Add more debug when Puppet fails during first and second run.
It will help people debugging CI to figure why job failed.
Change-Id: I9ef32afaa594a65be6ed49f40bd8fba735fe2c70
When running periodic jobs, ZUUL_BRANCH & ZUUL_REF are empty because we
don't have any reference or branch set for the jobs.
To not make zuul failing when cloning Puppet modules, let's set defaults
values:
ZUUL_BRANCH default to master (we are running tests against master now)
If we need to test stable branches, we'll revisit later but it would be
a bigger change (job names, etc).
ZUUL_REF default to None.
Change-Id: I34b298230bc174a3fdd431095043543c4c223296
Deploy Cinder API in WSGI with Apache instead of eventlet.
Change-Id: I6e6fb66cf79c43e063a2f4b3fea1017dd4ccd661
Depends-On: Ia1d3cfaa5ce6e4719b6c008bcc4f75d3c1db3487
In order to help be able to determine what part of the script a failure
occurs in, this change adds a print header function and updates the
shell scripts to print some header information for various actions.
Change-Id: I8e622c4d1c9c03aa3dbcb27712a346ce3284003f
Add an empty other-requirements.txt file to prevent the
openstack gate environment to pre-install dependencies for
us; these are installed by our Puppet modules or run_tests.sh script.
Change-Id: I96c733b03933cce04b0beee8952a2c49f83bb5b5
puppet-ovn is part of Puppet OpenStack project, and we'll start testing
it in puppet-openstack-integration.
Change-Id: If7329f5758819160b305c90241f56d3dea566496
Stop to run mistral tests for now, they fail too often.
We'll enable it as soon as we find what's wrong.
Change-Id: I041b6d35708c10df6bbddf672f56deb56ff86ab7
This patch will change the way we deployed Ceph on CentOS nodes.
* Use CentOS SIG by default
https://wiki.centos.org/SpecialInterestGroup/Storage/
* Stop deploying EPEL, which is conflicting with RDO packaging.
All our CI needs should be in RDO and SIG. Let's stop deploying EPEL at
all.
Depends-On: Ic42c19b2b5b4bc499565c69c3053395173c29a8e
Change-Id: I8d435a77c41c38354b20a40ab30dbc595cb6412c
A new version of python urllib send us ugly warnings because our SSL
certificates don't have SubjectAltNames.
I re-generated some SSL certs with it, for both ipv4 & ipv6 deployments.
Change-Id: Ibed9f23869de9d2871c3d25e9bd24df809aa4c16
Do not merge, testing with v3 auth
* Deploy Mistral on scenario003
* Run Tempest tests for Mistral
* Bump repo to recent RDO repo that contains a fix in Mistral to make
Tempest tests working.
Change-Id: Ia43427cc207ee5b7b95e9610f06a568992fb58cd
Switch package_provider to use $::package_provider fact present in
puppetlabs-stdlib.
It will avoid any collision later.
Change-Id: I2df1bff1b911c22eed8a2f6b9c0ab38edda43156
Run the tests in permissive mode but run sealert at the end to see if we
have some SElinux alerts that we need to fix.
We want to run catch_selinux_alerts function before exiting the script
(with failure or success) so we can detect alerts:
* right after first Puppet run
* after second Puppet run
* after Tempest run
If an AVC is detected, fails so we can report the bugs in
openstack-selinux and bring useful feedback.
Change-Id: Iaef45c88232505600cbb80f82b195aafcc615474
This commit will add explicit dependencies in some resources in the
provision profile.
Without this change the CI does not pass on Puppet 4 because of the
change in default run order.
Depends-On: Ifc3f40ae1cb71c215b8a0bf6cd5f6d05101a6168
Change-Id: Iedfa6f78936676ba54ebb888734a3b381a2378e7
This is a workaround until https://review.openstack.org/#/c/304399/ will
be merged and a new image will be built and uploaded to OpenStack Infra
Clouds.
Until then, we'll carry this code so we can move forward with Puppet4
testing and deploy Facter correctly.
Change-Id: I823087aed543c3cf43d247f7adf5f32c5c0fc31e
epel-release is installed by OpenStack Infra when building images.
The problem is RDO + EPEL does not work find, there are some
dependencies issues.
This patch makes sure EPEL is not installed when we start our CI jobs.
The next steps will be to run puppet-openstack-integration with EPEL
disabled in ceph::repo, and patch ceph::repo to use centos ceph repos [1]
[1] https://bugs.launchpad.net/puppet-ceph/+bug/1562998
This patch is part of the efforts to stop using EPEL at all.
Change-Id: I506754e2c5021ee728675a8771f2e24b8576f7d5
puppet-zaqar is freshly new and we want to test it.
* configure mongodb replicaset (required for zaqar): 'openstack'.
* deploy zaqar on scenario002 with mongodb backend.
* configure tempest to test zaqar.
* run TestManageQueue tempest tests in tempest.
Note about TestManageQueue:
Zaqar does not maintain Tempest tests, and run their own functional
testing tools. api.messaging.test_queues tests are known to work, so
they will be the only tests run in our CI to validate Zaqar is working.
Note:
* SSL does not work out of the box, we need to add support for WSGI
deployment with Apache, but Zaqa does not provide WSGI app, it's in
their roadmap.
* IPv6 binding on default wsgi does not work, so using IPv4 now.
Change-Id: I1aa563f37aa7a231b2218f79b72c73c668d968f3
If IPv6 is enabled (on centos7 now), deploy Swift using this binding.
Note: adding notes in TODO to also deploy memcached in IPv6 later.
Change-Id: I8c8e8e8870212765f8b8a0d639194e040963d6e0