jaeger

🛑 This library is being deprecated!

We urge all users to migrate to OpenTelemetry. Please refer to the notice in the documentation for details.

Jaeger Bindings for Go OpenTracing API

Instrumentation library that implements an OpenTracing Go Tracer for Jaeger (https://jaegertracing.io).

IMPORTANT: The library's import path is based on its original location under github.com/uber. Do not try to import it as github.com/jaegertracing, it will not compile. We might revisit this in the next major release.

• ✅ import "github.com/uber/jaeger-client-go"
• ❌ import "github.com/jaegertracing/jaeger-client-go"

How to Contribute

Please see CONTRIBUTING.md.

Installation

Preferred

Add github.com/uber/jaeger-client-go to go.mod.

Old way

We recommended using a dependency manager like dep and semantic versioning when including this library into an application. For example, Jaeger backend imports this library like this:

[[constraint]]
  name = "github.com/uber/jaeger-client-go"
  version = "2.17"

If you instead want to use the latest version in master, you can pull it via go get. Note that during go get you may see build errors due to incompatible dependencies, which is why we recommend using semantic versions for dependencies. The error may be fixed by running make install (it will install dep if you don't have it):

go get -u github.com/uber/jaeger-client-go/
cd $GOPATH/src/github.com/uber/jaeger-client-go/
git submodule update --init --recursive
make install

Initialization

See tracer initialization examples in godoc and config/example_test.go.

There are two ways to create a tracer:

• Using Configuration struct that allows declarative configuration. For example, you can populate that struct from a YAML/JSON config, or ask it to initialize itself using environment variables (see next section).
• Using NewTracer() function that allows for full programmatic control of configuring the tracer using TracerOptions.

Environment variables

The tracer can be initialized with values coming from environment variables, if it is built from a config that was created via FromEnv(). None of the env vars are required and all of them can be overridden via direct setting of the property on the configuration object.

By default, the client sends traces via UDP to the agent at localhost:6831. Use JAEGER_AGENT_HOST and JAEGER_AGENT_PORT to send UDP traces to a different host:port. If JAEGER_ENDPOINT is set, the client sends traces to the endpoint via HTTP, making the JAEGER_AGENT_HOST and JAEGER_AGENT_PORT unused. If JAEGER_ENDPOINT is secured, HTTP basic authentication can be performed by setting the JAEGER_USER and JAEGER_PASSWORD environment variables.

Closing the tracer via io.Closer

The constructor function for Jaeger Tracer returns the tracer itself and an io.Closer instance. It is recommended to structure your main() so that it calls the Close() function on the closer before exiting, e.g.

tracer, closer, err := cfg.NewTracer(...)
defer closer.Close()

This is especially useful for command-line tools that enable tracing, as well as for the long-running apps that support graceful shutdown. For example, if your deployment system sends SIGTERM instead of killing the process and you trap that signal to do a graceful exit, then having defer closer.Close() ensures that all buffered spans are flushed.

Metrics & Monitoring

The tracer emits a number of different metrics, defined in metrics.go. The monitoring backend is expected to support tag-based metric names, e.g. instead of statsd-style string names like counters.my-service.jaeger.spans.started.sampled, the metrics are defined by a short name and a collection of key/value tags, for example: name:jaeger.traces, state:started, sampled:y. See metrics.go file for the full list and descriptions of emitted metrics.

The monitoring backend is represented by the metrics.Factory interface from package "github.com/uber/jaeger-lib/metrics". An implementation of that interface can be passed as an option to either the Configuration object or the Tracer constructor, for example:

import (
    "github.com/uber/jaeger-client-go/config"
    "github.com/uber/jaeger-lib/metrics/prometheus"
)

    metricsFactory := prometheus.New()
    tracer, closer, err := config.Configuration{
        ServiceName: "your-service-name",
    }.NewTracer(
        config.Metrics(metricsFactory),
    )

By default, a no-op metrics.NullFactory is used.

Logging

The tracer can be configured with an optional logger, which will be used to log communication errors, or log spans if a logging reporter option is specified in the configuration. The logging API is abstracted by the Logger interface. A logger instance implementing this interface can be set on the Config object before calling the New method.

Besides the zap implementation bundled with this package there is also a go-kit one in the jaeger-lib repository.

Instrumentation for Tracing

Since this tracer is fully compliant with OpenTracing API 1.0, all code instrumentation should only use the API itself, as described in the opentracing-go documentation.

Features

Reporters

A "reporter" is a component that receives the finished spans and reports them to somewhere. Under normal circumstances, the Tracer should use the default RemoteReporter, which sends the spans out of process via configurable "transport". For testing purposes, one can use an InMemoryReporter that accumulates spans in a buffer and allows to retrieve them for later verification. Also available are NullReporter, a no-op reporter that does nothing, a LoggingReporter which logs all finished spans using their String() method, and a CompositeReporter that can be used to combine more than one reporter into one, e.g. to attach a logging reporter to the main remote reporter.

Span Reporting Transports

The remote reporter uses "transports" to actually send the spans out of process. Currently the supported transports include:

• Jaeger Thrift over UDP or HTTP,
• Zipkin Thrift over HTTP.

Sampling

The tracer does not record all spans, but only those that have the sampling bit set in the flags. When a new trace is started and a new unique ID is generated, a sampling decision is made whether this trace should be sampled. The sampling decision is propagated to all downstream calls via the flags field of the trace context. The following samplers are available:

1. RemotelyControlledSampler uses one of the other simpler samplers and periodically updates it by polling an external server. This allows dynamic control of the sampling strategies.
2. ConstSampler always makes the same sampling decision for all trace IDs. it can be configured to either sample all traces, or to sample none.
3. ProbabilisticSampler uses a fixed sampling rate as a probability for a given trace to be sampled. The actual decision is made by comparing the trace ID with a random number multiplied by the sampling rate.
4. RateLimitingSampler can be used to allow only a certain fixed number of traces to be sampled per second.

Delayed sampling

Version 2.20 introduced the ability to delay sampling decisions in the life cycle of the root span. It involves several features and architectural changes:

• Shared sampling state: the sampling state is shared across all local (i.e. in-process) spans for a given trace.
• New SamplerV2 API allows the sampler to be called at multiple points in the life cycle of a span:
  • on span creation
  • on overwriting span operation name
  • on setting span tags
  • on finishing the span
• Final/non-final sampling state: the new SamplerV2 API allows the sampler to indicate if the negative sampling decision is final or not (positive sampling decisions are always final). If the decision is not final, the sampler will be called again on further span life cycle events, like setting tags.

These new features are used in the experimental x.TagMatchingSampler, which can sample a trace based on a certain tag added to the root span or one of its local (in-process) children. The sampler can be used with another experimental x.PrioritySampler that allows multiple samplers to try to make a sampling decision, in a certain priority order.

Baggage Injection

The OpenTracing spec allows for baggage, which are key value pairs that are added to the span context and propagated throughout the trace. An external process can inject baggage by setting the special HTTP Header jaeger-baggage on a request:

curl -H "jaeger-baggage: key1=value1, key2=value2" http://myhost.com

Baggage can also be programatically set inside your service:

if span := opentracing.SpanFromContext(ctx); span != nil {
    span.SetBaggageItem("key", "value")
}

Another service downstream of that can retrieve the baggage in a similar way:

if span := opentracing.SpanFromContext(ctx); span != nil {
    val := span.BaggageItem("key")
    println(val)
}

Debug Traces (Forced Sampling)

Programmatically

The OpenTracing API defines a sampling.priority standard tag that can be used to affect the sampling of a span and its children:

import (
    "github.com/opentracing/opentracing-go"
    "github.com/opentracing/opentracing-go/ext"
)

span := opentracing.SpanFromContext(ctx)
ext.SamplingPriority.Set(span, 1)

Via HTTP Headers

Jaeger Tracer also understands a special HTTP Header jaeger-debug-id, which can be set in the incoming request, e.g.

curl -H "jaeger-debug-id: some-correlation-id" http://myhost.com

When Jaeger sees this header in the request that otherwise has no tracing context, it ensures that the new trace started for this request will be sampled in the "debug" mode (meaning it should survive all downsampling that might happen in the collection pipeline), and the root span will have a tag as if this statement was executed:

span.SetTag("jaeger-debug-id", "some-correlation-id")

This allows using Jaeger UI to find the trace by this tag.

Zipkin HTTP B3 compatible header propagation

Jaeger Tracer supports Zipkin B3 Propagation HTTP headers, which are used by a lot of Zipkin tracers. This means that you can use Jaeger in conjunction with e.g. these OpenZipkin tracers.

However it is not the default propagation format, see here how to set it up.

SelfRef

Jaeger Tracer supports an additional span reference type call Self, which was proposed to the OpenTracing Specification (https://github.com/opentracing/specification/issues/81) but not yet accepted. This allows the caller to provide an already created SpanContext when starting a new span. The Self reference bypasses trace and span id generation, as well as sampling decisions (i.e. the sampling bit in the SpanContext.flags must be set appropriately by the caller).

The Self reference supports the following use cases:

• the ability to provide externally generated trace and span IDs
• appending data to the same span from different processes, such as loading and continuing spans/traces from offline (ie log-based) storage

Usage requires passing in a SpanContext and the jaeger.Self reference type:

span := tracer.StartSpan(
    "continued_span",
    jaeger.SelfRef(yourSpanContext),
)
...
defer span.Finish()

License

Apache 2.0 License.