hurtn / datadomain

Enterprise Scale Analytics - Data Domain: Batch Processing

General disclaimer Please be aware that this template is in public preview. Therefore, expect smaller bugs and issues when working with the solution. Please submit an Issue in GitHub if you come across any issues that you would like us to fix.

Description

Enterprise Scale Analytics emphasizes self-service and follows the concept of creating landing zones for cross-functional teams. Operation and responsibility of these landing zones is handed over to the responsible teams inside the data node. The teams are free to deploy their own services within the guardrails set by Azure Policy. To scale across the landing zones more quickly and allow a shorter time to market, we use the concept of Data Domain and Data Product templates. Data Domain and Data Product templates are blueprints, which can be used to quickly spin up environments for these cross-functional teams. The teams can fork these repos to quickly spin up environments based on their requirements. This Data Domain template deploys a set of services, which can be used for batch processing of data. The template includes a set of different SQL engines, which allows the teams to choose their tools based on their requirements and preferences.

What will be deployed?

By default, all the services which come under Data Domain Batch are enabled, and you must explicitly disable services that you don't want to be deployed.

Note: Before deploying the resources, we recommend to check registration status of the required resource providers in your subscription. For more information, see Resource providers for Azure services.

For each Data Domain Batch template, the following services are created:

• Key Vault
• Data Factory
• Cosmos DB
• Synapse Workspace
• MySQL Database
• Azure SQL Database
• PostgresSQL Database
• MariaDB Database
• SQL Pool
• SQL Server
• SQL Elastic Pool
• BigData Pool

For more details regarding the services that will be deployed, please read the Domains guide in the Enterprise Scale Analytics documentation.

You have two options for deploying this reference architecture:

1. Use the Deploy to Azure button for an immediate deployment
2. Use GitHub Actions or Azure DevOps Pipelines for an automated, repeatable deployment

Prerequisites

The following prerequisites are required to make this repository work:

• A resource group within an Azure subscription
• User Access Administrator or Owner access to a resource group to be able to create a service principal and role assignments for it.
• Access to a subnet with privateEndpointNetworkPolicies and privateLinkServiceNetworkPolicies set to disabled. The Data Landing Zone deployment already creates a few subnets with this configuration.

If you don’t have an Azure subscription, create your Azure free account today.

Option 1: Deploy to Azure - Quickstart

Data Domain Batch

Option 2: GitHub Actions or Azure DevOps Pipelines

1. Create repository from a template

1. On GitHub, navigate to the main page of this repository.
2. Above the file list, click Use this template

3. Use the Owner drop-down menu and select the account you want to own the repository.

4. Type a name for your repository and an optional description.
5. Choose a repository visibility. For more information, see "About repository visibility."
6. Optionally, to include the directory structure and files from all branches in the template and not just the default branch, select Include all branches.
7. Click Create repository from template.

2. Setting up the required Service Principal

A service principal needs to be generated for authentication and authorization from GitHub or Azure DevOps to your Azure subscription. This is required to deploy resources to your environment. Just go to the Azure Portal to find the ID of your subscription. Then start the Cloud Shell or Azure CLI, login to Azure, set the Azure context and execute the following command to generate the required credentials:

NOTE: The purpose of this new Service Principal is to assign least-privilege rights. Therefore, it requires the Contributor role at a resource group scope in order to deploy the resources inside the resource group dedicated to a specific data domain. The Network Contributor role assignment is required as well in this repository in order to assign the resources to the dedicated subnet.

Azure CLI

# Replace {service-principal-name} and {subscription-id} with your 
# Azure subscription id and any name for your service principal.
az ad sp create-for-rbac \
  --name "{service-principal-name}" \
  --skip-assignment \
  --sdk-auth

This will generate the following JSON output:

{
  "clientId": "<GUID>",
  "clientSecret": "<GUID>",
  "subscriptionId": "<GUID>",
  "tenantId": "<GUID>",
  (...)
}

Take note of the output. It will be required for the next steps.

Now that the new Service Principal is created, as mentioned, role assignments are required for this service principal in order to be able to successfully deploy all services. Required role assignments which will be added on a later step include:

Role Name	Description	Scope
Private DNS Zone Contributor	We expect you to deploy all Private DNS Zones for all data services into a single subscription and resource group. The service principal needs to be Private DNS Zone Contributor on that resource group. This is required to deploy A-records for the respective private endpoints.	(Resource Group Scope) /subscriptions/{subscriptionId}/resourceGroups/{resourceGroupName}
Contributor	We expect you to deploy all data-domain-batch services into a single resource group within the Data Landing Zone subscription. The resource group and the role assignment will be setup in the parameter update process (4). The service principal requires a Contributor role-assignment on that resource group to deploy all services.	(Resource Group Scope) /subscriptions/{subscriptionId}/resourceGroups/{resourceGroupName}
Network Contributor	In order to deploy Private Endpoints to the specified subnet, the service principal requires Network Contributor access on the subnet.	(Child-Resource Scope) /subscriptions/{subscriptionId}/resourceGroups/{resourceGroupName} /providers/Microsoft.Network/virtualNetworks/{virtualNetworkName}/subnets/{subnetName}"

          

To add these role assignments, you can use the Azure Portal or run the following commands:

Azure CLI

# Get Service Principal Object ID
az ad sp list --display-name "{servicePrincipalName}" --query "[].{objectId:objectId}" --output tsv

# Add role assignment
# 1. Resource Scope
az role assignment create \
  --assignee "{servicePrincipalObjectId}" \
  --role "{roleName}" \
  --scopes "{scope}"

# 2. Resource group Scope
az role assignment create \
  --assignee "{servicePrincipalObjectId}" \
  --role "{roleName}" \
  --resource-group "{resourceGroupName}"

# 3. Child-Resource Scope
# TBD

Azure Powershell

# Get Service Principal Object ID
$spObjectId = (Get-AzADServicePrincipal -DisplayName "{servicePrincipalName}").id

# Add role assignment
# 1. Resource Scope
New-AzRoleAssignment `
  -ObjectId $spObjectId `
  -RoleDefinitionName "{roleName}" `
  -Scope "{scope}"

# 2. Resource group Scope
New-AzRoleAssignment `
  -ObjectId $spObjectId `
  -RoleDefinitionName "{roleName}" `
  -ResourceGroupName "{resourceGroupName}"

# 3. Child-Resource Scope
New-AzRoleAssignment `
  -ObjectId $spObjectId `
  -RoleDefinitionName "{roleName}" `
  -ResourceName "{subnetName}" `
  -ResourceType "Microsoft.Network/virtualNetworks/subnets" `
  -ParentResource "virtualNetworks/{virtualNetworkName}" `
  -ResourceGroupName "{resourceGroupName}

Now you can choose whether you would like to use GitHub Actions or Azure DevOps for your deployment.

3. a) GitHub Actions

If you want to use GitHub Actions for deploying the resources, add the previous JSON output as a repository secret with the name AZURE_CREDENTIALS in your GitHub repository:

To do so, execute the following steps:

1. On GitHub, navigate to the main page of the repository.
2. Under your repository name, click on the Settings tab.
3. In the left sidebar, click Secrets.
4. Click New repository secret.
5. Type the name AZURE_CREDENTIALS for your secret in the Name input box.
6. Enter the JSON output from above as value for your secret.
7. Click Add secret.

3. b) Azure DevOps

If you want to use Azure DevOps Pipelines for deploying the resources, you need to create an Azure Resource Manager service connection. To do so, execute the following steps:

1. First, you need to create an Azure DevOps Project. Instructions can be found here.
2. In Azure DevOps, open the Project settings.
3. Now, select the Service connections page from the project settings page.
4. Choose New service connection and select Azure Resource Manager.

5. On the next page select Service principal (manual).
6. Select the appropriate environment to which you would like to deploy the templates. Only the default option Azure Cloud is currently supported.
7. For the Scope Level, select Subscription and enter your subscription Id and name.
8. Enter the details of the service principal that we have generated in step 3. (Service Principal Id = clientId, Service Principal Key = clientSecret, Tenant ID = tenantId) and click on Verify to make sure that the connection works.
9. Enter a user-friendly Connection name to use when referring to this service connection. Take note of the name because this will be required in the parameter update process.
10. Optionally, enter a Description.
11. Click on Verify and save.

More information can be found here.

4. Parameter Update Process

In order to deploy the ARM templates in this repository to the desired Azure subscription, you will need to modify some parameters in the forked repository. As updating each parameter file manually is a time-consuming and potentially error-prone process, we have simplified the task with a GitHub Action workflow. After successfully executing the previous steps, please open the /.github/workflows/updateParameters.yml YAML file. In this file you need to update the environment variables. Just click on /.github/workflows/updateParameters.yml and edit the following section:

env:
  GLOBAL_DNS_RESOURCE_GROUP_ID: '/subscriptions/{subscriptionId}/resourceGroups/{resourceGroupName}'
  DATA_LANDING_ZONE_SUBSCRIPTION_ID: '{dataLandingZoneSubscriptionId}'
  DATA_LANDING_ZONE_NAME: '{dataLandingZoneName}' # Choose max. 11 characters. They will be used as a prefix for all services. If not unique, deployment can fail for some services.
  LOCATION: '{regionName}'                        # Specifies the region for all services (e.g. 'northeurope', 'eastus', etc.)
  SUBNET_ID: '/subscriptions/{subscriptionId}/resourceGroups/{resourceGroupName}/providers/Microsoft.Network/virtualNetworks/{vnetName}/subnets/{subnetName}'
  SYNAPSE_STORAGE_ACCOUNT_NAME: '{synapseStorageAccountName}'
  SYNAPSE_STORAGE_ACCOUNT_FILE_SYSTEM_NAME: '{synapseStorageAccountFileSystemName}'
  AZURE_RESOURCE_MANAGER_CONNECTION_NAME: '{resourceManagerConnectionName}'

The following table explains each of the parameters:

Parameter	Description	Sample value
GLOBAL_DNS_RESOURCE_GROUP_ID	Specifies the global DNS resource group resource ID which gets deployed with the Data Management Landing Zone	/subscriptions/xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx/resourceGroups/my-resource-group
DATA_LANDING_ZONE_SUBSCRIPTION_ID	Specifies the subscription ID of the Data Landing Zone where all the resources will be deployed	xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx
DATA_LANDING_ZONE_NAME	Specifies the name of your Data Landing Zone. The value should consist of alphanumeric characters (A-Z, a-z, 0-9) and should not contain any special characters like -, _, ., etc. Special characters will be removed in the renaming process.	mynode01
LOCATION	Specifies the region where you want the resources to be deployed.	northeurope
SUBNET_ID	Specifies the resource ID of the dedicated privatelink-subnet which was created during the Data Landing Zone deployment. The subnet should be configured with privateEndpointNetworkPolicies and privateLinkServiceNetworkPolicies, as mentioned in the Prerequisites	/subscriptions/xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx/resourceGroups/my-network-rg/providers/Microsoft.Network/virtualNetworks/my-vnet/subnets/{my}-privatelink-subnet
SYNAPSE_STORAGE_ACCOUNT_NAME	Specifies the name of the Azure Synapse Storage Account	synapsestorageaccount
SYNAPSE_STORAGE_ACCOUNT_FILE_SYSTEM_NAME	Specifies the name of the Synapse Account filesystem	fs
AZURE_RESOURCE_MANAGER_CONNECTION_NAME	Specifies the resource manager connection name in Azure DevOps. You can leave the default value if you want to use GitHub Actions for your deployment. More details on how to create the resource manager connection in Azure DevOps can be found in step 4. b) or here.	my-connection-name

            

After updating the values, please commit the updated version to the main branch of your repository. This will kick off a GitHub Action workflow, which will appear under the Actions tab of the main page of the repository. The Update Parameter Files workflow will update all parameters in your repository according to a certain naming convention. Once the process has finished, it will open a new pull request in your repository, where you can review the changes made by the workflow. Please follow the instructions in the pull request to complete the parameter update process. The instructions will guide towards the following steps:

• create a new resource group where all the resources specific to this Data Domain Batch will be deployed;
• add the required role assignments for the Service Principal created at step 2. Setting up the required Service Principal;
• change the environment variables in the deployment workflow file

Note: We are not renaming the environment variables in the workflow files because this could lead to an infinite loop of workflow runs being started.

After following the instructions in the pull request, you can merge the pull request back into the main branch of your repository by clicking on Merge pull request. Finally, you can click on Delete branch to clean up your repository.

5. (not applicable for GH Actions) Reference pipeline from GitHub repository in Azure DevOps Pipelines

A. Install Azure DevOps Pipelines GitHub Application

First you need to add and install the Azure Pipelines GitHub App to your GitHub account. To do so, execute the following steps:

1. Click on Marketplace in the top navigation bar on GitHub.
2. In the Marketplace, search for Azure Pipelines. The Azure Pipelines offering is free for anyone to use for public repositories and free for a single build queue if you’re using a private repository.

4. Select it and click on Install it for free.

5. If you are part of multiple GitHub organizations, you may need to use the Switch billing account dropdown to select the one into which you forked this repository.
6. You may be prompted to confirm your GitHub password to continue.
7. You may be prompted to log in to your Microsoft account. Make sure you log in with the one that is associated with your Azure DevOps account.

B. Configuring the Azure Pipelines project

As a last step, you need to create an Azure DevOps pipeline in your project based on the pipeline definition YAML file that is stored in your GitHub repository. To do so, execute the following steps:

1. Select the Azure DevOps project where you have setup your Resource Manager Connection.
2. Select Pipelines and then New Pipeline in order to create a new pipeline.

3. Choose GitHub YAML and search for your repository (e.g. "GitHubUserName/RepositoryName").

4. Select your repository.
5. Click on Existing Azure Pipelines in YAML file
6. Select main as branch and /.ado/workflows/dataNodeDeployment.yml as path.

7. Click on Continue and then on Run.

6. Follow the workflow deployment

Congratulations! You have successfully executed all steps to deploy the template into your environment through GitHub Actions or Azure DevOps.

If you are using GitHub Actions, you can navigate to the Actions tab of the main page of the repository where you will see a workflow with the name Data Domain Deployment running. Click on it to see how it deploys one service after another. If you run into any issues, please open an issue here.

If you are using Azure DevOps Pipelines, you can navigate to the pipeline that you have created as part of step 6 and monitor it as each service is deployed. If you run into any issues, please open an issue here.

Documentation

Code Structure

File/folder	Description
.ado/workflows	Folder for ADO workflows. The dataDomainDeployment.yml workflow shows the steps for an end-to-end deployment of the architecture.
.github/workflows	Folder for GitHub workflows. The updateParameters.yml workflow is used for the parameter update process, while the dataDomainDeployment.yml workflow shows the steps for an end-to-end deployment of the architecture.
code	Sample password generation script that will be run in the deployment workflow for resources that require a password during the deployment.
configs	Folder containing a script and configuration file that is used for the parameter update process.
docs	Resources for this README.
infra	Folder containing all the ARM templates for each of the resources that will be deployed (deploy.{resource}.json) together with their parameter files (params.{resource}.json).
CODE_OF_CONDUCT.md	Microsoft Open Source Code of Conduct.
LICENSE	The license for the sample.
README.md	This README file.
SECURITY.md	Microsoft Security README.

Enterprise Scale Analytics Documentation and Implementation

• Documentation
• Implementation - Data Management
• Implementation - Data Landing Zone
• Implementation - Data Domain - Batch
• Implementation - Data Domain - Streaming
• Implementation - Data Product - Reporting
• Implementation - Data Product - Analytics & Data Science

Known issues

Error: MissingSubscriptionRegistration

Error Message:

ERROR: Deployment failed. Correlation ID: ***
  "error": ***
    "code": "MissingSubscriptionRegistration",
    "message": "The subscription is not registered to use namespace 'Microsoft.DocumentDB'. See https://aka.ms/rps-not-found for how to register subscriptions.",
    "details": [
      ***
        "code": "MissingSubscriptionRegistration",
        "target": "Microsoft.DocumentDB",
        "message": "The subscription is not registered to use namespace 'Microsoft.DocumentDB'. See https://aka.ms/rps-not-found for how to register subscriptions."
 

Solution:

This error message appears, in case during the deployment it tries to create a type of resource which has never been deployed before inside the subscription. We recommend to check prior the deployment whether the required resource providers are registered for your subscription and if needed, register them through the Azure Portal, Azure Powershell or Azure CLI as mentioned here.

Error: PECsNotExistingToDenyPublicNetworkAccess

Error Message:

[error]PECsNotExistingToDenyPublicNetworkAccess: Unable to set Deny Public Network Access to Yes since there is no private endpoint enabled to access the server. Please set up private endpoints and retry the operation (https://docs.microsoft.com/azure/sql-database/sql-database-private-endpoint-overview#how-to-set-up-private-link-for-azure-sql-database).

Solution:

This error message appears during the deployment of a resource, in case the subnet associated with that specific resource has not been configured as per the Prerequisites. Please update the subnet configuration or set the correct SUBNET_ID parameter in /.github/workflows/updateParameters.yml.

Contributing

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 [e-mail] with any additional questions or comments.