From 3bbdb21a01f4c59a698001696d31c2b87302f668 Mon Sep 17 00:00:00 2001
From: Tahlia Richardson <3069029+tahliar@users.noreply.github.com>
Date: Fri, 22 Nov 2024 16:50:19 +1000
Subject: [PATCH] Add installation overview (WIP)
---
xml/ha_installation_overview.xml | 145 ++++++++++++++++++++++++++-----
xml/ha_log_in.xml | 2 +-
2 files changed, 122 insertions(+), 25 deletions(-)
diff --git a/xml/ha_installation_overview.xml b/xml/ha_installation_overview.xml
index f921b6ddf..8629ad4f8 100644
--- a/xml/ha_installation_overview.xml
+++ b/xml/ha_installation_overview.xml
@@ -16,7 +16,7 @@
-You can also use a combination of both setup methods, for example: set up one node with YaST cluster and then use one of the bootstrap scripts to integrate more nodes (or vice versa).
+ This chapter shows an overview of the steps required to set up a working and supported &ha; cluster. It includes all of the options for a full cluster. If you want to start with a basic two-node cluster with only the default options, see .
@@ -24,27 +24,124 @@ You can also use a combination of both setup methods, for example: set up one no
yes
-
- If you are setting up a &ha; cluster with &productnamereg; for the first time, the
- easiest way is to start with a basic two-node cluster. You can also use the
- two-node cluster to run some tests. Afterward, you can add more
- nodes by cloning existing cluster nodes with &ay;. The cloned nodes will
- have the same packages installed and the same system configuration as the
- original ones.
-
-
-
- Workflow options
-
-
-
-
-
-
- Preconfiguration options
-
-
-
-
-
+
+ Overview of installing a &ha; cluster
+
+
+ Review to make sure your nodes and other infrastructure meet the requirements for a &ha; cluster.
+
+
+
+
+ You can set up the cluster with either the &rootuser; user or a user with sudo privileges. Review to determine the appropriate user for your requirements.
+
+
+
+
+ Install the &ha; extension and packages on the nodes as described in .
+
+
+
+
+ Set up the cluster on the nodes. You can use either of the following methods:
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+ You can also use a combination of both methods. For example, you could set up one node with the &yast; cluster module and then use the bootstrap scripts to integrate more nodes (or vice versa).
+
+
+ The following table shows the components that are configured by each method:
+
+
+
+
+
+
+
+
+ Component
+ Bootstrap script
+ &yast; cluster module
+
+
+
+
+ Firewall ports
+ Yes
+ Yes
+
+
+ Passwordless SSH
+ Yes
+ No; configure before setup
+
+
+ &csync;
+ Yes
+ Yes
+
+
+ &corosync;
+ Yes
+ Yes
+
+
+ &stonith; (node fencing)
+ Optional (SBD only)
+ No; configure after setup
+
+
+ &qdevice;
+ Optional
+ Optional
+
+
+
+
+
+
+
+ Add more nodes. After setting up the initial cluster, you can clone nodes for mass deployment as described in .
+
+
+
+
+ To be supported, a &sleha; cluster must have &stonith; (node fencing) enabled. A node fencing mechanism can be one of the following:
+
+
+
+
+ A physical device (a power switch). To configure the cluster to use physical &stonith; devices, see .
+
+
+
+
+ SBD (&stonith; Block Device) in combination with a watchdog. To configure SBD devices and the watchdog, see .
+
+
+
+ Optionally, you can configure SBD during the cluster setup with the bootstrap scripts (but not &yast;). In this case, you should set up shared storage and the watchdog before you initialize the cluster.
+
+
+
+
+
+
+
+ Configure other things you declined to do before, e.g. QDevice
+
+
+
+
diff --git a/xml/ha_log_in.xml b/xml/ha_log_in.xml
index fdc04895f..34d0d7f71 100644
--- a/xml/ha_log_in.xml
+++ b/xml/ha_log_in.xml
@@ -69,7 +69,7 @@
A user with sudo privileges (with SSH agent forwarding)
- You can use SSH forwarding to pass your local SSH keys to the cluster nodes.
+ You can use SSH agent forwarding to pass your local SSH keys to the cluster nodes.
This can be useful if you need to avoid storing SSH keys on the nodes, but requires
additional configuration on your local machine and on the cluster nodes.