Queens Series Release Notes¶
Added new horizon::dropdown_max_items parameter that configures the DROPDOWN_MAX_ITEMS config option in local_settings.py. Default value is set to 30 like the Horizon default is.
The wsgi_processes and wsgi_threads options that is passed to horizon::wsgi::apache can now be set in ::horizon class.
Added a new parameter horizon::manage_memcache_package that can be used to determine if you want the python memcache library installed if the cache backend is set to using memcache.
Fixed a bug where having ssl and redirection enabled combined with a empty root_url did not redirect. If you are setting listen_ssl and ssl_redirect to true it will always redirect to https no matter what your root_url is.
Fixed a bug where the http vhost would not redirect to the custom root_url when setting listen_ssl to true and ssl_redirect to false.
Add the config option “HORIZON_IMAGES_UPLOAD_MODE”. now it is possible in puppet to configure the option with “off”, “legacy” and “direct”.
Add parameter to wsgi::apache to allow to overwrite and/or add additional wsgi process options.
Updated the cap on the puppetlabs-apache module to include the 2.x series.
Horizon has a global override mechanism available to perform customizations. This change adds customization_module key to HORIZON_CONFIG dictionary in local_settings.py. The value of the parameter should be a string with the path to the module containing modifications in dotted python path notation.
Added parameter to horizon class to allow configuring LAUNCH_INSTANCE_DEFAULTS.
The horizon_ca option is now optional. This implies that you could in theory use the CA configured in the global httpd SSL configuration (ssl.conf) or simply not using this value.