The tempest log can help developers analyze tests but
there is no option in run_tests.sh to enable logging output now.
This adds a logging configuration file and options in run_test.sh
and enables developers to get the tempest log more easily.
Implements: blueprint add-logging-configuration
Change-Id: Iee68a34f5771f1bff88110d95538a5b43103ced9
This patch fixes a typo of getopt's arguments. If the second 's' option letter
is not 'S', -S option is not recognized.
Fix bug 1161921
Change-Id: I0b5a372e7257b9319b350131303c1a9f51b57eb5
Both the run_tests.sh and tox.ini using the same check_source.sh for
style and basic lint testing.
Adding a pyflakes based unused import test.
Removing 2 unused imports from
tempest/tests/network/test_network_basic_ops.py.
Change-Id: I5b4c8b945495751ac68fc64ed5ebf1cf441a5a31
Change I7da0226a6621ed7120b00eafd8d0ce75c1c2a67b didn't add
the new options to the get options if statement. The option
wasn't recognized and would just print the script usage. This
commit corrects the behavior.
Change-Id: If5d4496a3003d5f5a7016749679a82960894d506
This commit adds an update option to run_tests.sh. This option
is used to rerun install_venv.py on an already installed venv.
This will then just update out of date packages with pip.
Change-Id: I7da0226a6621ed7120b00eafd8d0ce75c1c2a67b
* Ability to pass arbitrary argument to nosetests
* Ability to run the script from any cwd
* Stricter argument validation
Change-Id: I675f30d6faf4ac9cfb39cb43400ae6cb58b68931
The "-s" option in run_tests.sh is shared by
"--no-site-packages" and "--smoke".
Proposing "-n" for --no-site-packages.
Fixes: LP bug #1111748
Change-Id: I1784107994a0ddc50aa47608250092429f9a2764
This adds support for testing tempest's coverage on nova.
tools/tempest_coverage.py will enable coverage reporting in
nova to be started and stopped. It also performs coverage report
generation.
When it is enabled from run_tests with '-c' or '--nova_coverage'
a single text report file will be generated per nova service.
Implements: blueprint tempest-coverage-reporting
Change-Id: I00a52fb013c5a7a66a2317dbd5359a22d35bdb29
Signed-off-by: Matthew Treinish <treinish@linux.vnet.ibm.com>
Fix all the one line docstring errors in T401 and T402, now
we no longer ignore any of our own rules in hacking.py
Fix run_tests.sh to pick up the 'stress' directory, which tox
does for pep8. Additional fixes to the stress tests for T4* tests
Change-Id: Ie569a924e8eb355afbbd9b244f77dec34061b5cb
Because this was originally based on the nova hacking.py
the error codes were all N*. But now that we are owning it
in tempest they should be T*, as we're diverging.
Change the ignores to follow the errors that we're ignoring.
Change-Id: I7b23325da41a3942ae51ee3d29a49f54c75c57f8
add new rule for hacking.py to catch and prevent docstrings on
test_ functions in the tempest/test directory. Found 2 oddballs
that I missed on the last conversion.
update ./run_tests.sh to not ingore all N4* errors (basically all
our custom rules), 401 and 402 still ignored because we actually
have a lot of errors on both of those.
Change-Id: I894d181433b7dcd1a9f4429f1a463be75018f05e
Currently just running pep8 with ./run_test.sh -p is not
using the venv because the wrapper definition was occurring
after tools/hacking.py was run. This moves the pep8 check to
after the venv setup to fix this issue.
Change-Id: Ie9fd2a427d8e8875d0cb0db885ea442b2d29ae73
Currently when running Tempest tests with run_tests.sh a venv
isn't being used. This leaves tempest at the mercy of whatever
required package versions are installed on the system.
This patch takes install_venv.py and with_venv.sh from nova and
integrates them into run_tests to ensure that the correct
dependencies versions from pip-requires and test-requires are
being used.
Change-Id: I4bf4a02890a33c4034e4493d1763ed4019fdf46e
This adds hacking.py, based on the one in nova, to wrap around
pep8 and enforce more stringent checks on tempest. This adds
the rules that nova is currently enforcing on incoming commits.
The rules that pep8 is ignoring are copied from the pep8 check on
jenkins.
The HACKING file was also updated to indicate the inclusion
of the new rules that are being enforced.
Change-Id: I5c85829ee15d4264d8de2d9f0207d9dd3dfbbff2
* Try /etc/tempest DIR if the config file not found
* Exit immediately if the config file not found
* Remove pika from dependencies
* tempest named starter script (pep8 removed run_tests.sh)
Change-Id: I1f0b1fb061a265c5f8e39627efd26decdc61ef63
we were always adding 'tempest' to the list for tests in
run_tests.sh, which didn't let you use the wrapper for running
a single tempest test. This change should fix that.
Probably could be cleaner, but my bash-fu is weaksause.
Change-Id: I5a3600b14214dd38be505481dad5ab72998df386
* Splits out the "manager" class into its own file (at least
for now to keep the code change size down initially)
* Adds base manager classes for Fuzz and Default clients
* Adds base test case class and a derived base SmokeTest class
* Adds smoke test for basic server operations in compute
* Adds non-smoke test for advanced server operations in compute
* Adds Whitebox base test case class
* New basic db-checking whitebox tests for servers and images
This change builds upon the proposed refactoring to the Tempest Manager
and base test classes as per https://review.openstack.org/#/c/7069
Change-Id: I12125fffb725cad3a4fef3134c83e55437529252
* Allows a tenant/user to be created by the base test
class, allowing isolation of the tests. The base
compute test case class now controls what users
and tenants get created for the test case
* Moves identity admin stuff into appropriate directory
* Removes _multiprocess_shared_ to force setUpClass to
run for every test case.
* Adds setup.cfg for use with openstack.nose_plugin
* run_test.sh refactoring to make running smoketests easier
Change-Id: I3ea4c44a8c194d7ae2833b5443df7678924bcd5d
agnostic
Refactored some of the shell scripts to handle
being run from another directory, as well as to
not force the use of 2.6
edit: format the comments
Change-Id: Id92b9dc82743598b7c570e431cad423c8a2ee95d
