Temporarily move CNI deployment to postKubeadmCommands

Addons (CNI) currently fail to deploy, likely due to a timeout with
API server availability. Without a pre-built image, capz takes longer
for the control plane to come up. To work around this, we'll use a
`postKubeadmCommands` in controlplane-0
(examples/controlplane/controlplane.yaml). This should be removed once
we confirm the create-cluster target can successfully deploy CNI.

Signed-off-by: Stephen Augustus <[email protected]>

Author: justaugustus

Makefile

@@ -280,8 +280,13 @@ create-cluster: $(CLUSTERCTL) ## Create a development Kubernetes cluster on Azur
 		--namespace=default \
 		--cluster-name=$(CLUSTER_NAME)
 	# Apply addons on the target cluster, waiting for the control-plane to become available.
-	$(CLUSTERCTL) \
-		alpha phases apply-addons -v=3 \
+# TODO: Uncomment this once we've enabled image building: https://github.com/kubernetes-sigs/image-builder/pull/49
+#       Addons (CNI) currently fail to deploy, likely due to a timeout with API server availability.
+#       Without a pre-built image, capz takes longer for the control plane to come up.
+#       In the meantime, we'll use a `postKubeadmCommands` in controlplane-0 (examples/controlplane/controlplane.yaml).
+#       Please also remove that command once we confirm the create-cluster target works again.
+#	$(CLUSTERCTL) \
+		alpha phases apply-addons -v=10 \
 		--kubeconfig=./kubeconfig \
 		-a examples/addons.yaml
 	# Create a worker node with MachineDeployment.

examples/controlplane/controlplane.yaml

@@ -70,6 +70,8 @@ spec:
         readOnly: true
   preKubeadmCommands:
   - bash -c /tmp/kubeadm-bootstrap.sh
+  postKubeadmCommands:
+  - kubectl --kubeconfig /etc/kubernetes/admin.conf apply -f https://raw.githubusercontent.com/kubernetes-sigs/cluster-api-provider-azure/v1alpha2/examples/addons.yaml
   files:
   - path: /tmp/kubeadm-bootstrap.sh
     owner: "root:root"