The customer's OpenShift clusters are disconnected and reside in AWS. The Cost Management Metrics Operator is used to generate cost data for these clusters. They periodically download the node and usage CSV files from the CMMO operator and load them into tables in a database – joining the content from those tables on the "node name" column.

But as their AWS node names are based on IP addresses, those names are occasionally re-used as nodes are spun up and down over time. And the re-use of those node names affects the integrity of their database content (i.e. the same node name doesn't represent the same node over long periods).

As a solution, the customer would like to have the ARNs of the nodes included in the node CSV. The ARNs of the nodes are already included in the usage CSV. So by adding the much more unique ARN to the node CSV, that ARN could be used instead of the often-changing "node name" when joining the tables.