Age | Commit message (Collapse) | Author | Files | Lines |
|
After previous change introducing the usage of node labels for
Armband CI deploy jobs, we discovered that the old params were
not applied to labeled pods.
To complicate things, we had different DEFAULT_BRIDGE params
for each CI pod, which contradicts the Jenkins templating
paradigm.
Since we already have a mechanism in place for overriding
env vars based on the POD we are running against, move both
<DEPLOY_TIMEOUT> and <DEFAULT_BRIDGE> params to our local_env,
which is sourced by the deploy script and can apply POD specific
configuration.
Add missing LAB_CONFIG_URL to newly added baremetal and virtual
defaults for Armband.
While at it, also replace "brahmaputra" comments with "colorado".
v2 -> v3:
Source local_env before using DEFAULT_BRIDGE in <armband-deploy.sh>.
Previously, we (Armband) sourced this file in our <ci/deploy.sh> [1],
which will be dropped after this change is merged.
[1] https://git.opnfv.org/cgit/armband/tree/ci/deploy.sh#n24
Change-Id: Ibafea25b3df8c3118ce05e3f40934c8e8b9c8dcd
Signed-off-by: Alexandru Avadanii <Alexandru.Avadanii@enea.com>
|
|
Change-Id: Idb3cabe46082b8c0a160cbc156abe0fe8cafbc4a
Signed-off-by: Paul Vaduva <Paul.Vaduva@enea.com>
|
|
This change
- creates upstream daily job to trigger and control the execution of jobs
that provision nodes using bifrost, deployment using puppet-infracloud,
and simple smoke test job.
- fixes the hashbang placement in 2 of the scripts.
- removes sandbox dummy jobs as the infra jobs can serve as examples now.
Only the node provisioning job is functional at the moment.
Change-Id: I311d159956d8b7ce3f7a500fac5a7dcb459104cf
Signed-off-by: Fatih Degirmenci <fatih.degirmenci@ericsson.com>
|
|
It seems tardstick need the external-network variable defined in
this job so introducing the definition again.
Change-Id: I37bab57a9dda70554fb53de725a0124dc18181e9
Signed-off-by: Narinder Gupta <narinder.gupta@canonical.com>
|
|
Change-Id: Iaf83ba8bc96189e9ef7749bd367197857c3e656a
Signed-off-by: Fatih Degirmenci <fatih.degirmenci@ericsson.com>
|
|
Change-Id: I133b1dae993fa6b5e0fc52be411f68a06fbe72bb
Signed-off-by: David Blaisonneau <david.blaisonneau@orange.com>
|
|
Change-Id: I2e45ea68a30ec356e14d2162df85bf50f51faab2
Signed-off-by: Tim Rozet <trozet@redhat.com>
|
|
Daily jobs for Colorado branch for installer and test projects
have not been created yet and it needs to be done via separate
patches.
Change-Id: I34517e89dfc502ce5741733e01bf8425d513df02
Signed-off-by: Fatih Degirmenci <fatih.degirmenci@ericsson.com>
|
|
|
|
Change-Id: I61b68a7db60193f5525de7151e43b7752e5dd604
Signed-off-by: Fatih Degirmenci <fatih.degirmenci@ericsson.com>
|
|
While at it, also do the following:
- revert previous addition of "armband-defaults" in
opnfv/installer-params.yml, no longer needed;
- remove Josep from failure mail recipient list, armband@enea.com
is already forwarded to all of us;
- add CACHE_DIRECTORY variable for enabling build cache;
TODO (left for future rework):
- rename build and deploy jobs to include installer info, e.g.
"armband-fuel-build-...", preparing for more installers;
- move "installer" parameter to fuel jobs only instead of having
it as a global param;
Change-Id: I2e6cc43484576bd6569605a7c9e60f6fe64f435b
Signed-off-by: Josep Puigdemont <josep.puigdemont@enea.com>
Signed-off-by: Alexandru Avadanii <Alexandru.Avadanii@enea.com>
|
|
Due to migration of ZTE PODs, update some info and reschedule trigger.
Change-Id: Ie7040cd462c3afe3fb7f6d5b3b7b2ad0cc270196
Signed-off-by: wu.zhihui <wu.zhihui1@zte.com.cn>
|
|
Recent changes in armband jjb tooling rely on INSTALLER_TYPE
to be set, so define all default params, similar to Fuel.
Change-Id: Ie29a114bee77a9fbae5f33fc4e716f86a39c14ec
Signed-off-by: Alexandru Avadanii <Alexandru.Avadanii@enea.com>
|
|
JIRA:DAISY-1
Change-Id: I669a9916f561525f1bb49ffe0b7db83889b8172e
Signed-off-by: zhifeng.jiang <jiang.zhifeng@zte.com.cn>
|
|
Due to migrate ZTE-POD2 from Nanjing to Shanghai, update
zte-pod2 info.
JIRA:PHAROS-245
Change-Id: I17279d90076eb0c83a8747fba59221afb1df493d
Signed-off-by: wu.zhihui <wu.zhihui1@zte.com.cn>
|
|
This reverts commit 624b472448cdf9a3ee4e733ac0526d82e29d0127.
Change-Id: I900e6a3206331024cc9722de94a1237b9e1b89e0
Signed-off-by: Fatih Degirmenci <fatih.degirmenci@ericsson.com>
|
|
All the slaves can clone repos from OPNFV Gerrit via https. As a
result of this, the previously introduced GIT_BASE parameter can
now be removed from slave-params.yml file to keep it clean and tidy
and prevent questions.
See the build history of the below job for clone testing on all the slaves
labelled with opnfv-sysinfo.
https://build.opnfv.org/ci/job/repo_clone_test/
This change can only be submitted once the GIT_BASE env var on OPNFV Jenkins
Global Configuration is switched to https.
If any job fails on a slave due to issues with the cloning, it needs to
be fixed immediately.
Change-Id: I9620b9a6a9abed3696f1c917aa392a720f5b1cb0
Signed-off-by: Fatih Degirmenci <fatih.degirmenci@ericsson.com>
|
|
Add ZTE-POD3 to support whole procedure of CI, including:
deployment, yardstick and funectest.
Currently it supports os-nosdn-kvm-ha scenario at the initial step
JIRA: PHAROS-247
Change-Id: I70dde8765696ee08339e7cada2428f8461b5f3a2
Signed-off-by: Julien-zte <zhang.jun3g@zte.com.cn>
|
|
Currently BRIDGE is hardcode with 'pxebr'. If the lab uses another
bridge name, it can be use supported in releng. This parameter is
supported in Fuel project already.
JIRA: RELENG-134
Change-Id: Ie1df46a247eaebb1814506d555f73d03a1e5566e
Signed-off-by: Julien-zte <zhang.jun3g@zte.com.cn>
|
|
|
|
Change-Id: I0a9684ca93667924f34cbbd44c3e8e5111ebd8e8
Signed-off-by: Fatih Degirmenci <fatih.degirmenci@ericsson.com>
|
|
This change
- changes the kvmfornfv-verify job to multijob so job abortion works when a new
patch for the same change arrives
- creates 2 phase jobs for patchset verification, build and test
- enables artifact upload and download
Change-Id: I9de85e2f223e83e2256aaba831b88b0a879962b5
Signed-off-by: Fatih Degirmenci <fatih.degirmenci@ericsson.com>
|
|
Change-Id: I88733b01136e553181e3fbe17c84842b3a707f4f
Signed-off-by: Ryota MIBU <r-mibu@cq.jp.nec.com>
|
|
Change-Id: Ib0462e33381d1a1a454a845f1e8fd7dfe3a3e7d1
Signed-off-by: Ryota MIBU <r-mibu@cq.jp.nec.com>
|
|
Change-Id: I919b7ba04e573f0b386407681ab46f012359d19e
Signed-off-by: Ryota MIBU <r-mibu@cq.jp.nec.com>
|
|
All opnfv-docs jobs are not run, except yardstick, after [1] merged.
This patch fixes the bug in the verify job definition.
[1] https://gerrit.opnfv.org/gerrit/#/c/17479/
Change-Id: I9c0ac67d31f712d44030649aaec38b1c22fa3f67
Signed-off-by: Ryota MIBU <r-mibu@cq.jp.nec.com>
|
|
All opnfv-docs jobs are not run, except yardstick, after [1] merged.
This patch fixes bug in the job definition.
[1] https://gerrit.opnfv.org/gerrit/#/c/17479/
Change-Id: I75c1fac52a5a87a659fe0417f5da046b73a53b20
Signed-off-by: Ryota MIBU <r-mibu@cq.jp.nec.com>
|
|
- make 'opnfv-docs-*' jobs run against all repositories which has docs
directory.
- remove docs builder from yardstick-project-jobs
JIRA: RELENG-133
Change-Id: Id0cf6e600ed112c6f082f6bc924eaeccb0481e49
Signed-off-by: Ryota MIBU <r-mibu@cq.jp.nec.com>
|
|
Add '.tox' to list of directories to ignore since it contains unrelated
python code in virtualenvs.
JIRA: QTIP-89
Change-Id: I577c4ea61fe6b76f7f67ec7c9d4af544d0f73f71
Signed-off-by: Trevor Bramwell <tbramwell@linuxfoundation.org>
|
|
Change-Id: I99d7eba10f9adee3faccb1eb4a0e3bfd94fcaed2
Signed-off-by: Trevor Bramwell <tbramwell@linuxfoundation.org>
|
|
Change-Id: Ifdc9d00a882990c893d27599e2bae4f387e9bdf0
Signed-off-by: Fatih Degirmenci <fatih.degirmenci@ericsson.com>
|
|
Change-Id: I9f570cb5594d6b27e433fd5920f047209c404dcd
Signed-off-by: Fatih Degirmenci <fatih.degirmenci@ericsson.com>
|
|
This is the first step towards having alignment of the job structure, applying
commit gating and use of labels for all installer jobs.
Apart from the above, it also
- puts all CI baremetal and virtual PODs together
- gets rid of use of slaves directly for CI baremetal and virtual PODs
- switches labels for all CI baremetal and virtual PODs using below labels
{installer}-baremetal
{installer}-virtual
- puts all CI build slaves together and switches to labels - except huawei-build
- puts all none-CI PODs together
Please note that this change does not impact the overall job structure.
Change-Id: I2d8d15799a57fe7e063d39f81b96d19c218b100e
Signed-off-by: Fatih Degirmenci <fatih.degirmenci@ericsson.com>
|
|
Change-Id: I366332922a91b1e957ba104fe8d5cee07398787d
Signed-off-by: Fatih Degirmenci <fatih.degirmenci@ericsson.com>
|
|
|
|
Change-Id: I1ee2593445737da135ba52b07960d4c0f0d52a78
Signed-off-by: Fatih Degirmenci <fatih.degirmenci@ericsson.com>
|
|
Includes job to kick off retention script
Retention script will just list files, nothing is done at this time
does not yet dive into the brahmaputra directories.
Change-Id: I2e608262130326944880b9694fbb8d6a12c09be0
Signed-off-by: Aric Gardner <agardner@linuxfoundation.org>
|
|
gpg agent does not work in non interactive mode.
merging this so that I can reproduce and debug
Change-Id: I641e8988885aa2d0acf794c9ec7592cb6c90a3e9
Signed-off-by: Aric Gardner <agardner@linuxfoundation.org>
|
|
Change-Id: I127d4634448f32e6bf9df53c6dc441cd63d5f505
Signed-off-by: Paul Vaduva <Paul.Vaduva@enea.com>
|
|
We reconfigured centos build server now and it is important to
know which jobs require centos and which ones ubuntu. The machines
with ubuntu were labelled with opnfv-build previously, preventing
us from keeping track of which projects require what OS. This tries
to solve that.
Change-Id: I1fb2912ec49f5bc2781853e500508d9992d59fbb
Signed-off-by: Fatih Degirmenci <fatih.degirmenci@ericsson.com>
|
|
Change-Id: I33cc3ddd5971231fd6d188a8c6d8c48a5885e31b
Signed-off-by: Fatih Degirmenci <fatih.degirmenci@ericsson.com>
|
|
This patch will do the default dpeloyment with xenial than trusty.
Change-Id: I8c628d90490d45a81bc339783ca5c7120d2bf5b7
Signed-off-by: Narinder Gupta <narinder.gupta@canonical.com>
|
|
Change-Id: I317321bbaf1e74419b9508c6da3039b95d9d399d
Signed-off-by: Josep Puigdemont <josep.puigdemont@enea.com>
|
|
This patch to modify the default deployment to mitaka. Ans also
disable the creation of external network as now joid deploy.sh will
create the external network itself based on labconfig.yaml
Change-Id: I282eaa803af47f04348532ffd1a7a5990d50d380
signed-off-by: Narinder Gupta <narinder.gupta@canonical.com>
|
|
Change-Id: Ia56ceefe5bf93cf15f89133c40f0e23c678e11d3
Signed-off-by: Tim Rozet <trozet@redhat.com>
|
|
This change switches compass jobs to use labels instead of tying
jobs to slaves.
The changes in detail are:
- Compass jobs can roam on any pod that is labelled with compass-baremetal
and compass-virtual.
- Above means that there is no branch/pod allocation anymore.
- New jobs are created in order to get rid of pod names from them and just
use baremetal or virtual in names.
- Old jobs are removed.
- Related test jobs are updated.
- huawei-virtual label is renamed to compass-virtual since we still have
installer specific slaves, not company specific. In future the job
might end up on a none-Huawei slave too.
- huawei-pod2 is not a CI POD so it is left in none-CI PODs section.
- block relations and triggers might require fine tuning and left to compass
project to decide how to adjust
Job structure remains same; one parent job calling 3 downstream jobs:
deploy, functest, yardstick.
The OS to deploy (trusty or centos7) needs to be solved in a separate patch
in a sustainable way.
Change-Id: I22bfaff36640667b6764f0e303a3286af0eb41b2
Signed-off-by: Fatih Degirmenci <fatih.degirmenci@ericsson.com>
|
|
This change switches joid jobs to use labels instead of tying
jobs to slaves.
The changes in detail are:
- Joid jobs can roam on any pod that is labelled with joid-baremetal
and joid-virtual.
- Above means that there is no branch/pod allocation anymore.
- New jobs are created in order to get rid of pod names from them and
just use baremetal or virtual in names.
- Old jobs are removed.
- Related test jobs are updated.
- POD specific parameters are updated and instead of adding pod
specific parameters, they now become parameters for all the pods
labelled with same label. It is not a problem at the moment as
both Intel PODs (pod5, pod6) have same parameters but a different
pod with different parameters will require alignment between pods.
Job structure remains same; one parent job calling 3 downstream jobs:
deploy, functest, yardstick.
Change-Id: I30d270cd05ba3756f6f161d65ee5160eab6b80f4
Signed-off-by: Fatih Degirmenci <fatih.degirmenci@ericsson.com>
|
|
This change switches fuel jobs to use labels instead of tying
jobs to slaves.
The possibilities and changes introduced with this approach are
- If things do not work as expected, slave-label parameter can
be set to certain slave name to get back to old way of
tying jobs to slaves.
- If a scenario requires a different slave, it can still be
created using baremetal or virtual but the label can be
different, pointing to a slave that is capable of running
that scenario. (fuel-baremetal-multisite, etc.)
- The auto-trigger-name for brahmaputra is moved under the
branch anchor and can be used like this if triggers for all
all the jobs for a certain branch need not to be triggered
automatically but job is still kept enabled in case if some
crucial bug needs to be tested.
- No of max concurrent builds stepped to 4 in order to use
all the labelled slaves available. This will require some
adjustments before colorado jobs start to run in order to
share resources between branches fairly.
- All the fuel CI PODs except lf-pod1 have been removed.
Change-Id: I91eca6743bb3d34263e634f9b74550efa1e2ddbe
Signed-off-by: Fatih Degirmenci <fatih.degirmenci@ericsson.com>
|
|
Change-Id: I55255a33bb0c2cf88c709024f1712e5310082ae6
Signed-off-by: Fatih Degirmenci <fatih.degirmenci@ericsson.com>
|
|
JIRA: YARDSTICK-269
Change-Id: I78082402e3083fbab3e0b8974f23ccd4dad2e980
Signed-off-by: MatthewLi <matthew.lijun@huawei.com>
|