ACS Kubernetes, failing creation and uncertainty whether cluster is successfully created

[bremnes]

Is this a request for help?:
possible bug, otherwise user experience flaw

Is this a BUG REPORT or FEATURE REQUEST? (choose one):
Bug report

Orchestrator and version (e.g. Kubernetes, DC/OS, Swarm)
Kubernetes with Windows node pool, default version

What happened:
Attempted twice to create ACS Kubernetes through the portal with a windows node pool.

First attempt in UK West lead to the portal giving me a "deployment in progress" for about two hours before failing with The resource provision operation did not complete within the allowed timeout period. Please see https://aka.ms/arm-deploy for usage details. (Code: ResourceDeploymentFailure).

Second attempt, now in UK South, with a new resource group and ACS returned after ~10 minutes with deployment succeeded. But looking into the activity log of the managed resource group revealed a lot of ongoing activities. Refreshing the activites showed them finished one by one - almost. Still a few remaining with status Started or Accepted, and I'm not sure if that cluster is 100% or not:

What you expected to happen:
Following the portal experience with no customizations should lead to a cluster being successfully created. In addition I expect the activites to leave no doubt whether they are completed or not.

How to reproduce it (as minimally and precisely as possible):
Create a ACS Kubernetes cluster in UK West/UK South with a Windows agent pool.

Anything else we need to know:
If it helps you debug, correlationId for the first cluster's failed creation: 6468b9d7-99bb-4a8a-b65a-0af6df81fdbd.
For the second cluster which might've been successfully created: 1df142c1-6482-4a20-bc56-06c1ab81d111.