Emails Stuck in Queue

When campaigns show Sending forever or the queue table grows without delivery, the problem is almost always workers not running, a misconfigured queue driver, or a failed job blocking retries. This runbook walks through diagnosis in order.

Quick checks (2 minutes)

1. Cron alive? — Is * * * * * php artisan schedule:run configured? See cron setup.
2. Workers running? — On VPS: sudo supervisorctl status. Shared hosting: cron for queue:work --stop-when-empty.
3. Sending server active? — Admin → Sending → Sending Servers — status green, credentials valid.
4. Failed jobs — php artisan queue:failed or Admin → Monitor (if exposed).

Step 1 — Confirm queue connection

In .env:

QUEUE_CONNECTION=database
# or redis on VPS

After any change:

php artisan config:clear
php artisan queue:restart

Redis required? See Redis for queue processing.

Step 2 — Process one job manually

cd /path/to/senddera
php artisan queue:work --once -v

Watch output. Common errors:

• SMTP connection refused → SMTP troubleshooting
• Maximum execution time → raise max_execution_time in PHP-FPM/cli
• Memory exhausted → raise memory_limit to 512M for workers

Step 3 — Inspect the jobs table

For database driver, table jobs holds pending work. Large reserved_at timestamps stuck in the past can indicate zombie workers. After stopping all workers:

php artisan queue:restart
# Only if you understand the impact — releases reserved jobs:
php artisan queue:flush  # use with care; check docs for your version

Prefer retrying failed jobs: php artisan queue:retry all

Step 4 — Supervisor configuration

Workers must run as the same user that owns storage/. Logs: storage/logs/worker.log and laravel.log. Example config in cron & workers guide.

Step 5 — Campaign-specific issues

• Throttling — sending server rate limit pauses queue; check server dashboard
• Empty segment — campaign completes with zero sends (not a queue bug)
• Paused list / banned subscribers — reduces visible sends

Step 6 — Read logs

Full walkthrough: Reading application logs. Search for ProcessCampaign, SendMessage, or SMTP exceptions.

When to scale

Sustained backlog at high volume: add workers (numprocs in Supervisor), switch to Redis, tune MySQL — Scaling for 100K+ emails/day.