How to Update Prometheus Swamper Configurations in YugabyteDB Anywhere

This article explains how to update Prometheus Swamper configurations in YugabyteDB Anywhere when specific metrics are missing from Prometheus monitoring. These steps are typically required when remediating a Universe after a failed task.

Issue overview

In some cases, Prometheus may stop collecting specific metrics due to outdated Swamper configurations. This can happen after a failed YBA task on this Universe, like “Add node”, “YBDB software upgrade”, “GFlags upgrade” which fails to update the swamper target configuration for the changed state of the Universe. This can affect metric visibility in the YugabyteDB Anywhere dashboard.

Solution:

• Update Swamper Configurations
  • Connect to the YugabyteDB Anywhere Database:
  • Run the Following SQL Query by replacing universe_name.

UPDATE universe SET swamper_config_written = false WHERE name = 'universe_name';

• Verify the Update
  • Wait a few minutes for Yugaware to regenerate the Swamper configuration files.
  • Run the following query by replacing universe_name to confirm the update:

SELECT swamper_config_written FROM universe WHERE name = 'universe_name';

• If the query returns true, the Swamper configuration has been successfully updated.

How It Works:

• Setting swamper_config_written to false triggers Yugaware to regenerate the Swamper configuration files.
• Yugaware automatically sets swamper_config_written back to true after completing the update process.

Next Steps:

• Verify Metrics in Prometheus:
  • Ensure the missing metrics are now visible in Prometheus.
• Check the YugabyteDB Anywhere Dashboard:
  • Confirm that the metrics are displayed correctly in the YugabyteDB Anywhere UI.

For search purposes:

• Prometheus Swamper configurations
• Metrics missing from Prometheus monitoring
• Metrics missing in YugabyteDB Anywhere dashboard
• Update Swamper configurations in YugabyteDB Anywhere
• Failed YBA task
• Nodes are unreachable