951c6f5f02
* Added Blockstorage Datasets * Added per-volume-type min and max size configurability. * Added behavior method for getting min and max size of configured volume types * Added 'default' min and max volume type properties to the volumes_api config, which deprecate the older min and max volume_size properties. * Added volume_type_properties property in volumes_api config * Added Image, Flavor, and Volume Type filters and filter mode options to the volumes_api config. * Added configuration option for toggling cross-type snapshot restores (since this will soon be configurable in cinder) Change-Id: I8a4ba2de4dda66c61abf14938dfd14635ff68456
CloudCAFE, An Open CAFE Implementation for OpenStack
_ _ _
( ` )_
( ) `) _
(____(__.___`)__)
( (
) )
.........
| |___
| |_ |
| :-) |_| |
| |___|
|_______|
=== CloudCAFE ===
= An Open CAFE Implementation =
CloudCAFE is an implementation of the Open CAFE Framework specifically designed to test deployed versions of OpenStack. It is built using the Open CAFE Core.
Supported Operating Systems
CloudCAFE has been developed primarily in Linux and Mac environments, however it supports installation and execution on Windows.
Installation
CloudCAFE can be installed with pip from the git repository after it is cloned to a local machine.
- First follow the README instructions to install Open CAFE Core.
- Clone this repository to your local machine.
- CD to the cloned cloudcafe repository directory.
- Run
pip install . --upgradeso that pip will auto-install all other dependencies.
Configuration
CloudCAFE works in tandem with the Open CAFE Core cafe-runner. This
installation of CloudCAFE includes a reference configuration for each of the CloudCAFE supported OpenStack products.
Configurations will be installed to <USER_HOME>/.cloudcafe/configs/<PRODUCT>.
To use CloudCAFE you will need to create/install your own configurations based on the reference configs pointing to your deployment of OpenStack.
At this stage you will have the Open CAFE Core engine and the CloudCAFE Framework implementation. From this point you are ready to:
-
Write entirely new tests using the CloudCAFE Framework,
or...
-
Install the CloudRoast Test Repository, an open source body of OpenStack automated tests written with CloudCAFE that can be executed or extended.
Logging
CloudCAFE leverages the logging capabilities of the CAFE Core engine. If tests are executed with the built-in
cafe-runner, runtime logs will be output to <USER_HOME>/.cloudcafe/logs/<PRODUCT>/<CONFIGURATION>/<TIME_STAMP>.
In addition, tests built from the built-in CAFE unittest driver will generate csv statistics files in
<USER_HOME>/.cloudcafe/logs/<PRODUCT>/<CONFIGURATION>/statistics for each and ever execution of each and every test
case that provides metrics of execution over time for elapsed time, pass/fail rates, etc.
Basic CloudCAFE Package Anatomy
Below is a short description of the top level CloudCAFE Packages.
-
cloudcafe
This is the root package for all things CloudCAFE. -
common
Contains modules that extend the CAFE Core engine specific to OpenStack. This is the primary namespace for tools, data generators, common reporting classes, etc. -
identity
OpenStack Identity Service plug-in based on CAFE Core extensions. -
compute
OpenStack Compute plug-in based on CAFE Core extensions. -
blockstorage
OpenStack Block Storage plug-in based on CAFE Core extensions. -
objectstorage
OpenStack Object Storage plug-in based on CAFE Core extensions.