Integration / Apache Pulsar Interview questions
Explain the lifecycle of an alert from health rule violation to notification?
An AppDynamics alert isn't a single event; it's a short pipeline where a metric condition is evaluated, a status change is recorded, and only then — if configured — does an external action actually fire.
flowchart TD
A[Live metric arrives] --> B{Health Rule condition evaluated}
B -- Within normal range --> C[Status stays Normal]
B -- Violates threshold --> D[Status changes to Warning or Critical]
D --> E{Is a Policy attached to this Health Rule?}
E -- No --> F[Status visible on dashboard/flow map only]
E -- Yes --> G[Policy Action(s) execute]
G --> H[Email / Slack / PagerDuty / ServiceNow / Webhook]
D --> I[Status returns to Normal when condition clears]
The key nuance is that steps D and E are genuinely separate: a health rule violation always changes the visible health status on its own, but nothing gets sent anywhere unless a policy explicitly wires that violation to an action. Teams that only configure health rules and forget the policy step often get confused when a dashboard clearly shows red but nobody was paged — the color-coded status and the outbound notification are two distinct configuration layers, not one automatic package deal.
What must be attached to a health rule for a notification to actually be sent?
What is a common source of confusion in this pipeline?
More Related questions...
