Grafana dashboard "status" pie charts wrong with HAProxy 2.4+

[Michael Johnson]

HAProxy made some breaking changes to how the "status" of the load balancer components are reported via the Prometheus exporter:

MAJOR: contrib/prometheus-exporter: move health check status to labels
MAJOR: contrib/prometheus-exporter: move ftd/bkd/srv states to labels

This causes the "status" pie charts in the Octavia Grafana dashboard to report incorrectly when using an Amphora image with HAProxy 2.4 or newer.

The Octavia exporter help text should also be updated to reflect this change.

[OpenStack Infra (hudson-openstack)]

Fix proposed to branch: master
Review: https://review.opendev.org/c/openstack/octavia/+/919776

[OpenStack Infra (hudson-openstack)]

Reviewed: https://review.opendev.org/c/openstack/octavia/+/919776
Committed: https://opendev.org/openstack/octavia/commit/06fccc6db54c91f61cef7c2db11e52fed99c585a
Submitter: "Zuul (22348)"
Branch: master

commit 06fccc6db54c91f61cef7c2db11e52fed99c585a
Author: Michael Johnson <email address hidden>
Date: Wed May 15 20:46:09 2024 +0000

    Update the Grafana dashboard for HAProxy 2.4+

    HAProxy made some breaking changes to the metrics format for the
    listener, pool, and memeber status metrics in version 2.4 and newer.
    This patch updates the dashboard to use the new metric format and
    updates the associated help text in the Prometheus exporter content.

    This patch also corrects some typos for idle connections and SSL reuse.

    Closes-Bug: #2065793
    Change-Id: I6ba54e93f23961ede965dd1b4b3f58f73f7c1188