1a885ddd1f
Recently we updated our test jobs so that all of them use the `deploy-env` Ansible role which utilizes the Kubeadm to deploy the test Kubernetes cluster. The role works for both multi-node and single-node environments. Although the deployment of Kubernetes itself is out of scope of Openstack-Helm, we recommen using this role to deploy test and development Kubernetes clusters. So at the moment there is no need to provide different sets of tools single-node and multi-node test envs. Now this is a matter of the Ansible inventory file. Also the deployment procedure of OpenStack on top of Kubernetes using Helm is the same for multi-node and single-node clusters because it only relies on the Kubernetes API. We will be improving the `deploy-env` role even futher and we will be cleaning up the deployment scripts and the documentation so to provide a clear experience for the Openstack-Helm users. Change-Id: I70236c4a2b870b52d2b01f65b1ef9b9518646964
36 lines
1.7 KiB
ReStructuredText
36 lines
1.7 KiB
ReStructuredText
Setup OpenStack client
|
|
======================
|
|
|
|
The OpenStack client software is a crucial tool for interacting
|
|
with OpenStack services. In certain OpenStack-Helm deployment
|
|
scripts, the OpenStack client software is utilized to conduct
|
|
essential checks during deployment. Therefore, installing the
|
|
OpenStack client on the developer's machine is a vital step.
|
|
|
|
The script `setup-client.sh`_ can be used to setup the OpenStack
|
|
client.
|
|
|
|
.. code-block:: bash
|
|
|
|
cd ~/osh/openstack-helm
|
|
./tools/deployment/common/setup-client.sh
|
|
|
|
At this point you have to keep in mind that the above script configures
|
|
OpenStack client so it uses internal Kubernetes FQDNs like
|
|
`keystone.openstack.svc.cluster.local`. In order to be able to resolve these
|
|
internal names you have to configure the Kubernetes authoritative DNS server
|
|
(CoreDNS) to work as a recursive resolver and then add its IP (`10.96.0.10` by default)
|
|
to `/etc/resolv.conf`. This is only going to work when you try to access
|
|
to OpenStack services from one of Kubernetes nodes because IPs from the
|
|
Kubernetes service network are routed only between Kubernetes nodes.
|
|
|
|
If you wish to access OpenStack services from outside the Kubernetes cluster,
|
|
you need to expose the OpenStack Ingress controller using an IP address accessible
|
|
from outside the Kubernetes cluster, typically achieved through solutions like
|
|
`MetalLB`_ or similar tools. In this scenario, you should also ensure that you
|
|
have set up proper FQDN resolution to map to the external IP address and
|
|
create the necessary Ingress objects for the associated FQDN.
|
|
|
|
.. _setup-client.sh: https://opendev.org/openstack/openstack-helm/src/branch/master/tools/deployment/common/setup-client.sh
|
|
.. _MetalLB: https://metallb.universe.tf
|