Debian 11: OOM killer terminates Java service under memory pressure

📖 ~1 min read

Table of contents

1. Symptom & Impact
2. Environment & Reproduction
3. Root Cause Analysis
4. Quick Triage
5. Step-by-Step Diagnosis
6. Solution – Primary Fix
7. Solution – Alternative Approaches
8. Verification & Acceptance Criteria
9. Rollback Plan
10. Prevention & Hardening
11. Related Errors & Cross-Refs
12. References & Further Reading

Symptom & Impact

Kernel OOM killer stops a Java process when host memory is exhausted.

Environment & Reproduction

Service restarts unexpectedly and kernel logs contain Out of memory kill entries.

Root Cause Analysis

Check `dmesg -T | grep -i ‘out of memory’`, JVM heap settings, and cgroup memory limits.

Quick Triage

Heap sizing, memory leaks, or insufficient host capacity cause sustained pressure beyond safe thresholds.

Step-by-Step Diagnosis

Right-size JVM heap, set sensible memory limits, and reduce non-essential co-located workload pressure.

Solution – Primary Fix

Confirm stable RSS trends and absence of new OOM events during representative traffic periods.

Solution – Alternative Approaches

Baseline memory per service and enforce limits with proactive alerting before saturation.

Verification & Acceptance Criteria

Revert recent heap or deployment changes if they increased memory footprint unexpectedly.

Rollback Plan

Track OOM signals and memory growth slope through host and application telemetry pipelines.

Prevention & Hardening

`dmesg -T | grep -i ‘out of memory’`; `free -h`; `ps -eo pid,cmd,rss –sort=-rss | head`

Related Errors & Cross-Refs

Include heap dump strategy, recent release changes, and node memory graphs when escalating.

Related tutorial: View the step-by-step tutorial for debian-11.

View all debian-11 tutorials on the Tutorials Hub →

Browse all common problems & solutions on the Tutorials Hub.

References & Further Reading

Swap can reduce abrupt kills in some workloads but may degrade latency-sensitive applications.