View Transcript
Episode Description
A duty engineer showed me the page that woke her at 3 a.m. It said: Error rate elevated. Three words and a graph. ThenShe lay there guessing which service, how bad, the big one or the noisy one.
That gap is the problem. A context-free alert is an open loop, and open loops fill with fear. This episode is about treating an alert as what it actually is: a message to a tired human at theworst hour of their day. What broke, how bad, what to do
The book is out. Metrics & Mayhem: A CTO's Guide to Observability That Actually Works. Kindle is live now
Get your free chapter here:https://www.masteringobservability.com/metrics-and-mayhem/free-chapter
Newsletter:https://masteringobservability.com