To Reproduce Steps to reproduce the behavior:

1. Follow the doc to upgrade from 16.2.5 to 17.1.6.
2. Run "openstack overcloud upgrade run --yes --stack overcloud --debug --limit allovercloud,undercloud --playbook all"
3. Wait until task or near it:
   Create containers managed by Podman for /var/lib/tripleo-config/container-puppet-config/step_1
4. Observe that ovn database container will be created on only one controllers and 'container-puppet_ovn_controller' will be stuck in the others.

Expected behavior

• Upgrade run fine or a fatal occur with the error message.

Device Info (please complete the following information):

• OS Version:RHEL8.4
• RHOSP16.2.5

Bug impact

• Customer is not able to manage the environment while not end the upgrade or restore a backup.

Known workaround

• Only restore a backup.

Additional context

• Puppet container stuck on process "
  /usr/bin/ovs-vsctl set Open_vSwitch . external_ids:ovn-chassis-mac-mappings=backup:<mac1>,datacentre:<mac2>"