OpenShift is a cloud-based Kubernetes service by Red Hat. Azure Red Hat OpenShift, a.k.a. Azure OpenShift, is a product jointly developed by Red Hat and Azure to offer a seamless integration of OpenShift on the Azure platform.  In this article, we’ll cover:  

• Azure OpenShift benefits 
• Azure OpenShift cost
• How to deploy Azure OpenShift 
• Monitoring an OpenShift container with a monitoring script from Microsoft

Benefits of using Azure OpenShift

Red Hat and Microsoft jointly engineer, operate, and support Azure Red Hat OpenShift. The table below summarizes the key benefits of Azure Red Hat OpenShift over OpenShift as IaaS. Azure OpenShift Benefits

Azure OpenShift Cost

Azure OpenShift 4 has a minimum cluster size of three master nodes and three worker nodes. Both node types use Linux Azure VM pricing. Additionally, worker nodes have OpenShift license costs as well. With Azure OpenShift, these VM sizes are billed as part of the standard Azure subscription. Both master and worker nodes can use Azure's reduced on-demand and reserved instance pricing. For a complete list of supported VM sizes, see Azure Red Hat OpenShift pricing

Azure OpenShift Deployment Options

There are two ways to run OpenShift on Azure:

• OpenShift Container Platform (OCP) deployed on a virtual machine
• Azure Red Hat OpenShift deployed through Azure portal or Azure CLI.

OpenShift running on an Azure VM

OpenShift Container Platform on virtual machines is deployed through cloud.redhat.com/openshift. With this deployment model, OpenShift must be installed and set up on a VM, and you must bring your own license.  There are no Azure-specific operations, integrations, or billing benefits in this case. With this option, using OpenShift on Azure is the same as running it on your own hardware or another infrastructure as a service (IaaS) platform like GCP or AWS. 

Azure OpenShift 

With this deployment model, you can deploy Azure Red Hat OpenShift through the Azure portal or the Azure CLI. It is fully managed and supported by both Red Hat and Azure. Since this is part of the Azure service stack, the billing integrates into your Azure subscription, including a license.

How to Deploy Azure OpenShift 

Now, let’s walk through how you can deploy Azure Red Hat OpenShift using the Azure portal and Azure CLI. 

Prerequisites

Before we get started, you’ll need to:

• Ensure Azure Red Hat is available in your region. 
• Make sure you have a pay-as-you-go Azure account, not a free entry-level account. The free version does not cover third-party license costs.
• Have enough Azure quota for the standard dsv3 family of vCPUs. 

Note that the initial quota allowed is ten vCPUs. You’ll need at least 40. Use the normal Azure process to request a large quota by going to the “Usage + quotas” option available under the Azure portal and requesting an additional quota for the instance type you will use to deploy the Azure OpenShift cluster.  You can find a list of supported Quotas and other requirements for Azure OpenShift here. 

Begin the process in the Red Hat OpenShift management console

1. First, log in to your Red Hat OpenShift console using your Red Hat account at https://console.redhat.com/. Once logged in, you will find many options to interact with Red Hat software. The RedHat OpenShift console displays a variety of submenus.

   

2. Select Red Hat OpenShift Cluster Manager.
3. Then click the "Create Cluster"  which will display many different options to create a new cluster.
4. Click “Try it on Azure” for the Azure Red Hat OpenShift offering. You will see detailed instructions for quickly deploying a production-grade OpenShift cluster standing up on Azure. We’ll follow these instructions here.
5. Get your pull secret from the Red Hat management console for Azure to begin deployment. It is located under “Run it yourself” below the Managed services offering.

   

6. Choose the “Installer-provisioned infrastructure” option.

   

7. Click “Download pull secret” to download the pull secret. This step will save the pull secret to your local machine.

   

Prep Azure for installation

Next, we’ll prep Azure for installation. We can also do this through the Azure portal, but for this article, we’ll use the Azure CLI as it makes deployment easier. 

1. First, we need to register the Red Hat OpenShift resource on our subscription along with Microsoft’s compute, storage, and authorization resource providers. Run the following commands to register required resources:

   
       az provider register -n Microsoft.RedHatOpenShift –wait 
       az provider register -n Microsoft.Compute –wait
       az provider register -n Microsoft.Storage –wait
       az provider register -n Microsoft.Authorization –wait                            
                               

2. Next, run these commands to configure environment variables:

   
       LOCATION=eastus #location of your cluster
       RESOURCEGROUP=arorg #name of the resource group where you want to create your cluster
       CLUSTER=cluster #name of your cluster                                                      
                               

3. Now, we’ll create a resource group with a location. Use these commands to create the group:

   
       az group create 
       --name $RG
       --location $LOCATION                                                                              
                               

   You should see output similar to this:

   
       {
       "id": "/subscriptions/<guid>/resourceGroups/aro-rg",
       "location": "eastus",
       "name": "aro-rg",
       "properties": {
         "provisioningState": "Succeeded"
       },
       "type": "Microsoft.Resources/resourceGroups"
       }                                                                                                
                               

   Note that in Azure, a resource group is a logical group in which Azure resources are deployed and managed. We have to set a location for this. This location is where the resource group metadata is stored and sets the default location for the resources created in this resource group. The resource created in a resource group is not limited to the location of the resource group.
4. Next, we’ll configure networking for our cluster. We will create a virtual network that will have two empty subnets. The first subnet will be for the control pane, and the other for the workers. We will build all this networking inside the resource group we created earlier. Run these commands to configure the cluster networking:

   
       az network vnet create 
       --resource-group $RG 
       --name arovnet 
       --address-prefixes 20.0.0.0/22                                                                                                                       
                               

   You should see output similar to:

   
       {
       "newVNet": {
         "addressSpace": {
           "addressPrefixes": [
             "20.0.0.0/22"
           ]
         },
         "dhcpOptions": {
           "dnsServers": []
         },
         "id": "/subscriptions/<guid>/resourceGroups/aro-rg/providers/Microsoft.Network/virtualNetworks/aro-vnet",
         "location": "eastus",
         "name": "arovnet",
         "provisioningState": "Succeeded",
         "resourceGroup": "arorg",
         "type": "Microsoft.Network/virtualNetworks"
       }                                                                                                                                             
                               

5. After creating the virtual network, we need to create the two subnets we discussed earlier. We’ll also explicitly set a service endpoint, and that’s because Azure guarantees a secure and direct route when we do that. Use this command to create the master subnet:

   
       az network vnet subnet create 
         --resource-group $RG
         --vnet-name arovnet 
         --name master-subnet 
         --address-prefixes 20.0.0.0/23 
         --service-endpoints Microsoft.ContainerRegistry                                                                                                                                               
                               

6. Next, use the following command to create the worker subnet:

   
       az network vnet subnet create 
         --resource-group $RG 
         --vnet-name arovnet 
         --name workersubnet 
         --address-prefixes 20.0.2.0/23 
         --service-endpoints Microsoft.ContainerRegistry                                                                                                                                                                       
                               

7. Now, disable the subnet private endpoint policies for that master subnet. It is required for the service to be able to connect to and manage the cluster, and if we’re using the portal to do it, it would be the default. These are the CLI commands required.

   
       az network vnet subnet update 
         --name mastersubnet 
         --resource-group $RG 
         --vnet-name arovnet 
         --disable-private-link-service-network-policies true                                                                                                                                                                                               
                               

8. Create a cluster using the aro create command. You can run this command to see the options available for the aro create command.

   aro create –help | more                                                                                                                                                                                             
                               

   There are plenty of options available, including the ability to set the VM sizes, set up advanced network configuration, worker sizing, etc., to name a few. To create the cluster, run the commands below. It references the pull secret we downloaded earlier through the Red Hat management console.

   
       az network vnet subnet update 
         --name mastersubnet 
         --resource-group $RG 
         --vnet-name arovnet 
         --disable-private-link-service-network-policies true                                                                                                                                                                                                                
                               

   
       az aro create 
         --resource-group $RESOURCEGROUP 
         --name $CLUSTER 
         --vnet arovnet 
         --master-subnet mastersubnet 
         --worker-subnet workersubnet
         --pull-secret @pull-secret.txt                                                                                                                                                                                                                                    
                               

   Installation typically takes around 30-35 minutes, depending on your region. You can watch the deployment progress on the Azure portal or through the debug command in CLI. When the installation completes, you’ll see the following list of resources deployed on your Azure, as shown in the screenshot below.

   

9. Once the cluster is reporting complete, we need to connect to it. Use this command to retrieve the credentials for the cluster:

   
       az aro list-credentials \
         --name $CLUSTER 
         --resource-group $RG                                                                                                                                                                                                                                
                               

10. Running this command retrieves the details for the cluster.

    
        az aro show 
          --name $CLUSTER 
          --resource-group $RG 
          --query “consoleProfile.url” -o tsv                                                                                                                                                                                                                                                      
                                

    The output will include a URL to connect to the cluster we created.
11. Use the credentials extracted from the list-credentials command to log in to the Red Hat Management portal.

Now you have full cluster access for advanced customization and management. It also gives you complete control over upgrades and life cycle management.

You also get direct access to Azure storage, compute, and auto-scaling options, among others.

How to Configure Azure OpenShift Monitoring

Another excellent integration with Azure is being able to configure Azure RedHat OpenShift version 4 with container insight. After deployment, Azure Insights automatically displays the container. It is displayed under “Unmonitored clusters”, as shown in the screenshot below.

Prerequisites

Before we proceed, make sure:

• Your CLI has access to bash4
• Helm is installed
• The OC command is installed

Running the install script and configuring monitoring

1. Microsoft provides a monitoring script that makes it very easy to enable monitoring. We start by setting up the resource id, resource group, and cluster name, with the command below:

   export ResourceId=”subscriptions/xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx/resourceGroup/rg-name/providers/Microsoft.RedHatOpenShift/OpenShiftClusters/clustername” 
                                   bash enable-monitoring.sh -- resource-id $ResourceId                                                                                                                                                                                                                                                             
                               

2. After running the script, CLI will prompt you to grant access by authentication.

   

3. Go to the highlighted URL and enter the provided authentication code
4. Once the script runs, the cluster will be under monitored clusters.

   

5. Click on the cluster name to access all the various configured dashboards. Depending on the deployment region, it will take around 5-10 minutes for the data to become available.

   

That’s it! Now you can monitor your cluster, nodes, controllers, and containers and run detailed reports.

Conclusion

Azure OpenShift is an excellent choice for OpenShift deployment. The joint support and ease of deployment and use make it one of the best OpenShift deployment platforms available. It offers all the in-demand industry certifications for compliance out of the box. Joint support also makes troubleshooting issues far less complicated. Azure also provides comprehensive monitoring through Azure insights with a prebuilt script from Microsoft.