CDCgov / prime-simplereport

SimpleReport is a fast, free, and easy way for COVID-19 testing facilities to report results to public health departments.
https://simplereport.gov
Creative Commons Zero v1.0 Universal
56 stars 56 forks source link

Application Insights query for "HTTP Server 2xx Failed Requests" alert should return requests in results #4063

Closed nathancrtr closed 1 year ago

nathancrtr commented 2 years ago

Description

Underlying the Training HTTP Server 2xx (requests where (failed requests * 100.00 / total requests) >= 1) PagerDuty alert is an Application Insights query that returns 1 if the failed request threshold has been breached:

Screen Shot 2022-07-25 at 1.34.08 PM.png

This is valuable as it triggers an important alert and lets engineers begin an investigation promptly. However, since the failing requests themselves are not accessible from the log results, engineers must separately search for the failures.

Acceptance criteria

alismx commented 1 year ago

This has been as completed as possible. A comment in the alert definition outlines how to get the results.

https://github.com/CDCgov/prime-simplereport/blob/41dc26e99d99af52118837fd5551e26fbffb8efd/ops/services/alerts/app_service_metrics/main.tf#L92