Skip to content

Latest commit

聽

History

History
385 lines (297 loc) 路 12.1 KB

running-tracetest-with-tempo.mdx

File metadata and controls

385 lines (297 loc) 路 12.1 KB
Raw
id title description hide_table_of_contents keywords image
running-tracetest-with-tempo
Node.js and Grafana Tempo
Quick start on how to configure a Node.js app with OpenTelemetry traces, Grafana Tempo as a trace data store, and Tracetest for enhancing your E2E and integration tests with trace-based testing.
false
tracetest
trace-based testing
observability
distributed tracing
testing

:::note Check out the source code on GitHub here. :::

Tracetest is a testing tool based on OpenTelemetry that allows you to test your distributed application. It allows you to use data from distributed traces generated by OpenTelemetry to validate and assert if your application has the desired behavior defined by your test definitions.

Grafana Tempo is an open-source, high-scale distributed tracing data store. It utilizes object storage to minimize costs, allowing you to store more information at a lower cost. It is deeply integrated with Grafana, Prometheus, and Loki. Grafana Tempo is also available as a paid hosted service at Grafana Cloud.

Sample Node.js App with Tempo, OpenTelemetry and Tracetest

This is a simple quick start on how to configure a Node.js app to use OpenTelemetry instrumentation with traces and Tracetest for enhancing your E2E and integration tests with trace-based testing. The infrastructure will use Tempo as the trace data store, and OpenTelemetry Collector to receive traces from the Node.js app and send them to Tempo.

Prerequisites

You will need Docker and Docker Compose installed on your machine to run this quick start app!

Project Structure

The project is built with Docker Compose. It contains two distinct docker-compose.yaml files.

1. Node.js App

The docker-compose.yaml file and Dockerfile in the root directory are for the Node.js app.

2. Tracetest

The docker-compose.yaml file, collector.config.yaml, tracetest-provision.yaml, and tracetest-config.yaml in the tracetest directory are for the setting up Tracetest, Tempo, and the OpenTelemetry Collector.

The tracetest directory is self-contained and will run all the prerequisites for enabling OpenTelemetry traces and trace-based testing with Tracetest.

Docker Compose Network

All services in the docker-compose.yaml are on the same network and will be reachable by hostname from within other services. For example, tempo:4317 in the collector.config.yaml will map to the tempo service, where the port 4317 is the port where Tempo accepts traces. And, tempo:9095 in the tracetest-provision.yaml will map to the tempo service and port 9095 where Tracetest will fetch trace data from Tempo.

Node.js App

The Node.js app is a simple Express app, contained in the app.js file.

The OpenTelemetry tracing is contained in the tracing.otel.grpc.js or tracing.otel.http.js files, respectively. Traces will be sent to the OpenTelemetry Collector.

Here's the content of the tracing.otel.grpc.js file:

const opentelemetry = require('@opentelemetry/sdk-node')
const { getNodeAutoInstrumentations } = require('@opentelemetry/auto-instrumentations-node')
const { OTLPTraceExporter } = require('@opentelemetry/exporter-trace-otlp-grpc');

const sdk = new opentelemetry.NodeSDK({
  traceExporter: new OTLPTraceExporter({ url: 'http://otel-collector:4317' }),
  instrumentations: [getNodeAutoInstrumentations()],
})
sdk.start()

Depending on which of these you choose, traces will be sent to either the grpc or http endpoint.

The hostnames and ports for these are:

  • GRPC: http://otel-collector:4317
  • HTTP: http://otel-collector:4318/v1/traces

Enabling the tracer is done by preloading the trace file.

node -r ./tracing.otel.grpc.js app.js

In the package.json you will see two npm scripts for running the respective tracers alongside the app.js.

"scripts": {
  "with-grpc-tracer":"node -r ./tracing.otel.grpc.js app.js",
  "with-http-tracer":"node -r ./tracing.otel.http.js app.js"
},

To start the server, run this command:

npm run with-grpc-tracer
# or
npm run with-http-tracer

As you can see the Dockerfile uses the command above.

FROM node:slim
WORKDIR /usr/src/app
COPY package*.json ./
RUN npm install
COPY . .
EXPOSE 8080
CMD [ "npm", "run", "with-grpc-tracer" ]

And, the docker-compose.yaml contains just one service for the Node.js app.

version: '3'
services:
  app:
    image: quick-start-nodejs
    build: .
    ports:
      - "8080:8080"

To start it, run this command:

docker compose build # optional if you haven't already built the image
docker compose up

This will start the Node.js app. But, you're not sending the traces anywhere.

Let's fix this by configuring Tracetest and OpenTelemetry Collector.

Tracetest

The docker-compose.yaml in the tracetest directory is configured with four services.

  • Postgres - Postgres is a prerequisite for Tracetest to work. It stores trace data when running the trace-based tests.
  • OpenTelemetry Collector - A vendor-agnostic implementation of how to receive, process and export telemetry data.
  • Tempo - Grafana Tempo is an open source, easy-to-use, and high-scale distributed tracing backend.
  • Tracetest - Trace-based testing that generates end-to-end tests automatically from traces.
version: "3"
services:
  tracetest:
    image: kubeshop/tracetest:${TAG:-latest}
    platform: linux/amd64
    volumes:
      - type: bind
        source: ./tracetest/tracetest-config.yaml
        target: /app/tracetest.yaml
      - type: bind
        source: ./tracetest/tracetest-provision.yaml
        target: /app/provisioning.yaml
    ports:
      - 11633:11633
    command: --provisioning-file /app/provisioning.yaml
    depends_on:
      postgres:
        condition: service_healthy
      tempo:
        condition: service_started
      otel-collector:
        condition: service_started
    healthcheck:
      test: ["CMD", "wget", "--spider", "localhost:11633"]
      interval: 1s
      timeout: 3s
      retries: 60
    environment:
      TRACETEST_DEV: ${TRACETEST_DEV}

  postgres:
    image: postgres:14
    environment:
      POSTGRES_PASSWORD: postgres
      POSTGRES_USER: postgres
    healthcheck:
      test: pg_isready -U "$$POSTGRES_USER" -d "$$POSTGRES_DB"
      interval: 1s
      timeout: 5s
      retries: 60

  otel-collector:
    image: otel/opentelemetry-collector-contrib:0.59.0
    command:
      - "--config"
      - "/otel-local-config.yaml"
    volumes:
      - ./tracetest/collector.config.yaml:/otel-local-config.yaml
    depends_on:
      - tempo

  tempo:
    image: grafana/tempo:2.0.0
    command: ["-config.file=/etc/tempo.yaml"]
    volumes:
      - ./tracetest/tempo.config.yaml:/etc/tempo.yaml

Tracetest depends on Postgres, Tempo and the OpenTelemetry Collector. All three services require config files to be loaded via a volume. The volumes are mapped from the root directory into the tracetest directory and the respective config files.

To start both the Node.js app and Tracetest, we will run this command:

docker-compose -f docker-compose.yaml -f tracetest/docker-compose.yaml up # add --build if the images are not built already

The tempo.config.yaml file contains the initial config for running Tempo.

The key takeaway is the server block.

#...
server:
  http_listen_port: 3100
  grpc_listen_port: 9095
#...

We'll use the following code. Check out the full Tempo config for reference.

auth_enabled: false

server:
  http_listen_port: 3100
  grpc_listen_port: 9095

distributor:
  receivers:                           # this configuration will listen on all ports and protocols that tempo is capable of.
    jaeger:                            # the receives all come from the OpenTelemetry collector.  more configuration information can
      protocols:                       # be found there: https://github.com/open-telemetry/opentelemetry-collector/tree/master/receiver
        thrift_http:                   #
        grpc:                          # for a production deployment you should only enable the receivers you need!
        thrift_binary:
        thrift_compact:
    zipkin:
    otlp:
      protocols:
        http:
        grpc:
    opencensus:

ingester:
  trace_idle_period: 10s               # the length of time after a trace has not received spans to consider it complete and flush it
  max_block_bytes: 1_000_000           # cut the head block when it hits this size or ...
  #traces_per_block: 1_000_000
  max_block_duration: 5m               #   this much time passes

compactor:
  compaction:
    compaction_window: 1h              # blocks in this time window will be compacted together
    max_compaction_objects: 1000000    # maximum size of compacted blocks
    block_retention: 1h
    compacted_block_retention: 10m

storage:
  trace:
    backend: local                     # backend configuration to use
    wal:
      path: /tmp/tempo/wal            # where to store the the wal locally
      #bloom_filter_false_positive: .05 # bloom filter false positive rate.  lower values create larger filters but fewer false positives
      #index_downsample: 10             # number of traces per index record
    local:
      path: /tmp/tempo/blocks
    pool:
      max_workers: 100                 # the worker pool mainly drives querying, but is also used for polling the blocklist
      queue_depth: 10000

The tracetest-config.yaml file contains the basic setup of connecting Tracetest to the Postgres instance and defining the trace data store and exporter. The exporter is set to the OpenTelemetry Collector.

# tracetest-config.yaml

postgres:
  host: postgres
  user: postgres
  password: postgres
  port: 5432
  dbname: postgres
  params: sslmode=disable

telemetry:
  exporters:
    collector:
      serviceName: tracetest
      sampling: 100 # 100%
      exporter:
        type: collector
        collector:
          endpoint: otel-collector:4317

server:
  telemetry:
    exporter: collector

The tracetest-provision.yaml file defines the trace data store, set to Tempo, meaning the traces will be stored in Tempo and Tracetest will fetch them from Tempo when running tests.

How does Tracetest fetch traces?

Tracetest uses tempo:9095 to connect to Tempo and fetch trace data.

# tracetest-provision.yaml

---
type: PollingProfile
spec:
  name: Default
  strategy: periodic
  default: true
  periodic:
    retryDelay: 5s
    timeout: 10m

---
type: DataStore
spec:
  name: Tempo
  type: tempo
  tempo:
    type: grpc
    grpc:
      endpoint: tempo:9095
      tls:
        insecure: true

How do traces reach Tempo?

The collector.config.yaml explains that. It receives traces via either grpc or http. Then, exports them to Tempo's OTLP gRPC endpoint tempo:4317.

# collector.config.yaml
receivers:
  otlp:
    protocols:
      grpc:
      http:

processors:
  batch:
    timeout: 100ms

exporters:
  logging:
    loglevel: debug
  otlp/1:
    endpoint: tempo:4317
    tls:
      insecure: true

service:
  pipelines:
    traces:
      receivers: [otlp]
      processors: [batch]
      exporters: [otlp/1]

Run Both the Node.js App and Tracetest

To start both the Node.js app and Tracetest, run this command:

docker-compose -f docker-compose.yaml -f tracetest/docker-compose.yaml up # add --build if the images are not built already

This will start your Tracetest instance on http://localhost:11633/.

Open the URL and start creating tests! Make sure to use the http://app:8080/ URL in your test creation, because your Node.js app and Tracetest are in the same network.

Learn More

Feel free to check out our examples in GitHub and join our Slack Community for more info!