bed597363a
.coverage and cover are generated by tox's cover task. So we should ignore them. Change-Id: Ica14d566fcd7bd31fea87471178db882add7813a |
||
---|---|---|
doc/source | ||
etc | ||
tempest | ||
tools | ||
.gitignore | ||
.gitreview | ||
.mailmap | ||
.testr.conf | ||
HACKING.rst | ||
LICENSE | ||
openstack-common.conf | ||
README.rst | ||
requirements.txt | ||
run_tempest.sh | ||
run_tests.sh | ||
setup.cfg | ||
setup.py | ||
test-requirements.txt | ||
tox.ini |
Tempest - The OpenStack Integration Test Suite
This is a set of integration tests to be run against a live OpenStack cluster. Tempest has batteries of tests for OpenStack API validation, Scenarios, and other specific tests useful in validating an OpenStack deployment.
Design Principles ----------Tempest Design Principles that we strive to live by.
- Tempest should be able to run against any OpenStack cloud, be it a one node devstack install, a 20 node lxc cloud, or a 1000 node kvm cloud.
- Tempest should be explicit in testing features. It is easy to auto discover features of a cloud incorrectly, and give people an incorrect assessment of their cloud. Explicit is always better.
- Tempest uses OpenStack public interfaces. Tests in Tempest should only touch public interfaces, API calls (native or 3rd party), public CLI or libraries.
- Tempest should not touch private or implementation specific interfaces. This means not directly going to the database, not directly hitting the hypervisors, not testing extensions not included in the OpenStack base. If there is some feature of OpenStack that is not verifiable through standard interfaces, this should be considered a possible enhancement.
- Tempest strives for complete coverage of the OpenStack API and common scenarios that demonstrate a working cloud.
- Tempest drives load in an OpenStack cloud. By including a broad array of API and scenario tests Tempest can be reused in whole or in parts as load generation for an OpenStack cloud.
- Tempest should attempt to clean up after itself, whenever possible we should tear down resources when done.
- Tempest should be self testing.
Quickstart
To run Tempest, you first need to create a configuration file that will tell Tempest where to find the various OpenStack services and other testing behavior switches.
The easiest way to create a configuration file is to copy the sample
one in the etc/
directory :
$> cd $TEMPEST_ROOT_DIR
$> cp etc/tempest.conf.sample etc/tempest.conf
After that, open up the etc/tempest.conf
file and edit
the configuration variables to match valid data in your environment.
This includes your Keystone endpoint, a valid user and credentials, and
reference data to be used in testing.
Note
If you have a running devstack environment, tempest will be
automatically configured and placed in /opt/stack/tempest
.
It will have a configuration file already set up to work with your
devstack installation.
Tempest is not tied to any single test runner, but testr is the most
commonly used tool. After setting up your configuration file, you can
execute the set of Tempest tests by using testr
:
$> testr run --parallel
To run one single test :
$> testr run --parallel tempest.api.compute.servers.test_servers_negative.ServersNegativeTestJSON.test_reboot_non_existent_server
Alternatively, you can use the run_tempest.sh script which will create a venv and run the tests or use tox to do the same.
Configuration
Detailed configuration of tempest is beyond the scope of this document. The etc/tempest.conf.sample attempts to be a self documenting version of the configuration.
The sample config file is auto generated using the script: tools/generate_sample.sh
The most important pieces that are needed are the user ids, openstack endpoints, and basic flavors and images needed to run tests.
Common Issues
Tempest was originally designed to primarily run against a full OpenStack deployment. Due to that focus, some issues may occur when running Tempest against devstack.
Running Tempest, especially in parallel, against a devstack instance may cause requests to be rate limited, which will cause unexpected failures. Given the number of requests Tempest can make against a cluster, rate limiting should be disabled for all test accounts.
Additionally, devstack only provides a single image which Nova can use. For the moment, the best solution is to provide the same image uuid for both image_ref and image_ref_alt. Tempest will skip tests as needed if it detects that both images are the same.
Unit Tests
Tempest also has a set of unit tests which test the tempest code itself. These tests can be run by specifing the test discovery path:
$> OS_TEST_PATH=./tempest/tests testr run --parallel
By setting OS_TEST_PATH to ./tempest/tests it specifies that test discover should only be run on the unit test directory. The default value of OS_TEST_PATH is OS_TEST_PATH=./tempest/test_discover which will only run test discover on the tempest suite.
Alternatively, you can use the run_tests.sh script which will create a venv and run the unit tests. There are also the py26, py27, or py33 tox jobs which will run the unit tests with the corresponding version of python.
Python 2.6
Tempest can be run with Python 2.6 however the unit tests and the gate currently only run with Python 2.7, so there are no guarantees about the state of tempest when running with Python 2.6. Additionally, to enable testr to work with tempest using python 2.6 the discover module from the unittest-ext project has to be patched to switch the unittest.TestSuite to use unittest2.TestSuite instead. See: