upstream neutron-fwaas-dashboard has added/renamed files, update
osa horizon role to account for this
Change-Id: Iee185167fefdcb4c85006df07992ccbd1c501420
This is required after duplicate argument settings, fixed in
https://review.openstack.org/#/c/559071/
The branch must be overridable per item, and should default
to the master branch, and be the same for all the items
by default.
Change-Id: I656c8793714b5dc764d7aa0cef28e5eca65d927a
Co-Authored-By: Jean-Philippe Evrard <jean-philippe@evrard.me>
Heat is now a plugin that needs to be enabled in horizon [1].
This uses the same process as the other plugins.
[1]: https://review.openstack.org/#/c/523402/
Change-Id: Ic3ec4684d941df6343074d9275fcfaeff6724181
Closes-Bug: 1745400
If any user wants to provide its own dashboard, he can now
install its own pip package by overriding the proper variables,
and then do the linking directly by overriding the new variable,
``horizon_panels``.
If we wants to use any existing horizon_panels, and just add a
personal panel, a deployer must define the override like this:
``horizon_panels: "{{ my_horizon_panels + _horizon_panels }}"``
where ``my_horizon_panels`` is a list whose items are composed
of src, dest, and state.
Change-Id: I87324aaa3a685b1e65081b7c7dce767c080f22ee
Add support for the openSUSE Leap distributions. Additionally,
for openSUSE, we need to load some extra apache2 modules which are
not there by default. Finally, Apache on openSUSE configures some
of it's components in httpd.conf and some others in default-server.conf
so it's not possible to drop one of them as that will likely break
Apache. A future improvement would be to drop both and provide a good
httpd.conf template. Until then, we need to drop the default
configuration for the root directory as that breaks Horizon since
it does not allow symlinks and global access to it.
Change-Id: I51613df836c6a507f6f36967c0ce4b76ba9202a9
Restarting the apache service under ubuntu read the envvars file.
This envvars contains the folder for logrotate. If the folder
doesn't exist or hasn't the apache permissions, the restart of
the service will fail.
We ensure here the folder properly exists, with the right
permissions.
Change-Id: I041a044ef88aff917eeb3a618c2c08ac9ceb80fe
The security guide suggests that /etc/horizon/local-settings.py
should be owned by root and in the horizon group with 0640 permissions.
Change-Id: I4a49394cd2afb92e0da1b233e21bad862ddee059
In order to make it easier to differentiate between the lists of
python packages, distribution packages, downloaded packages,
package pins and other similar variables the variable names are
being changed to ensure that they have a more explicit suffix
that defines the purpose and makes the naming more consistent.
This is to facilitate a lookup plugin which will be able to look
up all the package lists and present them as a consolidated piece
of data which may be used for artifact preparation.
Change-Id: Ib4b177992753ad53b8bdb7a7652f1fbc6cc5267d
The Xenial gate is failing due to missing "gettext" package, which is
required for the "horizon-manage.py compilemessages" command to run.
This PR adds gettext as a package to install and should ensure the
Xenial gate now passes successfully.
Change-Id: Ib744121b002970e606b86dde4ed356184433cba7
These changes allow the role to be tested on Xenial hosts
while also ensuring that the tests are testing everything
within the container as expected.
Overview:
* Included in this PR is a manual-test.rc. This has been
added to allow developers to run tests locally without
having to invoke tox. This RC file was lifed from the
os_keystone role.
* The git package was added to the install list. This is
needed on the target when the role is running with developer
mode enabled.
* Connection settings within the container create prep playbook
were removed. A``connection: local`` setting forces all
commands to be executed against the host regardless of a
delegated task.
* A task has been added to the horizon role to when SSL is
enabled. This task ensures that that ca certificates are
updated. Without this change, when running on Xenial,
tempest will fail due to the following error:
[SSL: CERTIFICATE_VERIFY_FAILED] certificate verify failed (_ssl.c:590)
By updating the ca certificats on tempest is able to verify
the certificate and pass the tests. NOTE: This fix came from
the puppet-OpenStack community where they ran into the same
problems in Xenial as discussed here: [0]
[0] - https://irclogs.ubuntu.com/2016/05/18/%23ubuntu-server.html#t13:37
Change-Id: I1e6808c49a8faaba5b2748918be5d6b5a59fd3d1
Signed-off-by: Kevin Carter <kevin.carter@rackspace.com>
Debian-specific vars and logic have been moved to tasks
that will execute only on those distributions.
Change-Id: I16664a9c4364938a065bf07472e25e93ba1b828c
Implements: blueprint multi-platform-host
