Performance and Fault Management

StarlingX provides a number of tools to allow system administrators to manage performance and troubleshoot system issues.

Performance Management

StarlingX utilizes collectd ( https://collectd.org/ ) to capture the following platform statistics and to generate threshold events based on these statistics:

  • CPU Usage of Platform Cores of StarlingX hosts

  • Platform Memory Usage of StarlingX hosts

  • Platform File Systems Usage

  • Platform Interface Usage

  • PTP Clock Skew Monitor

Any collectd threshold events trigger StarlingX fault management Set/Clear Customer Alarms.

Fault Management

For an overview of StarlingX fault management, see Fault Management Overview.

For a listing of all StarlingX fault management resources, including alarm log messages, see ‘Alarm messages’ and ‘Log messages’ in the Fault Management Contents page.