Local SonarQube 7.9.1-Community LTS On Kubernetes

https://gitlab.com/afireinside/kubernetes-sonarqube

Versions

As of this writing, the following versions have been used:

• SonarQube 7.9.1-community LTS
• Minikube 1.5.2
• OSX 10.14.6 Mojave
• Kubernetes 1.16.2
• Docker 18.09.9

Description

SonarQube is a static analysis tool where it analyzes your code for bugs and code smells then shows you a nice report. The report includes why the lines chosen are bugs/code smells and what the recommended solution is.

Many coding languages are supported in the 7.9.1-community LTS version. I primarily use it for analyzing my Java code. The Java quality profile I prefer is FindBugs + FB Contrib.

Some of the benefits for having your own instance of SonarQube on your development machine include (but are not limited to)

• being able to make changes to the quality profiles and gates
• not using up the static analysis line limit if you're using an enterprise license
• experimenting with different plugins to find what works for you
• having a later version than what your enterprise version probably is
• privatizing all your code smells and bugs so no one ever has to know
• performing static analysis without making a commit or using up network bandwidth
• not relying on a small set of admins to make impactful changes to the instance
• etc

Installation

Lets hop into configuration and installation. This article assumes you have a Macbook laptop and Homebrew. Installation instructions are also available in the GitLab repository listed at the very top.

1. First, you'll need Minikube (local Kubernetes cluster) and sonar-scanner (to perform static analysis)
   • brew install sonar-scanner
   • brew install minikube
2. Configure Minikube to support the Sonar server and start the cluster (takes a few minutes)
   • minikube config set memory 4096
   • minikube config set cpus 2
   • minikube config set vm-driver hyperkit
   • minikube start
3. Deploy your instance of SonarQube (will take a few minutes for pods to fully warm up and load SonarQube)
   • kubectl create secret generic postgres-pwd --from-literal=password={some made up password} Remember to change the password. It can be anything you want
   • kubectl create -f sonar-pv-postgres.yml
   • kubectl create -f sonar-pvc-postgres.yml
   • kubectl create -f sonar-postgres-deployment.yml
   • kubectl create -f sonarqube-deployment.yml
   • kubectl create -f sonarqube-service.yml
   • kubectl create -f sonar-postgres-service.yml
4. Once all the pods are up and running, view your SonarQube instance (this will open a new browser tab with SonarQube). Default username/password is admin/amin
   • minikube service sonar

Perform static analysis

Java

1. In the root of your Java project, add an empty sonar-project.properties file. The sonar-scanner service will be looking for this file when performing static analysis.
2. Paste the following into the newly created sonar-project.properties file:

sonar.projectKey={name of project}
sonar.host.url=http://192.168.##.##:##### (url from minikube service sonar command)
sonar.login=${env.SONAR_TOKEN}
sonar.java.binaries=build/classes
sonar.sources=src/main/java

An example config would look like the following:

sonar.projectKey=notificationemailproc
sonar.host.url=http://192.168.64.9:31828
sonar.login=${env.SONAR_TOKEN}
sonar.java.binaries=build/classes
sonar.sources=src/main/java

1. Create and copy a new SonarQube token by going to the SonarQube instance in the browser and navigating to
   My Account -> Security tab -> Enter Token Name -> Generate -> Copy token generated

2. In your .bashrc or .zshrc file, add the following line:

   • export SONAR_TOKEN={SonarQube token that was just copied}

3. Reload your rc file

   • source ~/.bashrc or source ~/.zshrc

4. Run sonar-scanner in the project root directory. Once static analysis is finished, you can view the results in your SonarQube instance in the browser.

Useful plugins

• FindBugs
• Checkstyle
• Mutation Analysis

Thanks for reading!

Any questions, comments, or concerns, feel free to send me an email at philip.c.nguyen@gmail.com