Testcontainers Spring Boot

If you develop services using Spring Boot and maybe Spring Cloud and you do medium sized tests during build process, then this set of Spring Boot auto-configurations might be handy. By adding module into classpath, you will get stateful service, like Couchbase or Kafka, auto-started and available for connection from your application service w/o wiring any additional code. Docker and TestContainers are used to bootstrap stateful service using Spring Cloud bootstrap phase. Usage of Spring Cloud in your production code is optional, but you will need it in tests. See How to use below.

Versions compatibility

Spring Boot	Test Containers Spring Boot
2.5.X, 2.6.X, 2.7.X	2.X.X
3.0.X, 3.1.X	3.0.X
3.2.X	3.1.X

How to use

1. Install Docker on your machine

2. Make sure you have Spring Boot and Spring Cloud in classpath of your tests. In case if you need to pick version.

   pom.xml

   <project>
   ...
         <dependency>
               <groupId>org.springframework.cloud</groupId>
               <artifactId>spring-cloud-starter-bootstrap</artifactId>
               <!-- https://mvnrepository.com/artifact/org.springframework.cloud/spring-cloud-starter-bootstrap -->
               <version>[pick version]</version>
           </dependency>
   ...
   </project>

   Note

   testcontainers-spring-boot project migrated to Spring Boot 2.4 in 2.0.0 version. Please note, that in order to use this project with Spring Boot 2.4, you need to use spring-cloud-starter-bootstrap dependency. For earlier Spring Boot (prior to 2.4) users — you need to use spring-cloud-starter dependency instead.

3. If you do not use Spring Cloud - make it work for tests only:

   pom.xml

   <project>
   ...
           <dependency>
               <groupId>org.springframework.boot</groupId>
               <artifactId>spring-boot-starter</artifactId>
               <!-- https://mvnrepository.com/artifact/org.springframework.boot/spring-boot-starter -->
               <version>[pick version]</version>
           </dependency>
           <dependency>
               <groupId>org.springframework.cloud</groupId>
               <artifactId>spring-cloud-starter-bootstrap</artifactId>
               <!-- https://mvnrepository.com/artifact/org.springframework.cloud/spring-cloud-starter-bootstrap -->
               <version>[pick version]</version>
               <scope>test</scope>
           </dependency>
   ...
   </project>

4. Add data service library:

   pom.xml

   <project>
   ...
           <dependency>
               <groupId>com.playtika.testcontainers</groupId>
               <artifactId>embedded-kafka</artifactId>
               <!-- https://mvnrepository.com/artifact/com.playtika.testcontainers/ -->
               <version>[pick version]</version>
               <scope>test</scope>
           </dependency>
   ...
   </project>

5. Use produced properties in your configuration.

   Example:

   /src/test/resources/application.properties

   spring.kafka.bootstrap-servers=${embedded.kafka.brokerList}

   /src/test/resources/bootstrap.properties

   embedded.kafka.topicsToCreate=some_topic

In-depth guides, how-tos and samples

• Testing Strategies in a Microservice Architecture

• Functional Testing in Spring Boot Using Docker in Tests

• Functional tests best practices sample repo

• Spring Boot Microservices Integration Testing

• Easy Integration Testing With Testcontainers

• SpringBoot Integration Testing using TestContainers Starter

• Easy way to test Spring Boot microservices

Common configuration options

Shutdown of embedded containers on spring application shutdown immediately

/src/test/resources/application.properties

embedded.containers.forceShutdown=true #default is false

Note	Otherwise, it will be shutdown with delay, see https://github.com/testcontainers/moby-ryuk

Disable all embedded containers

/src/test/resources/bootstrap.properties

embedded.containers.enabled=true #default is true

Note	If you setup, for example embedded.kafka.enabled + embedded.containers.enabled, result will be same as using AND between two booleans.

Note	embedded.kafka.enabled=false will cause DockerNotPresentException if you don’t have docker installed. But embedded.containers.enabled=false won’t cause any exceptions in this case.

Setting1	Setting2	Outcome
embedded.containers.enabled=false	embedded.memsql.enabled=true	Memsql will not start
embedded.containers.enabled=true	embedded.memsql.enabled=false	Memsql will not start
embedded.containers.enabled=true	embedded.memsql.enabled=true	Memsql will start
embedded.containers.enabled is missing	embedded.memsql.enabled is missing	Memsql will start

Other specific container related properties

Setting name	Default value	Description
embedded.{module-name}.dockerImage	Depends on module	Full Docker image name for container setup. Most of the modules have default value already setup.
embedded.{module-name}.dockerImageVersion	N/A	Use this property if you want to override only Docker image’s version.
embedded.{module-name}.waitTimeoutInSeconds	60	Waiting time for a container to start in seconds
embedded.{module-name}.enabled	true	Enables a container to be started on startup
embedded.{module-name}.reuseContainer	false	Enables a reuse container Testcontainers feature. For more info please refer to: testcontainers/testcontainers-java#2555 and testcontainers/testcontainers-java#1781.
embedded.{module-name}.command	null	List of keywords which combines into command for container startup. Some modules ship container’s commands by default, so resetting this value may lead to incorrect work of container.
embedded.{module-name}.attachContainerLog	false	Attach embedded container output log.
embedded.{module-name}.env	null	key-value map of additional environment variables. Where key is name of variable and value is actual value of it.
embedded.{module-name}.label	null	key-value map of additional labels to the container. Where key is name of label and value is actual value of label.
embedded.{module-name}.filesToInclude	empty list	List of files to include objects. Each object should have two parameters: classpathResource (path to local file) containerPath (path in a container to where file needs to be copied) Example: embedded.redis.filesToInclude: classpathResource: "/my_local_file.txt" containerPath: "/etc/path_in_container.txt"
embedded.{module-name}.mountVolumes	empty list	List of mount volumes to persist between container restarts. Each object should have three parameters: hostPath (path to local file/directory) containerPath (path in container to mount file/directory onto) mode (access mode default READ_ONLY, or READ_WRITE) Example: embedded.postgresql.mountVolumes: hostPath: "pgdata" containerPath: "/var/lib/postgresql/data" mode: READ_WRITE
embedded.{module-name}.capabilities	empty list. NET_ADMIN is set for Aerospike, Couchbase, Elasticsearch, Kafka, Mariadb, Memsql, Minio, Mongodb, Mysql, Neo4j, Redis containers.	The Linux capabilities that should be enabled. You can disable all capabilities by providing empty value for this property. See: https://man7.org/linux/man-pages/man7/capabilities.7.html. Available values can be taken from com.github.dockerjava.api.model.Capability class.
embedded.{module-name}.tmpFs.mounts	empty list	A list of container directories which should be replaced by tmpfs mounts, and their corresponding mount options. Check TmpFs mount docs. For example, for MariaDb: embedded: mariadb: tmp-fs: mounts: - folder: /var/lib/mysql options: rw

Supported services

embedded-mariadb

embedded-couchbase

embedded-kafka

embedded-rabbitmq

embedded-aerospike

embedded-memsql

embedded-redis

embedded-neo4j

embedded-postgresql

embedded-elasticsearch

embedded-opensearch

embedded-dynamodb

embedded-voltdb

embedded-minio

embedded-mongodb

embedded-google-pubsub

embedded-google-storage

embedded-keycloak

embedded-keydb

embedded-influxdb

embedded-vault

embedded-oracle-xe

embedded-mysql

embedded-localstack

embedded-cassandra

embedded-clickhouse

embedded-pulsar

embedded-vertica

embedded-prometheus

embedded-grafana

embedded-consul

embedded-artifactory

embedded-azurite

embedded-toxiproxy

embedded-nats

embedded-k3s

embedded-mockserver

embedded-solr

embedded-cockroachdb

embedded-git

embedded-wiremock

embedded-mailhog

embedded-spicedb

How to contribute

Flow

• You need to fork project and create branch from develop

• You do not need to update project version in pom.xml files, this will be done by release job

• Once finished - create pull request to develop from your fork, pass review and wait for merge

• On release, ci job will update to next release version + publish artifacts to the Maven Central

Checklist for contributing new module

• Naming/formatting patterns match existing code

• Test for success scenario

• Test for negative scenario (autoconfiguration is disabled via properties). How to test autoconfiguration

• Add new module to testcontainers-spring-boot-bom

• Module provides documentation in README.adoc and this documentation is included in parent README.adoc (see an example in already existing modules). Documentation should include:

  • maven module declaration

  • consumed properties

  • produced properties

  • notes (if applicable)

  • example of usage

Release

• Release is done per each major change, critical bug

• Release can be done by contributor request

• Contacts to start release:

  • sstus@playtika.com

  • iyova@playtika.com

  • admitrov@playtika.com