MuleSoft Automate Deployment

Deploy Mule Applications to CloudHub using the Mule Maven Plugin and Bitbucket Pipelines

Credit: Atlasssian (BitBucket)

Hello and welcome. In this post I will explain how to automate MuleSoft Application deployment to CloudHub using Mule Maven Plugin and Bitbucket Pipeline.

Prerequisite

• MuleSoft CloudHub Account
• Bitbucket Atlasssian Account — for code repository.
• If you are using the HTTP Listener as source for your flow, you need to set its host to 0.0.0.0 and its port to ${http.port}
• Declare all external classes and resources in the exportedPackages and exportedResources fields on the mule-artifact.jsonfile.

Project Structure

Pipeline file should have the name — bitbucket-pipelines.yml. It should be at repository root-level as described below in a sample project structure. Once it is defined at the root-level, Bitbucket will automatically pick-up any changes to the code commit and will start to run the pipeline.

Pipeline Stages

1. Git Security Scan
2. Validate & Compile
3. Test
4. Deploy to Sandbox CloudHub

Alright, a big question — how to integrate this pipeline to a MuleSoft Project? Let’s check this out in the next section.

How to integrate pipeline to your own MuleSoft Project

1. Update pom.xml: open pom.xml — plugin element: Inside the plugin element in your project’s pom.xml file, search for plugin section with mule-maven-plugin artifactIdand replace the below pluginplaceholder.

2. Create Variable References in Bitbucket: In step 1, as you can see, we are using pipeline variables e.g. — {MULESOFT_ENVIRONMENT} . These pipeline variables are convenient as you can set these once and reuse in your pipeline at build time.

• So let’s create Workspace Variablesas described below;

Bitbucket Workspace Variables

The reason we’re using Workspace Variables is for the reusability. Once defined, can be reused within the workspace, i.e. for all repository; so you don’t have to create these variables for each and every repository within the workspace.

We’re usingMULESOFT_USERNAME and MULESOFT_PASSWORD variables as secured . Bitbucket supports variables stored as secured so it’s value is hidden and not visible in the variables list and in the build logs. Secure variables are stored as encrypted values.

• Next, Let’s create Deployment Variables ; these variables can only be used in a specific deployment environment. We will set MULESOFT_ENVIRONMENT variables with the value of a deployment environment in your MuleSoft CloudHub, e.g. Sandbox

Ensure that the pipeline is enabled before creating Deployment Variables

• Change the Test environment name to Sandbox ; [we’re going to use CloudHub’s Sandbox environment, you can use the one that’s available]
• Under Variables section for the Sandbox environment, create MULESOFT_ENVIRONMENT variable with value Sandbox

Bitbucket — Repository Deployment variables

For more details about Bitbucket variables, you can visit — https://support.atlassian.com/bitbucket-cloud/docs/variables-and-secrets/

3. Create bitbucket-pipelines.yml as described below

4. that’s it basically, all done.

5. Push the code to your remote master branch — and Bitbucket will trigger the deployment.

Successful Deployment — BitBucket pipeline (y)

Successful Deployment — CloudHub logs (y)

Demo Project

Remember — the objective is to deploy the application to CloudHub. I have used a sample MuleSoft Application — with a Listener that returns a “Hello World!” JSON response.

Mule Application — Listener returns a simple “Hello World!” message.

You can use this repository included with pipeline configurations here — https://bitbucket.org/kuldeepbhati/cloud-hub-deploy/src/master/

Common Issues

• 4.3.0–20210119 is not a valid value for muleVersion.version.

Caused by: org.mule.tools.client.core.exception.ClientException: 400 Bad Request: {“status”:400,”message”:”4.3.0–20210119 is not a valid value for muleVersion.version.”}

If you get this error; this is due to the incorrect <app.runtime>4.3.0–20210119</app.runtime> in your pom.xml file. Correct it as <app.runtime>4.3.0</app.runtime>

Important Notes

• We’re using a single environment in this pipeline — that’s Sandbox . You can always extend this pipeline and use as many upstream environments as it pleases you.
• Ensure that you have correct environment naming across — bitbucket-pipelines.yml , Deployment Variables and your CloudHub environment name.
• You may have noticed, that we’re using MULESOFT_USERNAME and MULESOFT_PASSWORD variables for deployment; although these are stored as secured and not visible in the variables list and also not in build logs, it is advised to use Connected App functionality, please visit this link for more information — https://help.mulesoft.com/s/article/How-to-deploy-an-application-to-Cloudhub-using-Connected-App-functionality
• We have not introduced any branching model in this post; so If you push code from any branch; the deployment will be triggered. You can use this link to configure branch specific deployment — https://support.atlassian.com/bitbucket-cloud/docs/pipeline-triggers/
• As of now; we kept this pipeline very simple; if there’s need for using Mule EE repositories you would need to create a Maven settings.xml file that will have Mule EE username and password . You can use this post — https://medium.com/@chris.shellenbarger/continuous-integration-a-complete-guide-to-using-bitbucket-pipelines-with-maven-repositories-e835bd0f5f11 that describes how to achieve that. Please leave a comment below if you need any help.