-
Book Overview & Buying
-
Table Of Contents
The Platform Engineer's Handbook
By :
Observability is no longer an afterthought, and is perhaps one of the backbones of a great engineering platform.
From a CI/CD point of view, the CI pipeline should fail if the services don't properly expose the endpoints, emit the right kind of logs or generate the appropriate trace spans. This will ensure that no uninstrumented code ever reaches production in an unintended manner. Even if the developers try to bypass this, the observability platform would then flag this to the security teams to remediate it. Staging environments should also demonstrate telemetry generated before promotion to production.
Pipeline performance itself should be observable. These are the simple but standard activities such as build duration, test execution time, deployment frequency and so on, that can help the platform teams optimize any CI/CD bottlenecks. Measuring mean time to build, test flakiness, and the build...