Server and machine metrics are retrieved from the cn=monitor backend of the monitored server.

The Stats Collector plugin is responsible for collecting performance data from the cn=monitor backend. Data includes server responses, replication activity, local database activity, and host system metrics.

The Stats Collector configuration defines:

  • Data sample and collection intervals
  • The granularity of data collected, such as basic, extended, or verbose
  • The types of host system data collected, such as CPU, disk, and network
  • The type of data aggregation that occurs for LDAP application statistics

For more information, see Tuning data collection.

Data Collection

A diagram illustrating the data collection process between the monitored server and the metrics engine.

Processing steps

  1. Data samples are taken and stored in time-contiguous blocks on the disk of the monitored server.
  2. The PingDataMetrics server collection service polls for new sample blocks.
  3. The new sample blocks are queued to disk on the PingDataMetrics server.
  4. The PingDataMetrics server import service loads new blocks into the database.