Age | Commit message (Collapse) | Author | Files | Lines |
|
Var checks for unset required proper parameter expansion to avoid
nounset.
Change-Id: Ide5db47e2243c8aee0f97369bf3cd0673e7f4bd3
Signed-off-by: Tim Rozet <trozet@redhat.com>
|
|
When we run deployments for snapshot creation we typically pass in the
OS_VERSION var so we know what openstack version we are deploying with.
However when we run a verify/gate job to ensure that the snapshot
deploys work we do not have this variable, and need to detect it based
on the OPNFV branch, or if the branch is master, the deploy scenario
naming scheme.
Change-Id: If1445dc564d2af59b9740391de56b821b78de0b5
Signed-off-by: Tim Rozet <trozet@redhat.com>
|
|
This patch changes snapshot related scripts to allow for more than just
the current snapshot type of "csit". This allows us to add other
snapshot types, including the functest type. The same
snapshot.properties file will be used to store information for all types
of snapshots. The snap cache folder structure will remain the same, but
hold different types of snapshots in the same folders.
These changes include renaming the URLs for the snapshot artifacts to
reflect the snapshot type, but remains backwards compatible until enough
promotions have occurred and other apex snapshot deployment scripts have
been migrated to the new format.
Change-Id: If3e47e700b5ad9c95418179042b85814fe389233
Signed-off-by: Tim Rozet <trozet@redhat.com>
|
|
With changes to deploy scenario files for gambia, we still need the
logic to pull in the right csit-env. That logic broke as it depended on
DEPLOY_SCENARIO with the changes to scenario names. This patch fixes it
to check openstack version (OS_VERSION) instead.
Change-Id: Ie130efd0fc3cb7f09a3fa4e896ad461d24394777
Signed-off-by: Tim Rozet <trozet@redhat.com>
|
|
With the move to upstream artifacts post-fraser the checks to use
upstream artifacts were relying on BRANCH == 'master'. This doesn't
scale as we preserve fraser functionality but add more branches like
gambia. Therefore inversing the logic to accomodate for new branches.
Change-Id: Ice17f6afd33744f44d253d320a194c3f6d02d8bf
Signed-off-by: Tim Rozet <trozet@redhat.com>
|
|
This patch allows us to run a gate job against csit environment, by
doing something like:
start-gate-scenario: os-nosdn-nofeature-noha-csit
Change-Id: I3041404cdfdffe159cff33ecd72be93fe9383e10
Signed-off-by: Tim Rozet <trozet@redhat.com>
|
|
This allows us to run a verify job on an all-in-one (control+compute)
type of deployment.
Change-Id: Ia4174c8eb5d55aaa05e63ddce59301e62b3a8a04
Signed-off-by: Tim Rozet <trozet@redhat.com>
|
|
Change-Id: I4f860f9b9adcc96d94b1daa333eec7a746bc91b1
Signed-off-by: Tim Rozet <trozet@redhat.com>
|
|
Changes-Include:
- Use correct csit env file with promote jobs
- Ensure odl container starts when snapshots brought up
- Use SDN_CONTROLLER_PASSWORD for ODL pass rather than
hardcoding admin
Change-Id: I910a4eefa2b4c621e3f82939bedee338fad91bf3
Signed-off-by: Tim Rozet <trozet@redhat.com>
|
|
This route was used as a hack for JIRA: APEX-112
Change-Id: Ia7afc95389ed4e3c7a4128318868bfe562b34b62
Signed-off-by: Tim Rozet <trozet@redhat.com>
|
|
Changes-Include:
- Use 2 computes as required by ODL team
- Get rid of checking JOB_NAME var for csit, and simply check if
PROMOTE var is True
- Pass PROMOTE all the way from top-down into deploy job
- Use CSIT network settings file when deploying CSIT
Change-Id: I56d953516ce690749634b474e553baad3a762364
Signed-off-by: Tim Rozet <trozet@redhat.com>
|
|
Change-Id: If50ad310049edca687da8e0acf412a7e1eacd2b6
Signed-off-by: Tim Rozet <trozet@redhat.com>
|
|
Change-Id: I92635178ddc354dd7773cd372a34d46f5ee0bf2a
Signed-off-by: Tim Rozet <trozet@redhat.com>
|
|
- Change scenario parsing to allow k8s scenarios in gate jobs.
- Remove scenario detection in shell scripts.
Also changes network setting file for fd.io scenarios as we
don't use separate config files any more.
Change-Id: Icbcc6cb7b6b66c7cc43f5d6c58dd97730e41ea5f
Signed-off-by: Feng Pan <fpan@redhat.com>
|
|
Change-Id: I37c7b8b6e4bd19ef94b9b42fe2e5e89cc3e2da21
Signed-off-by: Tim Rozet <trozet@redhat.com>
|
|
Scenario names with 'upstream' in them will not download any artifacts
and deploy from upstream. For now, this even applies to the apex
python RPM for daily deployments. We will only use git repo for
daily until after Fraser.
Change-Id: I0da16dfde117ba6c1e7597294d8e4afc8501dd53
Signed-off-by: Tim Rozet <trozet@redhat.com>
|
|
The tmp directory no longer holds large files, and removing this while
other jobs are running can cause build failures.
Change-Id: I504d06e2e114dd1be4fe3790fcefaf97c724552c
Signed-off-by: Tim Rozet <trozet@redhat.com>
|
|
Change-Id: I2b7ee35500da4523a6cb872f89225fad6dd8af7f
Signed-off-by: Tim Rozet <trozet@redhat.com>
|
|
Change-Id: I3246aa7d7b26b2d79ba9bd23aba4f305eb47fd94
Signed-off-by: Tim Rozet <trozet@redhat.com>
|
|
Change-Id: Icac677ac9e17b4b4ae4e7fc82139587cc201bcb6
Signed-off-by: Tim Rozet <trozet@redhat.com>
|
|
Change-Id: I59523b6233ca48c2eb2d4b4934263f0114d8e7a8
Signed-off-by: Tim Rozet <trozet@redhat.com>
|
|
We dont need to use the apex/ci directory anymore and the ci scripts
should be in releng. A follow up patch will remove the ci/clean script
usage.
Change-Id: Ib2c7bd3d2dfee360bb13b42852ff2010a691a018
Signed-off-by: Tim Rozet <trozet@redhat.com>
|
|
Change-Id: I4a1da6ba1efe2b704f0109f63817021435d283b8
Signed-off-by: Tim Rozet <trozet@redhat.com>
|
|
Change-Id: I36066bf003b1504524a5b8143cceef7829b23ca8
Signed-off-by: Tim Rozet <trozet@redhat.com>
|
|
Change-Id: Iba1b2f47e8ace4fb4a9eed80342fcb8950e848f8
Signed-off-by: Tim Rozet <trozet@redhat.com>
|
|
Change-Id: I0455283ba94501119786ffbc3b1217bb71cc35c7
Signed-off-by: Tim Rozet <trozet@redhat.com>
|
|
Change-Id: Ib71936bec70951dea7478c1dcf657c5f2581fdf6
Signed-off-by: Tim Rozet <trozet@redhat.com>
|
|
Change-Id: I5431e1c86d9f56b5192880dc9e7eb6f2f0ab61ee
Signed-off-by: Jamo Luhrsen <jluhrsen@redhat.com>
|
|
Change-Id: I7da324903df1cc0c26cfba4433748dab459bbeb1
Signed-off-by: Tim Rozet <trozet@redhat.com>
|
|
Change-Id: I515afe7a8f230ebbece8ee4d99ca7744a6a35fb8
Signed-off-by: Tim Rozet <trozet@redhat.com>
|
|
- Remove any build blockers on 'apex-build-*' jobs.
- Adds script to download latest build artifacts before deployment.
- Make each deploy download the needed artifacts before running.
- ARTIFACT_VERSION now set to dev for verify builds to indicate tarball
should be uploaded to artifacts repo
- apex.yml jjb is now generated by using apex-jjb-renderer
- scenarios are managed in scenarios.yaml
JIRA: APEX-328
Change-Id: I4a51565686bd2011a310c3926d8260cd4641dca5
Signed-off-by: Trevor Bramwell <tbramwell@linuxfoundation.org>
|
|
Change-Id: I6f42ffe5fad41ac90830784d8a4ec8ee04eae7d2
Signed-off-by: Feng Pan <fpan@redhat.com>
|
|
Change-Id: Ic482f75bb905082f503958c1be01c9cd7e68566a
Signed-off-by: Feng Pan <fpan@redhat.com>
|
|
Change-Id: Ifeea80111dd55785acbc6697ae216da4c8d3670a
Signed-off-by: Feng Pan <fpan@redhat.com>
|
|
Change-Id: I7df2c6a6a940f49268146334cb7f77d3fffeb665
Signed-off-by: Feng Pan <fpan@redhat.com>
|
|
|
|
|
|
Change-Id: I42f8836320e4bce2f890aa4a17d59f53efab7f97
Signed-off-by: Tim Rozet <trozet@redhat.com>
|
|
- CONFIG and RESOURCES have been replaced by BASE and IMAGES in Danube
Change-Id: I1b1d4e73b1f4e8a13d63bd199ec4c91e853d14a6
Signed-off-by: Dan Radez <dradez@redhat.com>
|
|
Change-Id: I9169434d3dc62c911d351f8238af2c5daf314d99
Signed-off-by: Dan Radez <dradez@redhat.com>
|
|
According to FDS folks, uio_pci_generic (default) should work with UCSB
baremetal. We also hit a bug in UCSB where vfio-pci isn't working with
VPP. Therefore falling back to uio_pci_generic.
Change-Id: Id9a6fe5be91aaf8476292a46c15c6e876616f4a6
Signed-off-by: Tim Rozet <trozet@redhat.com>
|
|
Change-Id: I298b09f654dfe5efa64df9dc287fd586fcdf131d
Signed-off-by: Tim Rozet <trozet@redhat.com>
|
|
ONOS on Apex will not be supported in Danube
Change-Id: I58979724df7f6fa359e9789458ac305965ce71b1
Signed-off-by: Dan Radez <dradez@redhat.com>
|
|
Change-Id: I6331f835ec20abeb5e001ccd7c29a76ad72c583c
Signed-off-by: Tim Rozet <trozet@redhat.com>
|
|
For now this is a manually triggered job, until there is a demand to
produce it automatically. Only the snapshots are uploaded for this job
and we are not updating snapshot properties file. We will move to
making that more dynamic in the future.
Change-Id: I3d8392c2c99a524fd20de82d5032c2d4557b045a
Signed-off-by: Tim Rozet <trozet@redhat.com>
|
|
Change-Id: I6c06ecd40b5c2d57ca8f62fd52fe78267b4a8735
Signed-off-by: Tim Rozet <trozet@redhat.com>
|
|
Change-Id: I997202032fd868802166fb84aa0f77b27abf28cd
Signed-off-by: Feng Pan <fpan@redhat.com>
|
|
Change-Id: I9bbcd1b801d10b6fab6b32b33372260ac2e65cae
Signed-off-by: Feng Pan <fpan@redhat.com>
|
|
Change-Id: I5aaf203f3db5078e462007c78aa9ff13d102c70c
Signed-off-by: Tim Rozet <trozet@redhat.com>
|
|
Apex needs a way to run a gate job after basic scenarios (nosdn,
odl-nofeature) pass. We need the capability to specify which scenario
should be the gate. This patch allows us via a comment to specify
'start-gate-scenario: <scenario>' which will start a gate job to
validate that scenario deploys and passes basic functest health check.
Some drawbacks of this approach include only being able to validate a
single scenario per comment + gerrit votes overriding (not true
workflow/pipeline plugin). The positive is this is a fairly small CI
change to enable some much needed verification. It will be the Apex
committers responsibility to issue this comment post Jenkins +1 approval
of pre-gate.
Change-Id: I9a1364bb1cece8b35488d496789a5e83c707b32d
Signed-off-by: Tim Rozet <trozet@redhat.com>
|