Age | Commit message (Collapse) | Author | Files | Lines |
|
Fuel enables SSL based on scenario type (only for HA), not based
on cluster type (baremetal or virtual).
Change-Id: I3dc94f953e9a8ee0e94028577270ac05c9318ad6
Signed-off-by: Alexandru Avadanii <Alexandru.Avadanii@enea.com>
|
|
Even though RC_FILE_PATH is defined in Jenkins as passed in as a
parameter, bash still sees it as unset. Removing the 'nounset' option
from these scripts should allow the check to success (RC_FILE_PATH set
to blank).
Change-Id: Ibcfdcf0d2a12c8119d2fe3ec3b354be782338bd5
Signed-off-by: Trevor Bramwell <tbramwell@linuxfoundation.org>
|
|
This patch is needed for the case where there are provided OS
credentials to functest, and there is no need to find the installer VM
and copy the credentials to the host. In the Apex snapshot case, there
is no installer VM on the slave, so we must provide the credentials
file.
Also includes a fix to use tempest_smoke_serial for the snapshot netvirt
verification until CSIT suite is working correctly.
Change-Id: Ic84955f527e3f84bb0d8579a377b4a1883fe3eb2
Signed-off-by: Tim Rozet <trozet@redhat.com>
|
|
Now that keystone V3 Identity is only one supported, OOO 'overcloudrc'
should be used in Queens and later as that contains V3 creds.
'overcloudrc.v3' will eventually be removed in OOO.
Note, this was causing tempest_smoke_serial to fail in Apex with Swift
test cases. This is due to the fact that in Apex we were only setting
the OS_REGION_NAME in overcloudrc, and functest was using
overcloudrc.v3. When Tempest does not have the region set it will simply
look for the first endpoint it can find for that service type, and was
not using the right keystone endpoint for Apex.
Change-Id: I154c9e762dfc5ba035f6e108606ecedcfb0d60d2
Signed-off-by: Tim Rozet <trozet@redhat.com>
|
|
The BUILD_TAG on ZTE baremetal PODs don't include the word "baremetal".[1]
Line 117 can't be correctly executed. Use a negative checking instead.
[1] https://build.opnfv.org/ci/job/functest-fuel-zte-pod1-daily-master/206/injectedEnvVars/
Change-Id: I1a36494768e0172bcf6a016fd8f2524854c9b876
Signed-off-by: wu.zhihui <wu.zhihui1@zte.com.cn>
|
|
This reverts commit 3f27c8b519e42295bd76ace5506903f265e48560.
Change-Id: I024ba8c0b960387138a71be9bc19f6d1b72030ca
|
|
Fuel baremetal deployments provide a certificate, which should be
copied over to test containers. Virtual deployments do not, hence
the need to determine the deployment type and decide whether the
certificate should be transferred in the first place.
However, the Docker args we contruct will try to map the certificate
file no matter what, so we will instead try to transfer the cert
from Salt Master in all deployments, and just create an empty file
in case it does not exist.
Change-Id: I3d3fe6cabf552a88841c529ce38446b2e54b3053
Signed-off-by: Alexandru Avadanii <Alexandru.Avadanii@enea.com>
|
|
Change-Id: I22df519d949fb28eaecc6af99347868023a1e2a8
Signed-off-by: Linda Wang <wangwulin@huawei.com>
|
|
Fuel@OPNFV mcpcontrol network (equiv to old admin/PXE net) has been
configured to use 192.168.10.0/24 during E release cycle.
However, due to overlapping with Ericsson-pod1 networks, we decided
to re-use the old CIDR Fuel used before E release (10.20.0.0/24).
This allows reusing the INSTALLER_IP param instead of recently added
SALT_MASTER_IP param.
Current change refactors all usage of SALT_MASTER_IP in favor of old
INSTALLER_IP, thus reducing complexity and providing a familiar env
to leagcy Fuel users.
JIRA: FUEL-285
Change-Id: I1e89cca08ce4cb9bf3a82a06043c91ee26aa0794
Signed-off-by: Alexandru Avadanii <Alexandru.Avadanii@enea.com>
|
|
|
|
Functest CI logs have been FAIL with compass but the CI logs are not
explicit. Remove a redirection to /dev/null to get something more
explicit
Change-Id: Icdf17595b051ccba6c58cbe86ab546e5693375ab
Signed-off-by: Morgan Richomme <morgan.richomme@orange.com>
|
|
- add the rc file retrieval in fetch_os.sh
- adapt alpine to split healtcheck from daily
(healthcheck + smoke + features)
- adapt the xci builder
Change-Id: I38836814240a1d769d379a10c9ef14def7e6df97
Signed-off-by: Morgan Richomme <morgan.richomme@orange.com>
|
|
Change-Id: I28e259bf8cdc7431a99227ba3e1eb8615744851e
Signed-off-by: Tim Rozet <trozet@redhat.com>
|
|
|
|
There is a bug where by default the INSTALLER_IP is set to 192.168.X.X
and in yardstick the IP is never found for Apex before trying to fetch,
so fetching fails to an invalid IP address.
This patch makes fetch os creds more robust to check if the INSTALLER_IP
is valid and if not, then attempt to find the correct IP for Apex.
Change-Id: Ie18d6e70f77ccff3592fb37c6d0b6e87b6435cf8
Signed-off-by: Tim Rozet <trozet@redhat.com>
|
|
Change-Id: I02503772dc5ed59562c2e0aba96280e2cae1da50
Signed-off-by: Michael Polenchuk <mpolenchuk@mirantis.com>
|
|
In case ctl01 node is down, the current mechanism will return
invalid data (the string "Minion") instead of a valid controller IP.
Since "keystonercv3" is present on all controller nodes at </root>,
adjust the credentials fetch to use the first online controller node.
Change-Id: I4ba15bb40d465b112c5e9fc827e3e2a69e671b26
Signed-off-by: Alexandru Avadanii <Alexandru.Avadanii@enea.com>
|
|
sometimes it can't copy os_cacert but copy opnfv_openrc.
it cause an issue and didn't execute the command chown.
so the opnfv_openrc will belong to root, which will cause errors next time.
Change-Id: I2955ba347d399d65c9e5cd566235be06fb7c30af
Signed-off-by: wutianwei <wutianwei1@huawei.com>
|
|
Now only Openstack https is supported by OSA repo LXC in compass
deployment, so certification file is required to enable SSL
verification.
After [1] is merged, the cacert file os_cacert will be provided
in the container compass-tasks.
[1]: https://gerrit.opnfv.org/gerrit/#/c/37451/
Change-Id: Id747cef1955733a1c93feb20a26d503a5de1fb93
Signed-off-by: Linda Wang <wangwulin@huawei.com>
|
|
The jenkins user is not permitted to view openrc file
with root owner and group.[1]
cat: /home/jenkins/opnfv-openrc.sh: Permission denied
[1]: https://build.opnfv.org/ci/view/functest/job/functest-compass-baremetal-daily-master/1293/console
Change-Id: I89d0597c8f0ece00f70fd6622048794b15510636
Signed-off-by: Linda Wang <wangwulin@huawei.com>
|
|
Compass has changed the mechanism of deployment (Containers) in master,
so we adjust the way to fetch the OS credentials file.
Change-Id: If126ac9b081589c21466b59eae05b107bf6dc425
Signed-off-by: Linda Wang <wangwulin@huawei.com>
|
|
Adjusts retrieving OS credentials for MCP (master branch in fuel repo)
Also removes legacy admin_ip and public_ip retrieving.
Change-Id: Ib7781a3701c7281b781872b18150332b94298260
Signed-off-by: Andrii Ostapenko <aostapenko@mirantis.com>
|
|
Change-Id: Ib8a3ff3a747be55d8e23513bd028d934fef02eb8
Signed-off-by: Tim Rozet <trozet@redhat.com>
|
|
Also allows unset and pipefail explicitly in set-functest-env.sh
Change-Id: I06ae718f563fed0596a37cec1b3a78e44ee7ea5d
Signed-off-by: Tim Rozet <trozet@redhat.com>
|
|
Change-Id: I086adca7f9671c2c5e6fd04a1ec339e8b22f1de4
Signed-off-by: Tim Rozet <trozet@redhat.com>
|
|
Soon, Functest will start using only public endpoints
and will remove admin connectivity to the deployments.
Change-Id: I2089ccd3d6e7f1acda7b46d3104a8a78a47af8ff
Signed-off-by: jose.lausuch <jose.lausuch@ericsson.com>
|
|
JIRA: FUNCTEST-743
Change-Id: Ic9153575bd43bbc67042cd69a996ad18175427a7
Signed-off-by: SerenaFeng <feng.xiaowei@zte.com.cn>
|
|
Change-Id: Idcf6739498837e828dd65897f599dada720fb48e
Signed-off-by: meimei <meimei@huawei.com>
|
|
Although Fuel default RC file sets export OS_IDENTITY_API_VERSION=3,
the AUTH_URL is poiting to ip:port.
By adding /v3 at the end of the URL, some API calls work.
Change-Id: I745e38524cef6105c23950f318d392e7473366bf
Signed-off-by: jose.lausuch <jose.lausuch@ericsson.com>
|
|
Change-Id: Ic6be1e5559653145e6a80442b205533901c0a2bd
Signed-off-by: jose.lausuch <jose.lausuch@ericsson.com>
|
|
Since the virtual deployments are now able to use
the internal mgmt network we remove the swap
Change-Id: I8eed4dd8218a78c697ffde4716c35170fa72d814
Signed-off-by: Nikolas Hermanns <nikolas.hermanns@ericsson.com>
|
|
The openstack.creds file is not aligned with commonly
used methodologies.
Change-Id: Iba0afc83ebbe35dd189cc35e5bf682680fc96535
Signed-off-by: Nikolas Hermanns <nikolas.hermanns@ericsson.com>
|
|
For newton we have to grep differently for the public ip.
Change-Id: I646358005b163829004e0016514e254575d03157
Signed-off-by: Nikolas Hermanns <nikolas.hermanns@ericsson.com>
|
|
Error log:
More than one endpoint exists with the name 'identity'.
fetch_os_creds.info: public_ip:
Change-Id: Ie7c8eeec5e84de4c38a17b4e718be65f36c562a7
Signed-off-by: QiLiang <liangqi1@huawei.com>
|
|
JIRA: COMPASS-457
please merge after https://gerrit.opnfv.org/gerrit/#/c/18665/
Change-Id: Ie73017dccc8e203fe14c75702e797cb63f20b462
Signed-off-by: MatthewLi <matthew.lijun@huawei.com>
|
|
Change-Id: I5082a6fb3a2d27d17753cf42dc1555cbb3d8eb9a
Signed-off-by: meimei <meimei@huawei.com>
|
|
'fuel --env $env_id' fails because a variable in single quotes
is not read, it is treated as a literal string.
When using double quotes, the variable is replaced by
its value
Change-Id: I39e777ec27c547e012d6f4674bc24e20d421889c
Signed-off-by: jose.lausuch <jose.lausuch@ericsson.com>
|
|
When retreiving an environment id use head to get the first available
environment
Signed-off-by: Dimitri Mazmanov <dimitri.mazmanov@ericsson.com>
|
|
Removed hardcoded value of fuel environment. The value is now fetched
from an environment variable FUEL_ENV.
Signed-off-by: Dimitri Mazmanov <dimitri.mazmanov@ericsson.com>
Change-Id: Iae8e5139f013ca7947e79a1a9a44efd82f48b95b
|
|
In case of multiregion fuel deployments (which is implemented using fuel
environments), always use a controller from the first environment
(RegionOne), as for multisite RegionOne is the master region.
This fix will not affect single region deployments, because the -env 1
property will result in the same single output.
Signed-off-by: Dimitri Mazmanov <dimitri.mazmanov@ericsson.com>
|
|
Commit 622002bea0a0eda42e9a76d14af4bdff47c692d6 is breaking the
Fuel Yardstick testing, this is an emergency fix.
The change introduced spaces in the controller_ip address,
breaking later ssh and scp commands!
Signed-off-by: Stefan K. Berg <stefan.k.berg@ericsson.com>
Change-Id: Ied329103a3568c523ec8ab16a7df5a4763ec14b3
|
|
Prior method of extracting the controller IP was relying on the fact
that default node names have spaces in them and hence generate 9 fields
before the IP address.
This change proposes to change awk tokenizing to divide string by
column separator in `fuel node` output.
Signed-off-by: Stanislaw Kardach <stanislaw.kardach@cavium.com>
|
|
the online state form "True" to "1".
JIRA:QTIP-85
Change-Id: Iaa07af84954bb590801336f2ca44e621c97baeb9
Signed-off-by: zhifeng.jiang <jiang.zhifeng@zte.com.cn>
|
|
NOT VERIFIED
Change-Id: Iabe82f34c6a538646469d72e612f63e51917ae98
Signed-off-by: Jonas Bjurel <jonas.bjurel@ericsson.com>
|
|
Change-Id: Ic0f9b66b1cec3d4249bf979e2a4ca5b8d803ca0e
Signed-off-by: Jo¶rgen Karlsson <jorgen.w.karlsson@ericsson.com>
|
|
Change-Id: I9acce6f1dabb5186375894cdcdc83b392e9d872e
Signed-off-by: meimei <meimei@huawei.com>
|
|
temp implementation: do nothing, rely on volume declared during the creation of the docker file
JIRA: FUNCTEST-97
Change-Id: I8c2e935ab8a72b5593d68236b9573b380a61dcfa
Signed-off-by: Morgan Richomme <morgan.richomme@orange.com>
|
|
Change-Id: I7cb34330dbfaa94ac5b8d2efd9df6f257eccfa6c
Signed-off-by: jose.lausuch <jose.lausuch@ericsson.com>
|
|
JIRA:RELENG-61
Change-Id: I879fa009d53b6ea586b1fd04c52a7c3c23918427
Signed-off-by: jose.lausuch <jose.lausuch@ericsson.com>
|
|
JIRA:RELENG-61
Change-Id: I9017118a42f0794ac34ef03e6788366322ffd550
Signed-off-by: jose.lausuch <jose.lausuch@ericsson.com>
|