59 / 81 · Module 16 — Production Engineering and Observability · filtered: interview← prev⊞ allnext →
16.50Health dashboard hierarchy
A useful dashboard might show:
BUSINESS
customer sync success %
sync lag
API
traffic
p95 latency
5xx
DEPENDENCIES
Salesforce latency/errors
Billing latency/errors
RUNTIME
CPU
memory
replicas
ASYNC
queue depth
oldest message
DLQ
This goes from:
business impact
↓
technical cause
which is usually how incidents should be investigated.