Skip to content

Azure Devops with Terraform and Cisco ACI

Estimated time to read: 6 minutes

  • Originally Written: October, 2020

The following posts will guide you through the steps to build the environment.

Azure Docker Agent Setup

  • Sign up for an Azure Devops account

https://azure.microsoft.com/en-us/services/devops/​​​​​​​

  • In the bottom left hand corner of the Azure Portal select Project Settings and then Agent Pools from the menu

  • Click the Add pool button

  • Provide a name and select Create

  • Make a note of the name of this pool

  • In the top right corner of the Azure portal select the person icon

  • Click Personal access tokens

  • Create a New Token
  • Provide a name for the token and set the scope to full access
    • This token will be used to register the agent on your laptop to your Azure DevOps Pipeline

  • Make a copy of the access token created

  • Follow the instructions outlined here to build a new Docker image which will run Terraform.

https://docs.microsoft.com/en-us/azure/devops/pipelines/agents/docker?view=azure-devops

Info

You should end up with two files on your local machine, a Dockerfile, and a start.sh script.

  • Copy and paste from the instructions above, no changes necessary.

  • Build the image as per the instructions

docker build -t azure_devops_agent:latest .

  • Run the container on your local machine.

    • You need to provide the following:
      • The URL to Azure (your organization name is found under the organization settings from the bottom left corner of the main Azure Devops homepage)
      • The personal access token just created
      • The name of the agent pool you created

    docker run -e AZP_URL= https://dev.azure.com/<your org here> -e AZP_TOKEN=<your access token> -e AZP_POOL=<your pool name> -e AZP_AGENT_NAME=azure_devops_terraform_aci azure_devops_agent:latest

Info

If everything is successful after a few moments of running the docker container you should see the agent registered in Azure.

​​​​​​​Azure Pipeline Setup

  • ​​​​​​​Login to your account and create a new project
  • On the left hand menu select Pipelines and ​​​​​​​Create Pipeline​​​​​​​.
  • Select Github and the new repo which you forked.

Info

You may be asked to login to Github to provide Azure access to your repos.


Info

If everything is successful you will see the azure-pipelines.yml displayed in the window.

  • Click the dropdown menu that shows Run and select Save

  • On the left hand menu select Library and then add a Variable Group

  • Name the new variable group, "shared", and enter the following variables.

    • APIC_URL: URL of the APIC cluster you will be configuring with Terraform. We are using the always on ACI demo from Cisco Devnet
    • APIC_USERNAME
    • APIC_PASSWORD: Select the padlock on the right hand side of the text box to make this a secure field (covered by asterisks)
    • AWS_ACCESS_KEY: AWS access key used to access the terraform.tfstate file from the S3 bucket. This was created previously when setting up the AWS User
    • AWS_SECRET_KEY: AWS secret key used to access the terraform.tfstate file from the S3 bucket. Select the padlock on the right hand side of the field to make this a secure field. This was created previously when setting up the AWS User
    • TERRAFORM_VERSION: 0.13.4 (This can be changed if required)
    • AWS_S3_BUCKET_NAME: The name of the S3 bucket you previously created
    • AWS_S3_REGION:​​​​​​​ The region in which the S3 bucket resides

  • Click the Save button at the top

​​​​​​Your First Test Run

If everything has been successful up until this point you should be ready for the first test. As previously mentioned, the Github repo contains the pipeline steps and the Terraform files. In the Terraform files you'll see that the following are created:

  • Tenant
  • VRF
  • Bridge Domain

  • EPG

  • Navigate to the pipeline you created earlier

  • Select Run Pipeline and then Run


  • You should now see the pipeline running with two stages, plan and apply. You can click into each stage to see the current jobs and output.


  • The final check is confirm that the pipeline has succeeded, you can see the changes in your ACI fabric, and you can also see the new terraform.tfstate file in your AWS S3 bucket



Next Steps

​​​​​​​​​Part 3: Explanation of the configuration files

Comments