Ussuri Series Release Notes¶
Added a new role for
tripleo_ovn_clusterto configure OVN DBs clustered for TripleO deployments.
A new variable named tripleo_podman_default_network_config has been added to be able to configure the default network used by podman. By default the configuration will use 10.255.255.0/24 as a network for podman managed containers. TripleO does not use podman’s networking, however removing this configuration (or making it invalid) breaks the podman network functions. The network is not applied if it’s already running which is indicated by the existance of a cni-podman0 interface.
Due to the use of a folding block operator instead of the literal block operator the check for existing namespaces does not work correct and namespaces get created on subsequent deploy runs even if they already exist. Now namespaces won’t get created if the are already there.
Stop using systemd to manage container healthchecks and use native podman healthchecks which are easier to consume than systemd. The tripleo_container_manage role will take care of cleaning up the old systemd healthchecks if they exist.
An incorrect lookup issue with respect to the StorageNFS network has been fixed. See Launchpad bug 1849393 for more details.
Fixed a configuration issue where required settings for Octavia services were missing.
The tripleo-ssh-known-hosts ansible role used the list of network names to create the different possible hostnames in the ssh_known_hosts file. The network names do not match the actual network hostnames, like internal_api vs. internalapi which results in ssh host verification to fail and e.g. live migration to fail.
Initial approach in a5bcbc8d015b792b3546ecbb139506f248dacfe8 had to be reverted as it resulted in errors when the _hostname inventory information was missing from the ansible inventory.
This syncs to use the same approach to create the hostname like in https://review.opendev.org/693010 to have entries in /etc/hosts and ssh_known_hosts created in the same way/source.