History

Gael Chamoulaud (Strider) 4db92ba2ca Convert all of the fedora28 container usage to CentOS8 This change moves all of the fedora28 container images to centos8 to ensure we're maximizing coverage in all of our roles with OS images that represent real deployments. While fedora28 was useful to exercise RHEL8 code path's while there was no RHEL/CentOS8 container image, we can now use the upstream CentOS8 image ensuring we're getting a true, representative, environment. The following molecule test has been moved to non-voting as it is facing some issues. A follow up patch will try to fix them ASAP. - check_network_gateway (facing an issue when Ansible is trying to discover the python interpreter used in CentOS8 Docker Image) Related-Bug: #1894882 Change-Id: I02f68b78fdf73a60443d8ea5d2dcb46e5ad67e25 Signed-off-by: Gael Chamoulaud (Strider) <gchamoul@redhat.com> (cherry picked from commit `4fc31e06dc`) Signed-off-by: Gael Chamoulaud (Strider) <gchamoul@redhat.com>		2020-09-09 14:05:56 +02:00
..
defaults	Add haproxy role	2019-03-15 10:41:53 +01:00
molecule/default	Convert all of the fedora28 container usage to CentOS8	2020-09-09 14:05:56 +02:00
tasks	Add haproxy role	2019-03-15 10:41:53 +01:00
vars	Add haproxy role	2019-03-15 10:41:53 +01:00
README.md	Add haproxy role	2019-03-15 10:41:53 +01:00

haproxy

An Ansible role to check if the HAProxy configuration has recommended values.

Requirements

This role requires an Up and Running Overcloud

config_file: '/var/lib/config-data/puppet-generated/haproxy/etc/haproxy/haproxy.cfg'
global_maxconn_min: 20480
defaults_maxconn_min: 4096
defaults_timeout_queue: '2m'
defaults_timeout_client: '2m'
defaults_timeout_server: '2m'
defaults_timeout_check: '10s'

No dependencies

- hosts: undercloud
  roles:
     - { role: haproxy }

Apache

Red Hat TripleO Validations Team.