AZ-305 Practice Question 43: Endpoint Availability Monitoring
Microsoft Azure Solutions Architect Expert (AZ-305) · Test 5 · question 3 of 10
A charity runs a public donation website on Azure App Service. Operations wants to know within minutes if the site stops responding to visitors anywhere in the world, wants warning before the site's TLS certificate expires, and does not want to be paged every time one network path has a transient blip. The website's code must not be modified.
Which monitoring solution should you recommend?
Show answer and explanation
Answer:
Availability tests send web requests to an application at regular intervals from points around the world and alert when it stops responding or responds too slowly, and they require no modification to the site being tested. A standard test also validates the TLS/SSL certificate and supports a proactive lifetime check that fails the test a configurable period before the certificate expires. Microsoft recommends a minimum of five test locations so that a problem in the website can be distinguished from a network problem, and an alert threshold of three out of five locations is the documented way to avoid paging on a single-location blip.
Option A only sees requests that reached the app. If DNS breaks, the certificate expires, or the front end is unreachable from a region, the server-side error metric stays flat and nobody is paged.
Option C reports on the health of the Azure resource as the platform sees it. That catches platform faults, but not an expired certificate, a broken deployment returning errors, or a regional network path failing, and it offers no view from the visitor's side of the internet.
Option D is the do-it-yourself version with weaker properties: one vantage point rather than five, no certificate validation, no retry logic to suppress transient failures, and a Function plus storage table the charity now has to keep running and alert on.
Reference: Application Insights availability tests