test", numThreads = 1, queueSize = 1000)
)
Await.result(database.run(sqlu"""CREATE ALIAS IF NOT EXISTS SLEEP FOR "java.lang.Thread.sleep(long)""""), Duration.Inf)
Await.result(
database.run(
sqlu"""CREATE ALIAS IF NOT EXISTS SLEEP FOR "java.lang.Thread.sleep(long)""""
),
Duration.Inf
)
def startQuery(query: String): Future[Vector[Int]] = {
val span = TRACER.spanBuilder("run query").startSpan()

8
instrumentation/jdbc/jdbc.gradle
View File

@ -1,5 +1,5 @@
plugins {
id 'com.intershop.gradle.javacc' version '4.0.0'
id 'com.intershop.gradle.javacc' version '4.0.0'
}
apply from: "$rootDir/gradle/instrumentation.gradle"
@ -21,9 +21,6 @@ javacc {
}
}
tasks.withType(com.github.sherter.googlejavaformatgradleplugin.VerifyGoogleJavaFormat).configureEach {
exclude '**/jdbc/normalizer/*.java'
}
tasks.withType(Checkstyle).configureEach {
exclude '**/jdbc/normalizer/*.java'
}
@ -36,7 +33,8 @@ testSets {
dependencies {
// jdbc unit testing
testCompile group: 'com.h2database', name: 'h2', version: '1.3.169' // first version jdk 1.6 compatible
testCompile group: 'com.h2database', name: 'h2', version: '1.3.169'
// first version jdk 1.6 compatible
testCompile group: 'org.apache.derby', name: 'derby', version: '10.6.1.0'
testCompile group: 'org.hsqldb', name: 'hsqldb', version: '2.0.0'

66
instrumentation/play/play-2.3/src/test/scala/server/AsyncServer.scala
View File

@ -25,18 +25,64 @@ import scala.concurrent.Future
object AsyncServer {
val routes: PartialFunction[(String, String), Handler] = {
case ("GET", "/success") => Action.async { request => HttpServerTest.controller(SUCCESS, new AsyncControllerClosureAdapter(Future.successful(Results.Status(SUCCESS.getStatus).apply(SUCCESS.getBody)))) }
case ("GET", "/redirect") => Action.async { request => HttpServerTest.controller(REDIRECT, new AsyncControllerClosureAdapter(Future.successful(Results.Redirect(REDIRECT.getBody, REDIRECT.getStatus)))) }
case ("GET", "/query") => Action.async { result => HttpServerTest.controller(QUERY_PARAM, new AsyncControllerClosureAdapter(Future.successful(Results.Status(QUERY_PARAM.getStatus).apply(QUERY_PARAM.getBody)))) }
case ("GET", "/error-status") => Action.async { result => HttpServerTest.controller(ERROR, new AsyncControllerClosureAdapter(Future.successful(Results.Status(ERROR.getStatus).apply(ERROR.getBody)))) }
case ("GET", "/exception") => Action.async { result =>
HttpServerTest.controller(EXCEPTION, new AsyncBlockClosureAdapter(() => {
throw new Exception(EXCEPTION.getBody)
}))
}
case ("GET", "/success") =>
Action.async { request =>
HttpServerTest.controller(
SUCCESS,
new AsyncControllerClosureAdapter(
Future.successful(
Results.Status(SUCCESS.getStatus).apply(SUCCESS.getBody)
)
)
)
}
case ("GET", "/redirect") =>
Action.async { request =>
HttpServerTest.controller(
REDIRECT,
new AsyncControllerClosureAdapter(
Future.successful(
Results.Redirect(REDIRECT.getBody, REDIRECT.getStatus)
)
)
)
}
case ("GET", "/query") =>
Action.async { result =>
HttpServerTest.controller(
QUERY_PARAM,
new AsyncControllerClosureAdapter(
Future.successful(
Results.Status(QUERY_PARAM.getStatus).apply(QUERY_PARAM.getBody)
)
)
)
}
case ("GET", "/error-status") =>
Action.async { result =>
HttpServerTest.controller(
ERROR,
new AsyncControllerClosureAdapter(
Future
.successful(Results.Status(ERROR.getStatus).apply(ERROR.getBody))
)
)
}
case ("GET", "/exception") =>
Action.async { result =>
HttpServerTest.controller(
EXCEPTION,
new AsyncBlockClosureAdapter(() => {
throw new Exception(EXCEPTION.getBody)
})
)
}
}
def server(port: Int): TestServer = {
TestServer(port, FakeApplication(withGlobal = Some(new Settings()), withRoutes = routes))
TestServer(
port,
FakeApplication(withGlobal = Some(new Settings()), withRoutes = routes)
)
}
}

6
instrumentation/play/play-2.3/src/test/scala/server/ControllerClosureAdapter.scala
View File

@ -29,10 +29,12 @@ class BlockClosureAdapter(block: () => Result) extends Closure[Result] {
override def call(): Result = block()
}
class AsyncControllerClosureAdapter(response: Future[Result]) extends Closure[Future[Result]] {
class AsyncControllerClosureAdapter(response: Future[Result])
extends Closure[Future[Result]] {
override def call(): Future[Result] = response
}
class AsyncBlockClosureAdapter(block: () => Future[Result]) extends Closure[Future[Result]] {
class AsyncBlockClosureAdapter(block: () => Future[Result])
extends Closure[Future[Result]] {
override def call(): Future[Result] = block()
}

5
instrumentation/play/play-2.3/src/test/scala/server/Settings.scala
View File

@ -22,7 +22,10 @@ import play.api.mvc.{RequestHeader, Result, Results}
import scala.concurrent.Future
class Settings extends GlobalSettings {
override def onError(request: RequestHeader, ex: Throwable): Future[Result] = {
override def onError(
request: RequestHeader,
ex: Throwable
): Future[Result] = {
Future.successful(Results.InternalServerError(ex.getCause.getMessage))
}
}

64
instrumentation/play/play-2.3/src/test/scala/server/SyncServer.scala
View File

@ -23,26 +23,54 @@ import play.api.test.{FakeApplication, TestServer}
object SyncServer {
val routes: PartialFunction[(String, String), Handler] = {
case ("GET", "/success") => Action { request =>
HttpServerTest.controller(SUCCESS, new ControllerClosureAdapter(Results.Status(SUCCESS.getStatus).apply(SUCCESS.getBody)))
}
case ("GET", "/redirect") => Action { request =>
HttpServerTest.controller(REDIRECT, new ControllerClosureAdapter(Results.Redirect(REDIRECT.getBody, REDIRECT.getStatus)))
}
case ("GET", "/query") => Action { request =>
HttpServerTest.controller(QUERY_PARAM, new ControllerClosureAdapter(Results.Status(QUERY_PARAM.getStatus).apply(QUERY_PARAM.getBody)))
}
case ("GET", "/error-status") => Action { request =>
HttpServerTest.controller(ERROR, new ControllerClosureAdapter(Results.Status(ERROR.getStatus).apply(ERROR.getBody)))
}
case ("GET", "/exception") => Action { request =>
HttpServerTest.controller(EXCEPTION, new BlockClosureAdapter(() => {
throw new Exception(EXCEPTION.getBody)
}))
}
case ("GET", "/success") =>
Action { request =>
HttpServerTest.controller(
SUCCESS,
new ControllerClosureAdapter(
Results.Status(SUCCESS.getStatus).apply(SUCCESS.getBody)
)
)
}
case ("GET", "/redirect") =>
Action { request =>
HttpServerTest.controller(
REDIRECT,
new ControllerClosureAdapter(
Results.Redirect(REDIRECT.getBody, REDIRECT.getStatus)
)
)
}
case ("GET", "/query") =>
Action { request =>
HttpServerTest.controller(
QUERY_PARAM,
new ControllerClosureAdapter(
Results.Status(QUERY_PARAM.getStatus).apply(QUERY_PARAM.getBody)
)
)
}
case ("GET", "/error-status") =>
Action { request =>
HttpServerTest.controller(
ERROR,
new ControllerClosureAdapter(
Results.Status(ERROR.getStatus).apply(ERROR.getBody)
)
)
}
case ("GET", "/exception") =>
Action { request =>
HttpServerTest.controller(EXCEPTION, new BlockClosureAdapter(() => {
throw new Exception(EXCEPTION.getBody)
}))
}
}
def server(port: Int): TestServer = {
TestServer(port, FakeApplication(withGlobal = Some(new Settings()), withRoutes = routes))
TestServer(
port,
FakeApplication(withGlobal = Some(new Settings()), withRoutes = routes)
)
}
}

8
instrumentation/servlet/README.md
View File

@ -30,7 +30,7 @@ In the example above this is `ApplicationFilterChain.doFilter(ServletRequest, Se
Let us call this first servlet specific method an "entry point".
This is the main target for `Servlet3Instrumentation` and `Servlet2Instrumentation` instrumenters:
`public void javax.servlet.FilterChain#doFilter(ServletRequest, ServletResponse)`
`public void javax.servlet.FilterChain#doFilter(ServletRequest, ServletResponse)`
`public void javax.servlet.http.HttpServlet#service(ServletRequest, ServletResponse)`.
@ -39,7 +39,7 @@ the second method as instrumentation entry point.
These instrumentations are located in two separate submodules `request-3.0` and `request-2.3`, respectively,
because they and corresponding tests depend on different versions of servlet specification.
Next, request passes several other methods from Servlet specification, such as
Next, request passes several other methods from Servlet specification, such as
`protected void javax.servlet.http.HttpServlet#service(HttpServletRequest, HttpServletResponse)` or
@ -70,5 +70,5 @@ Becaue their root spans will all have the same named based on common entry point
In order to alleviate this problem, instrumentations for specific frameworks, such as Spring MVC here,
_update_ name of the span corresponding to the entry point.
Each framework instrumentation can decide what is the best span name based on framework implementation details.
Of course, still adhering to OpenTelemetry
[semantic conventions](https://github.com/open-telemetry/opentelemetry-specification/blob/master/specification/trace/semantic_conventions/http.md).
Of course, still adhering to OpenTelemetry
[semantic conventions](https://github.com/open-telemetry/opentelemetry-specification/blob/master/specification/trace/semantic_conventions/http.md).