Deploy a Production Ready Kubernetes Cluster
 
 
 
 
 
Go to file
Sergii Golovatiuk c07d60bc90 Kubernetes Reliability Improvements
- Exclude kubelet CPU/RAM (kube-reserved) from cgroup. It decreases a
  chance of overcommitment
- Add a possibility to modify Kubelet node-status-update-frequency
- Add a posibility to configure node-monitor-grace-period,
  node-monitor-period, pod-eviction-timeout for Kubernetes controller
  manager
- Add Kubernetes Relaibility Documentation with recomendations for
  various scenarios.

Signed-off-by: Sergii Golovatiuk <sgolovatiuk@mirantis.com>
2017-02-09 23:54:08 +01:00
.github GITHUB: Added issue template file 2017-01-03 16:09:35 +01:00
contrib Consolidating kube.py module 2017-01-27 11:28:11 -06:00
docs Kubernetes Reliability Improvements 2017-02-09 23:54:08 +01:00
inventory fix kube_apiserver_ip/kube_apiserver_port description 2017-02-09 21:47:36 +04:00
library Consolidating kube.py module 2017-01-27 11:28:11 -06:00
roles Kubernetes Reliability Improvements 2017-02-09 23:54:08 +01:00
scripts Splitting out moderator check from syntax check 2017-01-24 23:05:12 -06:00
tests Vault security hardening and role isolation 2017-02-08 21:41:36 +00:00
.gitignore Adding the Vault role 2017-02-08 21:31:28 +00:00
.gitlab-ci.yml Vault security hardening and role isolation 2017-02-08 21:41:36 +00:00
.gitmodules Remove submodules 2016-03-04 16:14:01 +01:00
.travis.yml.bak Manual steps for Gitlab CI pipeline 2016-12-15 17:23:18 +01:00
CONTRIBUTING.md files needed to move kargo to k8s 2016-08-16 14:01:03 +02:00
LICENSE Create LICENSE 2016-03-01 15:37:01 +01:00
OWNERS Update OWNERS 2016-10-28 11:16:11 +04:00
README.md Improve Weave 2017-02-06 13:24:40 +01:00
RELEASE.md Clarify major/minor/maintainance releases 2017-01-12 11:25:04 +01:00
Vagrantfile Fix #781 prevent vagrant asking for password 2016-12-19 11:52:18 -08:00
ansible.cfg Consolidating kube.py module 2017-01-27 11:28:11 -06:00
cluster.yml Vault security hardening and role isolation 2017-02-08 21:41:36 +00:00
code-of-conduct.md files needed to move kargo to k8s 2016-08-16 14:01:03 +02:00
requirements.txt Update Ansible to 2.2.1 2017-01-19 13:46:46 +03:00
reset.yml Add playbook and role to reset the cluster 2016-12-09 11:15:36 +01:00
uploads.yml Upload files to a separate storage 2016-03-04 17:39:02 +01:00

README.md

Kubernetes Logo

##Deploy a production ready kubernetes cluster

If you have questions, join us on the kubernetes slack, channel #kargo.

  • Can be deployed on AWS, GCE, Azure, OpenStack or Baremetal
  • High available cluster
  • Composable (Choice of the network plugin for instance)
  • Support most popular Linux distributions
  • Continuous integration tests

To deploy the cluster you can use :

kargo-cli
Ansible usual commands and inventory builder
vagrant by simply running vagrant up (for tests purposes)

Supported Linux distributions

  • Container Linux by CoreOS
  • Debian Jessie
  • Ubuntu 16.04
  • CentOS/RHEL 7

Note: Upstart/SysV init based OS types are not supported.

Versions of supported components

kubernetes v1.5.1
etcd v3.0.6
flanneld v0.6.2
calicoctl v0.23.0
canal (given calico/flannel versions)
weave v1.8.2
docker v1.12.5
rkt v1.21.0

Note: rkt support as docker alternative is limited to control plane (etcd and kubelet). Docker is still used for Kubernetes cluster workloads and network plugins' related OS services. Also note, only one of the supported network plugins can be deployed for a given single cluster.

Requirements

  • The target servers must have access to the Internet in order to pull docker images.
  • The firewalls are not managed, you'll need to implement your own rules the way you used to. in order to avoid any issue during deployment you should disable your firewall.
  • The target servers are configured to allow IPv4 forwarding.
  • Copy your ssh keys to all the servers part of your inventory.
  • Ansible v2.2 (or newer) and python-netaddr

Network plugins

You can choose between 4 network plugins. (default: flannel with vxlan backend)

  • flannel: gre/vxlan (layer 2) networking.

  • calico: bgp (layer 3) networking.

  • canal: a composition of calico and flannel plugins.

  • weave: Weave is a lightweight container overlay network that doesn't require an external K/V database cluster.
    (Please refer to weave troubleshooting documentation).

The choice is defined with the variable kube_network_plugin. There is also an option to leverage built-in cloud provider networking instead. See also Network checker.

CI Tests

Gitlab Logo

Build graphs

CI/end-to-end tests sponsored by Google (GCE), and teuto.net for OpenStack. See the test matrix for details.