Queue Growing Continuously

Confirm the symptom

Record the campaign, customer, delivery server, timestamp, message identifiers, and the first observed error. Avoid changing multiple components before preserving evidence.

Immediate triage

1. Pause affected traffic if duplication or abuse is possible.
2. Check cron freshness and application logs.
3. Check SMTP, bounce, web-server, PHP, and database health.
4. Confirm DNS and credentials from the current host.

Useful checks

crontab -l
systemctl status php-fpm nginx mariadb
journalctl -u php-fpm --since "30 minutes ago"
mysqladmin processlist

Decision path

If the application cannot generate work, focus on PHP, database, permissions, and cron. If work is generated but not delivered, focus on delivery-server connectivity, authentication, routing, throttling, and provider responses.

Recovery and prevention

Recover with the smallest reversible change. Then add monitoring for the signal that would have detected the issue earlier.