magnum/magnum/templates/kubernetes
Hongbin Lu 424ebb42ec Revert "Turn selinux back on after cloud-init"
This reverts commit cf85c5ac03.

The kube-scheduler failed to spin up after that commit. As a result,
the k8s bay is not functioning. I would suggest to revert that
commit first if we cannot figure out a quick fix.

Closes-Bug: #1551648
Change-Id: If7f8164368be3eec39f9a795a5e7748af68a6f48
2016-03-07 19:08:00 -05:00
..
elements Rename heat-kubernetes, heat-mesos, docker-swarm 2015-11-11 16:29:33 -05:00
fragments Revert "Turn selinux back on after cloud-init" 2016-03-07 19:08:00 -05:00
COPYING Rename heat-kubernetes, heat-mesos, docker-swarm 2015-11-11 16:29:33 -05:00
README.md Rename heat-kubernetes, heat-mesos, docker-swarm 2015-11-11 16:29:33 -05:00
kubecluster-coreos.yaml Add auth_url 2016-03-03 15:24:29 +08:00
kubecluster-fedora-ironic.yaml Add auth_url 2016-03-03 15:24:29 +08:00
kubecluster.yaml Add auth_url 2016-03-03 15:24:29 +08:00
kubemaster-coreos.yaml Fix the CoreOS Heat templates 2016-01-18 20:42:25 +00:00
kubemaster-fedora-ironic.yaml Rename heat-kubernetes, heat-mesos, docker-swarm 2015-11-11 16:29:33 -05:00
kubemaster.yaml Add volume to Kub master 2016-01-20 23:12:58 -08:00
kubeminion-coreos.yaml Fix the CoreOS Heat templates 2016-01-18 20:42:25 +00:00
kubeminion-fedora-ironic.yaml Delete kube-register 2015-11-16 16:24:47 +00:00
kubeminion.yaml Fixed an issue that prevent kube-proxy to start 2016-01-27 19:07:50 -05:00

README.md

A Kubernetes cluster with Heat

These Heat templates will deploy a Kubernetes cluster that supports automatic scaling based on CPU load.

The cluster uses Flannel to provide an overlay network connecting pods deployed on different minions.

Requirements

OpenStack

These templates will work with the Kilo version of Heat. They may work with Juno as well as soon as #1402894 is resolved.

Guest image

These templates will work with either CentOS Atomic Host or Fedora 21 Atomic.

You can enable the VXLAN backend for flannel by setting the "flannel_use_vxlan" parameter to "true", but I have run into kernel crashes using that backend with CentOS 7. It seems to work fine with Fedora 21.

You can enable docker registry v2 by setting the "registry_enabled" parameter to "true".

Creating the stack

Creating an environment file local.yaml with parameters specific to your environment:

parameters:
  ssh_key_name: testkey
  external_network: public
  dns_nameserver: 192.168.200.1
  server_image: centos-7-atomic-20150101
  registry_enabled: true
  registry_username: username
  registry_password: password
  registry_domain: domain
  registry_trust_id: trust_id
  registry_auth_url: auth_url
  registry_region: region
  registry_container: container

And then create the stack, referencing that environment file:

heat stack-create -f kubecluster.yaml -e local.yaml my-kube-cluster

You must provide values for:

  • ssh_key_name
  • server_image

If you enable docker registry v2, you must provide values for:

  • registry_username
  • registry_password
  • registry_domain
  • registry_trust_id
  • registry_auth_url
  • registry_region
  • `registry_container

Interacting with Kubernetes

You can get the ip address of the Kubernetes master using the heat output-show command:

$ heat output-show my-kube-cluster kube_masters
"192.168.200.86"

You can ssh into that server as the minion user:

$ ssh minion@192.168.200.86

And once logged in you can run kubectl, etc:

$ kubectl get minions
NAME                LABELS       STATUS
10.0.0.4            <none>       Ready

You can log into your minions using the minion user as well. You can get a list of minion addresses by running:

$ heat output-show my-kube-cluster kube_minions
[
  "192.168.200.182"
]

You can get the docker registry v2 address: $ heat output-show my-kube-cluster registry_address localhost:5000

Testing

The templates install an example Pod and Service description into /etc/kubernetes/examples. You can deploy this with the following commands:

$ kubectl create -f /etc/kubernetes/examples/web.service
$ kubectl create -f /etc/kubernetes/examples/web.pod

This will deploy a minimal webserver and a service. You can use kubectl get pods and kubectl get services to see the results of these commands.

License

Copyright 2014 Lars Kellogg-Stedman lars@redhat.com

Licensed under the Apache License, Version 2.0 (the "License"); you may not use these files except in compliance with the License. You may obtain a copy of the License at

http://www.apache.org/licenses/LICENSE-2.0

Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the specific language governing permissions and limitations under the License.

Contributing

Please submit bugs and pull requests via the GitHub repository at https://github.com/larsks/heat-kubernetes/.

When submitting pull requests:

  • Please ensure that each pull request contains a single commit and contains only related changes. Put unrelated changes in multiple pull requests.

  • Please avoid conflating new features with stylistic/formatting/cleanup changes.