92b1b97885
At the moment services might have different MPM selected while all operating the same Apache setup, ie on metal setup. This results in failures to set selected MPMs, so eventually second run of roles after initial deployment will end up in failure (ie upgrade). This patch ensures that all except selected MPMs are disabled and do role get's the desired state of deployment. We also need to align selected MPM across all roles to avoid future conflicts. Depends-On: https://review.opendev.org/c/openstack/openstack-ansible-openstack_hosts/+/930272 Depends-On: https://review.opendev.org/c/openstack/openstack-ansible-repo_server/+/929690 Depends-On: https://review.opendev.org/c/openstack/openstack-ansible-rabbitmq_server/+/930446 Change-Id: I480222993a63af41cfba65464f5b5c8585b2d4fd
9 lines
233 B
YAML
9 lines
233 B
YAML
---
|
|
|
|
upgrade:
|
|
- |
|
|
In order to align used Apache MPM across the board, Horizon default
|
|
MPM is switched from ``worker`` to ``event``.
|
|
A variable ``horizon_apache_mpm_backend`` was introduced to define
|
|
the MPM in use.
|