E2E failure: LB GET failed after 5 attempts

[CecileRobertMichon]

/kind bug

What steps did you take and what happened:
[A clear and concise description of what the bug is.]

Seeing a lot of PR E2E flakes recently on the with a single control plane node and an AzureMachinePool with 2 nodes [It] spec. Test fails with:

/home/prow/go/src/sigs.k8s.io/cluster-api-provider-azure/test/e2e/azure_test.go:262
Unexpected error:
    <*fmt.wrapError | 0xc0000e2580>: {
        msg: "GET http://40.91.127.106 giving up after 5 attempt(s): Get http://40.91.127.106: dial tcp 40.91.127.106:80: i/o timeout",
        err: {
            Op: "Get",
            URL: "http://40.91.127.106",
            Err: {
                Op: "dial",
                Net: "tcp",
                Source: nil,
                Addr: {
                    IP: [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 255, 255, 40, 91, 127, 106],
                    Port: 80,
                    Zone: "",
                },
                Err: {},
            },
        },
    }
    GET http://40.91.127.106 giving up after 5 attempt(s): Get http://40.91.127.106: dial tcp 40.91.127.106:80: i/o timeout
occurred
/home/prow/go/src/sigs.k8s.io/cluster-api-provider-azure/test/e2e/azure_lb.go:186

https://prow.k8s.io/view/gcs/kubernetes-jenkins/pr-logs/directory/pull-cluster-api-provider-azure-e2e/1348697453653659648

What did you expect to happen:

Anything else you would like to add:
[Miscellaneous information that will assist in solving the issue.]

Environment:

• cluster-api-provider-azure version:
• Kubernetes version: (use kubectl version):
• OS (e.g. from /etc/os-release):

[CecileRobertMichon]

related #1123

[CecileRobertMichon]

Update: we root caused this to a platform regression. The fix is being rolled out, ETA: 2 weeks from now.