Configure Grafana to visualize metrics emitted from the managed instance cluster

When you deploy an Azure Managed Instance for Apache Cassandra cluster, the service provisions Metric Collector for Apache Cassandra agent software on each data node. The metrics can be consumed by Prometheus and visualized through Grafana. This article describes how to configure Prometheus and Grafana to visualize metrics emitted from your managed instance cluster.

The following tasks are required to visualize metrics:

• Deploy an Ubuntu Virtual Machine inside the Azure Virtual Network where the managed instance is present.
• Install the Prometheus Dashboards onto the VM.

Deploy an Ubuntu server

1. Sign in to the Azure portal.

2. Navigate to the resource group where your managed instance cluster is located. Select Add and search for Ubuntu Server 18.04 LTS image:

   

3. Pick the image and select Create.

4. In the Create a virtual machine blade, enter values for the following fields, you can leave default values for other fields:

   • Virtual machine name - Enter a name for you VM.
   • Region - Select the same region where your Virtual Network has been deployed.

   

5. In the Networking tab, select the Virtual Network in which your managed instance is deployed:

   

6. Finally select Review + Create to create your Metrics server.

Install Prometheus Dashboards

1. First, ensure the networking settings for your newly deployed Ubuntu server have inbound port rules allowing ports 9090 and 3000. These will be required later for Prometheus and Grafana respectively.

   

2. Connect to your Ubuntu server by using Azure CLI or your preferred client tool to connect via SSH.

3. After connecting to the VM, you have to install the metrics collector software. First, download and unzip the files:

    #install unzip utility (if not already installed)
    sudo apt install unzip
   
    #get dashboards
    wget https://github.com/datastax/metric-collector-for-apache-cassandra/releases/download/v0.3.0/datastax-mcac-dashboards-0.3.0.zip -O temp.zip
    unzip temp.zip
   

4. Next, navigate to the prometheus directory and use vi to edit the tg_mcac.json file:

    cd */prometheus
    vi tg_mcac.json    
   

5. Add the ip addresses of each node in your cluster in targets, each with port 9443. Your tg_mcac.json file should look like the below:

    [
      {
        "targets": [
          "10.9.0.6:9443","10.9.0.7:9443","10.9.0.8:9443"
        ],
        "labels": {
   
        }
      }
    ]  
   

6. Save the file. Next, edit the prometheus.yaml file in the same directory. Locate the following section:

    file_sd_configs:
      - files:
        - 'tg_mcac.json'
   

7. Directly below this section, add the following. This is required because metrics are exposed via https.

    scheme: https
    tls_config:
            insecure_skip_verify: true
   

8. The file should now look like the following. Ensure the tabs on each line are as below.

    file_sd_configs:
      - files:
        - 'tg_mcac.json'
    scheme: https
    tls_config:
            insecure_skip_verify: true
   

9. Save the file. You are now ready to start Prometheus and Grafana. First, install Docker:

   sudo apt install apt-transport-https ca-certificates curl software-properties-common
   curl -fsSL https://download.docker.com/linux/ubuntu/gpg | sudo apt-key add -
   sudo add-apt-repository "deb [arch=amd64] https://download.docker.com/linux/ubuntu `lsb_release -cs` test"
   sudo apt update
   sudo apt install docker-ce
   

10. Then install docker compose:

    sudo apt install docker-compose
    

11. Now navigate to the top level directory where docker-compose.yaml is located, and start the application:

    cd ..
    sudo docker-compose up
    

12. Prometheus should be available at port 9090, and Grafana dashboards on port 3000 on your metrics server:

    

Next steps

In this article, you learned how to configure dashboards to visualize metrics in Prometheus using Grafana. Learn more about Azure Managed Instance for Apache Cassandra with the following articles:

• Overview of Azure Managed Instance for Apache Cassandra
• Deploy a Managed Apache Spark Cluster with Azure Databricks