Three minutes later, your phone buzzes. Then again. Then the Slack channel erupts. “500 error on profile uploads.” “Jobs stuck in failed table.” “Why is S3 returning AccessDenied?” Welcome to Laravel after dark. Here’s your survival guide for this exact moment. Don’t open the controller. Don’t blame Redis.
It’s 9:58 PM. You just pushed what you thought was a tiny hotfix: “Update user avatar validation.”
php artisan optimize:clear php artisan config:cache # only if in production php artisan queue:restart 90% of “broken after deploy” is cached config pointing to the wrong env. Don’t tail -200 and panic. Do this: the laravel survival guide
php artisan config:clear php artisan config:cache php artisan queue:restart Then verify: php artisan tinker --execute="dd(config('queue.default'));" Check your Eloquent::saving() event listeners. Check your observers. One of them is returning false — which cancels the save silently.
After any .env change, run:
tail -50 storage/logs/laravel.log | grep -E "\[[0-9]4-[0-9]2-[0-9]2.*\].*ERROR" Or use Log::error() yourself next time with a unique marker like 🚨 DEPLOY . You set QUEUE_CONNECTION=database in .env . But config/queue.php says redis because someone committed it.
Use php artisan octane:start or sail up -d with APP_ENV=local but CACHE_DRIVER=redis and QUEUE_CONNECTION=database . Simulate production locally, or the Laravel gods will simulate failure for you at 2 AM. Three minutes later, your phone buzzes
Add this temporarily to AppServiceProvider@boot :