The managed-upgrade-operator
provides a configurable mechanism for retrieving and storing an UpgradeConfig
Custom Resource that can be reconciled against by the UpgradeConfig
controller.
Currently, the following sources are supported:
Source | Description |
---|---|
OCM |
Retrieve an UpgradeConfig from the OpenShift Cluster Manager upgrade_policies API |
LOCAL |
Using UpgradeConfig CR locally on the OpenShift Cluster |
The UpgradeConfig Manager is configured in a configManager
block in the managed-upgrade-operator-config
ConfigMap.
If this block is not present, the operator will not create a manager at all, but will still run.
For source-specific configuration, see the correpsonding section below.
The following configuration fields must be set:
Field | Description | Example |
---|---|---|
source |
Indicates the type of config manager being used | OCM |
ocmBaseUrl |
Base URL of the OpenShift Cluster Manager API | https://api.openshift.com/ |
watchInterval |
Frequency* in minutes with which the API will be polled | 60 |
The OCM UpgradeConfig Manager will intentionally apply a jitter factor of 10% to the watch interval, so the precise frequency may not always be the value specified.
Complete example:
configManager:
source: OCM
ocmBaseUrl: https://api.openshift.com
watchInterval: 60
If we set the OCM base URL to the URL of the local OCM agent service (http://ocm-agent.openshift-ocm-agent-operator.svc.cluster.local:8081
) we will activate the ocmAgent
client handler, which will use the OCM agent endpoints set out in the OCM Agent router
The following configuration fields must be set:
Field | Description | Example |
---|---|---|
source |
Indicates the type of config manager being used | LOCAL |
localConfigName |
Name of the Local config being used | managed-upgrade-config |
watchInterval |
Frequency* in minutes with which UpgradeConfig CR name being looked | 60 |
Complete example:
configManager:
source: LOCAL
localConfigName: managed-upgrade-config
watchInterval: 60