All issues
Ship Notes · SN #15·5 Apr
Observability before the incident
You can't debug what you can't see. Two notes on setting up the right visibility before you actually need it.
1. Logs, metrics, traces
Each answers a different question — what happened, how much, and where. You don't need all three on day one, but you do need to know which one you're missing the moment something breaks.
2. Alert on symptoms, not causes
Page people when users are hurting — errors, latency, failed checkouts — not on every CPU blip. Good alerts are rare, actionable, and trusted. Noisy ones get muted, and then the real one gets missed.
Previous issue
SN #14 — Designing the data model