Rancher Rodeo XII - RKE - Helm3

For testing purposes, Rancher can integrate with Wildcard DNS services. This allows you to quickly create a publicly resolvable DNS entry that points to the ingress controller of a cluster that is managed by Rancher.

To configure the Wildcard DNS service:

• Go to Settings
• Edit the ingress-ip-domain setting
• Change the value to sslip.io
  
  

  Creating A Kubernetes Lab Cluster within Rancher

  In this step, we will be creating a Kubernetes Lab environment within Rancher. Normally, in a production case, you would create a Kubernetes Cluster with multiple nodes; however, with this lab environment, we will only be using one virtual machine for the cluster.

  1. Hover over the top left dropdown, then click Global
     • The current context is shown in the upper left, and should say 'Global'
  2. Click Add Cluster
     • Note the multiple types of Kubernetes cluster Rancher supports. We will be using Existing nodes for this lab, but there are a lot of possibilities with Rancher.
  3. Click on the Existing nodes Cluster box
  4. Enter a name in the Cluster Name box
  5. Set the Kubernetes Version to a v1.18 version
  6. Click Next at the bottom.
  7. Make sure the boxes etcd, Control Plane, and Worker are all ticked.
  8. Click Show advanced options to the bottom right of the Worker checkbox
  9. Enter the Public Address (3.236.182.92) and Internal Address (172.31.0.15)
     • IMPORTANT: It is VERY important that you use the correct External and Internal addresses from the Cluster01 machine for this step, and run it on the correct machine. Failure to do this will cause the future steps to fail.
  10. Click the clipboard to Copy to Clipboard the docker run command
  11. Proceed to the next step of this scenario
      
      

      Start the Rancher Kubernetes Cluster Bootstrapping Process

      IMPORTANT NOTE: Make sure you have selected the Cluster01 tab in HobbyFarm in the window to the right. If you run this command on Rancher01 you will cause problems for your scenario session.

      1. Take the copied docker command and run it on Cluster01
      2. Once the docker run command is complete, you should see a message similiar to 1 node has registered
      3. Within the Rancher UI click on <YOUR_CLUSTER_NAME> which is the name you entered during cluster creation.
      4. You can watch the state of the cluster as your Kubernetes node Cluster01 registers with Rancher here as well as the Nodes tab
      5. Your cluster state on the Global page will change to Active
      6. Once your cluster has gone to Active you can click on it and start exploring.
         
         

         Interacting with the Kubernetes Cluster

         In this step, we will be showing basic interaction with our Kubernetes cluster.

         1. Click into your newly active cluster.
         2. Note the three dials, which illustrate cluster capacity.
         3. Click the Launch kubectl button in the top right corner of the Cluster overview page, and enter kubectl get pods --all-namespaces and observe the fact that you can interact with your Kubernetes cluster using kubectl.
         4. Also take note of the Kubeconfig File button which will generate a Kubeconfig file that can be used from your local desktop or within your deployment pipelines.
         5. Click the Ellipses in the top right corner and note the various operational options available for your cluster. We will be exploring these in a later step.
            
            

            Enable Rancher Monitoring

            To deploy the Rancher Monitoring feature, we will need to navigate to the Cluster Explorer.

            1. On your newly-created cluster, click the "Explorer" button to open the Cluster Explorer.
            2. Once the Cluster Explorer loads, use the dropdown in the upper-left section of the page, and navigate to "Apps & Marketplace."
            3. Locate the "Monitoring" chart, and click on it
            4. Select "Chart Options" on the left. Change Resource Limits > Requested CPU from 750m to 250m. This is required because our scenario virtual machine has limited CPU available.
            5. Click "Install" at the bottom of the page, and wait for the helm install operation to complete.

            Once Monitoring has been installed, you can click on that application under "Installed Apps" to view the various resources that were deployed.
            
            

            Working with Rancher Monitoring

            Once Rancher Monitoring has been deployed, we can view the various components and interact with them.

            1. In the dropdown in the upper-left corner of the Cluster Explorer, select "Monitoring"
            2. On the Monitoring Dashboard page, identify the "Grafana" link. Clicking this will proxy you to the installed Grafana server

            Once you have opened Grafana, feel free to explore the various dashboard and visualizations that have been setup by default.

            These options can be customized (metrics and graphs), but doing so is out of the scope of this scenario.
            
            

            Create a Deployment And Service

            In this step, we will be creating a Kubernetes Deployment and Kubernetes Service for an arbitrary workload. For the purposes of this lab, we will be using the docker image rancher/hello-world:latest but you can use your own docker image if you have one for testing.

            When we deploy our container in a pod, we probably want to make sure it stays running in case of failure or other disruption. Pods by nature will not be replaced when they terminate, so for a web service or something we intend to be always running, we should use a Deployment.

            The deployment is a factory for pods, so you'll notice a lot of similairities with the Pod's spec. When a deployment is created, it first creates a replica set, which in turn creates pod objects, and then continues to supervise those pods in case one or more fails.

            Note: These steps will need to be executed within the Cluster Manager. To access, click on the Cluster Manager button at the top of the page.

            ---

            1. Hover over the Dropdown next to the Rancher logo in the top left corner, hover over your cluster name, then select Default as the project.
            2. Under the Workloads tab press Deploy in the top right corner and enter the following criteria:
               • Name - helloworld
               • Docker Image - rancher/hello-world:latest
               • Click Add Port and enter 80 for the container port
               • ** NOTE: ** Note the other capabilities you have for deploying your container. We won't be covering these in this Rodeo, but you have plenty of capabilities here.
            3. Scroll down and click Launch
            4. You should see one pod get deployed and a TCP endpoint under your workload name exposing the NodePort service that has been created.
            5. Click the Arrow next to your workload that you just created, then note the + and - buttons under the replica count to the right. You can click these correspondingly and refresh your browser on the nodeport to see the changes in the pod name.
               
               

               Create a Kubernetes Ingress

               In this step, we will be creating a Layer 7 ingress to access the workload we just deployed in the previous step. For this example, we will be using xip.io as a way to provide a DNS hostname for our workload. Rancher will automatically generate a corresponding workload IP.

               1. Hover over the Dropdown next to the Rancher logo in the top left corner, hover over your cluster name, then select Default as the project.
               2. In the Default project in the Workloads section, click on the Load Balancing tab
               3. Click Add Ingress and enter the following criteria:
                  • Name - helloworld
                  • Leave 'Automatically generate a .sslip.io hostname' selected
                  • Click the minus button on the right to remove the empty backend rule
                  • Click the + Service button
                  • Pick the helloworld-nodeport service from the dropdown under Target
               4. Click Save and wait for the sslip.io hostname to register, you should see the rule become Active within a few minutes.
               5. Click on the hostname and browse to the workload.

               ** Note: ** You may receive transient 404/502/503 errors while the workload stabilizes. This is due to the fact that we did not set a proper readiness probe on the workload, so Kubernetes is simply assuming the workload is healthy.
               
               

               Upgrading your Kubernetes Cluster

               This step shows how easy it is to upgrade your Kubernetes clusters within Rancher.

               1. Hover over the Dropdown next to the Rancher logo in the top left corner, then select your cluster.
               2. Click the ellipses (...) next to the Kubeconfig file button
               3. Click Edit
               4. Scroll down and select the dropdown under the Kubernetes Version
               5. Select a newer version of Kubernetes
               6. Scroll down and hit Save

               Observe that your Kubernetes cluster will now be upgraded.
               
               

               Congratulations

               Congratulations, you have finished the Scenario. If you would like to tear down your lab environment, you can click the "Finish" button, otherwise, continue to work with your Kubernetes cluster while keeping HobbyFarm in the background.

            6.