There is a tricky problem that happens when queued job tries to get the URL (via URL helper or from config) – it doesn’t seem to “notice” the .env / config app.url setting.
The reason is that somehow the queue worker misses the current environment name (might only happen if it is managed system wide via upstart or similar service).
The solution would be to include the –env flag when running queue worker like that:
php /path/to/project/artisan queue:listen --env=production
Here is an example upstart queue-listener.conf that can be put in /etc/init
description "Laravel queue listener"
start on filesystem or runlevel [2345]
stop on runlevel [!2345]
setuid www-data
setgid www-data
respawn
respawn limit 5 2
script
# Start Listener
php /path/to/project/artisan queue:listen --tries=60 --env=production
end script
It can then be started / restarted like any other process using cli (file name is the name of the process):
sudo service queue-listener status sudo service queue-listener start sudo service queue-listener stop sudo service queue-listener restart