Age | Commit message (Collapse) | Author | Files | Lines |
|
|
|
Scenarios may not support all XCI flavors so record that information
per scenario basis. This will resolve failures when we try to deploy
a scenario that doesn't support the selected flavor.
JIRA: RELENG-338
Change-Id: I5f73b139ee6c1831896aa32147c4dc1368673d92
Signed-off-by: Markos Chandras <mchandras@suse.de>
|
|
The configure-opnfvhost playbook which is used for all flavors except
AIO can also be used for AIO if it's adapted to not configure networking
on the host and also do not fail if there aren't any OSA variables files
available so lets do all that in order to avoid duplicating code across
flavors.
Change-Id: I58379d8b52094294b9349494753ffcdd44982013
Signed-off-by: Markos Chandras <mchandras@suse.de>
|
|
|
|
The OPENSTACK_OSA_PATH only makes sense on localhost. As such, when we
use it on playbooks that operate on remote hosts, the result is not
predictable. However, we rsync the entire releng-xci repository to the
opfnv host so we can make everything predictable by simply clone
everything in advance in the .cache directory. That directory is then
rsync'd to the opnfv host. As such, we can repurpose the
OPENSTACK_OSA_PATH to point to the path into the OPNFV host. Moreover,
all external repositories are being cloned to .cache/repos so we can
eliminate some variables in order to simplify the code. Finally, we
bring back the ability to use an external OSA repository for
development purposes.
Change-Id: Ieef3e22ae2085f6735185634d555cfc0d4b69b39
Signed-off-by: Markos Chandras <mchandras@suse.de>
|
|
There is too much noise when XCI is cloning repositories, installing
packages etc so lets make the console output somewhat more readable.
Change-Id: I15667e0349f943ad5488daf0e3cea9336c9166d1
Signed-off-by: Markos Chandras <mchandras@suse.de>
|
|
|
|
This change fixes the scenario os-nosdn-nofeature by
moving directories/files from os-nosdn-ovs to os-nosdn-nofeature.
The contents of the files and variables are also either adjusted
or removed in order to prevent impacts on CI.
Change-Id: Icfde27f413c8f93d097f9f262c8cb1230b7fe59d
Signed-off-by: Fatih Degirmenci <fdegir@gmail.com>
|
|
Change-Id: I5a3f533f61b6572d8070fe5f0a306ef9d418ced8
Signed-off-by: Manuel Buil <mbuil@suse.com>
|
|
Previously, we used to clone the releng-xci repository under a directory
in /tmp, copy our changes to that repository and then run the
xci-deploy.sh script from it. However, this made things far too complex
for deployers and developers since some playbooks were used from the
local repo whereas others were used from teh /tmp checkout. By running
everything from our local repository simplifies things a lot since we
can directly test our changes and also reduces the code we have in our
playbooks.
Change-Id: If16aa51b2846c170676df82d25cb90e26b1568b2
Signed-off-by: Markos Chandras <mchandras@suse.de>
|
|
The provision-vm-nodes playbook is only responsible for running bifrost
to provision the XCI virtual machines. As such, drop all the extra tasks
that this playbook performed in order to simplify it. This also drops
the XCI_EXTRA_VARS_PATH variable since it had a rather obscure usage
and there is not much value in passing inventory information outside of
XCI. All these will be handled in the scenarios themselves.
Change-Id: If5cb381a3d1e101100eb04478d80fb3045cdfaf8
Signed-off-by: Markos Chandras <mchandras@suse.de>
|
|
|
|
|
|
All scenarios are being cloned to XCI_SCENARIOS_CACHE so look
there for the various override files. This will allow external
scenarios to influence the XCI environment.
Change-Id: I39a48ce55baaa29d09737ce6232867ef1165f099
Signed-off-by: Markos Chandras <mchandras@suse.de>
|
|
|
|
Change-Id: I4ea6b421a24f41ea3ca925ea98d961e1a34e7876
Signed-off-by: Periyasamy Palanisamy <periyasamy.palanisamy@ericsson.com>
|
|
Previously, the scenarios were symlinked from the checked out location
to {{ playbook_dir }}/roles. However, the symlinks were pointless since
they only exist in our local releng-xci directory. The roles much also
be present in OPNFV_RELENG_PATH which is the place where the deployment
happens. We use 'rsync' to copy our local releng-xci directory to the
remote one so our roles can be automatically be placed there if we copy
them instead of symlinking them.
Change-Id: I735db71eedfaa51d7196761aa2af8329d86cb775
Signed-off-by: Periyasamy Palanisamy <periyasamy.palanisamy@ericsson.com>
Signed-off-by: Markos Chandras <mchandras@suse.de>
|
|
|
|
We are installing the docker package but not checking if it is started
The service name for the three distros is the same but I still added the
variable in each distro variables file to keep best practices
Change-Id: I0c73069ea7edc366e824cf39d14d24d1416fd6c3
Signed-off-by: Manuel Buil <mbuil@suse.com>
|
|
The OPNFV_RELENG_DEV_PATH variable was used to point to a releng-xci
development repository. However, people normally set the current
directory as the development one and they almost always want to
test the current code in XCI. Using an secondary releng-xci tree
as development repo is a very obscure case and it normally complicates
things. As such, let drop this option and always use the current
repository for development purposes.
Change-Id: If111bf29a32a5f6ea28694f191645af0c6a87abc
Signed-off-by: Markos Chandras <mchandras@suse.de>
|
|
Change-Id: I9253edf028fce571e04f9f82103a94952e05d2d4
Signed-off-by: Periyasamy Palanisamy <periyasamy.palanisamy@ericsson.com>
|
|
|
|
The RUN_TEMPEST global environment has been defined in *user-vars*
but never used in the playbooks. This change pretends to enable the
use of that value.
Change-Id: I49ca092546494c0cdcb015a549828bf79fa5f889
Signed-off-by: Victor Morales <victor.morales@intel.com>
|
|
br-vlan-eth was created but not added to any bridge, therefore floating ips
were not working as it was impossible for the OVS in the compute to get the
ARP messages coming from the gateway of the public network
Change-Id: Ia0828b7e96359dc10012ac52bf0e5d4f2c5419cb
Signed-off-by: Manuel Buil <mbuil@suse.com>
|
|
When creating the external network for functest testing purposes, it does not
need to be shareable between tenants because functest uses routers and floating
ips. Therefore, better remove it to avoid confusion
Change-Id: I18a1fd34a52ddae2527ee3580a9d1a2b649e61b6
Signed-off-by: Manuel Buil <mbuil@suse.com>
|
|
The flavors may contain deployment specific user_variables.yml files
so we should only copy the generic one in the common tasks and leave
the rest for scenarios to copy them if necessary.
Change-Id: I39e3e090dc40d4f13bb42028a24e8574e19cf6dd
Signed-off-by: Markos Chandras <mchandras@suse.de>
|
|
|
|
|
|
We were not getting the eth12 interface in the computes because the
network-config-suse file did not have executable rights:
-rw-r--r-- 1 root root 300 Nov 9 14:40 network-config-suse
Change-Id: Ia796123b331b67bc442d3506b944f0430f7a5f32
Signed-off-by: Manuel Buil <mbuil@suse.com>
|
|
Openstack ansible support to deploy ceph.
The purpose of this patch is to configure the ceph,
just like we configure other openstack components.
The default is to not deploy ceph.
If you want to deploy ceph you just need to
export XCI_CEPH_ENABLED=true before running xci-deploy.sh.
When deployed successfully, the openstack storage will use ceph.
Change-Id: Ifd8d16fdce2914b6316842e72bbfd93228ea059d
Signed-off-by: wutianwei <wutianwei1@huawei.com>
|
|
The new bootstrap-scenarios.yml playbook is aimed to be used by
scenarios to configure themselves right before we proceed with
the bootstrapping. This way we can avoid having scenario specific
stuff into the generic code.
Change-Id: I3026013719a50458c65405439712270f25f569ff
Signed-off-by: Markos Chandras <mchandras@suse.de>
|
|
In order to plug the scenarios' roles properly, we need to have all
roles physically present in advance. As such, add a
opnfv-scenario-requirements.yml file which can be used to populate the
roles directory with all the scenarios.
Change-Id: I0cdadb63849e4565c31559817660d23217879053
Signed-off-by: Markos Chandras <mchandras@suse.de>
|
|
|
|
This playbook should be executed in case we want to use the osa deployment
as platform for testing a scenario. The playbook currently includes a call
to the prepare-functest role which installs and sets-up everything to run
functest based tests
Change-Id: Ibb59825e64c5889f36fd3bfb38a4d23b67b11866
Signed-off-by: Manuel Buil <mbuil@suse.com>
|
|
OpenStack-Ansible can make use of all the user_*.yml files so scenarios
can make use of this facility to override the default ones we provide
for every flavor. As such, lets copy all the user_*.yml files to the
deployment host.
Change-Id: I5ddce48677f66a3ff3ec127f0fc0163f9f45b200
Signed-off-by: Markos Chandras <mchandras@suse.de>
|
|
If we leave the quotes, functest will search for a neutron network with the
name "ext-net" instead of ext-net and will fail
Change-Id: I4ef62fa6ea69573f5872b864402ddb8644ce432d
Signed-off-by: Manuel Buil <mbuil@suse.com>
|
|
docker-py package is required by the docker_container module in ansible:
http://docs.ansible.com/ansible/latest/docker_container_module.html#docker-container
Change-Id: Ib051ae09c84cfa973ef814852e78626499471d0f
Signed-off-by: Manuel Buil <mbuil@suse.com>
|
|
|
|
Change-Id: I5a8ff4c358ab9f1d548c9cc834faa2465ec63d41
Signed-off-by: Tapio Tallgren <tapio.tallgren@nokia.com>
|
|
It is important to have the client installed for the ease of the
users so they can use the client for trying out their new deployment.
Change-Id: I091c1e23ff320cb9fcdb2743998483a6bc1fcdad
Signed-off-by: Fatih Degirmenci <fatih.degirmenci@ericsson.com>
|
|
In CentOS and Suse, /etc/ssl/certs is a symbolic link to
/etc/ssl/pki/tls/certs. The Ansible module "file" will fail if it is asked to
create a directory which is already a symbolic link.
This patch will check if /etc/ssl/certs exists before trying to create it.
The same check is done both on the host and the opnfv guest VM.
Also, /etc/certs is only writable by root, so we need a "become: true" clause
to be able to modify it in localhost (but not in opnvf VM).
Change-Id: Iab6c3c162548f84ad6082829e4a7c2ab63d2cfa0
Signed-off-by: Tapio Tallgren <tapio.tallgren@nokia.com>
|
|
* changes:
xci: xci-deploy.sh: Apply workaround for checking db cluster on SUSE
xci: playbooks: synchronize-time: Fix service name for openSUSE
xci: configure-opnfvhost: Do not run 'remove-folders'
xci: scripts: build-dib-os.sh: Pin diskimage-builder
xci: configure-opnfvhost: Do not check /etc/ssl/certs on SUSE
|
|
In openSUSE, the chrony service is 'chronyd'.
Change-Id: Ifb946b0ba49783108a84dd5b998d9d45f5c9df51
Signed-off-by: Markos Chandras <mchandras@suse.de>
|
|
The versions for the OpenStack services and global requirements
need to be pinned alongside with the OSA SHA1 and the role versions
to ensure we pinned everything.
This change is required for osa-periodic jobs as well since that
job will use sources-branch-updater.sh script and the script updates
these files to ensure things are pinned correctly.
modified: ansible-role-requirements.yml
modified: global-requirement-pins.txt
modified: playbooks/defaults/repo_packages/openstack_services.yml
modified: releasenotes/notes/glance-init-config-overrides-d1c8c3dcc50c109a.yaml
modified: releasenotes/notes/neutron-init-config-overrides-9d1d2b3b908705ed.yaml
modified: releasenotes/notes/trove-init-config-overrides-a78ed428a32adef8.yaml
By doing this change, we do not need to have any
magic to capture updates to those files. Just updating the files
we keep in xci/file folder will be sufficient and they get copied
over during the playbook execution.
Change-Id: Iae0db22574a0368e896132469a8587d1457ce177
Signed-off-by: Fatih Degirmenci <fatih.degirmenci@ericsson.com>
|
|
The OPNFV node is brand new and it doesn't have any traces
of XCI directories so there is no point in running the
'remove-folders' role.
Change-Id: Ic9b9203cc14abda2dab406de6a5feeef6a1b7e2a
Signed-off-by: Markos Chandras <mchandras@suse.de>
|
|
This is similar to Ibc188b76f47c4f7d1c5aa452a21e838420d65e6a
/etc/ssl/certs is a symlink so skip this task on SUSE.
Change-Id: Iae38640501748dc3dd802ce795acfaeefd836c97
Signed-off-by: Markos Chandras <mchandras@suse.de>
|
|
Functest complains that it cannot connect to the installer because it does
not understand the ip or hostname. The cause for this is the quotes in the
template
Change-Id: I75c30b7f1bbcee3f968692b4347ceb13ab1131d2
Signed-off-by: Manuel Buil <mbuil@suse.com>
|
|
Functest requires xci deployment to have:
- A public neutron network
- A subnet attached to the public neutron network
- The gateway from the subnet attached to one interface
The network and subnet creation are done via a script
instead of using Ansible OpenStack module in order to
get things working asap.
These three tasks are implemented in a role
This patch depends on a pending patch:
https://gerrit.opnfv.org/gerrit/#/c/39517/2
Change-Id: I0635227153087fd9e25b76f0859f82bf48bfb930
Signed-off-by: Manuel Buil <mbuil@suse.com>
Signed-off-by: Fatih Degirmenci <fatih.degirmenci@ericsson.com>
|
|
|
|
* changes:
xci: configure-localhost: Do not check /etc/ssl/certs on SUSE
xci: xci-deploy.sh: Ensure env_reset is not present
|