fa-ci-platform

The CI platform for FA consist in a Jenkins X installation in an AWS EKS cluster that defines CI/CD pipelines for different apps within the FA company. Note: This project is part of a Jenkins X, Terraform and EKS sample assessment.

Platform features

• The platform is mounted in a Kubernetes environment with EBS persistent volumes for its high availability and scalability.
• The pipelines are defined in Jenkinsfiles.
• The infrastructure for the platform and the apps is defined in Terraform.
• For testing purposes we will use a personal domain: somosaltum.com.
• Code coverage tests with Sonarqube.
• Notifications in Github by Jenkins (GitOps).
• Monitoring with Prometheus and Grafana.

Project contents

• terraform/: contains the infrastructure as code in Terraform for the FA apps and FA ci-platform.
• helm/: contains the values for the deployed charts (in our case, the nginx controller via Terraform).

Jenkins X installation

Prerequisites:

• Install JX CLI brew tap jenkins-x/jx brew install jx
• Install EKSCTL tool brew tap weaveworks/tap brew install eksctl

Installation commands

jx install --provider=eks --default-environment-prefix=fa-jx --skip-ingress --ingress-deployment=nginx-ingress-controller --ingress-service=nginx-ingress-controller --domain=somosaltum.com

Testing and monitoring tools installation

Prometheus installation commands

helm upgrade prometheus stable/prometheus -f helm/prometheus.yaml --namespace monitoring --set rbac.create=true --install

Grafana installation commands

helm upgrade grafana stable/grafana --set=ingress.enabled=True,ingress.hosts={grafana.somosaltum.com} --namespace monitoring --set rbac.create=true --install

Sonarqube installation commands

helm upgrade sonarqube stable/sonarqube --set=ingress.enabled=True,ingress.hosts={sonar.somosaltum.com} --namespace jx --install

Pipelines definitions

When working in a PR branch:

1. Application tests
2. Quality code analysis
3. CI Build and push snapshot

When the PR is merged to master:

1. Integration tests
2. Quality code analysis in master
3. Build Release
4. Promote to Staging
5. Promote to Production (manual step)

Further improvements

• Get a SSL certificate to configure the HTTPS communication.
• Different EKS clusters for different environments.
• Notifications to Github and Slack if stages fail.