Environment-Specific Thresholds
This troubleshooting guide explains how to use environment thresholds for production monitoring and alert tuning with the Nodinite JMX Monitoring Agent.
Understanding Threshold Settings
The JMX Configuration provides four threshold controls per monitored metric:
| Setting | Description |
|---|---|
| Enabled | Activates monitoring for this metric |
| Warning Threshold % | Percentage at which a Warning alert fires |
| Error Threshold % | Percentage at which an Error alert fires |
| Duration (seconds) | How long the metric must exceed the threshold before an alert fires |
The Duration setting is critical for reducing alert noise caused by temporary spikes during garbage collection, batch jobs, or deployments.
Recommended Thresholds by Environment
CPU Thresholds
| Environment | Warning % | Error % | Duration (s) | Rationale |
|---|---|---|---|---|
| Development | 80% | 95% | 30 | Allow high CPU for builds and debugging |
| Test / QA | 75% | 90% | 60 | Catch regressions without excessive noise |
| Staging | 70% | 85% | 120 | Mirror production behavior |
| Production | 65% | 80% | 120 | Early warning for sustained high CPU |
Heap Size Thresholds
| Environment | Warning % | Error % | Duration (s) | Rationale |
|---|---|---|---|---|
| Development | 80% | 92% | 60 | Allow exploratory testing |
| Test / QA | 75% | 88% | 120 | Catch memory leaks early |
| Staging | 72% | 85% | 180 | Validate production behavior |
| Production | 70% | 82% | 180 | Maximize early detection |
Tip
For batch-processing JVMs that intentionally use high heap, increase the Duration to 300–600 seconds to avoid alerting during expected batch execution peaks.
Configuration Steps
Configure thresholds for each Application Host in the JMX Monitoring Agent Configuration:
- Open the Nodinite Administration panel.
- Navigate to the Monitoring Agent Configuration.
- Click Configuration to open Remote Configuration.
- Select the JMX tab and locate the Application Host to configure.
- Open the Heap Size tab for the host.
- Set Enabled, Warning Threshold %, Error Threshold %, and Duration.
- Open the CPU tab for the host.
- Repeat the threshold settings.
- Click Save to apply the monitoring configuration.
Environment Identification
Use the Environment setting in the Settings tab of Remote Configuration to label each monitored JVM with its environment, for example: DEV, TEST, QA, or PROD. This label appears in Nodinite alerts and helps operators quickly identify which environment triggered an alert.
Note
The Environment name is cosmetic. It labels alerts but does not automatically apply different threshold profiles. Configure thresholds explicitly for each Application Host.
Next Step
Polling Frequency
Multi-JVM Monitoring
Related Topics
Heap Used vs Committed
Correlate Heap Spikes
JMX Troubleshooting Overview