Professional Cloud Developer
Professional Cloud Developer
Intermediate
Gauge your current knowledge
Intermediate
Gauge your current knowledge
Executing automated integration tests in Cloud Build
...
Professional Cloud Developer
Gauge your current knowledge
Gauge your current knowledge
Cloud Build uses a build configuration file to manage the steps needed to test and build applications. Each step in the process runs inside an isolated Docker container to provide a clean, independent environment for running test suites. This setup ensures that every test run is reproducible and consistent, regardless of where the code was originally written or local environment differences.
Developers use builders, which are container images pre-loaded with common languages and tools. These builders allow the pipeline to perform tasks like installing dependencies or running complex integration tests. Common builders include language-specific tools like Go and Python to run test suites, followed by Docker to containerize the application after tests pass.
The order of tasks is managed using the id and waitFor fields within the configuration file. While steps usually run one after another, these fields allow for orchestration, which is the coordination of multiple tasks running in parallel or sequence. This process is vital for complex pipelines that require certain integration tests to finish successfully before the final application is built.
Once the tests are complete, Cloud Build can save the results and logs to Cloud Storage or Artifact Registry. This allows teams to review build history and manage artifacts, which are the deployable files produced during the build. To maintain high security, Cloud Build supports provenance, which is verifiable metadata that describes how an artifact was built, using a service account with specific permissions to protect the software supply chain.
Cloud Build executes automated integration testing by running a series of build steps within secure containers. To protect sensitive data during this process, developers should integrate Secret Manager to handle sensitive information like API keys and database passwords. In the cloudbuild.yaml configuration file, the availableSecrets field allows these values to be passed securely to the build steps so credentials are never hard-coded in the source repository.
Proper security requires configuring IAM roles to follow the principle of least privilege for the build service account. You must grant the Secret Manager Secret Accessor role specifically to the service account executing the build so it can retrieve necessary secrets. This restricts unauthorized access and keeps the automated pipeline secure.
To establish this secure access, follow these steps:
Automating the enforcement of quality gates is essential for maintaining high code standards within the CI/CD pipeline. Cloud Build evaluates exit codes from each test step to determine if the build should proceed or fail immediately. Developers can analyze Cloud Logging data to identify specific failures and security anomalies that occur during the integration phase.
Following administrative best practices helps maintain a robust and secure testing lifecycle over time. It is recommended to use version pinning for secrets rather than the "latest" alias to ensure that builds remain predictable and repeatable.
Important management tasks include:
To run integration tests successfully, the system must manage dependencies, which are the external tools or libraries your code needs to function. Each task in the build process runs inside a separate Docker container connected through a local Docker network named cloudbuild. This local network allows different steps to communicate and share data seamlessly during execution.
When testing complex applications, developers often need backing services, such as databases or external APIs, to simulate a real-world environment. This setup involves service linking, where the main application container connects to these temporary helper services during the test. These are considered ephemeral dependencies because they are created only for the test and are deleted once the process is finished to save resources.
Automated tests are executed as specific build steps using specialized images for languages like Go or Python. Developers can define custom entrypoints to run scripts that trigger the testing framework and check for errors. This structure ensures that any failures are caught early before code is committed.
Key tasks during this phase include:
If the integration tests are successful, the final application is stored in the Artifact Registry as a ready-to-use container image. Cloud Build also generates provenance to prove the code has not been tampered with and is safe to use in a production environment. This step marks the transition from testing to deployment.