High memory pressure with kswapd CPU spikes

📖 ~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

System latency increases and application throughput drops while `kswapd` uses significant CPU.

Environment & Reproduction

RHEL 8 host with memory-heavy workload; sustained pressure triggers swap reclaim storms.

Root Cause Analysis

Insufficient RAM for workload, memory leak, aggressive swappiness, or overcommitted container limits.

Quick Triage

Capture memory snapshot and identify top consumers before applying kernel tunables.

Step-by-Step Diagnosis

Use `free -m`, `vmstat 1`, `top`, and `systemd-cgtop`; correlate with `journalctl` OOM and reclaim messages.

Solution – Primary Fix

Reduce workload memory footprint, tune `vm.swappiness`, add RAM or adjust cgroup limits, then restart impacted services.

Solution – Alternative Approaches

Enable zram where suitable, shard workloads horizontally, or move memory-intensive jobs off the node.

Verification & Acceptance Criteria

CPU usage by `kswapd` normalizes and swap-in/swap-out rates stabilize during peak periods.

Rollback Plan

Revert kernel tunables and service limits if tuning introduces unexpected performance regressions.

Prevention & Hardening

Implement capacity planning alerts and memory leak detection in pre-production testing.

Related Errors & Cross-Refs

OOM killer events, major page fault spikes, and service timeout cascades under memory pressure.

Related tutorial: View the step-by-step tutorial for rhel-8.

View all rhel-8 tutorials on the Tutorials Hub →

Browse all common problems & solutions on the Tutorials Hub.

References & Further Reading

RHEL performance tuning docs and Linux VM memory management references.