Troubleshooting Prometheus for SMM
Troubleshooting Prometheus for SMM requires being able to diagnose and debug problems related to performance, network connectivity, out-of-memory conditions, disk space usage, and crash or non-responsive conditions.
This section provides some troubleshooting solutions for Prometheus issues for SMM.
- Issue: If no metrics are available.
- Solution: Examine if all configured Prometheus targets are running.
- Issue: You observe incorrect metrics in SMM.
- Solution: Examine
upmetrics to determine if scraping failed at certain points in time.
- Issue: If Prometheus does not start up or scrape right after Kafka startup, or Prometheus is down for some time, scrapped metric values might be unusually high.
- Solution: This situation can be avoided by querying the endpoints manually with the
update=trueparameter and then starting Prometheus. If you already encountered this situation, you can delete that particular time series with the delete API of Prometheus.