1. Proposed title of this feature request

Require no reboot when adding/removing containerRuntimeSearchRegistries entries in image.config.openshift.io/cluster object

2. What is the nature and description of the request?

When adding/removing entries to/from containerRuntimeSearchRegistries section of image.config.openshift.io/cluster object as per the below doc, all nodes get drained and rebooted. This is incovenient for the customers as the pods get restarted in all the nodes causing outage for some apps.

https://docs.openshift.com/container-platform/4.11/openshift_images/image-configuration.html

This RFE is to have a feature added to avoid reboot in such case.

As per https://github.com/openshift/machine-config-operator/pull/2259, it was implented to avoid reboot for updates in /etc/containers/registries.conf file. The containerRuntimeSearchRegistries entries update files in /etc/containers/registries.conf.d/ directory though and still requires reboot. A similar feature is needed for this to avoid reboots.

3. Why does the customer need this? (List the business requirements here)

Updating the containerRuntimeSearchRegistries enties in image.config.openshift.io/cluster object causes drain reboot of all the nodes hence pods get restarted causing outage

4. List any affected packages or components.

image.config.openshift.io/cluster object, Files in /etc/containers/registries.conf.d/ directory on nodes