.. _deployment:

Deployment Commands
-------------------

.. _21.1|VOSS-669:

Architecturally, VOSS Automate offers two main deployment topologies:

* Unified Node Cluster Topology
* Modular Cluster Deployment

For details:

.. raw:: html

   see <a href="../best-practices/best-prac-architecture.html">VOSS Automate Deployment Topologies</a>.

.. raw:: latex

   see the VOSS Automate Deployment Topologies chapter in the Architecture and Hardware Specification Guide.


Also refer to:

* the Install Guide
* :ref:`cluster-commands`

.. note::

   From release 21.1 onwards, a standalone topology is considered a cluster-of-one.
   This means that commands such as **cluster list** and **cluster status** should
   be run on a standalone topology.

In a clustered topology, a number of nodes with different roles are 
clustered together and provisioned to form a networked system.  When nodes 
are clustered together, High Availability and Disaster Recovery can be 
achieved by ensuring that there are redundant services.
Nodes can be deployed in any order.

.. index:: cluster;cluster add
.. index:: cluster;cluster del
.. index:: cluster;cluster list
.. index:: cluster;cluster status
.. index:: cluster;cluster provision

.. important::

   * On a Unified node topology, the commands below
     should be run on the primary database node.
   * On a Modular topology, the commands below should be run on the database node.


Once two or more nodes have been deployed, the nodes can be grouped into a 
cluster by executing **cluster add <ip>**.  Note that a node already in one 
cluster cannot be added to another cluster.  Likewise, nodes can be removed 
from a cluster with the **cluster del <ip>** command.  The nodes in the cluster 
can be displayed using **cluster list**.  

Cluster roles cannot be changed after installation because it is dependent 
on installed software and other configuration at time of deployment.  

The status of the cluster can be viewed using **cluster status**.

If the node topology needs to be changed, the following procedure can be 
followed:

* A node can be removed from the cluster with **cluster del <ip>**
* The node can be redeployed with the correct parameters
* Add the new node to the cluster with **cluster add <ip>**
* Provision the cluster with **cluster provision**. This command should only 
  be run on one node in the cluster, usually an application server.

.. |VOSS Automate| replace:: VOSS Automate
.. |Unified CM| replace:: Unified CM