FreeBSD 14 – Disk I/O Latency Spikes from Queue Saturation – Throughput Recovery

📖 ~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

Read/write latency surges and applications time out during peak I/O windows.

Environment & Reproduction

Common on mixed workloads where backup and transactional I/O overlap.

iostat -x 1 10
zpool iostat -v 1 10

Root Cause Analysis

Queue depth saturation and random I/O contention overwhelm storage throughput.

Quick Triage

Identify hottest devices and defer non-critical heavy write jobs.

gstat -a
ps -axo pid,command | grep -E 'backup|rsync|zfs send'

Step-by-Step Diagnosis

Correlate latency with process-level I/O and storage error counters.

procstat -r 
dmesg | egrep -i 'timeout|reset|I/O'

Solution – Primary Fix

Reschedule batch jobs and tune dataset/workload settings to reduce contention.

renice +15 -p 
zfs set recordsize=128K zroot/data
service  restart

Solution – Alternative Approaches

Move heavy sequential jobs to separate pool or faster storage tier.

Verification & Acceptance Criteria

Sustained latency drops and SLA response times return to target.

iostat -x 1 10
/usr/bin/time -p 

Rollback Plan

Undo tuning values if workload regressions appear in other applications.

Prevention & Hardening

Establish I/O budgets and isolate backup windows from latency-sensitive traffic.

Related Errors & Cross-Refs

device timeout, high await, transaction timeout due to storage delay.

Related tutorial: View the step-by-step tutorial for freebsd-14.

View all freebsd-14 tutorials on the Tutorials Hub →

Browse all common problems & solutions on the Tutorials Hub.

References & Further Reading

FreeBSD and OpenZFS performance analysis resources.