Question

Your company runs a web application behind an Azure Application Gateway v2. The backend pool contains three VMs that host an API. The application's health endpoint at /api/health returns an HTTP 200 status but includes a JSON body, and under a partial failure it returns HTTP 503. Recently, the Application Gateway marked all backend servers as unhealthy even though the site returned HTTP 200, because the health endpoint responded with a 200 status but a body of '{"status":"degraded"}'. You need the probe to consider a backend healthy ONLY when it returns HTTP 200 and the response body contains the text 'healthy'. What should you configure?

Accepted Answer

Create a custom health probe and configure match conditions for the accepted HTTP status codes and a matching response body string.

Answer

Change the HTTP setting to use the default probe and set the request timeout to a higher value.

Answer

Add a rewrite rule set to modify the response body returned by the backend so it always contains 'healthy'.

Answer

Configure a listener with a custom error page that maps HTTP 503 responses to HTTP 200.