Age | Commit message (Collapse) | Author | Files | Lines |
|
Some places when create image is referred, they check if image exist,
if exist use it directly, else create a new one.
Abstract a method to integrate the process
JIRA: FUNCTEST-432
Change-Id: I83d1112aa5cb0d3cdfab92fd49cd5f2c1cceff82
Signed-off-by: SerenaFeng <feng.xiaowei@zte.com.cn>
(cherry picked from commit 5f5af6b1d8d0af19db143e67f04f67a643af863c)
|
|
Skip TestServerBasicOps which fails when deleting server
and floating ip almost at the same time.
JIRA: FUEL-173
JIRA: FUEL-176
Change-Id: I24f3397096757028bed9d32d6353aa689e8eaa9a
Signed-off-by: Michael Polenchuk <mpolenchuk@mirantis.com>
(cherry picked from commit 44819aec4a70a75abc9fbcff23edd9d904170bf7)
|
|
In openstack_util.py, when create/get/update network or other resources failed,
False is returned, which is incompatible with success return value such as id, network_nic.
JIRA: FUNCTEST-427
Change-Id: Id2f55d8524e5aff150ba6bfb799085377a63baa4
Signed-off-by: SerenaFeng <feng.xiaowei@zte.com.cn>
(cherry picked from commit 3e95a1cd0721924e89e2cfd2960d4736bc6c5588)
|
|
In rally, image_exists is used to decide whether the image
resource should be cleaned or not. But unlike rally,
image_exists in vping is only assigned but not never be actually used.
JIRA: FUNCTEST-426
Change-Id: Id2089dd2a9734ba957cc1c80d4dc31bc27907b0a
Signed-off-by: SerenaFeng <feng.xiaowei@zte.com.cn>
(cherry picked from commit 73053cd38fed875e3e5d64cabfb7fbfeb2839a80)
|
|
During generating tempest.conf file rally tries to download
cirros image from the internet if no suitable images exist in
the configuration. This change ensures that cirros image is
created before generating tempest.conf file.
JIRA: FUNCTEST-195
Change-Id: I9e2df8af40bd02db9b78b8b6b730b4f4168a8b2d
Signed-off-by: vitikkan <viktor.tikkanen@nokia.com>
(cherry picked from commit 93f4a6576d58522dfbb458f184399767ef8764aa)
|
|
Change-Id: I91681736817f881ca53e5124328bd4399aa54740
Signed-off-by: Michael Polenchuk <mpolenchuk@mirantis.com>
|
|
There are lots of common processes in vPing_userdata.py
and vPing_ssh.py, abstract and refactor them.
JIRA: FUNCTEST-414
Change-Id: I4cd2c635318c063319d2a3c31f9a1e512eeea6c8
Signed-off-by: SerenaFeng <feng.xiaowei@zte.com.cn>
|
|
Change-Id: I0ae03b1f13bbc8801d8c14fa6ff1b93117c80e91
Signed-off-by: wuwb1989 <wuwenbin2@huawei.com>
|
|
Only last part of blacklist.txt file was loaded by yaml.safe_load()
function. File format is corrected.
JIRA: FUNCTEST-418
Change-Id: I9c922b1a7b739e7d0c0d5afa3e0ea700749f9d50
Signed-off-by: vitikkan <viktor.tikkanen@nokia.com>
|
|
Change-Id: If722e0d3376b8d04766b1eb871caadc5bc6e7c18
Signed-off-by: CNlucius <lukai1@huawei.com>
|
|
Change-Id: Idda3a3a5c8fb20f6894675eb087417e150f8646d
Signed-off-by: meimei <meimei@huawei.com>
|
|
|
|
JIRA: FUNCTEST-268
Change-Id: I8c8a3eff4a7e94b963c0307010a60723dc13a537
Signed-off-by: boucherv <valentin.boucher@orange.com>
|
|
Change-Id: I004225131de57f1597edc62558a3a3e42f31b504
Signed-off-by: meimei <meimei@huawei.com>
|
|
Change-Id: I23ff06b53e45e9aa226f430e7d2fb75025281b9e
Signed-off-by: meimei <meimei@huawei.com>
|
|
|
|
Change-Id: I07f0f5e253b04c9ac70e6dc4ea210e7cecf22c6a
Signed-off-by: meimei <meimei@huawei.com>
|
|
Adding support of LXD scenario
JIRA: FUNCTEST-268
Change-Id: I5891ff9de674d99d32d8db747be501ee8a2e6609
Signed-off-by: boucherv <valentin.boucher@orange.com>
|
|
Earlier, there was external URL for accessing KB services
in tempest.conf which is not accessible from functest docker.
Use bind host URL for KB service to access from the docker.
Change-Id: I147ffc2cd9883a6913080f9e1ab245807e0ca707
Signed-off-by: ashish singh <ashishsingh2k8@gmail.com>
|
|
Some tempest test cases try to establish SSH connection to
Cirros VMs using default 300s timeout. Since Cirros VMs send
only three DHCP requests with 1 minute interval, future SSH
connection attempts will most probably fail if DHCP response
will not be received right after third request. For that reason
the optimal timeout is slightly above 2 minutes.
JIRA: FUNCTEST-378
Change-Id: I57415ad37a85cd84887d3cac4f7a44f7e145d820
Signed-off-by: vitikkan <viktor.tikkanen@nokia.com>
|
|
"rally verify genconfig" command doesn't generate tempest.conf
file if it already exists in the used deployment. This can cause
troubles if some resources (images, flavors etc.) change between
consecutive runs of tempest. The change ensures that tempest.conf
is deleted before executing "rally verify genconfig".
JIRA: FUNCTEST-403
Change-Id: Icf5e380da3cb8757f4177b49028937355e2cf1d7
Signed-off-by: vitikkan <viktor.tikkanen@nokia.com>
|
|
Blacklist file handling is changed so that also partial test case
names are accepted. For example, "tempest.api.object_storage"
entry will exclude all the test cases containing this string.
JIRA: FUNCTEST-401
Change-Id: I620be7dfd5d980a343d2f7cc7c0dbe40598bf65c
Signed-off-by: vitikkan <viktor.tikkanen@nokia.com>
|
|
|
|
|
|
Currently, there is internal IP address which is getting populated
in tempest.conf. Since the tempest are run inside functest docker container
we need to use public IP address for accessing Kingbird services.
Now, there will not be any installer specific endpoints as we had earlier.
All should communicate with public endpoints.
Change-Id: Id22e7e4d9ece9a0000a1186113627b82d6ba400a
Signed-off-by: ashish singh <ashishsingh2k8@gmail.com>
|
|
when checking tempest test results using success_rate
configured in testcases.yaml rather than written it in the code
JIRA: FUNCTEST-400
Change-Id: I2a34e3e032a0d94c6a4605fcf09a66c8f8f9b78c
Signed-off-by: SerenaFeng <feng.xiaowei@zte.com.cn>
|
|
Change-Id: Ic9f6b9592202c3bed9d6835eeea7c14971e1c01f
Signed-off-by: jose.lausuch <jose.lausuch@ericsson.com>
|
|
In case of joid we implement ceph-radosgw for object storage which creates
swift api for object storage. Manually i can verify we can upload the
object. But functest is failing other installer removed the API but we
need in JOID for juju based deployment on Openstack. So this patch will
mute the object storage test cases for functest.
Change-Id: I21d2bdc665a9589630cfcacb278cfb65a1c15f11
Signed-off-by: Narinder Gupta <narinder.gupta@canonical.com>
|
|
JIRA: FUNCTEST-382
Change-Id: Iaf475ecd3dbfcf6fb639ff146fe96e54c5393a1e
Signed-off-by: Markus Kullberg <markus.kullberg@nokia.com>
|
|
|
|
add method check_success_rate() to unify criteria check by success_rate
JIRA: FUNCTEST-398
Change-Id: I1b3f0ccd7061b80ac88356206f29844e31e4101b
Signed-off-by: SerenaFeng <feng.xiaowei@zte.com.cn>
|
|
JIRA: FUNCTEST-396
Added example for SDNVPN test cases
Change-Id: I623130fc330c95239d982097552ca09f2a7ea34a
Signed-off-by: jose.lausuch <jose.lausuch@ericsson.com>
|
|
The healthcheck tier used hard-coded x86 image for testing glance and nova.
This breaks functest for ARM based pods, because x86 instances cannot run on
ARM and even more, if healthcheck fails no other testcases are executed.
Change-Id: I8a4cc754569a26d98bff57470574ed28eebbf2da
Signed-off-by: Ciprian Barbu <ciprian.barbu@enea.com>
|
|
By default public endpoint was used but now you can use internal or admin
JIRA: FUNCTEST-387
Change-Id: If1adaa88248d379ad6d74b20f2dffd76278b67b9
Signed-off-by: boucherv <valentin.boucher@orange.com>
|
|
JIRA: FUNCTEST-393
Change-Id: I94158531739e1e05c694a785a8bbca7f80684af5
Signed-off-by: SerenaFeng <feng.xiaowei@zte.com.cn>
|
|
|
|
Signed-off-by: Dimitri Mazmanov <dimitri.mazmanov@ericsson.com>
|
|
JIRA: FUNCTEST-381
Change-Id: I533bb50e26a3b97da9fd3bfa007f8ac7ce752a5b
Signed-off-by: jose.lausuch <jose.lausuch@ericsson.com>
|
|
Change-Id: I351311688742c04c2c2b7278015e221178267fda
Signed-off-by: meimei <meimei@huawei.com>
|
|
JIRA: FUNCTEST-390
Change-Id: I35a2f72016573fc9e0a8403e436d5568937c5b14
Signed-off-by: SerenaFeng <feng.xiaowei@zte.com.cn>
|
|
stop_time is assigned twice, delete the first one
JIRA:FUNCTEST-384
Change-Id: I380113474e66a73c516f4411d8d7c5832e33921a
Signed-off-by: SerenaFeng <feng.xiaowei@zte.com.cn>
|
|
format total_success to float when compared with numeric 90
JIRA: FUNCTEST-365
Change-Id: I473582053c107bb026bf0e8d76dab46b2e5e3eb6
Signed-off-by: SerenaFeng <feng.xiaowei@zte.com.cn>
|
|
|
|
JIRA: FUNCTEST-376
Also some fixes to get rid of flake8 violations in onos-sfc
Change-Id: I52c2fd30a6b81af20ea779db2d73b9386524f7ce
Signed-off-by: jose.lausuch <jose.lausuch@ericsson.com>
|
|
The patch multihops to the fuel node and dynamically retrieves kingbird
endpoint information. It also resolves the controller ip instead of
using the hardcoded value from the config ONLY if the installer type is
Fuel.
Signed-off-by: Dimitri Mazmanov <dimitri.mazmanov@ericsson.com>
Change-Id: I5161df5df3da7283f6374efa6bd1108246c1095a
|
|
JIRA: FUNCTEST-375
Change-Id: I4f0cb68639b927f2980389d9331ec260605c8caf
Signed-off-by: jose.lausuch <jose.lausuch@ericsson.com>
|
|
Also, turn off debug info from paramiko python module in vping_ssh
JIRA: FUNCTEST-373
JIRA: FUNCTEST-374
Change-Id: I6168cae85db91b2dd6ceac591f12dd9295786b59
Signed-off-by: jose.lausuch <jose.lausuch@ericsson.com>
|
|
Instead of repeating the code in all the scripts and
importing the openstack libraries, the clients shall
be given by openstack utils, which is its purpose
JIRA: FUNCTEST-163
Change-Id: I1ccc05a3af44ee1ab5938ea9e4e01dbe55f4816d
Signed-off-by: jose.lausuch <jose.lausuch@ericsson.com>
|
|
JIRA: FUNCTEST-179
Change-Id: I4a8ae43e2814b3d859c32a6278671f87155aa855
Signed-off-by: meimei <meimei@huawei.com>
|
|
In some small virtual environments, booting 4 VMs using the
flavor m1.small can fail because of insufficient disk space
or available hosts.
Check if the flavor exists, if it doesnt, create it.
Change-Id: Ib7a3f337ea5e0686bcb9af442be175db436e4e8a
Signed-off-by: jose.lausuch <jose.lausuch@ericsson.com>
|