A GitHub Action to deploy ARM templates. With this action you can automate your workflow to deploy ARM templates and manage Azure resources.
This action can be used to deploy Azure Resource Manager templates at different deployment scopes - resource group deployment scope, subscription deployment scope and management group deployment scopes.
By default, the action only parses the output and does not print them out. In order to get the values of outputs
use this.
scope
: Provide the scope of the deployment. Valid values are: resourcegroup
(default) , tenant
, subscription
, managementgroup
. resourceGroupName
: Conditional Provide the name of a resource group. Only required for Resource Group ScopesubscriptionId
: Conditional Provide a value to override the subscription ID set by Azure Login.managementGroupId
: Conditional Specify the Management Group ID, only required for Management Group Deployments.region
: Conditional Provide the target region, only required for Tenant, Management Group or Subscription deployments.template
: Required Specify the path or URL to the Azure Resource Manager template.parameters
: Specify the path or URL to the Azure Resource Manager deployment parameter values file (local / remote) and/or specify local overrides. deploymentMode
: Incremental
(default) (only add resources to resource group) or Complete
(remove extra resources from resource group) or Validate
(only validates the template). deploymentName
: Specifies the name of the resource group deployment to create.failOnStdErr
: Specify whether to fail the action if some data is written to stderr stream of az cli. Valid values are: true, false. Default value set to true.additionalArguments
: Specify any additional arguments for the deployment. These arguments will be ignored while validating
the template.
A good way to use additionalArguments would be to send optional parameters like --what-if
or --what-if-exclude-change-types
. Read more about this here
Every template output will either be exported as output if output is a json object else will be consoled out where output is not a json object.
- uses: azure/arm-deploy@v2
with:
subscriptionId: <YourSubscriptionId>
resourceGroupName: <YourResourceGroup>
template: <path/to/azuredeploy.json>
on: [push]
name: AzureARMSample
jobs:
build-and-deploy:
runs-on: ubuntu-latest
steps:
- uses: actions/checkout@master
- uses: azure/login@v2
with:
creds: ${{ secrets.AZURE_CREDENTIALS }}
- uses: azure/arm-deploy@v2
with:
resourceGroupName: github-action-arm-rg
template: ./azuredeploy.json
parameters: examples/template/parameters.json storageAccountType=Standard_LRS sqlServerPassword=${{ secrets.SQL_SERVER }}
additionalArguments: "--what-if --rollback-on-error --what-if-exclude-change-types Create Ignore"
In the preceeding example there is a pre-requisite that an existing Azure Resource Group named github-action-arm-rg
must already exist.
The below example makes use of the Azure CLI Action to ensure the resource group is created before doing an ARM deployment. Note that the command az group create
is idempotent, so it will run sucessfully even if the group already exists.
When generating your credentials (in this example we store in a secret named AZURE_CREDENTIALS
) you will need to specify a scope at the subscription level.
az ad sp create-for-rbac --name "{sp-name}" --sdk-auth --role contributor --scopes /subscriptions/{subscription-id}
See Configure deployment credentials.
on: [push]
name: AzureARMSample
jobs:
build-and-deploy:
runs-on: ubuntu-latest
env:
ResourceGroupName: github-action-arm-rg
ResourceGroupLocation: "australiaeast"
steps:
- uses: actions/checkout@master
- uses: azure/login@v2
with:
creds: ${{ secrets.AZURE_CREDENTIALS }}
- uses: Azure/CLI@v2
with:
inlineScript: |
#!/bin/bash
az group create --name ${{ env.ResourceGroupName }} --location ${{ env.ResourceGroupLocation }}
echo "Azure resource group created"
- uses: azure/arm-deploy@v2
with:
resourceGroupName: ${{ env.ResourceGroupName }}
template: ./azuredeploy.json
parameters: storageAccountType=Standard_LRS
In this example, our template outputs containerName
.
- uses: azure/arm-deploy@v2
id: deploy
with:
resourceGroupName: azurearmaction
template: examples/template/template.json
parameters: examples/template/parameters.json
deploymentName: github-advanced-test
Here we see a normal use of the Action, we pass the template as json file as well as the parameters. If we look into the template.json
File we can see at the very bottom the defined outputs:
{
...
"outputs": {
...
"containerName": {
"type": "string",
"value": "[parameters('containerName')]"
}
}
}
And we know our Action writes this output(s) to an action output variable with the same name, we can access it using ${{ steps.deploy.outputs.containerName }}
(Note: deploy
comes from the id
field from above.)
If we now add a Shell script with a simple echo from that value, we can see that on the console the containername to be printed.
- run: echo ${{ steps.deploy.outputs.containerName }}
ARM Deploy Actions is supported for the Azure public cloud as well as Azure government clouds ('AzureUSGovernment' or 'AzureChinaCloud') and Azure Stack ('AzureStack') Hub. Before running this action, login to the respective Azure Cloud using Azure Login by setting appropriate value for the environment
parameter.
In this example, we are setting failOnStdErr
to false.
- uses: azure/arm-deploy@v2
id: deploy
with:
resourceGroupName: azurearmaction
template: examples/template/template.json
parameters: examples/template/parameters.json
deploymentName: github-advanced-test
failOnStdErr: false
failOnStdErr
equals true implied that if some data is written to stdErr and exit code from az-cli is 0, then action execution will fail.
failOnStdErr
equals false implies that if some data is written to stdErr and return code from az-cli is 0, then action will continue execution. This input is added to support cases where stdErr is being used to stream warning or progress info.
Non zero Exit code will always lead to failure of action irrespective the value of failOnStdErr
.
For more examples, refer : Example Guide
Internally in this action, we use azure CLI and execute az login
with the credentials provided through secrets. In order to validate the new az CLI releases for this action, canary test workflow is written which will execute the action on az CLI's edge build which will fail incase of any breaking change is being introduced in the new upcoming release. The test results can be posted on a slack or teams channel using the corresponding integrations. Incase of a failure, the concern will be raised to azure-cli for taking a necessary action and also the latest CLI installation will be postponed in Runner VMs as well for hosted runner to prevent the workflows failing due to the new CLI changes.
This project welcomes contributions and suggestions. Most contributions require you to agree to a Contributor License Agreement (CLA) declaring that you have the right to, and actually do, grant us the rights to use your contribution. For details, visit https://cla.opensource.microsoft.com.
When you submit a pull request, a CLA bot will automatically determine whether you need to provide a CLA and decorate the PR appropriately (e.g., status check, comment). Simply follow the instructions provided by the bot. You will only need to do this once across all repos using our CLA.
This project has adopted the Microsoft Open Source Code of Conduct. For more information see the Code of Conduct FAQ or contact opencode@microsoft.com with any additional questions or comments.