Instance configuration
======================

This page presents an overview of the various controls that TPA offers
to customise the deployment process on cluster instances, with links to
more detailed documentation.

Before you dive into the details of deployment, it may be helpful to
read :ref:`an overview of configuring a cluster <Cluster configuration>`  to understand how cluster and instance variables and
the other mechanisms in config.yml work together to allow you to write a
concise, easy-to-review configuration.

System-level configuration
--------------------------

The first thing TPA does is to ensure that Python is bootstrapped and
ready to execute Ansible modules (a distribution-specific process). Then
it completes various system-level configuration tasks before moving on
to :ref:`Postgres configuration <Postgres configuration>`  below.

- :ref:`Distribution support <Distribution support>`  

- :ref:`Setting up the TPA Python environment <Setting up the TPA Python environment>`  (`preferred_python_version` ) 

- :ref:`Environment variables <Environment>`  (e.g., `https_proxy` ) 

Package repositories
^^^^^^^^^^^^^^^^^^^^

You can use the :ref:`pre-deploy hook <TPA hooks>` 

to execute tasks before any package repositories are configured.

- :ref:`Configure YUM repositories <Configuring YUM repositories>`  

(for RHEL, Rocky and AlmaLinux)

- :ref:`Configure APT repositories <Configuring APT repositories>`  

(for Debian and Ubuntu)

- :ref:`Configure 2ndQuadrant and EDB repositories <How TPA uses 2ndQuadrant and EDB repositories>`  

(on any system)

- :ref:`Configure a local package repository <Creating and using a local repository>`  

(to ship packages to target instances)

You can use the :ref:`post-repo hook <TPA hooks>` 

to execute tasks after package repositories have been configured (e.g.,
to correct a problem with the repository configuration before installing
any packages).

Package installation
^^^^^^^^^^^^^^^^^^^^

Once the repositories are configured, packages are installed at various
stages throughout the deployment, beginning with a batch of system
packages:

- :ref:`Install non-Postgres packages <Installing packages>`  

(e.g., acl, openssl, sysstat)

Postgres and other components (e.g., Barman, repmgr, pgbouncer) will be
installed separately according to the cluster configuration; these are
documented in their own sections below.

Other system-level tasks
^^^^^^^^^^^^^^^^^^^^^^^^

- :ref:`Create and mount filesystems <Filesystem configuration>`  (including RAID, LUKS setup) 

- :ref:`Upload artifacts <Uploading artifacts>`  (files, directories, tar archives) 

- :ref:`Set sysctl values <Setting sysctl values>`  

- :ref:`Configure /etc/hosts <Configuring /etc/hosts>`  

- :ref:`Manage ssh_known_hosts entries <Managing SSH host keys>`  

- :ref:`Add system locale <Locale>`  

Skipping deployment completely
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

To prevent TPA from doing any part of the deployment process on an
instance - in other words, if you want TPA to provision the instance and
then leave it alone - set the ``provision_only`` setting for the
instance to ``true`` in ``config.yml`` . This setting will make TPA omit
the instance entirely from the inventory which ``tpaexec deploy`` sees.

Postgres
--------

Postgres configuration is an extended process that goes hand-in-hand
with setting up other components like repmgr and pgbouncer. It begins
with installing Postgres itself.

Version selection
^^^^^^^^^^^^^^^^^

Use the :ref:`configure options <Cluster configuration>`  to select a Postgres flavour and version, or set
``postgres_version`` in config.yml to specify which Postgres major
version you want to install.

That’s all you really need to do to set up a working cluster. Everything
else on this page is optional. You can control every aspect of the
deployment if you want to, but the defaults are carefully tuned to give
you a sensible cluster as a starting point.

Installation
^^^^^^^^^^^^

The default ``postgres_installation_method`` is to install packages for
the version of Postgres you selected, along with various extensions,
according to the architecture’s needs:

- :ref:`Install Postgres and Postgres-related packages <Installing Postgres-related packages>`  

(e.g., pglogical, BDR, etc.)

- :ref:`Build and install Postgres and extensions from source <Postgres source installation>`  

(for development and testing)

Whichever installation method you choose, TPA can give you the same
cluster configuration with a minimum of effort.

Configuration
^^^^^^^^^^^^^

- :ref:`Configure the postgres Unix user <The postgres Unix user>`  

- :ref:`Run initdb to create the PGDATA directory <Running initdb>`  

- :ref:`Configure pg_hba.conf <pg_hba.conf>`  

- :ref:`Configure pg_ident.conf <pg_ident.conf>`  

- :ref:`Configure postgresql.conf <postgresql.conf>`  

You can use the :ref:`postgres-config hook <TPA hooks>` 

to execute tasks after the Postgres configuration files have been
installed (e.g., to install additional configuration files).

Once the Postgres configuration is in place, TPA will go on to install
and configure other components such as Barman, repmgr, pgbouncer, and
haproxy, according to the details of the architecture.

Other components
----------------

- :ref:`Configure Barman <Barman>`  

- :ref:`Configure pgbouncer <Configuring pgbouncer>`  

- :ref:`Configure haproxy <Configuring haproxy>`  

- :ref:`Configure HARP <Configuring HARP>`  

- :ref:`Configure EFM <Configuring EFM>`  

Configuring and starting services
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

TPA will now install systemd service unit files for each service. The
service for Postgres is named ``postgres.service`` , and can be started
or stopped with ``systemctl start postgres`` .

In the first deployment, the Postgres service will now be started. If
you are running ``tpaexec deploy`` again, the service may be reloaded or
restarted depending on what configuration changes you may have made. Of
course, if the service is already running and there are no changes, then
it’s left alone.

In any case, Postgres will be running at the end of this step.

After starting Postgres
-----------------------

- :ref:`Create Postgres users <The postgres Unix user>`  

- :ref:`Create Postgres tablespaces <Creating Postgres tablespaces>`  

- :ref:`Create Postgres databases <Creating Postgres databases>`  

- :ref:`Configure pglogical replication <pglogical configuration>`  

- :ref:`Configure .pgpass <Configuring .pgpass>`  

You can use the :ref:`postgres-config-final hook <TPA hooks>` 

to execute tasks after the post-startup Postgres configuration has been
completed (e.g., to perform SQL queries to create objects or load data).

- :ref:`Configure BDR <EDB Postgres Distributed configuration>`  

You can use the :ref:`post-deploy hook <TPA hooks>` 

to execute tasks after the deployment process has completed.