Skip to content
A stats collection and distributed tracing framework
Branch: master
Clone or download
Type Name Latest commit message Commit time
Failed to load latest commit information.
buildscripts Exporter/Stats/Stackdriver: Support Exemplar. (#1771) Mar 14, 2019
contrib Fix a bunch of errorprone warnings. (#1806) Mar 19, 2019
gradle/wrapper gradle wrapper --gradle-version 4.9 (#1360) Aug 7, 2018
impl Remove redundant dependencies between compile and testCompile. (#1660) Jan 15, 2019
testing Remove redundant dependencies between compile and testCompile. (#1660) Jan 15, 2019
.travis.yml Remove build for java7. (#1690) Jan 30, 2019
LICENSE Fixes license file formatting. Oct 4, 2016 elasticsearch trace exporter (#1774) Mar 11, 2019
findbugs-exclude.xml Avoid doing string formatting when calling checkArgument. (#1394) Aug 28, 2018
gradlew update to gradle 4.0.1 (#451) Jul 12, 2017
settings.gradle elasticsearch trace exporter (#1774) Mar 11, 2019

OpenCensus - A stats collection and distributed tracing framework

Gitter chat Maven Central Javadocs Build Status Windows Build Status Coverage Status

OpenCensus is a toolkit for collecting application performance and behavior data. It currently includes 3 apis: stats, tracing and tags.

The library is in Beta stage and APIs are expected to be mostly stable. The library is expected to move to GA stage after v1.0.0 major release.

Please join gitter for help or feedback on this project.

OpenCensus Quickstart for Libraries

Integrating OpenCensus with a new library means recording stats or traces and propagating context. For application integration please see Quickstart for Applications.

The full quick start example can also be found on the OpenCensus website.

Add the dependencies to your project

For Maven add to your pom.xml:


For Gradle add to your dependencies:

compile 'io.opencensus:opencensus-api:0.19.2'

Hello "OpenCensus" trace events

Here's an example of creating a Span and record some trace annotations. Notice that recording the annotations is possible because we propagate scope. 3rd parties libraries like SLF4J can integrate the same way.

import io.opencensus.common.Scope;
import io.opencensus.trace.Tracer;
import io.opencensus.trace.Tracing;
import io.opencensus.trace.samplers.Samplers;

public final class MyClassWithTracing {
  private static final Tracer tracer = Tracing.getTracer();

  public static void doWork() {
    // Create a child Span of the current Span. Always record events for this span and force it to
    // be sampled. This makes it easier to try out the example, but unless you have a clear use
    // case, you don't need to explicitly set record events or sampler.
    try (Scope ss =
            .startScopedSpan()) {
      tracer.getCurrentSpan().addAnnotation("Finished initial work");

  private static void doInitialWork() {
    // ...
    // ...

  private static void doFinalWork() {
    // ...
    tracer.getCurrentSpan().addAnnotation("More important.");
    // ...

Hello "OpenCensus" stats events

Here's an example on

  • defining TagKey, Measure and View,
  • registering a view,
  • putting TagKey and TagValue into a scoped TagContext,
  • recording stats against current TagContext,
  • getting ViewData.

For the complete example, see here.

import io.opencensus.common.Scope;
import io.opencensus.stats.Aggregation;
import io.opencensus.stats.BucketBoundaries;
import io.opencensus.stats.Measure.MeasureLong;
import io.opencensus.stats.Stats;
import io.opencensus.stats.StatsRecorder;
import io.opencensus.stats.View;
import io.opencensus.stats.ViewData;
import io.opencensus.stats.ViewManager;
import io.opencensus.tags.TagKey;
import io.opencensus.tags.TagValue;
import io.opencensus.tags.Tagger;
import io.opencensus.tags.Tags;
import java.util.Arrays;
import java.util.Collections;

public final class MyClassWithStats {
  private static final Tagger tagger = Tags.getTagger();
  private static final ViewManager viewManager = Stats.getViewManager();
  private static final StatsRecorder statsRecorder = Stats.getStatsRecorder();

  // frontendKey allows us to break down the recorded data
  private static final TagKey FRONTEND_KEY = TagKey.create("myorg_keys_frontend");

  // videoSize will measure the size of processed videos.
  private static final MeasureLong VIDEO_SIZE =
      MeasureLong.create("", "size of processed videos", "By");

  // Create view to see the processed video size distribution broken down by frontend.
  // The view has bucket boundaries (0, 256, 65536) that will group measure values into
  // histogram buckets.
  private static final View.Name VIDEO_SIZE_VIEW_NAME = View.Name.create("");
  private static final View VIDEO_SIZE_VIEW =
          "processed video size over time",
              BucketBoundaries.create(Arrays.asList(0.0, 256.0, 65536.0))),

  public static void initialize() {
    // ...

  public static void processVideo() {
    try (Scope scopedTags =
            .put(FRONTEND_KEY, TagValue.create("mobile-ios9.3.5"))
            .buildScoped()) {
      // Processing video.
      // ...

      // Record the processed video size.
      statsRecorder.newMeasureMap().put(VIDEO_SIZE, 25648).record();

  public static void printStats() {
    ViewData viewData = viewManager.getView(VIDEO_SIZE_VIEW_NAME);
        String.format("Recorded stats for %s:\n %s", VIDEO_SIZE_VIEW_NAME.asString(), viewData));

OpenCensus Quickstart for Applications

Besides recording tracing/stats events the application also need to link the implementation, setup exporters, and debugging Z-Pages.

Add the dependencies to your project

For Maven add to your pom.xml:


For Gradle add to your dependencies:

compile 'io.opencensus:opencensus-api:0.19.2'
runtime 'io.opencensus:opencensus-impl:0.19.2'

How to setup exporters?

Trace exporters

Stats exporters

How to setup debugging Z-Pages?

If the application owner wants to export in-process tracing and stats data via HTML debugging pages see this link.


This library follows Semantic Versioning.

GA: Libraries defined at a GA quality level are stable, and will not introduce backwards-incompatible changes in any minor or patch releases. We will address issues and requests with the highest priority. If we were to make a backwards-incompatible changes on an API, we will first mark the existing API as deprecated and keep it for 18 months before removing it.

Beta: Libraries defined at a Beta quality level are expected to be mostly stable and we're working towards their release candidate. We will address issues and requests with a higher priority. There may be backwards incompatible changes in a minor version release, though not in a patch release. If an element is part of an API that is only meant to be used by exporters or other opencensus libraries, then there is no deprecation period. Otherwise, we will deprecate it for 18 months before removing it, if possible.

You can’t perform that action at this time.