[ACA] Add guidance to health probes article.

Author: v-jaswel

articles/container-apps/health-probes.md

@@ -167,8 +167,12 @@ If ingress is enabled, the following default probes are automatically added to t
 | Probe type | Default values |
 | -- | -- |
 | Startup | Protocol: TCP<br>Port: ingress target port<br>Timeout: 3 seconds<br>Period: 1 second<br>Initial delay: 1 second<br>Success threshold: 1<br>Failure threshold: 240 |
-| Readiness | Protocol: TCP<br>Port: ingress target port<br>Timeout: 5 seconds<br>Period: 5 seconds<br>Initial delay: 3 seconds<br>Success threshold: 1<br>Failure threshold: 48 |
 | Liveness | Protocol: TCP<br>Port: ingress target port |
+| Readiness | Protocol: TCP<br>Port: ingress target port<br>Timeout: 5 seconds<br>Period: 5 seconds<br>Initial delay: 3 seconds<br>Success threshold: 1<br>Failure threshold: 48 |
+
+If you're running your container app in [multiple revision mode](revisions.md#revision-modes), after you deploy a revision, you should wait until your readiness probes indicate success before you shift traffic to that revision. In single revision mode, this happens automatically (that is, traffic is not shifted to the new revision until its readiness probes indicate success.)
+
+A revision state appears as unhealthy if any of its replicas fails its readiness probe check, even if all other replicas in the revision are healthy. Container Apps restarts the replica in question until it is healthy again or the failure threshold is exceeded. If the failure threshold is exceeded, it might help to restart the revision, but it might also mean the revision is not configured correctly.
 
 If your app takes an extended amount of time to start (which is common in Java) you often need to customize the probes so your container doesn't crash.