Files
openstack-ansible/deploy-guide/source/configure.rst

372 lines
11 KiB
ReStructuredText

.. _configure:
========================
Configure the deployment
========================
.. figure:: figures/installation-workflow-configure-deployment.png
:width: 100%
Ansible references some files that contain mandatory and optional
configuration directives. Before you can run the Ansible playbooks, modify
these files to define the target environment. Configuration tasks include:
* Target host networking to define bridge interfaces and
networks.
* A list of target hosts on which to install the software.
* Virtual and physical network relationships for OpenStack
Networking (neutron).
* Passwords for all services.
Initial environment configuration
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
OpenStack-Ansible (OSA) depends on various files that are used to build an
inventory for Ansible. Perform the following configuration on the deployment
host.
#. Copy the contents of the
``/opt/openstack-ansible/etc/openstack_deploy`` directory to the
``/etc/openstack_deploy`` directory.
.. code-block:: shell-session
# cp -a /opt/openstack-ansible/etc/openstack_deploy /etc/openstack_deploy
#. Change to the ``/etc/openstack_deploy`` directory.
#. Copy the ``openstack_user_config.yml.example`` file to
``openstack_user_config.yml``.
.. code-block:: shell-session
# cp openstack_user_config.yml.example openstack_user_config.yml
#. Review the ``openstack_user_config.yml`` file and make changes
to the deployment of your OpenStack environment.
.. note::
This file is heavily commented with details about the various options.
See our :dev_docs:`User Guide <user/index.html>` and
:dev_docs:`Reference Guide <reference/index.html>` for more details.
#. Review the ``user_variables.yml`` file to configure global and role specific
deployment options. The file contains some example variables and comments but
you can get the full list of variables in each role's specific documentation.
.. note::
One important variable is the ``install_method`` which configures the installation
method for the OpenStack services. The services can either be deployed from
source (default) or from distribution packages. Source based deployments are closer
to a vanilla OpenStack installation and allow for more tweaking and customizations. On
the other hand, distro based deployments generally provide a package combination which
has been verified by the distributions themselves. However, this means that updates are
being released less often and with a potential delay. Moreover, this method offer
fewer opportunities for deployment customizations and is supported only by selected
services. The ``install_method`` variable is set during the initial deployment and
you **must not** change it as OpenStack-Ansible is not able to convert itself from
one installation method to the other. As such, it's important to judge your needs
against the pros and cons of each method before making a decision.
The configuration in the ``openstack_user_config.yml`` file defines which hosts
run the containers and services deployed by OpenStack-Ansible. For
example, hosts listed in the ``shared-infra_hosts`` section run containers for
many of the shared services that your OpenStack environment requires. Some of
these services include databases, Memcached, and RabbitMQ. Several other
host types contain other types of containers, and all of these are listed
in the ``openstack_user_config.yml`` file.
Some services, such as glance, heat, horizon and nova-infra, are not listed
individually in the example file as they are contained in the os-infra hosts.
You can specify image-hosts or dashboard-hosts if you want to scale out in a
specific manner.
For examples, please see our :dev_docs:`User Guides <user/index.html>`
For details about how the inventory is generated, from the environment
configuration and the variable precedence, see our
:dev_docs:`Reference Guide <reference/index.html>` under the inventory
section.
Configure target hosts
~~~~~~~~~~~~~~~~~~~~~~
Modify the ``/etc/openstack_deploy/openstack_user_config.yml`` file
to configure the target hosts.
Do not assign the same IP address to different target hostnames.
Unexpected results may occur. Each IP address and hostname must be
a matching pair. To use the same host in multiple roles, for example
infrastructure and networking, specify the same hostname and IP in
each section.
Unless otherwise stated, replace ``*_IP_ADDRESS`` with the IP address of
the br-mgmt container management bridge on each target host.
.. note::
If the SSH access to the host is via a different network than the
br-mgmt interface, please, refer to the `guide <https://docs.openstack.org/openstack-ansible/latest/reference/inventory/configure-inventory.html#having-ssh-network-different-from-openstack-management-network>`_.
#. Configure a list containing at least three infrastructure
target hosts in the ``shared-infra_hosts`` section:
.. code-block:: yaml
shared-infra_hosts:
infra01:
ip: INFRA01_IP_ADDRESS
infra02:
ip: INFRA02_IP_ADDRESS
infra03:
ip: INFRA03_IP_ADDRESS
infra04: ...
#. Configure a list of at least one keystone target host in the
``identity_hosts`` section:
.. code-block:: yaml
identity_hosts:
infra01:
ip: INFRA01_IP_ADDRESS
infra02: ...
#. Configure the appropriate set of hosts responsible for network-related
roles in your deployment:
.. code-block:: yaml
network-infra_hosts:
infra01:
ip: INFRA01_IP_ADDRESS
infra02: ...
network-northd_hosts:
infra01:
ip: INFRA01_IP_ADDRESS
infra02: ...
When deploying OpenStack with OVN, it's essential to properly configure
``network-gateway_hosts`` depending on your network architecture. There are
two typical scenarios:
Scenario 1: DVR with gateway on compute nodes:
.. code-block:: yaml
network-gateway_hosts:
compute01:
ip: COMPUTE01_IP_ADDRESS
compute02: ...
Scenario 2: standalone network nodes:
.. code-block:: yaml
network-gateway_hosts:
network01:
ip: NETWORK01_IP_ADDRESS
network02: ...
#. Configure a list containing at least one compute target host in
the ``compute_hosts`` section:
.. code-block:: yaml
compute_hosts:
compute01:
ip: COMPUTE01_IP_ADDRESS
compute02: ...
#. Configure a list containing at least one repository target host in
the ``repo-infra_hosts`` section:
.. code-block:: yaml
repo-infra_hosts:
infra01:
ip: INFRA01_IP_ADDRESS
infra02:
ip: INFRA02_IP_ADDRESS
infra03:
ip: INFRA03_IP_ADDRESS
infra04: ...
The repository typically resides on one or more infrastructure hosts.
#. Optionally configure storage host in the ``storage_hosts`` section:
.. code-block:: yaml
storage_hosts:
storage01:
ip: STORAGE01_IP_ADDRESS
storage02: ...
Each storage host requires additional configuration to define the
back end driver. The default configuration includes an optional
storage host. To install without storage hosts, comment out the
stanza beginning with the ``storage_hosts:`` line.
Installing additional services
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
To install additional services, the files in
``etc/openstack_deploy/conf.d`` provide examples showing
the correct host groups to use. To add another service, add the host group,
allocate hosts to it, and then execute the playbooks.
Advanced service configuration
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
OpenStack-Ansible has many options that you can use for the advanced
configuration of services. Each role's documentation provides information
about the available options.
.. important::
This step is essential to tailoring OpenStack-Ansible to your needs
and is generally overlooked by new deployers. Have a look at each
role documentation, user guides, and reference if you want a tailor
made cloud.
Infrastructure service roles
----------------------------
- :ansible_role_docs:`PKI <pki>`
- :role_docs:`galera_server <galera_server>`
- :role_docs:`haproxy_server <haproxy_server>`
- :role_docs:`memcached_server <memcached_server>`
- :role_docs:`rabbitmq_server <rabbitmq_server>`
- :role_docs:`repo_server <repo_server>`
- :ansible_role_docs:`Zookeeper <zookeeper>`
OpenStack service roles
-----------------------
- :role_docs:`os_adjutant <os_adjutant>`
- :role_docs:`os_aodh <os_aodh>`
- :role_docs:`os_barbican <os_barbican>`
- :role_docs:`os_ceilometer <os_ceilometer>`
- :role_docs:`os_cinder <os_cinder>`
- :role_docs:`os_cloudkitty <os_cloudkitty>`
- :role_docs:`os_designate <os_designate>`
- :role_docs:`os_glance <os_glance>`
- :role_docs:`os_gnocchi <os_gnocchi>`
- :role_docs:`os_heat <os_heat>`
- :role_docs:`os_horizon <os_horizon>`
- :role_docs:`os_ironic <os_ironic>`
- :role_docs:`os_keystone <os_keystone>`
- :role_docs:`os_magnum <os_magnum>`
- :role_docs:`os_manila <os_manila>`
- :role_docs:`os_masakari <os_masakari>`
- :role_docs:`os_mistral <os_mistral>`
- :role_docs:`os_neutron <os_neutron>`
- :role_docs:`os_nova <os_nova>`
- :role_docs:`os_octavia <os_octavia>`
- :role_docs:`os_placement <os_placement>`
- :role_docs:`os_rally <os_rally>`
- :role_docs:`os_swift <os_swift>`
- :role_docs:`os_tacker <os_tacker>`
- :role_docs:`os_tempest <os_tempest>`
- :role_docs:`os_trove <os_trove>`
- :role_docs:`os_zun <os_zun>`
Other roles
-----------
- :role_docs:`apt_package_pinning <apt_package_pinning>`
- :role_docs:`ceph_client <ceph_client>`
- :role_docs:`lxc_container_create <lxc_container_create>`
- :role_docs:`lxc_hosts <lxc_hosts>`
- :role_docs:`openstack_hosts <openstack_hosts>`
- :role_docs:`openstack_openrc <openstack_openrc>`
- :role_docs:`plugins <plugins>`
- :ansible_role_docs:`python_venv_build <python_venv_build>`
- :ansible_role_docs:`systemd_service <systemd_service>`
- :ansible_role_docs:`systemd_mount <systemd_mount>`
- :ansible_role_docs:`systemd_networkd <systemd_networkd>`
- :ansible_role_docs:`unbound <unbound>`
- :ansible_role_docs:`uWSGI <uwsgi>`
Configuring service credentials
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Configure credentials for each service in the
``/etc/openstack_deploy/user_secrets.yml`` file. Consider using the
`Ansible Vault <http://docs.ansible.com/playbooks_vault.html>`_ feature to
increase security by encrypting any files that contain credentials. You
can leverage `osa_ops.encrypt_secrets collection <https://docs.openstack.org/openstack-ansible-ops/latest/encrypt_secrets.html>`_
to automate simplify the process of encryption and further management
of secrets.
Adjust permissions on these files to restrict access by non-privileged
users.
The ``keystone_auth_admin_password`` option configures the ``admin`` tenant
password for both the OpenStack API and Dashboard access.
We recommend that you use the ``pw-token-gen.py`` script to generate random
values for the variables in each file that contains service credentials:
.. code-block:: shell-session
# cd /opt/openstack-ansible
# ./scripts/pw-token-gen.py --file /etc/openstack_deploy/user_secrets.yml
To regenerate existing passwords, add the ``--regen`` flag.
For information on how to rotate passwords, please refer to the `Password Rotation <https://docs.openstack.org/openstack-ansible/latest/user/security/password-rotation.html>`_
documentation.