increase machine health check node unready timeout to 15m (Azure#3133)

* increase machine health check node unready timeout to 15m

* update mhc docs

* increase machine health check node startup timeout to 25m

pkg/operator/controllers/machinehealthcheck/doc.go

@@ -19,11 +19,11 @@ aro.machinehealthcheck.managed
 - When set to false, the controller will attempt to remove the aro-machinehealthcheck CR and the MHC Remediation alert from the cluster.
   This should effectively disable the MHC we deploy and prevent the automatic reconciliation of nodes.
 - When set to true, the controller will deploy/overwrite the aro-machinehealthcheck CR and the MHC Remediation alert to the cluster.
-  This enables the cluster to self heal when at most 1 worker node goes not ready for at least 5 minutes and alert when remediation
+  This enables the cluster to self heal when at most 1 worker node goes not ready for at least 15 minutes and alert when remediation
   occurs 2 or more times within an hour.
 
 The aro-machinehealth check is configured in a way that if 2 worker nodes go not ready it will not take any action.
 More information about how the MHC works can be found here:
-https://docs.openshift.com/container-platform/4.9/machine_management/deploying-machine-health-checks.html
+https://docs.openshift.com/container-platform/4.12/machine_management/deploying-machine-health-checks.html
 
 */

pkg/operator/controllers/machinehealthcheck/staticresources/machinehealthcheck.yaml

@@ -15,10 +15,10 @@ spec:
       operator: Exists
   unhealthyConditions:
   - type: "Ready"
-    timeout: "300s"
+    timeout: "15m"
     status: "False"
   - type: "Ready"
-    timeout: "300s"
+    timeout: "15m"
     status: "Unknown"
   maxUnhealthy: "1"
-  nodeStartupTimeout: "20m"
+  nodeStartupTimeout: "25m"