Demo Request Quota

In this guide, you'll learn how to create and enforce a resource quota in an OpenShift project. By following these steps, you will set up a project, define a resource quota using a YAML manifest, and deploy a pod that respects the quota limitations.

Step 1: Creating a New Project

Begin by logging in to the OpenShift console. Navigate to Home → Projects and create a new project with the following details:

• Project Name: Limit Tester
• Display Name: Same as the project name
• Description: Namespace for Resource Testing

Once the project is created, it will appear in your project list.

Step 2: Defining the Resource Quota

Open your code editor (for example, VS Code) and create a YAML manifest file named limit.yaml. This manifest defines a resource quota that limits the total memory requests for the namespace "limittester" to 512 MiB. In this case, the resource quota is part of the core API group.

apiVersion: v1
kind: ResourceQuota
metadata:
  name: memlimit
  namespace: limittester
spec:
  hard:
    requests.memory: 512Mi

In the manifest:

• The metadata section specifies the quota's name and the namespace where it applies.
• The spec.hard section enforces a memory request limit of 512 MiB for all pods in the project.

Apply the manifest with the following command:

oc apply -f limit.yaml

After running the command, you should see a confirmation message indicating that the resource quota was created. To verify, run:

oc get resourcequota -n limittester

The expected output is similar to:

PS C:\Users\mike\Desktop> oc get resourcequota -n limittester
NAME      AGE   REQUEST               LIMIT
memlimit  35s   requests.memory: 0/512Mi

At this stage, no memory has yet been requested by any pods in the project.

Step 3: Deploying a Pod with a Memory Request

Next, test the quota by deploying a pod that requests memory. Create or open a deployment manifest file named deployment.yaml. In this file, the namespace is explicitly set to "limittester" to ensure the deployment is part of the correct project. Combine the metadata and pod specification in one deployment configuration:

apiVersion: apps/v1
kind: Deployment
metadata:
  name: carts-db
  namespace: limittester
  labels:
    app: carts-db
spec:
  replicas: 1
  selector:
    matchLabels:
      app: carts-db
  template:
    metadata:
      labels:
        app: carts-db
    spec:
      containers:
        - name: carts-db
          image: centos/mongodb-34-centos7
          resources:
            requests:
              memory: "100Mi"
          ports:
            - name: mongo

Key points in the deployment:

• The metadata.namespace field ensures that the deployment is created within the "limittester" project.
• The container carts-db is configured to request 100 MiB of memory.

Apply the deployment with:

oc apply -f deployment.yaml

Verify that the deployment is running by executing:

oc get deployments -n limittester

Then, check the resource quota again to see the updated memory usage:

oc get resourcequota -n limittester

Expected output:

PS C:\Users\mike\Desktop> oc get resourcequota -n limittester
NAME       AGE     REQUEST                      LIMIT
memlimit   3m14s  requests.memory: 100Mi/512Mi

This output confirms that 100 MiB of the allowed 512 MiB has been requested by the deployment, indicating that the resource quota is functioning as expected.

With these steps, you have successfully created a project, defined a resource quota, and deployed a pod with a memory request within the specified limits.

For further details and advanced configurations, consider exploring additional OpenShift and Kubernetes documentation.