Homelab Academy / Modules
Orient, prior check, guided practice, required Feynman, reflect.
Module: Requests & Media Servers
Activity type: Lesson (Orient → Recall → Learn → Explain → Reflect)
Learning objective: Separate service reachability from application health state.
Bloom level: Understand / Apply
Last reviewed: 2025-03-08
Separate service reachability from application health state.
Build and verify a synthetic health exporter that converts ARR-style reachability, declared health, queue depth, and storage data into deterministic Prometheus text metrics without requiring access to a live Sonarr or Radarr instance.
Complete the module reading first:
Explain the monitor to someone who knows only that Sonarr and Radarr are applications.
The monitor asks two different questions. First, can it talk to the application? That answer becomes arr_monitor_up, using 1 for yes and 0 for no. Second, what condition is the application in? That becomes a state number: 2 healthy, 1 warning, or 0 critical. These questions must stay separate because an application can answer successfully while reporting a serious internal problem. Unreachable and declared critical conditions win over warnings. If neither critical condition exists, a queue over 25 or free space below 15 causes warning. Two applications multiplied by two metrics produces exactly four samples.
Can you explain why reachable and healthy are not synonyms?
Can you state why reachable declared critical produces values 1 and 0 rather than 0 and 0?
Can you explain why queue depth 25 and free space 15 are healthy boundaries?
Can you account for all four expected samples without looking at the code?
Answers save in this browser (localStorage). Completing all five fields is part of the mastery unlock.