Managing Workload Scaling Policies
Scenario
After a workload scaling policy is created, you can update and delete the policy, as well as edit the YAML file.
Procedure
You can view the rules, latest status, and events of a workload scaling policy and handle exceptions based on the error information displayed.
- Log in to the CCE console and click the cluster name to access the cluster console.
- In the navigation pane, choose Policies. On the Scaling Policies tab page, click the HPA Policies/CronHPA Policies tab based on the scaling policy type.
- Check the latest status, rules, and associated workloads of a scaling policy.
You can also check a created scaling policy on the workload details page.
- Log in to the CCE console and click the cluster name to access the cluster console.
- In the navigation pane, choose Workloads. Click the workload name to view its details.
- On the workload details page, switch to the Auto Scaling tab page to obtain the scaling policies. You can also obtain the scaling policies you configured on the Policies page.
- Manage scaling policies.
Scaling Policy Type
Operation
HPA
- View Events: Check HPA policy events. If an error occurred, locate and rectify the fault based on the error message displayed on the page.
- Edit YAML: In the dialog box displayed, edit, copy, or download the YAML file.
- Edit: On the Edit HPA Policy page, configure policy parameters listed in Table 1.
- Clone: Duplicate an existing auto scaling policy and modify the parameter settings as required.
- Delete: In the dialog box displayed, click Yes.
CronHPA
- View YAML: In the dialog box displayed, copy or download the YAML file but you are not allowed to modify it.
- Delete: In the dialog box displayed, click Yes.
Feedback
Was this page helpful?
Provide feedbackThank you very much for your feedback. We will continue working to improve the documentation.