From aae3b1f10cc4666bdbb0c5dc2b74b70a2d95660f Mon Sep 17 00:00:00 2001
From: Michael Vincerra <michael.vincerra@intel.com>
Date: Fri, 25 Jan 2019 15:43:17 -0800
Subject: [PATCH] Adds kubernetes-bp and organizes in new kubernetes folder.

Signed-off-by: Michael Vincerra <michael.vincerra@intel.com>
---
 .../tutorials/kubernetes/kubernetes-bp.rst    | 104 ++++++++++++++++++
 .../tutorials/{ => kubernetes}/kubernetes.rst |   0
 2 files changed, 104 insertions(+)
 create mode 100644 source/clear-linux/tutorials/kubernetes/kubernetes-bp.rst
 rename source/clear-linux/tutorials/{ => kubernetes}/kubernetes.rst (100%)

diff --git a/source/clear-linux/tutorials/kubernetes/kubernetes-bp.rst b/source/clear-linux/tutorials/kubernetes/kubernetes-bp.rst
new file mode 100644
index 00000000..3c241cb2
--- /dev/null
+++ b/source/clear-linux/tutorials/kubernetes/kubernetes-bp.rst
@@ -0,0 +1,104 @@
+.. _kubernetes-bp:
+
+Kubernetes Best Practices on |CL|
+#################################
+
+This tutorial shows you how to manage your Kubernetes cluster while using
+:command:`swupd` to update |CL-ATTR|.
+
+In our tutorial :ref:`kubernetes`, we explain how to set up a Kubernetes
+cluster on |CL| using `kubeadm`. `Kubeadm documentation`_ often builds on the
+assumption that the distribution uses a traditional package manager (e.g.,
+RPM/DEB). In contrast, |CL| manages packages in bundles. This document
+describes best practices to manage cluster upgrades with kubeadm on a |CL|-based cluster.
+
+Prerequisites
+**************
+
+Let's review some basic assumptions:
+
+* |CL| includes Kubernetes support with the `cloud-native-basic` bundle.
+
+* Execute :command:`sudo swupd update` **on each node** to do the update.
+
+* Repeat the above command for both master and worker nodes.
+
+Executing this command updates all of the kubernetes node and client
+binaries simultaneously, which are part of `cloud-native-basic` bundle
+(including kubectl, kubeadm, kubelet).
+
+.. note::
+
+   Oher Linux\* distros shown in Kubernetes upgrade documentation reflect
+   `apt-get update`, `apt-mark hold kubeadm`, and similar commands; however, such commands **aren not valid** on |CL|.
+
+
+Update the cluster
+******************
+#. Read kubernetes documentation `before you begin`_.
+
+#. On your master node, run the command:
+
+   .. code-block:: bash
+
+      sudo swupd update
+
+#. Run this command to check the message-of-the-day, or `motd`:
+
+   .. code-block:: bash
+
+      cat /run/motd
+
+#. Next, ....[Explain resultant actions/choices from above]
+
+   .. note::
+
+      Swupd post install script adds a message-of-the-day to warn/notify
+      users if the OS update installs a new kubernetes version that is not
+      compatible (according to kubernetes API guarantee) with the previous
+      version. If the `motd` appears, a kubelet restart on master and nodes
+      must be postponed until the control plane is properly updated.
+      :command:`swupd update` does not restart services automatically unless
+      explicitly configured to do so (be warned about making kubelet
+      automatically restartable).
+
+   .. note:: Do not restart the kubelet before the control plane is updated.
+
+#. Now follow these instructions in kubernetes documentation.
+
+   * `Upgrade control plane`_
+   * `Drain control plane node`_
+   * `Restart Kubelet and undrain node`_
+
+
+Update worker nodes
+*******************
+
+#. On each worker node, run the command:
+
+   .. code-block:: bash
+
+      sudo swupd update
+
+#. Now follow these instructions in kubernetes documentation:
+
+   * `Drain node`_
+   * `Update kubelet configuration`_
+   * `Restart Kubelet and undrain node`_
+
+
+.. _Restart Keubelet and undrain node: https://kubernetes.io/docs/tasks/administer-cluster/kubeadm/kubeadm-upgrade-1-13/#restart-the-kubelet-for-all-nodes
+
+.. _Update kubelet configuration: https://kubernetes.io/docs/tasks/administer-cluster/kubeadm/kubeadm-upgrade-1-13/#upgrade-the-kubelet-config-on-worker-nodes
+
+.. _Drain node: https://kubernetes.io/docs/tasks/administer-cluster/kubeadm/kubeadm-upgrade-1-13/#drain-control-plane-and-worker-nodes
+
+ .. _Restart kubelet and undrain node: https://kubernetes.io/docs/tasks/administer-cluster/kubeadm/kubeadm-upgrade-1-13/#restart-the-kubelet-for-all-nodes
+
+.. _Upgrade control plane: https://kubernetes.io/docs/tasks/administer-cluster/kubeadm/kubeadm-upgrade-1-13/#upgrade-the-control-plane-node
+
+.. _Drain control plane node: https://kubernetes.io/docs/tasks/administer-cluster/kubeadm/kubeadm-upgrade-1-13/#drain-control-plane-and-worker-nodes
+
+.. _Kubeadmn documentation: https://kubernetes.io/docs/reference/setup-tools/kubeadm/kubeadm/
+
+.. _before you begin: https://kubernetes.io/docs/tasks/administer-cluster/kubeadm/kubeadm-upgrade-1-13/#before-you-begin
\ No newline at end of file
diff --git a/source/clear-linux/tutorials/kubernetes.rst b/source/clear-linux/tutorials/kubernetes/kubernetes.rst
similarity index 100%
rename from source/clear-linux/tutorials/kubernetes.rst
rename to source/clear-linux/tutorials/kubernetes/kubernetes.rst