Working with Elysian IT to integrate a secure VM image into Jenkins deployment in AWS and co-author ARM templates to deploy new Azure resources for a customer migration
In March 2019 MPFE Consultancy were engaged by Elysian IT to further support the project with their Financial Technology client and to contribute to a separate Azure migration project. During this engagement we worked with Jenkins, Gradle and AWS CloudFormation templates as well as Azure Blueprints and ARM templates for the migration project.
In March 2019 MPFE were engaged by Elysian IT for a further 3 month period as part of a second phase for their Financial Technology client. We also provided assistance with an Azure migration, developing an Azure blueprint, ARM (Azure Resource Manager) templates and PowerShell scripts to deploy the infrastructure for a customer’s new cloud environment.
As part of the Financial Technology client engagement MPFE worked to understand the clients complex build and deployment strategy for the product utilizing the secure Windows Server 2016 image we developed in the previous engagement as well as modifying the security configuration for the deployment, including the implementation of a Group Managed Service Account (GMSA). We then worked with the end client to modify their Jenkins and Gradle scripts in order to implement the new image and ensure the product stack was successfully deployed and tested. At the end of the engagement we travelled with Elysian IT to the client’s office in Germany and worked with them for three days to handover the solution.
In addition MPFE provided assistance with a separate Azure migration project. We created an Azure blueprint and ARM templates to deploy infrastructure and Virtual Machines and to secure and configure the Azure subscription. We then created PowerShell scripts that could be used to trigger deployments and infrastructure updates and also created a Proof of Concept for ARM template deployments via Azure DevOps pipelines, a solution being considered for the future.
Technologies
Jenkins
Jenkins is a widely used automation server product used to build, test, and deploy software. It can be used for CI (continuous integration) and CD (continuous deployment) and is highly extensible via plugins.
Jenkins was an existing technology choice for the client. They had implemented multiple pipelines that built their solution in layers. MPFE assisted by modifying the tasks in these pipelines to ensure the new image was utilized as part of the automated build and deployment process.
Gradle
Gradle is an open source build and automation tool that runs on JVM (Java Virtual Machine).
Gradle was also an existing technology choice for the client. MPFE did not have prior experience working with Gradle, but were able to quickly become familiar with the tool so that we could implement the required changes.
AWS CloudFormation Templates
CloudFormation allows infrastructure as code deployments to AWS. A CloudFormation template can be written as either JSON or YAML and describes the AWS resources you want to deploy. AWS then takes care of provisioning and configuring the resources, ensuring that the resultant state matches the definition of the template.
CloudFormation was another existing technology choice for the client. MPFE made changes to the customers AWS CloudFormation templates in order to implement the new image and security configuration and test the deployment of the solution.
ARM (Azure Resource Manager) Templates
ARM templates are the infrastructure as code technology for Azure. An ARM template is written in JSON and describes the Azure resources you want to deploy. When a template is executed Azure ensures that the resultant state matches what is defined in the template.
MPFE worked with Elysian IT to implement a set of new ARM templates that could be used to deploy the resources specified in Elysian ITs design for their clients new Azure environment. MPFE created PowerShell scripts for the deployments so they could be used in a repeatable fashion to stand up multiple environments (i.e UAT, Prod). We then tested and validated the deployments.
