Age | Commit message (Collapse) | Author | Files | Lines |
|
|
|
It was removed by "Move rally_full in benchmarking" [1]
[1] https://gerrit.opnfv.org/gerrit/#/c/66911/1
Change-Id: I7ade6f474364646bfe0f3ae1b2718749bd0ff26e
Signed-off-by: Cédric Ollivier <cedric.ollivier@orange.com>
|
|
It takes into account the wget exit values and avoids manipulating a
falsy empty file.
It won't help fixing the network outages in LF network which have been
highlighted by false failures in api_check (the wrong snaps package
is installed [1])
[1] https://build.opnfv.org/ci/job/functest-opnfv-functest-healthcheck-latest-api_check-run/100/console
Change-Id: Ic9b75965b1ffb1f7bb342ff216bda9933ca75f0b
Signed-off-by: Cédric Ollivier <cedric.ollivier@orange.com>
|
|
This patch adds success/failure information of rally test cases to
summary output.
Change-Id: Ie4f9b4972a26e187bb7afdddfd5ce51b16f3db12
Signed-off-by: Stamatis Katsaounis <mokats@intracom-telecom.com>
|
|
It adds xunit reports for rally-based and tempest-based testcases.
It completes the reports provided by snaps (thanks to Xtesting).
All rally related operations are moved to rally.
It allows removing the rally dependency to tempest which was false.
Change-Id: Ia7d2476f58f4f68b7c88442e50cad844037a36e9
Signed-off-by: Cédric Ollivier <cedric.ollivier@orange.com>
|
|
|
|
It switches from tempest.log to rally.log to conform with rally/stestr.
Change-Id: I00c2a268a2bc312a22a8d70278fad618c1c25235
Signed-off-by: Cédric Ollivier <cedric.ollivier@orange.com>
|
|
This patch adds the ability to specify which rally test to run by
setting the appropriate configuration value.
Change-Id: I6cbc9d12aada4f8eab68d2219d94ae1391a79021
Signed-off-by: Stamatis Katsaounis <mokats@intracom-telecom.com>
|
|
Rally_jobs were moved into benchmarking.
It completes "Move testcases in the right tiers" [1]
[1] https://gerrit.opnfv.org/gerrit/#/c/66875/
Change-Id: I1c2da77eff6682f85ab8504750077c9ee542edb8
Signed-off-by: Cédric Ollivier <cedric.ollivier@orange.com>
|
|
It eases sharing one single report including all task results.
For the time being, rally_jobs only generates neutron.html.
Then it only adds one report for rally_full and rally_sanity.
We may consider generating only one json file but it asks for
a refactoring (result parsing) which can be done in a second
change.
Change-Id: I8fa75ae80750336950f355c2aedc3dd4dc621cc5
Signed-off-by: Cédric Ollivier <cedric.ollivier@orange.com>
|
|
It increases the timeout in cloudify-based and clearwater-based
testcases to allow running them in parallel (functest-virtual) [1].
[1] https://build.opnfv.org/ci/job/functest-latest-gate/54/
Change-Id: Ic637cb46a1380a683cb1e2b633cb97ac3be6affa
Signed-off-by: Cédric Ollivier <cedric.ollivier@orange.com>
|
|
ci/testcases.yaml was copied/pasted when moving the rally testcases.
It completes "Set Gnocchi scenario as optional in rally tests" [1].
[1] https://gerrit.opnfv.org/gerrit/#/c/66653/
Change-Id: Ia23ae9764750775fe8983660c56ced06f3de977a
Signed-off-by: Cédric Ollivier <cedric.ollivier@orange.com>
|
|
It saves about 290M.
Change-Id: I603c689be989a603e933970f52510026026c7001
Signed-off-by: Cédric Ollivier <cedric.ollivier@orange.com>
|
|
The pull request #4 fixing the blueprint [1] was merged [1].
[1] https://github.com/oolorg/opnfv-vnf-vyos-blueprint/commit/9b76d46a388d32d4985797620e67c2ed3315b3e4
[2] https://build.opnfv.org/ci/job/functest-opnfv-functest-vnf-hunter-gate/4/console
Change-Id: I3ddd86ee8c9fd30e68de74f06f40f782b1b04fd2
Signed-off-by: Cédric Ollivier <cedric.ollivier@orange.com>
|
|
It conforms with "Move rally_full in benchmarking" [1]
[1] https://gerrit.opnfv.org/gerrit/#/c/66911/
Change-Id: I6382f6e612343896a1d817675e122d2c1e0b1a38
Signed-off-by: Cédric Ollivier <cedric.ollivier@orange.com>
|
|
|
|
It switches from the Cloudify virtual machine to the community
container which allows deploying vyos_vrouter and cloudify_ims.
Last stable Cloudify OpenStack plugin (2.4.17) is now deployed.
It also updates the Cloudify python package to latest release.
It adds a delay before deleting the blueprint which may be
improved in a second change (it allows enabling the testcase in
the stable release).
Change-Id: Ie48a2c9d2badab0d7ba0b9e7845f453fad335ef7
Signed-off-by: Cédric Ollivier <cedric.ollivier@orange.com>
|
|
It completes "Move testcases in the right tiers" [1]
[1] https://gerrit.opnfv.org/gerrit/#/c/66863/
Change-Id: Ic2e6ebf9371de87caafbfdb54029a24440e6fd5b
Signed-off-by: Cédric Ollivier <cedric.ollivier@orange.com>
|
|
It allows merging all benchmarking testcases in one tiers and saves
lots of time as they could all run in parallel.
Change-Id: I551179709d2ce862056c1f5fcc1e16ef0b776dd4
Signed-off-by: Cédric Ollivier <cedric.ollivier@orange.com>
|
|
It fails 4 times in functest-virtual (out of 34) due to a possible slow
bandwidth (e.g. [1]).
[1] https://build.opnfv.org/ci/job/functest-opnfv-functest-vnf-latest-juju_epc-run/24/console
Change-Id: Ie1aa69464f8dd0fa1a15900e7d49576368187a44
Signed-off-by: Cédric Ollivier <cedric.ollivier@orange.com>
|
|
Here are the moves:
- tempest_smoke in healthcheck
- tempest_full and tempest_scenario in smoke
- rally_jobs in benchmarking
It avoids conflicts between benchmarking tools and few functional
tests (short timeouts).
Then it should allow running in parallel all testcases in one tiers.
It also allows cleaning the additional steps defined in releng.
rally_full could move in benchmarking in a second step.
Change-Id: I0bc657557a61c7d9bf7679b4f2dbda08c36728ca
Signed-off-by: Cédric Ollivier <cedric.ollivier@orange.com>
|
|
The former image can't be downloaded and then breaks all gates.
It will be completed by changes in vnf if needed.
Change-Id: Ie364aa2315257157a5e09eb4ba121926a60e7b26
Signed-off-by: Cédric Ollivier <cedric.ollivier@orange.com>
|
|
snaps package is sometimes downloaded from PyPI which leads to runtime
failures (the package is very obsolete) [1].
[1] https://build.opnfv.org/ci/job/functest-opnfv-functest-healthcheck-latest-api_check-run/51/console
Change-Id: I1fd9b17339182f13d7455fb34b100ffc2ca2c21f
Signed-off-by: Cédric Ollivier <cedric.ollivier@orange.com>
|
|
The volume and the virtual machine created are not deleted.
Change-Id: I238fcb0839a2f609333080bc8c88b84327d51475
Signed-off-by: Cédric Ollivier <cedric.ollivier@orange.com>
|
|
Devstack allocates a 24G disk image by default [1].
It can be easily reached by setting volume size to 5G or 10G.
[1] https://github.com/openstack-dev/devstack/blob/master/stackrc#L779
Change-Id: I7c35d8f333718d6c192a9f12a714e626d60b1895
Signed-off-by: Cédric Ollivier <cedric.ollivier@orange.com>
|
|
|
|
|
|
It seems that juju-wait doesn't wait for all apps.
Juju_epc passes in functest-virtual thanks to that additional tries.
Change-Id: I161902c6308cfa6323c21e09140d16e036768ece
Signed-off-by: Cédric Ollivier <cedric.ollivier@orange.com>
|
|
It allows running test_user_update_own_password [1]
[1] http://artifacts.opnfv.org/functest/functest-opnfv-functest-components-latest-tempest_full-run-8/results/tempest_full/tempest-report.html
Change-Id: I9f273ba58871f1e16886af623dcbb91f5c20c2f6
Signed-off-by: Cédric Ollivier <cedric.ollivier@orange.com>
|
|
Change-Id: I88a58516a3620cc6b19013313672ad1a80791e70
Signed-off-by: Cédric Ollivier <cedric.ollivier@orange.com>
|
|
|
|
|
|
It has been fixed in Snaps [1] but we can't sync its requiremets [2].
Then we have to remove the test.
[1] https://gerrit.opnfv.org/gerrit/#/c/66625/
[2] https://gerrit.opnfv.org/gerrit/#/c/66627/1
Change-Id: Ied405092cb4992bc04ca13c1536d1e8cee056ace
Signed-off-by: Cédric Ollivier <cedric.ollivier@orange.com>
(cherry picked from commit 17b788c5340e7d76f6c37397cd8aa35c8db3f4c2)
|
|
|
|
urljoin works here only if url ends by /.
It replaces identity by v3 when running vs devstack.
Change-Id: Id896f2b54dfcabadaad1137deef91b1cb07c697d
Signed-off-by: Cédric Ollivier <cedric.ollivier@orange.com>
|
|
It checks all 3 endpoints for image but nothing forces to define them.
Here is the only endpoint created by devstack.
glance | image | RegionOne
| | public: http://192.168.1.10/image
[1] keystoneauth1.exceptions.catalog.EndpointNotFound: internal endpoint for image service in RegionOne region not found
Change-Id: If501b849fbbed6df98742db21e924094c1df9c7f
Signed-off-by: Cédric Ollivier <cedric.ollivier@orange.com>
|
|
The execution of an optional scenario is skipped if the corresponding
openstack service does not exist.
Change-Id: I8721f26ebc253ce02138b85d6c88e659c5afae80
Signed-off-by: Juha Kosonen <juha.kosonen@nokia.com>
|
|
NOTE: pip versions from 19.0 and later are producing an error when
used with option --no-cache-dir.
Change-Id: I3f4fb0b77046cda45d98673e7a5c8d42c72fb557
Signed-off-by: Stamatis Katsaounis <mokats@intracom-telecom.com>
Signed-off-by: Cédric Ollivier <cedric.ollivier@orange.com>
|
|
It increases timeouts which could be reached if shared bandwidth
(functest-virtual [1]).
[1] http://artifacts.opnfv.org/functest/functest-opnfv-functest-vnf-latest-heat_ims-run-4/results/functest.log
Change-Id: I29adcd9ea88bbccbaef9ba7f8e155d37a3217433
Signed-off-by: Cédric Ollivier <cedric.ollivier@orange.com>
|
|
It may select a falsy external network [1].
[1] https://build.opnfv.org/ci/job/functest-opnfv-functest-vnf-latest-juju_epc-run/4/console
Change-Id: Ic403d0cfb4bc829275617d6b1338311e8e7ea894
Signed-off-by: Cédric Ollivier <cedric.ollivier@orange.com>
|
|
Change-Id: I46053c92660139f4004d9cec8d1cfb6817cc0e07
Signed-off-by: Cédric Ollivier <cedric.ollivier@orange.com>
|
|
Rally selects the first external network [1] which could raise side
effects if:
- another testcase is creating a dummy external network in //
- all external subnets are not reachable from jumphost
It's now overriden by EXTERNAL_NETWORK as defined in Functest
Rally creates a new shared network if one already exists [2].
As juju now allows other shared networks, it can be safely created by
Functest to allow only one logic and to handle provider networks,
different name servers, etc...
[1] https://github.com/openstack/rally-openstack/blob/master/rally_openstack/verification/tempest/config.py#L146
[2] https://github.com/openstack/rally-openstack/blob/master/rally_openstack/verification/tempest/context.py#L85
Change-Id: Icf8c08077d4b0a9eb1c2e1b7309c62957b0a3b63
Signed-off-by: Cédric Ollivier <cedric.ollivier@orange.com>
|
|
|
|
It conforms with best programming practices.
It allows muting console and increasing log level (debug).
Change-Id: I0e0cfef6abe3b354a43817fcb242c71da701e237
Signed-off-by: Cédric Ollivier <cedric.ollivier@orange.com>
|
|
It's also available on [1]
[1] http://testresults.opnfv.org/functest/functest2019/
Change-Id: I6eed7bd49f20df86ec4316f1d691001ad904379f
Signed-off-by: Cédric Ollivier <cedric.ollivier@orange.com>
|
|
They can now be generated via the new ansible role.
Change-Id: I7746d08b940e8d6906cb023c39bc87d8dee441cc
Signed-off-by: Cédric Ollivier <cedric.ollivier@orange.com>
|
|
FDS didn't participate to Gambia.
Removing it in all branches avoids compexifying the jjbs.
This patch will be reverted as soon as the support is ready.
Change-Id: I9fb293b2972e0deb318ac9d3d08377039ea247e4
Signed-off-by: Cédric Ollivier <cedric.ollivier@orange.com>
|
|
It seems falsy (see line 72)
2019-01-07 11:50:32,550 - functest.core.cloudify - DEBUG - try 9: Cloudify Manager isn't up and running
Traceback (most recent call last):
File "/usr/lib/python2.7/site-packages/functest/core/cloudify.py", line 72, in execute
"foo", "bar", update_if_exists=True)
File "/usr/lib/python2.7/site-packages/cloudify_rest_client/secrets.py", line 86, in create
response = self.api.put('/secrets/{0}'.format(key), data=data)
File "/usr/lib/python2.7/site-packages/cloudify_rest_client/client.py", line 263, in put
timeout=timeout)
File "/usr/lib/python2.7/site-packages/cloudify_rest_client/client.py", line 224, in do_request
verify=self.get_request_verify(), timeout=timeout)
File "/usr/lib/python2.7/site-packages/cloudify_rest_client/client.py", line 158, in _do_request
self._raise_client_error(response, request_url)
File "/usr/lib/python2.7/site-packages/cloudify_rest_client/client.py", line 119, in _raise_client_error
response=response)
File "/usr/lib/python2.7/site-packages/cloudify_rest_client/client.py", line 131, in _prepare_and_raise_exception
status_code, error_code=error_code, response=response)
CloudifyClientError: 409: <Secret id=`foo` tenant=`default_tenant`> already exists on <Tenant name=`default_tenant`>
Change-Id: I6f4e2719ad2e55e997ebbd158dce4c6a204b2c36
Signed-off-by: Cédric Ollivier <cedric.ollivier@orange.com>
|
|
The traceback is printed as debug to avoid priting the first
unsuccessful tries in console. It will ease debugging issues as for [1]
[1] https://gerrit.opnfv.org/gerrit/#/c/66253/
Change-Id: I89992ea4452b24788cc06dec4c384a263b704efb
Signed-off-by: Cédric Ollivier <cedric.ollivier@orange.com>
|
|
It has rarely occured in Orange Openlab [1].
Traceback (most recent call last):
File "/usr/lib/python2.7/site-packages/functest/core/cloudify.py", line 70, in execute
self.cfy_client.secrets.create("foo", "bar")
File "/usr/lib/python2.7/site-packages/cloudify_rest_client/secrets.py", line 86, in create
response = self.api.put('/secrets/{0}'.format(key), data=data)
File "/usr/lib/python2.7/site-packages/cloudify_rest_client/client.py", line 263, in put
timeout=timeout)
File "/usr/lib/python2.7/site-packages/cloudify_rest_client/client.py", line 224, in do_request
verify=self.get_request_verify(), timeout=timeout)
File "/usr/lib/python2.7/site-packages/cloudify_rest_client/client.py", line 158, in _do_request
self._raise_client_error(response, request_url)
File "/usr/lib/python2.7/site-packages/cloudify_rest_client/client.py", line 119, in _raise_client_error
response=response)
File "/usr/lib/python2.7/site-packages/cloudify_rest_client/client.py", line 131, in _prepare_and_raise_exception
status_code, error_code=error_code, response=response)
CloudifyClientError: 409: <Secret id=`foo` tenant=`default_tenant`> already exists on <Tenant name=`default_tenant`>
Change-Id: Ic61d0c729f13caa552a0b0a00dad35e50d130e01
Signed-off-by: Cédric Ollivier <cedric.ollivier@orange.com>
|