docker-zipkin

This repository contains the Docker build definition and release process for openzipkin/zipkin. It also contains test images for transport and storage backends such as Kafka or Cassandra.

Automatically built images are available on Quay.io under the OpenZipkin organization, and are mirrored to Docker Hub.

Regarding production usage

The only images OpenZipkin provides for production use are:

• openzipkin/zipkin: The core server image that hosts the Zipkin UI, Api and Collector features.
• openzipkin/zipkin-dependencies: pre-aggregates data such that http://your_host:9411/dependency shows links between services.

If you are using these images and run into problems, please raise an issue or join gitter.

The other images here, and docker-compose, are for development and exploration purposes. For example, they aim to help you integrate an entire zipkin system for testing purposes, without having to understand how everything works, and without having to download gigabytes of files.

For example, openzipkin/zipkin-cassandra was not designed for real usage. You'll notice it has no configuration available to run more than one node sensibly, neither does it handle file systems as one would "in real life". We expect production users to use canonical images for storage or transports like Kafka, and only those testing or learning zipkin to use the ones we have here.

Running

Zipkin has no dependencies, for example you can run an in-memory zipkin server like so: docker run -d -p 9411:9411 openzipkin/zipkin

See the ui at (docker ip):9411

In the ui - click zipkin-server, then click "Find Traces".

Configuration

Configuration is via environment variables, defined by zipkin-server. Notably, you'll want to look at the STORAGE_TYPE environment variables, which include "cassandra", "mysql" and "elasticsearch".

When in docker, the following environment variables also apply

• JAVA_OPTS: Use to set java arguments, such as heap size or trust store location.
• STORAGE_PORT_9042_TCP_ADDR -- A Cassandra node listening on port 9042. This environment variable is typically set by linking a container running zipkin-cassandra as "storage" when you start the container.
• STORAGE_PORT_3306_TCP_ADDR -- A MySQL node listening on port 3306. This environment variable is typically set by linking a container running zipkin-mysql as "storage" when you start the container.
• STORAGE_PORT_9200_TCP_ADDR -- An Elasticsearch node listening on port 9200. This environment variable is typically set by linking a container running zipkin-elasticsearch as "storage" when you start the container. This is ignored when ES_HOSTS or ES_AWS_DOMAIN are set.
• KAFKA_PORT_2181_TCP_ADDR -- A zookeeper node listening on port 2181. This environment variable is typically set by linking a container running zipkin-kafka as "kafka" when you start the container.

docker-compose

This project is configured to run docker containers using docker-compose. Note that the default configuration requires docker-compose 1.6.0+ and docker-engine 1.10.0+. If you are running older versions, see the Legacy section below.

To start the default docker-compose configuration, run:

$ docker-compose up

View the web UI at $(docker ip):9411.

To see specific traces in the UI, select "zipkin-server" in the dropdown and then click the "Find Traces" button.

MySQL

The default docker-compose configuration defined in docker-compose.yml is backed by MySQL. This configuration starts zipkin, zipkin-mysql and zipkin-dependencies (cron job) in their own containers.

Cassandra

The docker-compose configuration can be extended to use Cassandra instead of MySQL, using the docker-compose-cassandra.yml file. That file employs docker-compose overrides to swap out one storage container for another.

To start the Cassandra-backed configuration, run:

$ docker-compose -f docker-compose.yml -f docker-compose-cassandra.yml up

Elasticsearch

The docker-compose configuration can be extended to use Elasticsearch instead of MySQL, using the docker-compose-elasticsearch.yml file. That file employs docker-compose overrides to swap out one storage container for another.

To start the Elasticsearch-backed configuration, run:

$ docker-compose -f docker-compose.yml -f docker-compose-elasticsearch.yml up

Elasticsearch 5+ and Host setup

The zipkin-elasticsearch5 and zipkin-elasticsearch6 images are more strict about virtual memory. You will need to adjust accordingly (especially if you notice elasticsearch crash!)

# If docker is running on your host machine, adjust the kernel setting directly
$ sudo sysctl -w vm.max_map_count=262144

# If using docker-machine/Docker Toolbox/Boot2Docker, remotely adjust the same
$ docker-machine ssh default "sudo sysctl -w vm.max_map_count=262144"

Elasticsearch Service on Amazon

If you are using Elasticsearch against Amazon, it will search for credentials including those in the ~/.aws directory. If you want to try Zipkin against Amazon Elasticsearch Service, the easiest start is to share your credentials with Zipkin's docker image.

For example, if you are able to run aws es list-domain-names, then you should be able to start Zipkin as simply as this:

$ docker run -d -p 9411:9411 \
  -e STORAGE_TYPE=elasticsearch -e ES_AWS_DOMAIN=your_domain \
  -v $HOME/.aws:/root/.aws:ro \
  openzipkin/zipkin

Kafka

The docker-compose configuration can be extended to host a test Kafka broker and activate the Kafka 0.10 collector using the docker-compose-kafka10.yml file. That file employs docker-compose overrides to add a Kafka+ZooKeeper container and relevant settings.

To start the MySQL+Kafka configuration, run:

$ docker-compose -f docker-compose.yml -f docker-compose-kafka10.yml up

Then configure the Kafka 0.10 sender or Kafka 0.8 sender using a bootstrapServers value of 192.168.99.100:9092.

By default, this assumes your Docker host IP is 192.168.99.100. If this is not the case, adjust KAFKA_ADVERTISED_HOST_NAME in docker-compose-kafka10.yml and the bootstrapServers configuration of the kafka sender to match your Docker host IP.

If you prefer to activate the Kafka 0.8 collector use docker-compose-kafka.yml instead of docker-compose-kafka10.yml:

$ docker-compose -f docker-compose.yml -f docker-compose-kafka.yml up

UI

The docker-compose configuration can be extended to host the experimental new UI on port 80 using docker-compose-ui.yml. That file employs docker-compose overrides to add an NGINX container and relevant settings.

To start the NGINX configuration, run:

$ docker-compose -f docker-compose.yml -f docker-compose-ui.yml up

This container doubles as a skeleton for creating proxy configuration around Zipkin like authentication, dealing with CORS with zipkin-js apps, or terminating SSL.

Prometheus

Zipkin comes with a built-in Prometheus metric exporter. The main docker-compose.yml file starts Prometheus configured to scrape Zipkin, exposes it on port 9090. You can open $DOCKER_HOST_IP:9090 and start exploring the metrics (which are available on the /prometheus endpoint of Zipkin).

docker-compose.yml also starts a Grafana container with authentication disabled, exposing it on port 3000. On startup it's configured with the Prometheus instance started by docker-compose as a data source, and imports the dashboard published at https://grafana.com/dashboards/1598. This means that, after running docker-compose up, you can open $DOCKER_IP:3000/dashboard/db/zipkin-prometheus and play around with the dashboard.

If you want to run the zipkin-ui standalone against a remote zipkin server, you need to set ZIPKIN_BASE_URL accordingly:

$ docker run -d -p 80:80 \
  -e ZIPKIN_BASE_URL=http://myfavoritezipkin:9411 \
  openzipkin/zipkin-ui

Legacy

The docker-compose files described above use version 2 of the docker-compose config file format. There is a legacy version 1 configuration also available, in docker-compose-legacy.yml. That configuration relies on container linking.

To start the legacy configuration, run:

$ docker-compose -f docker-compose-legacy.yml up

Notes

All images share a base image, openzipkin/jre-full, built on the Alpine image delitescere/java:8, which is much smaller than the previously used debian:sidd image.

If using a provided MySQL server or image, ensure schema and other parameters match the docs.