ערוך

שתף באמצעות


Quickstart: Create an Azure Managed Instance for Apache Cassandra cluster using Azure CLI

Azure Managed Instance for Apache Cassandra is a fully managed service for pure open-source Apache Cassandra clusters. The service also allows configurations to be overridden, depending on the specific needs of each workload, allowing maximum flexibility and control where needed.

This quickstart demonstrates how to use the Azure CLI commands to create a cluster with Azure Managed Instance for Apache Cassandra. It also shows to create a datacenter, and scale nodes up or down within the datacenter.

Prerequisites

Important

This article requires the Azure CLI version 2.30.0 or higher. If you are using Azure Cloud Shell, the latest version is already installed.

Create a managed instance cluster

  1. Sign in to the Azure portal

  2. Set your subscription ID in Azure CLI:

    az account set -s <Subscription_ID>
    
  3. Next, create a Virtual Network with a dedicated subnet in your resource group:

    az network vnet create -n <VNet_Name> -l eastus2 -g <Resource_Group_Name> --subnet-name <Subnet Name>
    

    Note

    The Deployment of a Azure Managed Instance for Apache Cassandra requires internet access. Deployment fails in environments where internet access is restricted. Make sure you aren't blocking access within your VNet to the following vital Azure services that are necessary for Managed Cassandra to work properly:

    • Azure Storage
    • Azure KeyVault
    • Azure Virtual Machine Scale Sets
    • Azure Monitoring
    • Microsoft Entra ID
    • Azure Security
  4. Apply some special permissions to the Virtual Network, which are required by the managed instance. Use the az role assignment create command, replacing <subscriptionID>, <resourceGroupName>, and <vnetName> with the appropriate values:

    az role assignment create \
      --assignee a232010e-820c-4083-83bb-3ace5fc29d0b \
      --role 4d97b98b-1d4f-4787-a291-c67834d212e7 \
      --scope /subscriptions/<subscriptionID>/resourceGroups/<resourceGroupName>/providers/Microsoft.Network/virtualNetworks/<vnetName>
    

    Note

    The assignee and role values in the previous command are fixed values, enter these values exactly as mentioned in the command. Not doing so will lead to errors when creating the cluster. If you encounter any errors when executing this command, you may not have permissions to run it, please reach out to your admin for permissions.

  5. Next create the cluster in your newly created Virtual Network by using the az managed-cassandra cluster create command. Run the following command the value of delegatedManagementSubnetId variable:

    Note

    The value of the delegatedManagementSubnetId variable you will supply below is exactly the same as the value of --scope that you supplied in the command above:

    resourceGroupName='<Resource_Group_Name>'
    clusterName='<Cluster_Name>'
    location='eastus2'
    delegatedManagementSubnetId='/subscriptions/<subscription ID>/resourceGroups/<resource group name>/providers/Microsoft.Network/virtualNetworks/<VNet name>/subnets/<subnet name>'
    initialCassandraAdminPassword='myPassword'
    cassandraVersion='3.11' # set to 4.0 for a Cassandra 4.0 cluster
    
    az managed-cassandra cluster create \
      --cluster-name $clusterName \
      --resource-group $resourceGroupName \
      --location $location \
      --delegated-management-subnet-id $delegatedManagementSubnetId \
      --initial-cassandra-admin-password $initialCassandraAdminPassword \
      --cassandra-version $cassandraVersion \
      --debug
    
  6. Finally, create a datacenter for the cluster, with three nodes, Standard D8s v4 VM SKU, with 4 P30 disks attached for each node, by using the az managed-cassandra datacenter create command:

    dataCenterName='dc1'
    dataCenterLocation='eastus2'
    virtualMachineSKU='Standard_D8s_v4'
    noOfDisksPerNode=4
    
    az managed-cassandra datacenter create \
      --resource-group $resourceGroupName \
      --cluster-name $clusterName \
      --data-center-name $dataCenterName \
      --data-center-location $dataCenterLocation \
      --delegated-subnet-id $delegatedManagementSubnetId \
      --node-count 3 \
      --sku $virtualMachineSKU \
      --disk-capacity $noOfDisksPerNode \
      --availability-zone false
    

    Note

    The value for --sku can be chosen from the following available SKUs:

    • Standard_E8s_v4
    • Standard_E16s_v4
    • Standard_E20s_v4
    • Standard_E32s_v4
    • Standard_DS13_v2
    • Standard_DS14_v2
    • Standard_D8s_v4
    • Standard_D16s_v4
    • Standard_D32s_v4

    Note also that --availability-zone is set to false. To enable availability zones, set this to true. Availability zones increase the availability SLA of the service. For more details, review the full SLA details here.

    Warning

    Availability zones are not supported in all regions. Deployments will fail if you select a region where Availability zones are not supported. See here for supported regions. The successful deployment of availability zones is also subject to the availability of compute resources in all of the zones in the given region. Deployments may fail if the SKU you have selected, or capacity, is not available across all zones.

  7. Once the datacenter is created, if you want to scale up, or scale down the nodes in the datacenter, run the az managed-cassandra datacenter update command. Change the value of node-count parameter to the desired value:

    resourceGroupName='<Resource_Group_Name>'
    clusterName='<Cluster Name>'
    dataCenterName='dc1'
    dataCenterLocation='eastus2'
    
    az managed-cassandra datacenter update \
      --resource-group $resourceGroupName \
      --cluster-name $clusterName \
      --data-center-name $dataCenterName \
      --node-count 9
    

Connect to your cluster

Azure Managed Instance for Apache Cassandra does not create nodes with public IP addresses. To connect to your newly created Cassandra cluster, you must create another resource inside the virtual network. This resource can be an application, or a virtual machine with Apache's open-source query tool CQLSH installed. You can use a Resource Manager template to deploy an Ubuntu virtual machine.

Connecting from CQLSH

After the virtual machine is deployed, use SSH to connect to the machine and install CQLSH as shown in the following commands:

# Install default-jre and default-jdk
sudo apt update
sudo apt install openjdk-8-jdk openjdk-8-jre

# Install the Cassandra libraries in order to get CQLSH:
echo "deb http://archive.apache.org/dist/cassandra/debian 311x main" | sudo tee -a /etc/apt/sources.list.d/cassandra.sources.list
curl https://downloads.apache.org/cassandra/KEYS | sudo apt-key add -
sudo apt-get update
sudo apt-get install cassandra

# Export the SSL variables:
export SSL_VERSION=TLSv1_2
export SSL_VALIDATE=false

# Connect to CQLSH (replace <IP> with the private IP addresses of a node in your Datacenter):
host=("<IP>")
initial_admin_password="Password provided when creating the cluster"
cqlsh $host 9042 -u cassandra -p $initial_admin_password --ssl

Connecting from an application

As with CQLSH, connecting from an application using one of the supported Apache Cassandra client drivers requires SSL encryption to be enabled, and certification verification to be disabled. See samples for connecting to Azure Managed Instance for Apache Cassandra using Java, .NET, Node.js and Python.

Disabling certificate verification is recommended because certificate verification will not work unless you map I.P addresses of your cluster nodes to the appropriate domain. If you have an internal policy which mandates that you do SSL certificate verification for any application, you can facilitate this by adding entries like 10.0.1.5 host1.managedcassandra.cosmos.azure.com in your hosts file for each node. If taking this approach, you would also need to add new entries whenever scaling up nodes.

For Java, we also highly recommend enabling speculative execution policy where applications are sensitive to tail latency. You can find a demo illustrating how this works and how to enable the policy here.

Note

In the vast majority of cases it should not be necessary to configure or install certificates (rootCA, node or client, truststores, etc) to connect to Azure Managed Instance for Apache Cassandra. SSL encryption can be enabled by using the default truststore and password of the runtime being used by the client (see Java, .NET, Node.js and Python samples), because Azure Managed Instance for Apache Cassandra certificates will be trusted by that environment. In rare cases, if the certificate is not trusted, you may need to add it to the truststore.

Configuring client certificates (optional)

Configuring client certificates is optional. A client application can connect to Azure Managed Instance for Apache Cassandra as long as the above steps have been taken. However, if preferred, you can also additionally create and configure client certificates for authentication. In general, there are two ways of creating certificates:

  • Self signed certs. This means a private and public (no CA) certificate for each node - in this case we need all public certificates.
  • Certs signed by a CA. This can be a self-signed CA or even a public one. In this case we need the root CA certificate (refer to instructions on preparing SSL certificates for production), and all intermediaries (if applicable).

If you want to implement client-to-node certificate authentication or mutual Transport Layer Security (mTLS), you need to provide the certificates via Azure CLI. The below command will upload and apply your client certificates to the truststore for your Cassandra Managed Instance cluster (i.e. you do not need to edit cassandra.yaml settings). Once applied, your cluster will require Cassandra to verify the certificates when a client connects (see require_client_auth: true in Cassandra client_encryption_options).

resourceGroupName='<Resource_Group_Name>'
clusterName='<Cluster Name>'

az managed-cassandra cluster update \
  --resource-group $resourceGroupName \
  --cluster-name $clusterName \
  --client-certificates /usr/csuser/clouddrive/rootCert.pem /usr/csuser/clouddrive/intermediateCert.pem

Troubleshooting

If you encounter an error when applying permissions to your Virtual Network using Azure CLI, such as Cannot find user or service principal in graph database for 'e5007d2c-4b13-4a74-9b6a-605d99f03501', you can apply the same permission manually from the Azure portal. Learn how to do this here.

Note

The Azure Cosmos DB role assignment is used for deployment purposes only. Azure Managed Instanced for Apache Cassandra has no backend dependencies on Azure Cosmos DB.

Clean up resources

When no longer needed, you can use the az group delete command to remove the resource group, the managed instance, and all related resources:

az group delete --name <Resource_Group_Name>

Next steps

In this quickstart, you learned how to create an Azure Managed Instance for Apache Cassandra cluster using Azure CLI. You can now start working with the cluster: