Merge pull request #178349 from dlepow/apimsf

[APIM] Update policy example for APIM-SF integration

Author: PRMerger8

articles/api-management/how-to-configure-service-fabric-backend.md

@@ -112,17 +112,21 @@ Add the [`set-backend-service`](api-management-transformation-policies.md#SetBac
 
 1. On the **Design** tab, in the **Inbound processing** section, select the code editor (**</>**) icon. 
 1. Position the cursor inside the **&lt;inbound&gt;** element
-1. Add the following policy statement. In `backend-id`, substitute the name of your Service Fabric backend.
+1. Add the `set-service-backend` policy statement. 
+      * In `backend-id`, substitute the name of your Service Fabric backend.
 
-   The `sf-resolve-condition` is a retry condition if the cluster partition isn't resolved. The number of retries was set when configuring the backend.
+      * The `sf-resolve-condition` is a condition for re-resolving a service location and resending a request. The number of retries was set when configuring the backend. For example:
 
-    ```xml
-    <set-backend-service backend-id="mysfbackend" sf-resolve-condition="@(context.LastError?.Reason == "BackendConnectionFailure")"  />
+      ```xml
+      <set-backend-service backend-id="mysfbackend" sf-resolve-condition="@(context.LastError?.Reason == "BackendConnectionFailure")"/>
     ```
 1. Select **Save**.
 
     :::image type="content" source="media/backends/set-backend-service.png" alt-text="Configure set-backend-service policy":::
 
+> [!NOTE]
+> If one or more nodes in the Service Fabric cluster goes down or is removed, API Management does not get an automatic notification and continues to send traffic to these nodes. To handle these cases, configure a resolve condition similar to: `sf-resolve-condition="@((int)context.Response.StatusCode != 200 || context.LastError?.Reason == "BackendConnectionFailure" || context.LastError?.Reason == "Timeout")"`
+
 ### Test backend API
 
 1. On the **Test** tab, select the **GET** operation you created in a previous section.