How to monitor Laravel in production (without Datadog)

You do not need a consumption-priced observability platform to keep Laravel online. You need signals that predict revenue loss and a workflow that fixes known failures.

Start with the Linux layer

1. PHP-FPM — pool saturation, slow requests, worker crashes
2. Queues — Horizon or supervisor restart loops, backlog depth
3. Deploy correlation — did 502s start after a release?
4. Disk & memory — logs and releases fill partitions quietly

What to skip early

Exhaustive distributed tracing across microservices you do not run yet. Start with host + process health.

Where Reflex fits

reflexd ships metrics and Brain playbooks for the failures Laravel teams see weekly. Start a trial or compare vs Datadog.