Release r11.0.2

Changelog: r11.0.1...r11.0.2

Notable Changes

• This release uses the run-upgrade-old.sh script while the new run-upgrade.sh script is tested.
• Neutron L3 HA should now work correctly
• RabbitMQ SSL fixes
• Updated RabbitMQ server version to 3.5.6-1

r11.0.1

Release r11.0.1

For full change log please go here r11.0.0...r11.0.1.

Notables:

• Several MaaS bug fixes and enhancements
• Additional monitoring for HP gear
• Bash auto-completion for openstack clients

r11.0.0

Release 11.0.0!

This release is based on the rc18 release candidate.

r11.0.0rc18

Release candidate 18 changes references to the upstream file from os-ansible-deployment to openstack-ansible.

This change includes comments as well as directories and files. Thus, the new location for upstream playbooks, roles, and scripts is now ./openstack-ansible in the root of rpc-openstack.

No other changes have been introduced.

r11.0.0rc14

This release candidate has introduced the following changes:

• Add role to allow rpc-openstack to patch OSAD files (GitHub issue)

r11.0.0rc13

This release provides fixes for the following major upgrade issues from os-ansible-deployment:

• Duplicate IP addresses created during upgrade (Upstream bug report)
  • This was causing issues that were random and hard to explain, since servers that existed since Juno were clashing with new repo server IPs.
• Kilo upgrade creates a temporary variable with a couple bugs (Upstream bug report.
  • This caused new Kilo python packages to fail to install, since the wrong repo URL was distributed to containers.

Two other small fixes were introduced to os-ansible-deployment between r11.0.0rc12 and r11.0.0rc13

• lxc_hosts role presumes irqbalance is installed (Upstream bug)
• lxc image download is sometimes corrupt (Upstream bug)

In addition, the following rpc-openstack upgrade bugs were fixed:

• Ensure getting the correct Keystone v3 endpoint in maas_common (GitHub issue)
• after juno -> kilo upgrade, the auth_ref on disk that maas-common uses is in the old keystone apiv2 format (GitHub issue)

r11.0.0rc12

This release contains some fixes to last minute blocking issues around upgrades identified by QE/Dev, including:

• https://github.com/rcbops/rpc-openstack/issues/308 : on upgrades, the galera restart fails due to the new cluster name in kilo
• https://github.com/rcbops/rpc-openstack/issues/310 : the rpc-openstack upgrade.sh script does not merge in the rpcd specific user_variables

r11.0.0rc11

This release contains some fixes to last minute blocking issues around upgrades identified by QE/Dev, including:

• https://bugs.launchpad.net/openstack-ansible/+bug/1479916 apt-transport-https not getting installed correctly during upgrades
• https://bugs.launchpad.net/openstack-ansible/+bug/1480342 playbook fails in run-upgrade.sh, failure is not reported
• https://bugs.launchpad.net/openstack-ansible/+bug/1480988 upgrade fails on juno log rotate playbook
• https://github.com/rcbops/rpc-openstack/issues/259 upgrade.sh has wrong path for deploy.sh

r11.0.0rc10

This RC addresses a problem with the upstream, OSAD upgrade process that prevented upgrades from succeeding.

For more details, see issue #299.

r11.0.0rc9

This RC of rpc-openstack addresses the following major issues identified with rc8:

• Rejoining galera clusters fails due to SST setting Launchpad bug
• Update MaaS checks to use Keystone v3 GitHub bug