Standardize API health check endpoints

Configure dedicated health check endpoints for Load Balancers

Standardize health check paths in OpenStack and Grafana
Ansible defaults to improve Load Balancer (HAProxy) reliability.

The 'httpchk' configuration for the modified roles now uses
dedicated health check endpoints (e.g., /healthcheck or
/api/health) and enforces an HTTP 200 OK expectation.
This ensures faster and more robust failure detection.

Additionally, by moving health checks away from generic
endpoints (such as OPTIONS /), application logs are now
much cleaner. The log files will no longer be flooded
every second with HAProxy probe entries like:

"OPTIONS / HTTP/1.0" 200 94 0.000954

NOTE: The following services are exceptions to this
standardization and still use the root endpoint (`/`)
for health checks due to the lack of a dedicated health
check middleware in their default API paste configuration:
**cyborg_api**, **masakari**, **mistral_api**, **nova_api**,
**octavia_api**, **placement**, **tacker**, **venus**, and **watcher**.

Updated roles:
* aodh
* barbican
* cinder
* cloudkitty
* cyborg
* designate
* etcd
* glance
* gnocchi
* grafana
* heat
* horizon
* influxdb
* ironic
* keystone
* loadbalancer
* magnum
* manila
* masakari
* mistral
* neutron
* nova
* nova-cell
* octavia
* opensearch
* placement
* rabbitmq
* skyline
* tacker
* trove
* venus
* watcher
* zun

Change-Id: Ibbb9fda45a98565ab12e784bbef80be787bd6638
Signed-off-by: Piotr Milewski <vurmil@gmail.com>

0 files changed, 0 insertions, 0 deletions