Show appropriate status message via self monitor if a target is close to scrape limit #1037
Labels
area/manager
Manager or module changes
area/metrics
MetricPipeline
kind/feature
Categorizes issue or PR as related to a new feature.
Description
As part of #976 the metric-agent should expose under a dedicated port the diagnostic metrics relevant for scraping metric targets via the prometheusreceiver. The new port should be protected by a networkpolicy so that user cannot reach it. For now the plain exposure is enough in order to enable operational diagnostic.
Criterias
Implementation details
Example configuration for metrics agent
Probably we should not expose just all metrics but also filter out business metrics. So instead we should use a new dedicated pipeline having the same receivers as input, but doing a specialized filtering and exporting
The text was updated successfully, but these errors were encountered: