MustGather: Performance, hanging, or high CPU issues with HCL Commerce
Problem
If you are experiencing performance degradation, system hangs, or high CPU usage in HCL Commerce, use this MustGather guide to collect the necessary diagnostic data to help identify and resolve the issue.
Procedure
- Grafana Monitoring
-
HCL Commerce version 9.1 and later includes robust monitoring features based on Prometheus-style metrics. The Commerce pods expose internal endpoints that are scraped by Prometheus for metric collection.
Due to the widespread adoption of Prometheus, you can use it with Grafana or other third-party tools that support Prometheus metrics.
HCL Commerce also offers pre-configured Grafana dashboards that highlight key metrics, help identify issues, and support performance troubleshooting.
For more information, refer to: Understanding the Grafana graphs
- Thread Monitoring
-
Javacores and thread stack dumps are essential for debugging performance issues. Analyzing and comparing thread stacks across different threads and timeframes can help isolate the root cause of slowdowns.
To configure thread monitoring tools, refer to: Thread monitoring tools- Collect all the logs and zip them together.
- Collect all the javacores and zip them together.
- JDBC Monitor
-
The JDBC Monitor tool captures and examines SQL statements executed by the application, making it valuable for performance and troubleshooting analysis.
For installation and configuration instructions, see: JDBC_MONITOR_ENABLED parameter
- Collect all the logs and zip them together.