Closed
Description
How to categorize this issue?
/area metering
/kind enhancement
/priority 3
What would you like to be added:
Presently, the mcm_machine_info
metric only exposes the each machine's
- name
- machineClass
- providerID
However, it's missing the node's name, which can be different from the machine's name (especially in the case of AWS where the node_name is the AWS hostname aka ip-10-0-0-1.<region>.compute.internal
).
Why is this needed:
For troubleshooting and RCA reasons. Presently, we can find this information by digging through the MCM's logs (if we are lucky, if the node is old sometimes the specific log line is not available anymore), but exposing that as a metric would make this much more convenient!