Aria Automation 8.18.1
Load Balancing of the polling between the Aria Automation Nodes. In some cases the minimum threshold for accounts to poll can be 0 and in such a case, if an account is not doesn't get polled from other nodes, it will never be picked up.
Issue will be fixed on Aria Automation 8.18.1 PATCH 3
Workaround for the issue :
The properties need to be added inside the JAVA_OPTS environment variable inside provisioning-service-app deployment in k8s, Setting these properties will stop the load balancing for 6 months.
kubectl edit deployments -n prelude provisioning-service-app-Debs.connect.upstream.service.unavailability.threshold=200-Debs.connect.node.availability.check.initialDelay=P180D-Debs.connect.node.availability.check.interval=P180D kubectl get pods -n prelude --selector=app=provisioning-service-app
Note:
In Some Cases vSphere account lost connection with Aria Automation the polling can be interrupted and the account will have to be validated and saved, to resume the polling.
Need to apply the Workaround steps again if you apply the PATCH or Upgrade Aria Automation.