Node memory high: JVM heap usage approaching limit

Memory leaks in custom business rules, client scripts, or scheduled jobs that accumulate objects without proper cleanup or use inefficient coding patterns like unclosed GlideRecord queries. (high likelihood)

Large cached objects in the system cache or user session cache, often caused by caching oversized datasets or images in memory rather than using proper file storage. (high likelihood)

Unexpectedly high concurrent user load exceeding the instance's provisioned capacity, particularly during peak usage periods or after go-live events. (medium likelihood)

Long-running batch operations or data imports that load large datasets into memory simultaneously without proper pagination or streaming. (medium likelihood)

JVM heap size configuration insufficient for the instance workload, or garbage collection settings that are not optimized for the current usage patterns. (low likelihood)

Step 1: Navigate to System Diagnostics > Stats > Memory to view current heap usage, garbage collection statistics, and identify which memory pools are consuming the most space.

Step 2: Check System Diagnostics > Log > Application Log for recent errors or warnings that might indicate memory-intensive operations or script failures occurring around the time of the memory spike.

Step 3: Review active sessions and concurrent users in System Diagnostics > Sessions > Active Sessions to determine if user load is contributing to memory pressure.

Step 4: Examine running scheduled jobs in System Definition > Scheduled Jobs and cancel or reschedule any memory-intensive batch operations that may be contributing to heap usage.

Step 5: Use System Diagnostics > Cache Statistics to identify oversized cache entries and clear problematic caches using the flush cache options or by restarting affected nodes.

Step 6: Review custom scripts in recently modified business rules, script includes, and scheduled scripts for memory leaks such as unclosed GlideRecord objects or infinite loops.

Step 7: If the issue persists, contact ServiceNow support to request heap dump analysis or consider scaling up instance resources if current capacity is insufficient for workload demands.