This PS adds support for CentOS, Fedora and running the gate locally
on a developers machine. It also expands on the original gate scripts
by setting up dns access to the OpenStack cluster from the host's
running OpenStack-Helm using a daemonset.
Change-Id: I99c573a44a73452d558e612fc020e22e7f41ade7
This PS adds a default pod disruption budget for mistral-api
to ensure at least one replica is running.
Change-Id: I45d89cacc15a41f9eb3663dfd1f250be1567c3c0
This PS adds a default pod disruption budget for magnum-api
to ensure at least one replica is running.
Change-Id: Idaff44d28960f4fe62a911758d3d75de1ad2cb85
This PS adds a default pod disruption budget for glance-api and
glance-registry to ensure at least one replica of each is running.
Change-Id: I06364bc2bb816fd4877353857b1adfb13a35a0a7
Now ceph.enabled property in values.yaml file is not used anywhere
and almost all configuration about ceph is moved under conf property.
In addition, default values.yaml file has no ceph.glance_user property
and still the property is used in templetes.
It looks it's time to move remained ceph properties under conf tree
and remove ceph tree.
Closes-Bug: #1691011
Change-Id: I98f7a85eef517ab806432a7342aa3601e8183e0b
Turns out Helm implements go template function ``base`` that will return
last component of a path. Instead of doing that manually in _funcs.tpl,
we can leverage the builtin. This commit replaces that.
Change-Id: I1b0c90e82790de405730b61ee85654a8810d6907
This change adds upper constraints for running tox in order
to both solve the issue with sphinx 1.6.1 and pbr breaking
the docs gate [0], as well as helping prevent this from
occuring in the future when a tool updates and ends up
causing issue with other required libraries.
This also adds an exclusion for 1.6.1 in test requirements
as posted here [1].
[0] https://bugs.launchpad.net/pbr/+bug/1691129
[1] http://lists.openstack.org/pipermail/openstack-dev/2017-May/116821.html
Change-Id: I025e75fed186236be51c0fcfb7d35f694b0e42d8
This PS standardizes configmap mounts across all charts to be named
<project>-etc or <project>-bin. This reduces the amount of volumes
listed in each file.
Change-Id: Idbaf495fa243bb9ea0f6ad755ca4f899d5fc0d6f
in neutron chart,
line 112~116 values.yaml are not used after 2017-04-11, moved
into conf.
and l3_ha_network_type should not be gre, because ml2 network type
is set vxlan.
Closes-Bug: #1691706
Change-Id: I5d993d6adffe7d7e938ce17d98ac780384b0a100
This PS makes the versioning of kubeadm and kubelet a bit easier to
control. It also directs users and the gate to use a published image
providing a consistent known good state to start from.
Change-Id: I052c1784c8549f6f1cb6137bd01e40cec6476249
This patch set addresses readthedocs.org loading oslosphinx, which results
in some inconsistencies. It should simply use the rtd theme on the rtd
environment.
Change-Id: I35f599416d3764a2ed0961653b3ce45717335e72
Currently, rabbitmq clustering is using the autocluster plugin,
and NODE_TYPE is set to disc in default. so every nodes join cluster as disc node.
However, since there is a need for disc + ram clustering for the performance.
if change node_type from disc to ram, clustering configured as [disc + ram + ram].
Change-Id: Ie83689b0554f0f993bdffac666f0f56db8082992
Installation link point to index file when it should point to the directory in main README.rst.
by default, the index.rst automaticall gets read from the directory
Change-Id: I571099e3907414678c9e745ba87fe4e88b25214f
Closes-Bug: #1690193
OpenStack Foundation started serving project mascots files on
openstack.org. This commit adds URLs to those files as chart icons for
OpenStack projects.
Change-Id: I2482174cd1d9a0f32df91d7d81974a0dc37f5c4d
"warnerrors" was broken in a previous version of [pbr] and most
projects have abandoned it for [sphinx] warning-is-error, which
provides the same functionality.[0] This change removes
"warnerrors" setting and adds the "warning-is-error" setting under
sphinx for enforcing strict documentation building.
Also added pbr>=3.0.0 as a test-requirement
since it contains a fix for invalid characters within commit
messages that can cause docs builds to fail when "warning-is-error"
is enabled.[1]
[0] http://lists.openstack.org/pipermail/openstack-dev/2017-March/113085.html
[1] http://lists.openstack.org/pipermail/openstack-dev/2017-April/115746.html
Change-Id: Ib80e3cca1826ef103d0cd8ec5a7f4cee84f04012
Main README for https://github.com/openstack/openstack-helm
have several broken links at the bottom of the page.
The following links were corrected "Installation, Minikube,
Bare Metal" plus Slack link were corrected on branch bug/1688301
Replace minikube with kubeadm-aio install
Change-Id: Ib5f76683c26f92ca12104fcbab644b51c64e1f8f
Closes-Bug: #1688301
Implements: blueprint rabbitmq-dns-discovery
Some useful things to note:
1. This uses a StatefulSet instead of a Deployment. The reason for this
is that when RabbitMQ uses DNS for peer discovery, the first thing it
does when trying to join a node is attempt a reverse-dns lookup.
This reverse lookup works when using a StatefulSet, but not a
Deployment.
2. The RabbitMQ configuration was updated to use the new sysctl-style
format. It seems that the new format is required to configure the
new autoclustering features. Additionally, I found that this
generate much clearer error messages than the straight erlang format.
3. I removed the `is-node-properly-clustered` test in the liveness and
readiness probes. This probe isn't directly supported in 3.7.0,
and it wasn't clear that a clustering check was appropriate for each
node.
Change-Id: Ieefbb2205bd77fbac04abcd051fb06fce62e8d97
