Skip to the content.

Most weeks I send one short letter about the alerting layer: a failure mode I hit, why the stack stayed green through it, and the change that would have caught it. No roundups, no tool news, no "10 best Grafana plugins".

What you get

Recent subjects: alert payloads that report active: [] while the metrics pipeline is down, per-path SLOs that site averages compress away, no_data handling nobody configured.

Read the archive first if you'd rather see the shape of it before handing over an address.

If you’d rather I just look at yours

The letters come out of the same work as the free alert audit: you send me your stack, I come back with the handful of alerts that actually matter and one thing you almost certainly have wrong.

Get your free alert audit