This role is no longer used, and was a long ago replaced with tempest.
Right now, there are still some pieces of code that overwrite the
tempest testrepository.subunit with the pingtest, and so, the best
approach right now is ensure that the validate-simple is not being
called anywhere.
Change-Id: I5a2993fa93c08f842e936c80b0adf4f9fb5c3b0f
In order to disable os_tempest any where, we define
the condition in featureset file, then that condition
gets updated at playbook level, due to missing |bool
condition, it is not able to disable it.
Adding the same will fix the issue.
Signed-off-by: Chandan Kumar (raukadah) <chkumar@redhat.com>
Change-Id: Iab00ea386cb3dd2df7932c11119057ff121f4bcb
This will avoid all the tasks under tempest.yml to be executed, since
those are only required when run os_tempest.
Related-Bug: #1911696
Change-Id: If546256da1e79d63c102590884e73db49c59e7ab
Remove the logic to choose between validate-tempest and os_tempest.
From now, only os_tempest should be executed.
Also other changes in order to make it work properly:
* Disable stackviz for CentOS-7 based distros.
* In Queens, Auth_url ends with //v3 leading to auth issue, replacing
it with /v3 fixes the issue.
* Generate clouds.yaml only for overcloud.
* Setting the CURL_CA_BUNDLE in order to avoid invalid ssl certificate
errors
Related-Bug: #1891372
Co-authored-by: Chandan Kumar (raukadah) <chkumar@redhat.com>
Signed-off-by: Arx Cruz <arxcruz@redhat.com>
Depends-On: https://review.opendev.org/#/c/759830/
Change-Id: I96808cc1d6c01391775a5f38d6dc27b6cf214afa
Today we have two roles that run tempest, the first one that is
deprecated, validate-tempest, and the new one os_tempest.
The problem right now is that, in order to deprecate validate-tempest,
we need to ensure a smooth transition, and centralize everything in one
single playbook, is the best way to do it.
Furthermore, move old jobs running on centos 7 require some workarounds
to create the clouds.yaml file, that today would be harder.
This patch consolidate both os_tempest and validate-tempest in one
single playbook, and so, in the future, it's easier to just remove the
calls from validate-tempest in one single place rather than keep
searching for all the places we use it.
Change-Id: I9991f6c0ea51a3bf537bdcafd4220a5a025cc026
In order to run tempest using os_tempest on undercloud,
we need to add undercloud-validate tag for the same.
Donot load skip list for undercloud as it is taken care
in featureset file itself in order to avoid breakage
against mistral project.
https://review.opendev.org/694719
Change-Id: Id80070fc6fa68e95ad111baee421779a76f67e02
Signed-off-by: Chandan Kumar (raukadah) <chkumar@redhat.com>
Once tempest run finishes using os_tempest role,
python_venv_build noop handler[1.] is called and the playbook
fails with tempest_install_method undefined var as it's scope
is only for that task when role is called.
Moving the var at the top of the play keeps the scope of the
var for the whole play.
Closes-Bug: #https://bugs.launchpad.net/tripleo/+bug/1838496
Change-Id: I58082b307cb5b3f9ff908c29c8e24b9b970561de
Signed-off-by: Chandan Kumar (raukadah) <chkumar@redhat.com>
It modifies multinode-validate playbook to use os_tempest
for running tempest and also pass required vars for the same.
Since os_tempest runs as a root user so we need clouds.yaml
in /root/.config/openstack/dir. For that two tasks were
add for the same.
Change-Id: I1f07fc996990273bfb8ef41cdcf7cb9a183a6d3c
Signed-off-by: Chandan Kumar (raukadah) <chkumar@redhat.com>
Some of the actions done at tempest like create a vm instance restart
some sidecar containers and they appear as failed containers since the
purpose of validate_services is to check them after deploy running
before tempest would be good enough.
Change-Id: I750e4d2ffc139433f7ec2a9a5c4adc6467ad19c9
Closes-Bug: #1824977
Sometimes container or service does not start, and this
doesn't make the CI fail. Until now, the failed containers
are listed in the /var/log/extras/ tree, but it's not
checked on a regular basis.
This patch intends to make a hard failure in case either
a service or a container doesn't start as expected.
Co-Authored-By: Cédric Jeanneret <cjeanner@redhat.com>
Related-Bug: #1816523
Change-Id: I001e2f27d2b562bb0be87c8eaadcf3622e530498
Makes those files conformant with current linting rules and avoids
linting errors when we need to toch them again.
Previous doing "pre-commit run -a" uncovered these errors, now is no
longer reporting any other errors.
Change-Id: Ie4cf229c8f11c2b55b323eac23c89483b26d3781
dstat async task with poll=0 should fire and forget, with 1 sec
max limit for the shell to run and send its process to bg.
The change includes:
- Ignore errors on dstat play.
- Enable dstat only when tempest runs.
- Kills dstat before parsing logs and creating charts.
- Remove dstat_count.
Closes-Bug: #1760917
Change-Id: I9ee407e8274f3ab6c3aea3650a25cd31772a81d6
Add undercloud-validate tag for tempest role to support running
tempest for undercloud.
Depends-On: I8f1d350e5e74ea77191255b7b086404b1db8ff00
Change-Id: I6b1ece53d6ff90874de9d09b3eaa731c8d582ae1
OVB playbooks are already in extras repository, so move multinode
playbook there too to keep them in one place.
Change-Id: Ie778d8893d0a92798dfada33260a656234d57350