What to watch,and why it matters.
Short, specific, and honest about what each one will not catch. Written for someone running production without a monitoring team.
Guides
5 min · no prior setup
Get a real alert on your phone in five minutes
Signup to a verified alert on your own phone, using one URL and the free plan. The point is to prove the delivery path before you depend on it.
10 min · you have a Stripe account
Know when Stripe payments start failing
Monitor payment availability, account capability, billing, payouts, fraud, and webhook delivery without handing over a full-access Stripe key.
8 min · you have a Supabase project
What to watch on a Supabase project
Supabase is several services behind one URL. Which parts fail independently, what its own dashboard tells you, and what to do about the rest.
10 min · you run Lambdas
Monitoring AWS Lambda without alarm sprawl
What is worth alerting on for a Lambda, what is only worth collecting, and how to avoid maintaining alarms per function per metric.
7 min · conceptual
The failures that return HTTP 200
Most damaging outages never touch your status code. A field guide to the failure modes uptime monitoring structurally cannot see, and what does catch them.
9 min · you have an Atlas cluster
What to watch on MongoDB Atlas
Atlas rarely goes down. It degrades — slowly, on one node, in ways your application reports as its own fault. What to watch, and what Atlas already tells you.
Something missing?
These are written from the questions people actually ask. If yours is not here, say so — it is a short list and it moves.