Age | Commit message (Collapse) | Author | Files | Lines |
|
It completes "min_computes_nodes is also used as max values".
Change-Id: I72c67e17015bc35ac5e14f20839cf74353695373
Signed-off-by: Cédric Ollivier <cedric.ollivier@orange.com>
|
|
Change-Id: Ica6455e3467ef2695d489170bd8f94553a6fdf06
Signed-off-by: Cédric Ollivier <cedric.ollivier@orange.com>
|
|
Its max value is now 10 according to the default quota.
Change-Id: Ie7bfcb224e821c15d38d0cba1d9f080548613e79
Signed-off-by: Cédric Ollivier <cedric.ollivier@orange.com>
|
|
It adds the remaining services which didn't work before.
It may be needed for CNTT (see security discussion in vevent).
Change-Id: I36b3ad8bbfc8ffdcc427c3d97b92c78da27915c0
Signed-off-by: Cédric Ollivier <cedric.ollivier@orange.com>
|
|
The regexes will be updated in a second change (chicken and egg
between Functest and Releng)
Change-Id: I4273a80799730fe30c941a7b2966cfee46f24d4a
Signed-off-by: Cédric Ollivier <cedric.ollivier@orange.com>
|
|
It takes latest updates into account [1][2].
[1] http://artifacts.opnfv.org/functest/4W4YF3954M7H/functest-opnfv-functest-smoke-cntt-latest-tempest_neutron_cntt-run-85/tempest_neutron_cntt/tempest-report.html
[2] http://artifacts.opnfv.org/functest/K4DV91F2Y9V0/functest-ovn-opnfv-functest-smoke-cntt-latest-tempest_neutron_cntt-run-76/tempest_neutron_cntt/tempest-report.html
Change-Id: Ia9b57e708522926fde9f8c44e6a0e936e97f304f
Signed-off-by: Cédric Ollivier <cedric.ollivier@orange.com>
|
|
It completes the previous change "Allow listing hypervisors in VMTP" [1].
[1] https://gerrit.opnfv.org/gerrit/c/functest/+/70359
Change-Id: I12a19fd473076cfc3262f096cac1615c183710b1
Signed-off-by: Cédric Ollivier <cedric.ollivier@orange.com>
|
|
It's useless from a CNTT point of view.
It remains in the classical IaaS verification to ease creating
Functest Jenkins Jobs.
Change-Id: Icb13ab7d0e43b4cf968cab4041b6827c11c20cbc
Signed-off-by: Cédric Ollivier <cedric.ollivier@orange.com>
|
|
It allows listing the compute under tests via VMTP_HYPERVISORS.
They are no additional log in Functest and the note in VMTP is still
valid [1].
[1] https://opendev.org/x/vmtp/src/branch/master/vmtp/vmtp.py#L833
Change-Id: Ic9fbda3901d796c0ed94e42e2a14c468598f4311
Signed-off-by: Cédric Ollivier <cedric.ollivier@orange.com>
|
|
Running them in parallel of rally full and jobs asks for a increase.
Change-Id: Ib2369d384ac839fb46031411e1985ac402da0c1e
Signed-off-by: Cédric Ollivier <cedric.ollivier@orange.com>
|
|
https://github.com/openstack/requirements/commit/c2f72e26a42eceb4b64f80d04db8253a4fbdb793
Change-Id: I0318c8614b09dc7cff1ffba4b813e55483df7eb3
Signed-off-by: Cédric Ollivier <cedric.ollivier@orange.com>
|
|
Neutron leverages on the upstream Rally task in Stein and newer.
Leveraging on the upstream task may fix the side effects which
sometimes occured on Vmtp and Shaker [1] and filled better the CNTT
model.
[1] https://build.opnfv.org/ci/view/functest/job/functest-hunter-daily/574/
Change-Id: I5ee2e3a3ca956cf57365bf4579534177e87dc08b
Signed-off-by: Cédric Ollivier <cedric.ollivier@orange.com>
|
|
Change-Id: I5cfd3e370ef16785eeceaafefa18b45d2221c04b
Signed-off-by: Cédric Ollivier <cedric.ollivier@orange.com>
|
|
Else cross compilation raises travis-ci.org's timeouts [1]
[1] https://travis-ci.org/github/collivier/functest/jobs/698227192
Change-Id: I985321ac68a7a94cb8483d831f03222d44b3be57
Signed-off-by: Cédric Ollivier <cedric.ollivier@orange.com>
|
|
It stops leveraging on wget to get constraints as it fails since
opendev was created. Then functest and requirements are no longer
removed for functest-core which simplifies all child Dockerfile.
It also fix the current incompatibilities in OpenStack constraints
ERROR: cliff 3.2.0 has requirement cmd2!=0.8.3,<0.9.0,>=0.8.0, but you'll have cmd2 1.1.0 which is incompatible.
https://github.com/openstack/requirements/blob/master/upper-constraints.txt#L230
https://github.com/openstack/requirements/blob/master/upper-constraints.txt#L416
Change-Id: I7c044208ef26ad047fb841710b422acb6c9f0320
Signed-off-by: Cédric Ollivier <cedric.ollivier@orange.com>
|
|
Attaching an encryted volume is optional in CNTT [1].
This test fails if the volume type already exists and if it runs
twice in parallel [2]. As this single test doesn't fullfil the Functest
requirements, it's skiped in the classical verification as well.
[1] https://github.com/cntt-n/CNTT/blob/master/doc/ref_arch/openstack/chapters/chapter05.md
[2] https://storyboard.openstack.org/#!/story/2007804
Change-Id: Ib6fc843d71704aab1bfc536133544eb3fa4f3e72
Signed-off-by: Cédric Ollivier <cedric.ollivier@orange.com>
|
|
Trusts redelegation is supported by Train [1][2] which forces to skip
RemoteStack as CNTT Baldy is designed for Pike (the support is
incomplete in case of trust [3]).
[1] https://docs.openstack.org/releasenotes/heat/train.html#relnotes-13-0-0-stable-train
[2] https://github.com/openstack/heat/commit/e377658586c737150dad1bfd80b7b2267d46be51
[3] https://bugs.launchpad.net/heat/+bug/1701498
Change-Id: Ib4f66170b9b1f50b8224705edb4820e5a4cb0167
Signed-off-by: Cédric Ollivier <cedric.ollivier@orange.com>
|
|
Change-Id: I0fae7487955895bf8f41046043425849f8fd0bf2
Signed-off-by: Cédric Ollivier <cedric.ollivier@orange.com>
|
|
It mostly takes into account the last changes in test case names and
new cntt-related test cases.
Change-Id: I271fd02095b57f3b19b5ff95299fb4b257163c1f
Signed-off-by: Cédric Ollivier <cedric.ollivier@orange.com>
|
|
Change-Id: Ib53b9f97f30bf86d33857910ab9bffa1ce0d1e9e
Signed-off-by: Cédric Ollivier <cedric.ollivier@orange.com>
|
|
|
|
|
|
It takes all CNTT changes into account.
Change-Id: Ib6ddbd12695266fc78fd8ae4cce451bf8a4b95ad
Signed-off-by: Cédric Ollivier <cedric.ollivier@orange.com>
|
|
It seems useless and then it limits the diff to Rally.
Change-Id: Ib745c7aad20ab5ab7fd4f8a5c7deb9d4aff30881
Signed-off-by: Cédric Ollivier <cedric.ollivier@orange.com>
|
|
https://alpinelinux.org/posts/Alpine-3.12.0-released.html
Change-Id: Ie7c72abd10d70b90d65f260d595c0ca47b69a8c0
Signed-off-by: Cédric Ollivier <cedric.ollivier@orange.com>
|
|
http://artifacts.opnfv.org/functest/Q3047D5AS3HJ/functest-ovn-opnfv-functest-smoke-cntt-latest-tempest_neutron_cntt-run-40/tempest_neutron_cntt/tempest-report.html
http://artifacts.opnfv.org/functest/03PRDE6TZC14/functest-opnfv-functest-smoke-cntt-latest-tempest_neutron_cntt-run-49/tempest_neutron_cntt/tempest-report.html
Change-Id: I4be6d2f58e50572a3e8b90ede0055f36bc1d6886
Signed-off-by: Cédric Ollivier <cedric.ollivier@orange.com>
|
|
3 seems good enough and stable which allows trying 4 as before
Change-Id: Id1f17ebf73f9faa13ed3ceb327ccb4ecf88c2bf2
Signed-off-by: Cédric Ollivier <cedric.ollivier@orange.com>
|
|
They pass in parallel if the bandwidth is good enough.
https://build.opnfv.org/ci/view/functest/job/functest-latest-daily/679/
Change-Id: I2f466f7f25898d8af3004da1d3ef2208e4410f7c
Signed-off-by: Cédric Ollivier <cedric.ollivier@orange.com>
|
|
The debug logs works only vs py3.8 and breaks backport to Iruya.
Change-Id: Ie91a8a876e263a83b6ae3fbfb5bb05644b87b970
Signed-off-by: Cédric Ollivier <cedric.ollivier@orange.com>
|
|
Change-Id: Id5059a06447357f4c9b058bad374ed6cbe4d742c
Signed-off-by: Cédric Ollivier <cedric.ollivier@orange.com>
|
|
http://artifacts.opnfv.org/functest/KJSOWXFF5CIQ/functest-opnfv-functest-smoke-cntt-latest-tempest_neutron_cntt-run-41/tempest_neutron_cntt/tempest-report.html
Change-Id: I04f0c3ae695f17e2bd56add11410e24ee25967d2
Signed-off-by: Cédric Ollivier <cedric.ollivier@orange.com>
|
|
It's still failing.
Change-Id: I20fdf5a8e05bccc30a760934f6c0e3cac16d763c
Signed-off-by: Cédric Ollivier <cedric.ollivier@orange.com>
|
|
Change-Id: I1f3cb636813b9c3c10b5c829e35bbdeea02a318c
Signed-off-by: Cédric Ollivier <cedric.ollivier@orange.com>
|
|
Change-Id: Id93da8c0fc710e1491adeb2b42e6da99974a0580
Signed-off-by: Cédric Ollivier <cedric.ollivier@orange.com>
|
|
http://artifacts.opnfv.org/functest/1NCC8YUFKE7X/functest-opnfv-functest-smoke-cntt-latest-tempest_slow_cntt-run-64/tempest_slow_cntt/tempest-report.html
Change-Id: I196421d96ada3ec1eb8fc718f72c3e796a0bd3b4
Signed-off-by: Cédric Ollivier <cedric.ollivier@orange.com>
|
|
Change-Id: I61cb70a74179dea6a02e81989695325fe6331e10
Signed-off-by: Cédric Ollivier <cedric.ollivier@orange.com>
|
|
As detected by Orange's RC Field trial, Ceph doesn't support this
feature before Rocky [1].
"AttributeError: 'RBDDriver' object has no attribute 'unmanage_snapshot'"
manage_snapshot testing is fully passing in Functest SUT because CNTT
is verified vs OpenStack Rocky and newer.
Ceph is selected by RI as backend and is widely reused by the ecosytem.
It's similar to [2].
[1] https://bugs.launchpad.net/cinder/+bug/1645288
[2] https://gerrit.opnfv.org/gerrit/c/functest/+/69888
Change-Id: I3b7f88f5bed9a47a1e26e16bbbf74fea24ca76c2
Signed-off-by: Cédric Ollivier <cedric.ollivier@orange.com>
|
|
It sometimes fails in our gates and it seems unverified in OpenStack [1].
"Archive policy gabbilive is still in use"
[1] http://artifacts.opnfv.org/functest/DET32CBF0L0F/functest-opnfv-functest-smoke-jerma-tempest_telemetry-run-29/tempest_telemetry/tempest-report.html
Change-Id: Id5e5ad5ff0f7bade7932a085df28cc35b0ac7984
Signed-off-by: Cédric Ollivier <cedric.ollivier@orange.com>
|
|
http://artifacts.opnfv.org/functest/R1927JRCFJR5/functest-opnfv-functest-smoke-cntt-latest-tempest_full_cntt-run-62/tempest_full_cntt/tempest-report.html
Change-Id: Ia4b5bbdb92e7877d81acdcf9c4f3a9745badfb36
Signed-off-by: Cédric Ollivier <cedric.ollivier@orange.com>
|
|
ServerActionsTestJSON.test_reboot_server_hard [1] sometimes fails in
gates. We may increase timeouts if it stills fails.
[1] http://artifacts.opnfv.org/functest/LIQH38CPR30X/functest-opnfv-functest-smoke-hunter-tempest_full-run-649/tempest_full/tempest-report.html
Change-Id: Ia27c0faead1df5c2b33a3e1a272e5a1010a5a83e
Signed-off-by: Cédric Ollivier <cedric.ollivier@orange.com>
(cherry picked from commit 7ccc1b0793e32cad3912e63bbca971d5c3ecd724)
|
|
CNTT selects Pike as baseline and Hunter matches Rocky.
Change-Id: I8a1aa8385adce1ebb9d74bcb897effaf5b434e5a
Signed-off-by: Cédric Ollivier <cedric.ollivier@orange.com>
(cherry picked from commit 82ceefd8f1609508a476aa2e4573e59926653165)
|
|
It would allow debugging our Rally failures.
Change-Id: I749a040fe76261e997d3a8dff4965867d54f7c42
Signed-off-by: Cédric Ollivier <cedric.ollivier@orange.com>
|
|
http://artifacts.opnfv.org/functest/V0A7B0J1PIRL/functest-opnfv-functest-smoke-cntt-latest-tempest_neutron_cntt-run-17/tempest_neutron_cntt/tempest-report.html
Change-Id: I95dc0d8dec370dedb501878cbedfe5567ab86afd
Signed-off-by: Cédric Ollivier <cedric.ollivier@orange.com>
|
|
It updates pyshaker to 1.3.4 and adds fio which should be useful for
the next disk benchmarking testcases.
Change-Id: I662599745590eae71922ea5ef3f3876ad1019387
Signed-off-by: Cédric Ollivier <cedric.ollivier@orange.com>
|
|
Change-Id: I362b70cdfe753f9c57f30d61662e9b08564ebfe9
Signed-off-by: Cédric Ollivier <cedric.ollivier@orange.com>
|
|
req.inf.ntw.16 is a MUST requirement.
Fixes cntt-n/CNTT#1556
Change-Id: Id48595f5f4710ea9cbf31edaa4ae0ed31f5278ba
Signed-off-by: Cédric Ollivier <cedric.ollivier@orange.com>
|
|
Change-Id: If69f74031c218e513464fda5be4e6fb86266292a
Signed-off-by: Cédric Ollivier <cedric.ollivier@orange.com>
|
|
It stops using noop drivers and adds all scenarios except:
- .*api.v2.test_availability_zone
- .*api.v2.test_availability_zone_profile
- .*api.v2.test_member.MemberAPITest.test_member_ipv4_create
- .*api.v2.test_member.MemberAPITest.test_member_ipv6_create
It doesn't run ipv6 which asks for a deep dive (ipv6 from containers?)
It increases threads to 3 as the other testcases.
Change-Id: I36d9d764a5ab92efe89196181bffa3ad0a57227c
Signed-off-by: Cédric Ollivier <cedric.ollivier@orange.com>
|
|
It's useless as they are skipped by regexes (vnc_console).
Change-Id: If3cea451c95179e7e200708e1f3b5e9534be9e1d
Signed-off-by: Cédric Ollivier <cedric.ollivier@orange.com>
|
|
scp + ssh calls are replaced by the right scp call.
Change-Id: Iac2be63fb380486393af65dda8f7fd80bdd41d5f
Signed-off-by: Cédric Ollivier <cedric.ollivier@orange.com>
|