Metric `karpenter_interruption_received_messages` includes non-karpenter instance events

[shailendher]

Description

How can the docs be improved?

I was quite surprised the first time when the metric karpenter_interruption_received_messages{message_type="SpotInterruptionKind"} didn't match karpenter_nodeclaims_terminated{reason="interruption"}.

Looking through the eventbridge configuration, it's quite clear why the karpenter_interruption_received_messages metric is aggregated at the account level. Perhaps the documentation could be updated to make it clear?

karpenter-provider-aws/website/content/en/docs/reference/metrics.md

Lines 101 to 102 in e326781

	### `karpenter_interruption_received_messages`
	Count of messages received from the SQS queue. Broken down by message type and whether the message was actionable.

Alternative would be to filter out non-Karpenter-related events from the karpenter_interruption_received_messages metric.

[kantajp]

I'd like to work on this issue. Could you please assign it to me?

[jonathan-innis]

I think we could probably think about updating the label that we are using on this metric. Right now we are using message_type=NoOp to indicate that a message doesn't apply to the current cluster. Maybe we can add a label like managed=true/false or something similar to better-indicate that this message does or doesn't apply to the current cluster.