This is basically revert of the [1] which was revert of the [2]
but now it should not break our CI jobs.
In the configure_for_func_testing script openvswitch is installed
from source. We need to set proper flag (Q_BUILD_OVS_FROM_GIT) which
is used in Devstack to tell Devstack to install it from source and
not from packages.
This patch also removes flag BUILD_OVS_FROM_SOURCE from the
configure_for_func_testing file as it was only used in that file
and was actually duplicating the Q_BUILD_OVS_FROM_GIT option used also
in Devstack.
[1] https://review.opendev.org/c/openstack/neutron/+/824750
[2] https://review.opendev.org/c/openstack/neutron/+/824750
Change-Id: I35715a047d23ed87312afd294cc898de7c164583
For functional and fullstack jobs tools/ is in the irrelevant list, thus
even if configure_for_func_testing.sh is changed none of these jobs are
executed.
Change-Id: I3fb7e162c92b0316b16c8af92831b7050a9727d8
Related-Bug: #1957936
We were using devstack-tobiko-faults-centos in our periodic queue
but this job is broken and it's better to use job
'devstack-tobiko-neutron' which is kind of dedicated to test neutron
code.
It is also single node job, so will be faster and easier for the infra
resources :)
Change-Id: Ic3009bd98e3cf1e7a88d2cef02c178044e5478d1
Without this port binding fails with below error:-
Network <nw> is type of vxlan but agent <host> or mechanism
driver only support ['gre', 'local', 'flat', 'vlan'].
Also fix permissions of /opt/stack/devstack in ml2 ovs testing
documentation and added these files to irrelevant-files to skip
running functional jobs as these files are not used in those jobs.
Related-Bug: #1934466
Change-Id: I3ca2ea19bf5e316e580669caab4c607447034a11
It's not needed at all so if we avoid installation, especially from
sources, it may save us few minutes of the job's execution time.
It may also save some resources on the node.
To save a bit more time in the fullstack job's execution this patch also
disables compilation of the OVS from source in that job. We can use OVS
installed from packages provided by the distro instead.
Change-Id: Ic4b6740671e51f0d306967013e3d500f4d0cd6a5
This patch adds definition of the functional and fullstack jobs
with enabled support for FIPS [1].
Jobs are based on the Centos 8 stream as this disto allows to enable
FIPS support.
Jobs are added to the experimental queue for now.
This patch also makes some changes in the bindep and
configure_functional_tests role to make functional/fullstack tests
working on the Centos.
[1] https://csrc.nist.gov/publications/detail/fips/140/3/final
Co-Authored-By: Ade Lee <alee@redhat.com>
Change-Id: I582495826155740ad2660ee2a8717696b0393d26
After discussing this topic again during the PTG I spent some time
checking our scenario jobs which runs in the check and gate queues.
After analysis this patch proposes to:
* remove neutron-ovs-tempest-slow job from both check and gate queue as
slow tests are already run also in the
neutron-ovs-tempest-multinode-full job,
* remove neutron-ovn-tempest-slow job from both check and gate queue as
slow tests are already run also in the
neutron-ovn-tempest-ipv6-only job - of course this job is using IPv6
instead of IPv4 but I don't really think it's big issue in that case,
neutron-ovn-tempest-slow job was multinode job, unfortunately
neutron-ovn-tempest-ipv6-only is single node job and for now it isn't
possible to make ipv6-only job to be multinode job so we will keep it
like single node job and hopefully move to be multinode job when zuul
will provide required data in the job's inventory,
* move neutron-ovn-tempest-ovs-release and
neutron-ovn-tempest-ovs-release-ipv6-only jobs to periodic queue - I
think that running those tests once per day should be enough.
Additionally this patch removes definition of the neutron-ovs-tempest-slow
and neutron-ovn-tempest-slow jobs are those jobs aren't used anywhere now.
Change-Id: I657881c319d425470277885545240d6a8b66a1f6
Increased openstack-tox-py38 timeout in gate CI up to 1 hour.
Increased openstack-tox-py39 timeout in check CI up to 1 hour.
Closes-Bug: #1949704
Closes-Bug: #1949476
Change-Id: I415c1d67757354243855ac44862f245eb08887c8
The original timeout is 3660 for both openstack-tox-py38-arm64 and
openstack-tox-py39-arm64, change it to 4800.
Related-Bug: #1949476
Related-Bug: #1949704
Change-Id: I2b0d2c8b60af0ab7398f38151ff0f62e451e7e3e
CI job "neutron-ovs-rally-task" needs an admin endpoint. Since
[1] devstack does not create it by default. This patch enables it
in this job now.
[1]https://review.opendev.org/c/openstack/devstack/+/777345
Closes-Bug: #1949945
Change-Id: Ia88d70dd142c919eebc982c82ca583d220d96ac4
Add the following jobs to the experimental queue to test with
neutron-lib master:
- neutron-ovs-tempest-with-neutron-lib-master
- neutron-fullstack-with-uwsgi-with-neutron-lib-master
- neutron-functional-with-uwsgi-with-neutron-lib-master
Change-Id: I12c2381eef365f1249a3779685112cb682d752ee
In the OVN repo [1] there is no "master" branch anymore.
There is "main" instead and that patch reflects that change in
the Neutron repo.
[1] https://github.com/ovn-org/ovn
Change-Id: Ia666e78b748567a63484752c99856c46c22630ec
The CI job "neutron-ovn-tempest-ovs-master-fedora" should be the same
as "neutron-ovn-tempest-ovs-master" but executed in the Fedora nodeset.
Related-Bug: #1942913
Change-Id: I89700030dedab33c28baf95993206cf5cdd1e6d6
After I13fd2dc023b806ac74ab5c8d86cc99f807f0f702 the base job
tempest-multinode-full-py3 does not have the MinBwAllocationPlacementTest
tempest tests enabled by default. So enable it in
neutron-ovs-tempest-multinode-full not to lose any test covereage
Change-Id: I6967b55e8ce8d422422e943818145b718cd10f02
DVR-HA check job has become pretty stable after latest DVR fix [1].
Let's make it voting to prevent further DVR-HA regressions.
[1] https://review.opendev.org/c/openstack/neutron/+/803211
Change-Id: Ieac9a8bf9f01a18069fc419cf52ca7a0624fec32
This jobs is almost the same as tempest-slow-py3 since we switched
OVN to be default backend in Neutron. And that tempest-slow-py3 job
is used by many projects. So to avoid potential breaks of the gate for
other projects (like we did recently, see related bug for details)
let's make this job voting and gating.
As it is really used in many different projects as voting and gating
job already I don't think there is any issue with doing the same in
the Neutron gate.
Related-bug: #1936983
Related-bug: #1930402
Change-Id: I85d3830e9cc65162db846e4858871e1db547a04b
Only OVS agent uses os-ken library, makes sense that the CI job
testing it uses this backend.
Closes-Bug: #1933626
Change-Id: I8b2eb11dfae5bc67ee9c3629f609e4b461e0ad7c
All Neutron CI jobs (except for unit, functional and fullstack jobs),
have explicitly defined the network backend used:
- linuxbridge
- ovs
- ovn
That was discussed and approved during the Neutron CI meetings [1].
[1]https://meetings.opendev.org/meetings/neutron_ci/2021/neutron_ci.2021-06-15-15.00.log.html
Change-Id: I036924f13b77dd5957ad5e60b9f536562deb9f05
It is set to true by default in base jobs. Also e.g. in
functional job for neutron-dynamic-routing it is set to True so
it should be fine to use default value in the neutron-functional job as
well.
Related-bug: #1932483
Change-Id: I2c8392628dd63ca612c6d8428daab66724e88c20
It wasn't enabled for neutron-tempest-iptables_hybrid and
neutron-ovn-tempest-ovs-master-fedora and that patch
enables it in both of those jobs.
Change-Id: I73240679055b943caf46646440f9c325cf21faf3
Added "nova:live_migration_events" flag to subnode neutron.conf file in
"neutron-tempest-multinode-full-py3" CI job. That flag was missing in
the patch implementing this feature [1].
[1]https://review.opendev.org/c/openstack/neutron/+/766277
Change-Id: Idc938a1dc9de3ad77f558df4f4fb4ae5c3de3d21
Related-Bug: #1901707
After switch of the default Neutron backend to ML2/OVN
in devstack we need to set ML2/OVS for proper multinode
jobs explicitly.
This patch configures ML2/OVS as a backend in the
'neutron-tempest-slow-py3', 'neutron-tempest-multinode-full-py3',
'neutron-tempest-dvr-ha-multinode-full' jobs.
Change-Id: I4d59a42db3e50cb1cb156a46cac708425f641b25
After switch of the default Neutron backend to the ML2/OVN
in the Devstack we had 2 rally jobs running actually with the
same Neutron backend.
This patch configures explicitly ML2/OVS as a backend in the
'neutron-rally-task' job so it is like it was before the change in
the Devstack repo.
In order to fix the CI, this patch also skips "test_models_sync" test
cases. Once [1] is fixed and released, we'll be able to create SQL
tables specifying the "_constraints_included" flag. When creating a
table, this flag should be set to "False" in order to be able to set
the "index" and "unique" parameters for each column in the table
definition, as we do now.
"test_update_router_admin_state" is excluded from
"neutron-ovn-tempest-slow" until LP#1890445 is fixed.
Set "test_restart_wsgi_on_sighup_multiple_workers" as unstable.
CI job "neutron-ovn-tempest-slow" has been set to non-voting until
LP#1930402 is fixed.
Related-Bug: #1929518
Related-Bug: #1890445
Related-Bug: #1930402
Related-Bug: #1930367
[1]https://github.com/sqlalchemy/alembic/issues/848
Change-Id: I38ed7931169434fecbbefa3762f51a522ef3c509