This patch does the following: - Simplifies the sphinx configuration introduced in https://review.openstack.org/371722 to reduce the number of variables involved. The variables are also ordered in the same way everywhere to make it easier to read and troubleshoot. - Simplifies some of the CLI guides to be more explicit about the tag to checkout when cloning the git repo. - Cleaned up some references which went to non-existant documents. - Added a link to the networking appendix. - As per https://review.openstack.org/369650 the backup directory for the upgrade process is now the name of the source version the upgrade process is working with. Change-Id: Iee30a32f99a66d9facb049311cadf1b9a8b2170e
1.2 KiB
Overview
An OpenStack-Ansible environment can be upgraded between minor versions, and between major versions.
Upgrades between minor versions of OpenStack-Ansible require updating
the repository clone to the latest minor release tag, then running
playbooks against the target hosts. For more information, see minor-upgrades
.
For major upgrades, the OpenStack-Ansible repository provides
playbooks and scripts used to upgrade an environment. The
run-upgrade.sh
script runs each upgrade playbook in the
correct order, or playbooks can be run individually if necessary.
Alternatively, a deployer can upgrade manually. A major upgrade process
performs the following actions:
- Modifies files residing in
/etc/openstack_deploy
in order to reflect new configuration values. - Some flag files are created by the migration scripts in order to achieve idempotency. These files are placed in the directory.
- Upgrade the RabbitMQ server during an OpenStack-Ansible upgrade
process. See
setup-infra-playbook
for details.
For more information on the major upgrade process, see script-upgrade
and manual-upgrade
.
Note
You can only upgrade between sequential releases.