From 96922f3b730995a94e0f113e0fd3a291a5927f50 Mon Sep 17 00:00:00 2001
From: "Lubomir I. Ivanov" <lubomirivanov@vmware.com>
Date: Thu, 18 Jul 2019 13:28:10 +0300
Subject: [PATCH] kubeadm: modify the step for upgrading control plane kubelets
 (#15452)

When upgrading controlplane nodes, upgrading the kubelets should be done
as the last step after "kubeadm upgrade" was run on all of the control
plane nodes already.

Also apply consistency related cleanup to sections and section names.
---
 .../kubeadm/kubeadm-upgrade-1-15.md           | 44 +++++++++++--------
 1 file changed, 25 insertions(+), 19 deletions(-)

diff --git a/content/en/docs/tasks/administer-cluster/kubeadm/kubeadm-upgrade-1-15.md b/content/en/docs/tasks/administer-cluster/kubeadm/kubeadm-upgrade-1-15.md
index 895ba679c9..20896a678f 100644
--- a/content/en/docs/tasks/administer-cluster/kubeadm/kubeadm-upgrade-1-15.md
+++ b/content/en/docs/tasks/administer-cluster/kubeadm/kubeadm-upgrade-1-15.md
@@ -56,7 +56,9 @@ The upgrade workflow at high level is the following:
     {{% /tab %}}
     {{< /tabs >}}
 
-## Upgrade the first control plane node
+## Upgrading control plane nodes
+
+### Upgrade the first control plane node
 
 1.  On your first control plane node, upgrade kubeadm:
 
@@ -220,7 +222,27 @@ The upgrade workflow at high level is the following:
     Check the [addons](/docs/concepts/cluster-administration/addons/) page to
     find your CNI provider and see whether additional upgrade steps are required.
 
-1.  Upgrade the kubelet and kubectl on the control plane node:
+    This step is not required on additional control plane nodes if the CNI provider runs as a DaemonSet.
+
+### Upgrade additional control plane nodes
+
+1.  Same as the first control plane node but use:
+
+```
+sudo kubeadm upgrade node
+```
+
+instead of:
+
+```
+sudo kubeadm upgrade apply
+```
+
+Also `sudo kubeadm upgrade plan` is not needed.
+
+### Upgrade kubelet and kubectl
+
+1.  Upgrade the kubelet and kubectl on all control plane nodes:
 
     {{< tabs name="k8s_install_kubelet" >}}
     {{% tab name="Ubuntu, Debian or HypriotOS" %}}
@@ -241,22 +263,6 @@ The upgrade workflow at high level is the following:
     sudo systemctl restart kubelet
     ```
 
-## Upgrade additional control plane nodes
-
-1.  Same as the first control plane node but use:
-
-```
-sudo kubeadm upgrade node
-```
-
-instead of:
-
-```
-sudo kubeadm upgrade apply
-```
-
-Also `sudo kubeadm upgrade plan` is not needed.
-
 ## Upgrade worker nodes
 
 The upgrade procedure on worker nodes should be executed one node at a time or few nodes at a time,
@@ -305,7 +311,7 @@ without compromising the minimum required capacity for running your workloads.
 
 ### Upgrade kubelet and kubectl
 
-1.  Upgrade the Kubernetes package version by running the Linux package manager for your distribution:
+1.  Upgrade the kubelet and kubectl on all worker nodes:
 
     {{< tabs name="k8s_kubelet_and_kubectl" >}}
     {{% tab name="Ubuntu, Debian or HypriotOS" %}}