How to get alerted when a cron job fails

Method 1

Built-in cron email (MAILTO)

Cron has a built-in notification system that almost nobody uses correctly. Set MAILTO at the top of your crontab and cron will email you all output from every job.

# At the top of your crontab (crontab -e):
[email protected]

# This job's output (stdout + stderr) gets emailed:
0 2 * * * /usr/local/bin/backup.sh

The trick: if your job produces no output on success, you only get emails when something goes wrong. Design your scripts this way deliberately:

#!/bin/bash
# backup.sh — silent on success, loud on failure
set -euo pipefail

mysqldump mydb > /backup/mydb.sql 2>&1

# Only output on failure:
if [ ! -s /backup/mydb.sql ]; then
  echo "ERROR: backup file is empty" >&2
  exit 1
fi

Good for: catching errors inside your script. Bad for: knowing whether the job ran at all.

Prerequisite: your server needs a working MTA (postfix, sendmail, msmtp). On many cloud VMs, mail isn't configured by default. Test with: echo "test" | mail -s "cron test" [email protected]

Method 2

Wrapper script with Slack/email on failure

Wrap your job in a script that checks the exit code and sends a notification on failure. More control than MAILTO, works with any notification channel.

#!/bin/bash
# cron-alert.sh — run a command, alert on failure
# Usage: cron-alert.sh "backup" /usr/local/bin/backup.sh

JOB_NAME="$1"
shift

OUTPUT=$("$@" 2>&1)
EXIT_CODE=$?

if [ $EXIT_CODE -ne 0 ]; then
  # Slack webhook:
  curl -s -X POST "$SLACK_WEBHOOK" \
    -H 'Content-type: application/json' \
    -d "{\"text\":\"Cron failed: ${JOB_NAME} (exit ${EXIT_CODE})\n\`\`\`${OUTPUT}\`\`\`\"}"

  # Or email:
  echo "$OUTPUT" | mail -s "Cron failed: ${JOB_NAME}" [email protected]
fi

exit $EXIT_CODE

# In your crontab:
0 2 * * * /usr/local/bin/cron-alert.sh "nightly backup" /usr/local/bin/backup.sh

Better than MAILTO for routing alerts. Same blind spot: can't detect jobs that never started.

Method 3

Log monitoring (syslog + grep)

Cron logs every job it runs to syslog. You can monitor these logs for missing entries or error patterns.

# Check if cron ran your job in the last hour:
grep "CRON.*backup.sh" /var/log/syslog | tail -5

# With journald:
journalctl -u cron --since "1 hour ago" | grep backup

For automated monitoring, you'd set up a separate cron job that checks whether the main job's log entries exist:

# Check every hour that the backup ran today:
0 * * * * grep -q "$(date +\%Y-\%m-\%d).*backup" /var/log/syslog || \
  curl -s -X POST "$SLACK_WEBHOOK" -d '{"text":"No backup log entry today"}'

Clever but fragile. You're monitoring infrastructure with more infrastructure. If syslog breaks, you lose both the job and the alert.

Method 4

Heartbeat monitoring

Different approach: instead of detecting failure, detect the absence of success. Your job pings an external endpoint after each successful run. If the ping stops arriving, something broke — and the alert comes from outside your infrastructure.

# Your crontab becomes:
0 2 * * * /usr/local/bin/backup.sh && curl -fsS --max-time 10 https://ping.trebben.dk/p/YOUR_TOKEN

The monitoring service tracks when it last received a ping. If the interval between pings exceeds your configured threshold, you get an alert via email, Slack, or webhook.

The only method that catches jobs which never start. The alert comes from outside your server, so it works even when everything on the server is broken.