ConfigClarity › Glossary › Cron Job Collision

Cron Job Collision

A cron job collision occurs when multiple scheduled cron jobs run simultaneously and compete for the same resources — CPU, disk I/O, database connections, or lock files — causing performance degradation or data corruption.
CronSchedulingflockLinuxDevOps

Cron job collisions are silent failures. The individual jobs may complete without errors, but the concurrent execution degrades performance, corrupts data, or causes cascading failures that appear unrelated. A backup job and a database dump running at the same time is a classic example — both appear to succeed but the backup captures a mid-write database state.

The collision problem compounds on low-spec servers where scheduled tasks like log rotation, backup, cache clearing, and report generation are all set to run at midnight or hourly boundaries (0 * * * *).

Types of Collision

Resource collision — two jobs competing for CPU/disk causes both to run slower. Common with backup jobs on I/O-limited VPS instances. Lock collision — two jobs trying to acquire the same file lock or database advisory lock. One waits, one times out. Data collision — two jobs reading and writing the same dataset. Most dangerous — can cause partial writes and corrupted state.

flock as the Standard Fix

flock is the standard Linux tool for preventing concurrent cron job execution. Wrapping a cron command with flock -n /tmp/job.lock ensures only one instance runs at a time. If the lock cannot be acquired (a previous instance is still running), the new invocation exits immediately without error.

Related Tools

Cron Visualiser →

Fix Guides

Overlapping cron jobs fixCron flock safety fixServer load spike fix

Frequently Asked Questions

How do I detect overlapping cron jobs?
Paste your crontab -l output into ConfigClarity's Cron Visualiser. It renders a 24-hour timeline showing every job's execution window and flags overlapping windows in red with a collision report.
What is flock safety for cron jobs?
flock safety means wrapping your cron command with flock -n /tmp/jobname.lock command. If the previous run is still executing when the next scheduled run starts, the new invocation exits immediately instead of running concurrently.
Why do all my cron jobs run at midnight?
It's the most common default. When a task is set up with no specific time in mind, developers often choose 0 0 * * *. When multiple tasks pile up at the same minute, the server CPU and I/O spike simultaneously. Spread jobs across off-peak minutes — 0:05, 0:15, 0:30, 0:45.