Skip to content

Latest commit

 

History

History
206 lines (136 loc) · 5.35 KB

ops-quota.md

File metadata and controls

206 lines (136 loc) · 5.35 KB
Raw

Quotas and Limits

A resource quota, defined by a ResourceQuota object, provides constraints that limit aggregate resource consumption per project. It can limit the quantity of objects that can be created in a project by type, as well as the total amount of compute resources and storage that may be consumed by resources in that project.

It can be used for resource protection, capacity platform and governance of the environment.

In this lab, we will create a simple resource quota in the frontend project.

Step 1 - Apply Resource Limits

In this lab, we will apply resource limits on a pod. Create a new project using the UI or the command line.

oc new-project myperf-user01

In the Web Console, Click on Add to Project. Add a JavaScript/Node.js application.

Use the following parameters: Name: myapp Git URL: https://github.com/sylvoie/testoc

Expand the Advanced Options

image

Set the following scaling options:

Scaling Mode: Automatic
Min Pod: 1
Max Pod: 4
CPU Target: 80%

image

Set the following resource limits:

CPU request: 100
CPU limit: 500

MEM request: 256 MiB
MEM limit: 512 MiB

image

Create the application

Navigate to the pod and open the "Metrics" tab in the UI. The resource utilization is capped.

image

Step 2 - Performance test

This step is optional. You will see how OpenShift autoscale applications and respond to additional load. Limits and scaling maximums will be enforced to protect the other applications and tenants of the platform.

Deploy the load testing container

In this step, we will deploy a container with Centos and HTTperf (an open source load testing tool) installed.

In the same project create above, deploy the httperf container.

To deploy this container, from the command line use the following command:

oc new-app https://github.com/OSE3Sandbox/httperf

In the OpenShift UI, wait for the httperf container to deploy and go to the httperf pod terminal.

In the home directory, edit the httperf.sh file to configure the endpoint specific to your deployment (myapp.myperf-user01.svc). Start the performance test by executing the script: ./httperf.sh The script will output the results when completed.

image

In another browser window, you can monitor the behaviour of OpenShift (metrics, autoscale). image

Step 3- Apply project level limit and quota

In this section, we will create a complete example of a project using limits and quotas. Limits are applied to resource consumption and resource request, quotas are applied to the number of objects that can be created.

Create a new project:

oc new-project myapp-dev

in that project, create a resource limits and resource quota using the following files:

oc create -f https://raw.githubusercontent.com/OSE3Sandbox/openshiftv3-workshop/master/limits.yaml

oc create -f https://raw.githubusercontent.com/OSE3Sandbox/openshiftv3-workshop/master/quota.yaml

Create the application:

oc new-app docker.io/openshift/hello-openshift:v1.4.1

In the OpenShift UI, observe the quotas and limits for the myapp-dev project.

Step 4 - Apply Quota automatically to new projects

In this lab, we will learn how to apply automatically a quota when a new project is created.

OpenShift automatically provisions projects based on the template that is identified by the projectRequestTemplate parameter of the master-config.yaml file. If the parameter is not defined, the API server creates a default template that creates a project with the requested name, and assigns the requesting user to the "admin" role for that project.

Create a project template

On the OpenShift server export the default project template to a file:

oc adm create-bootstrap-project-template -o yaml > template.yaml

Edit the template.yaml to add the following section under the "objects:" section

- apiVersion: v1
  kind: LimitRange
  metadata:
    name: limits
  spec:
    limits:
    - type: Pod
      max:
        cpu: 400m
        memory: 200M
      min:
        cpu: 100m
        memory: 50M
    - type: Container
      max:
        cpu: 200m
        memory: 100M
      min:
        cpu: 50m
        memory: 10M
      default:
        cpu: 200m
        memory: 100M
      defaultRequest:
        cpu: 200m
        memory: 100M
      maxLimitRequestRatio:
        cpu: '4'

Be careful when editing YAML file. Spaces and indentation must be preserved. Tabs are not allowed.

Save the file

Import the template in the OpenShift server:

oc create -f template.yaml -n default

We need to modify the OpenShift configuration file to use this new project template as the default. This operation must be performed as root. On the OpenShift server:

sudo -i
cd /etc/origin/master/
cp master-config.yaml master-config.yaml.backup

Edit the master-config.yaml. Locate the projectRequestTemplate attribute and change to:

  projectRequestTemplate: "default/project-request"

Again, it is YAML, be careful about spaces and indentation.

Restart the OpenShift master:

systemctl restart atomic-openshift-master

In the OpenShift console, create a new project. Navigate to the Resources->Quota.

image