helm-service consistently consuming more than 100% cpu in TCA-CP
Article ID: 345740
Updated On:
Products
Issue/Introduction
helm-service consistently consumes more than 100% CPU in TCA-CP
Environment
VMware Telco Cloud Automation 2.1
VMware Telco Cloud Automation 2.2
Cause
Enabling ECR repositories (Elastic Container Registry) causing 100% in of the helm-service in the TCA-CP
Resolution
This issue was identified in the extensions platform and fixed in the TCA 2.3 release.
The workaround is to turn off the extensions platform, however, this can be used as a workaround only if the customer is NOT using any ECR repositories (Elastic Container Registry).
The ECR repos can be reenable after upgrading it to TCA 2.3
Workaround:
The workaround is to disable the extensions platform entirely. This will not have an impact on any current harbor based CNFs. The only impact is if the customer wants to use charts/images from ECR repositories.
To disable the extensions platform via the API:
First, obtain the authorization token with the following API call:
curl -v -k -X POST https://<;TCA-CP-IP-address>/hybridity/api/sessions -H "accept: application/json" -H "Accept: application/json" -H "Content-Type: application/json" -d "
{ \"username\": \[[email protected]|mailto:[email protected]]\, \"password\": \"somepassword\"}
"
Use the value for x-hm-authorization token obtained from the previous API call in the next API to enable or disable the extension platform group.
curl -v -k -X PUT -H "Accept:application/json" -H "Content-Type:application/json" -H "x-hm-authorization:8ded3cdf:2866:49ef:a2ed:293f8ca23a33" https:// <TCA-CP-IP-address>/admin/hybridity/api/global/settings/ExtensionPlatform/disableRepoExtensionGroup -d '{"value":"true"}'
helm-service restart will be required after disabling the extensions platform.
These API changes do persist across reboots and upgrades.
Additional Information
TCA 2.1
Feedback
