⏰ Linux at Command

at> /home/user/scripts/backup.sh
at> echo "Backup started at $(date)" | mail -s "Backup Notice" admin@example.com
at> <EOT>
job 2 at Thu Nov  2 22:00:00 2025

at now + 5 minutes << END
cd /var/log
gzip *.log
mv *.gz /backup/logs/
echo "Log compression complete" | logger
END

📚 Additional Information

Time Specification Examples

Common at Command Options

Related Commands

Access Control

The at command uses two files to control access:

• /etc/at.allow - If exists, only users listed here can use at
• /etc/at.deny - If at.allow doesn't exist, users listed here cannot use at
• If neither file exists, only root can use at (most restrictive)
• If only at.deny exists and is empty, all users can use at (most permissive)

Real-World Use Cases

• Scheduled Reboots: Restart systems during maintenance windows
• Delayed Processing: Start resource-intensive tasks during off-hours
• Reminders: Send yourself notifications at specific times
• Temporary Tasks: One-time backups or data processing
• Testing: Schedule test scripts to run at specific times
• Auto-cleanup: Delete temporary files after a delay

Best Practices

• Use absolute paths: at jobs may have different PATH variables
• Test first: Schedule test jobs in 1-2 minutes to verify they work
• Check atq regularly: Monitor pending jobs to avoid surprises
• Use batch for heavy tasks: Prevents system overload
• Mail notifications: at will email output by default (if mail is configured)
• Document your jobs: Add comments in scripts for future reference

Troubleshooting

Environment Variables

at jobs inherit the environment from when they were scheduled, including:

• Current directory (cd to specific directory if needed)
• Environment variables (SHELL, PATH, HOME, etc.)
• User permissions (jobs run as the user who scheduled them)

at vs. cron: When to Use Which?

Advanced Examples

# Schedule script with output logging
echo "/home/user/backup.sh >> /var/log/backup.log 2>&1" | at 2:00 AM

# Chain multiple at jobs
echo "echo '/second/script.sh' | at now + 2 hours" | at now + 1 hour

# Conditional execution
echo "[ -f /tmp/ready ] && /usr/local/bin/process.sh" | at now + 30 minutes

# Schedule with multiple queues for prioritization
echo "critical_task.sh" | at -q a now + 10 minutes    # High priority
echo "normal_task.sh" | at -q b now + 10 minutes      # Normal priority
echo "low_priority.sh" | at -q c now + 10 minutes     # Low priority

# Schedule system maintenance
cat <<EOF | at 3:00 AM Sunday
apt update
apt upgrade -y
apt autoremove -y
systemctl reboot
EOF

# Monitor and alert if disk space low
echo "df -h | awk '\$5 > 90 {print}' | mail -s 'Disk Alert' admin@example.com" | at now + 1 hour

# Delayed cleanup after deployment
echo "rm -rf /tmp/deployment_files/*" | at now + 24 hours

# Schedule database backup at specific time
echo "mysqldump -u root -p'password' mydb > /backup/mydb_\$(date +%Y%m%d).sql" | at 11:30 PM

Security Considerations

• File permissions: Ensure scripts called by at have appropriate permissions
• Path safety: Use absolute paths to avoid PATH hijacking
• Credential management: Don't hardcode passwords in at jobs; use credential files with proper permissions
• Review jobs: Regularly check atq to ensure no unauthorized jobs are scheduled
• Audit logging: Monitor /var/log/syslog for at job execution

Practical Sysadmin Scenarios

# Delayed service restart after configuration change
echo "systemctl restart nginx" | at now + 5 minutes

# Schedule certificate renewal for 3 AM
echo "certbot renew --quiet" | at 3:00 AM tomorrow

# Test website response time
echo "curl -o /dev/null -s -w '%{time_total}\n' https://example.com" | at now + 15 minutes

# Schedule firewall rule removal (temporary access)
echo "iptables -D INPUT -s 192.168.1.100 -j ACCEPT" | at now + 2 hours

# Automated log rotation
echo "gzip /var/log/application.log && mv /var/log/application.log.gz /backup/logs/" | at midnight

# System health check before maintenance
echo "/usr/local/bin/pre-maintenance-check.sh" | at 11:00 PM Friday

Integration with Other Tools

# Schedule with notify-send (desktop notification)
echo "notify-send 'Backup Complete' 'Your scheduled backup has finished'" | at now + 3 hours

# Integration with logger
echo "logger -t at-job 'Scheduled maintenance complete'" | at 4:00 AM

# Send Slack notification
echo "curl -X POST -H 'Content-type: application/json' --data '{\"text\":\"Scheduled task complete\"}' SLACK_WEBHOOK_URL" | at now + 1 hour

# Update monitoring system
echo "curl -X POST https://monitoring.example.com/checkin/job-123" | at now + 30 minutes

Fun Facts

• The 'at' command dates back to Unix Version 7 (1979)
• "teatime" as a time specification is a nod to the British tradition of afternoon tea
• The 'batch' command is essentially 'at' with load-aware scheduling
• at jobs survive reboots - they're stored in /var/spool/cron/atjobs/
• The output of at jobs is emailed to you automatically (if mail is configured)