Linux Cron Manager

Linux Cron Manager

Overview

Generates production-ready cron entries and /etc/cron.d/ files with proper environment setup, flock-based locking to prevent overlapping runs, structured logging, and failure alerting. Includes a full cron syntax reference and a conversion path to systemd timers for complex scheduling needs.

Language: Respond in the user's language. Cron files and scripts in English.

When to Use

• User wants to schedule any recurring task
• User wants a cron job that logs and alerts on failure
• User wants to prevent a cron job from running twice at once
• User wants to convert a cron job to a systemd timer
• User asks "my cron job isn't running" — debugging
• User wants to understand cron syntax or cron environment behavior

When NOT to Use

• User needs complex dependencies between jobs, socket activation, or resource limits → linux-systemd-manager
• User wants to write the script that cron will run → linux-shell-scriptor

Step 1 — Understand the Request

Cron Syntax Reference

⚠️ Timezone: Cron uses the system timezone (from /etc/timezone). To override per-job, add TZ=America/New_York at the top of the /etc/cron.d/ file — this affects all entries in that file only.

┌──────────── minute       (0–59)
│ ┌────────── hour         (0–23)
│ │ ┌──────── day of month (1–31)
│ │ │ ┌────── month        (1–12 or JAN–DEC)
│ │ │ │ ┌──── day of week  (0–7, 0 and 7 = Sunday, or SUN–SAT)
│ │ │ │ │
* * * * *  user  command

Common expressions:

Validate any expression at: https://crontab.guru

Production Cron Entry Template

Never write a bare cron entry. Always wrap with locking, logging, and error handling.

/etc/cron.d/ file format (preferred — survives user crontab edits)

/etc/cron.d/myapp-task:

# myapp-task — describe what this does
# Runs: daily at 2 AM
# Logs: /var/log/cron/myapp-task.log
# Owner: appuser

SHELL=/bin/bash
PATH=/usr/local/sbin:/usr/local/bin:/sbin:/bin:/usr/sbin:/usr/bin
MAILTO=""
HOME=/home/appuser   # cron doesn't source profile — set HOME explicitly to avoid "~" surprises

0 2 * * *  appuser  /usr/bin/flock -n /var/lock/myapp-task.lock /opt/myapp/tasks/run-task.sh

Wrapper script (/opt/myapp/tasks/run-task.sh):

Always generate the wrapper script alongside the cron entry:

#!/usr/bin/env bash
set -euo pipefail

# ── Config ──────────────────────────────────────────────────
SCRIPT_NAME="myapp-task"
LOG_DIR="/var/log/cron"
LOG_FILE="${LOG_DIR}/${SCRIPT_NAME}.log"
LOCK_FILE="/var/lock/${SCRIPT_NAME}.lock"
MAX_LOG_SIZE_MB=50

# ── Logging ──────────────────────────────────────────────────
mkdir -p "$LOG_DIR"
log()   { echo "[$(date '+%Y-%m-%d %H:%M:%S')] [INFO]  $*" | tee -a "$LOG_FILE"; }
warn()  { echo "[$(date '+%Y-%m-%d %H:%M:%S')] [WARN]  $*" | tee -a "$LOG_FILE" >&2; }
error() { echo "[$(date '+%Y-%m-%d %H:%M:%S')] [ERROR] $*" | tee -a "$LOG_FILE" >&2; }

# ── Log rotation ─────────────────────────────────────────────
if [[ -f "$LOG_FILE" ]]; then
  LOG_SIZE_MB=$(du -m "$LOG_FILE" | cut -f1)
  if [[ "$LOG_SIZE_MB" -ge "$MAX_LOG_SIZE_MB" ]]; then
    ARCHIVED_LOG="${LOG_FILE}.$(date +%Y%m%d_%H%M%S)"
    mv "$LOG_FILE" "$ARCHIVED_LOG"
    gzip "$ARCHIVED_LOG"
  fi
fi

# ── Lock — prevent concurrent runs ───────────────────────────
exec 9>"$LOCK_FILE"
if ! flock -n 9; then
  warn "Another instance is already running. Exiting."
  exit 0
fi

# ── Trap — cleanup on exit ───────────────────────────────────
cleanup() {
  local exit_code=$?
  if [[ $exit_code -ne 0 ]]; then
    error "Task failed with exit code $exit_code"
    # Optional: send alert
    # echo "Cron job $SCRIPT_NAME failed" | mail -s "CRON FAILURE: $SCRIPT_NAME" admin@example.com
  fi
  flock -u 9
}
trap cleanup EXIT

# ── Main ─────────────────────────────────────────────────────
log "=== Starting $SCRIPT_NAME ==="

# --- YOUR TASK CODE HERE ---
# Example:
# log "Running backup..."
# /opt/backup/backup.sh
# log "Backup complete"
# ---------------------------

log "=== $SCRIPT_NAME complete ==="

Email / Alert on Failure

Add to the cron file to capture output and email on non-zero exit:

# Option 1: Send all output to email (requires working MTA)
MAILTO="admin@example.com"
0 2 * * *  appuser  /opt/myapp/tasks/run-task.sh

# Option 2: Only email on failure using a wrapper
0 2 * * *  appuser  /opt/myapp/tasks/run-task.sh || echo "myapp-task failed" | mail -s "CRON FAILURE" admin@example.com

# Option 3: Log to file and send to a webhook (Slack/Discord) on failure
0 2 * * *  appuser  /opt/myapp/tasks/run-task.sh 2>&1 | tee -a /var/log/cron/myapp.log | \
  grep -i "error\|fail" | while read line; do \
  curl -s -X POST -H 'Content-type: application/json' \
  --data "{\"text\":\"Cron failure: $line\"}" "${SLACK_WEBHOOK_URL}"; done

Environment Rules

Cron runs in a minimal environment — this is the #1 cause of cron jobs that "work manually but not in cron."

Always include in /etc/cron.d/ files:

SHELL=/bin/bash
PATH=/usr/local/sbin:/usr/local/bin:/sbin:/bin:/usr/sbin:/usr/bin

Common environment pitfalls:

Debugging Checklist

When a cron job isn't running or is failing silently:

# 1. Verify cron daemon is running
systemctl status cron    # Debian/Ubuntu
systemctl status crond   # RHEL/Rocky

# 2. Check cron logs
grep CRON /var/log/syslog | tail -30          # Debian/Ubuntu
journalctl -u cron --since "1 hour ago"       # systemd systems
grep cron /var/log/messages | tail -30        # RHEL/Rocky

# 3. Check the cron entry is valid
crontab -l -u appuser                         # list user crontab
ls -la /etc/cron.d/                           # list system cron jobs

# 4. Test the command manually as the cron user
sudo -u appuser /path/to/script.sh

# 5. Test with same minimal environment as cron
sudo -u appuser env -i HOME=/home/appuser \
  PATH=/usr/local/sbin:/usr/local/bin:/sbin:/bin:/usr/sbin:/usr/bin \
  /path/to/script.sh

# 6. Verify file permissions
ls -la /path/to/script.sh        # must be executable
stat /etc/cron.d/myfile          # must be owned by root, mode 644

# 7. Check for syntax errors in /etc/cron.d/ files
# File must NOT have a .sh extension — cron ignores them
# File must NOT be world-writable

The #1 reason cron jobs "silently do nothing": output (stdout + stderr) is discarded unless captured. Always write logs inside the wrapper script using the log() function above, or add >> /var/log/cron/myapp.log 2>&1 to the cron entry.

Common reasons cron jobs silently do nothing:

Cron → Systemd Timer Conversion

When to switch to systemd timers instead of cron:

Anacron (/etc/anacrontab) — runs jobs that were missed while the system was off:

# period  delay  job-id         command
1         5      daily-backup   /opt/backup/backup.sh
7         10     weekly-report  /opt/reports/weekly.sh

Install: apt install anacron / dnf install cronie-anacron

Conversion example:

Cron entry:

0 2 * * *  appuser  /opt/myapp/backup.sh

Equivalent systemd timer — see /linux-systemd-manager for full output.

Output Format

Write files to ./cron/ in the current working directory:

cron/
  myapp-task          ← drop into /etc/cron.d/ (no extension!)
  run-task.sh         ← wrapper script with locking + logging

Then print ONLY:

✅ Cron job written to ./cron/

▶ Install:
  # ⚠️  /etc/cron.d files MUST be owned by root with mode 644.
  #     Cron silently ignores files with wrong ownership or permissions.
  sudo cp cron/myapp-task /etc/cron.d/myapp-task
  sudo chmod 644 /etc/cron.d/myapp-task
  sudo chown root:root /etc/cron.d/myapp-task
  sudo cp cron/run-task.sh /opt/myapp/tasks/run-task.sh
  sudo chmod 755 /opt/myapp/tasks/run-task.sh
  sudo chown appuser:appuser /opt/myapp/tasks/run-task.sh
  sudo systemctl reload cron          # tell cron to pick up the new /etc/cron.d/ file

▶ Verify cron picked it up:
  grep myapp-task /var/log/syslog
  # or: journalctl -u cron | grep myapp-task

▶ Test manually as cron user:
  sudo -u appuser /opt/myapp/tasks/run-task.sh

💡 For complex scheduling with dependencies or resource limits → /linux-systemd-manager