Feature Request: Separate metrics for management vs. resource conflicts and use gauge aggregation #2024
Description
Checklist
- I did not find a related open enhancement request.
- I understand that enhancement requests filed in the GitHub repository are by default low priority.
- If this request is time-sensitive, I have submitted a corresponding issue with GCP support.
Describe the feature
Description: Currently, Config Sync records both management conflicts and resource conflicts under the same metric. This is confusing because resource conflicts are often benign (e.g., multiple controllers managing the same resource type but different objects), whereas management conflicts consistently require user intervention. The current metric uses a "total" count that only increments. It does not decrease when a conflict is resolved, making it impossible for users to know the current number of active conflicts in the cluster.
Feature request:
- Separate Metrics: specific metrics should be available to distinguish "management conflicts" from "resource conflicts," allowing users to alert only on the actionable items.
- Gauge Aggregation: Change the metric type to use "last value" (Gauge) aggregation. The metric should report the total number of currently active conflicts (deduplicated by object ID) so that the value goes down when conflicts are fixed.
Alternatives considered:
- Stop reporting "resource conflicts" entirely to reduce noise, though this removes visibility into controller fights.
- Removing both metrics and relying solely on generic error metrics.
Importance
No response