Insights Module

The insights module collects and exposes system information to the Insights Coredata analysis framework. It is intended to replace explicit interrogation ofCeph CLIs and daemon admin sockets, reducing the API surface that Insightsdepends on. The insights reports contains the following:

  • Health reports. In addition to reporting the current health of thecluster, the insights module reports a summary of the last 24 hours of healthchecks. This feature is important for catching cluster health issues that aretransient and may not be present at the moment the report is generated. Healthchecks are deduplicated to avoid unbounded data growth.

  • Crash reports. A summary of any daemon crashes in the past 24 hours isincluded in the insights report. Crashes are reported as the number of crashesper daemon type (e.g. ceph-osd) within the time window. Full details of acrash may be obtained using the crash module.

  • Software version, storage utilization, cluster maps, placement group summary,monitor status, cluster configuration, and OSD metadata.

Enabling

The insights module is enabled with:

  1. ceph mgr module enable insights

Commands

  1. ceph insights

Generate the full report.

  1. ceph insights prune-health <hours>

Remove historical health data older than <hours>. Passing 0 for <hours> willclear all health data.

This command is useful for cleaning the health history before automated nightlyreports are generated, which may contain spurious health checks accumulatedwhile performing system maintenance, or other health checks that have beenresolved. There is no need to prune health data to reclaim storage space;garbage collection is performed regularly to remove old health data frompersistent storage.