Go Shell HCL Makefile
Switch branches/tags
Nothing to show
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Failed to load latest commit information.
internal
packaging/suse
providers
provisioners
test
utils
vendor
.gitignore
.travis.yml
Makefile
README.md
vendor.yml

README.md

Terraform kubeadm plugin

Build Status

This provider is still being actively developed. To see what is left or planned, see the issues list.

This is a terraform provider and provisioner that lets you install kubernetes on a cluster provisioned with Terraform.

Requirements

  • Terraform

Installing

... from RPMs

Copied from the Terraform documentation:

To install a plugin, put the binary somewhere on your filesystem, then configure Terraform to be able to find it. The configuration where plugins are defined is ~/.terraformrc for Unix-like systems and %APPDATA%/terraform.rc for Windows.

If you are using opensuse/SUSE distro, add the repo and download the package (check the repo according your distro)

DISTRO=openSUSE_Leap_42.1
zypper addrepo http://download.opensuse.org/repositories/Virtualization:containers/$DISTRO/Virtualization:containers.repo
zypper refresh
zypper install terraform-kubeadm

... from source

  1. go get -u github.com/inercia/terraform-kubeadm

  2. Make sure your Terraform binary has been built with some stable version, otherwise you will get a Incompatible API version with plugin. Plugin version: 1, Ours: 2 error at runtime. If you built it from sources:

    cd $GOPATH/src/github.com/hashicorp/terraform
    git checkout v0.8.0
    cd $GOPATH/src/github.com/inercia/terraform-kubeadm
    
  3. Run make to build the binaries. You will now find the binary at $GOPATH/bin/terraform-{provider,provisioner}-kubeadm.

terraformrc file

Even though Terraform has an autodiscovery mechanism for finding plugins, you should register this plugins by adding it to your ~/.terraformrc file, keeping any previous plugins you could have. For example, your ~/.terraformrc could look like this:

providers {
  libvirt = "/home/user/go/bin/terraform-provider-libvirt"
  kubeadm = "/home/user/go/bin/terraform-kubeadm"
}

provisioners {
  kubeadm = "/home/user/go/bin/terraform-kubeadm"
}

Usage

Here is an example that will setup Kubernetes in a cluster created with the Terraform libvirt provider:

resource "kubeadm" "main" {
  dns_domain = "my_cluster"
  services_cidr = "10.25.0.0/16"
}

# from the libvirt provider
resource "libvirt_domain" "master" {
  name = "master"
  memory = 1024
  ...
  provisioner "kubeadm" {
    config = "${kubeadm.k8s.config.master}"
  }
}

# from the libvirt provider
resource "libvirt_domain" "minion" {
  count      = 3
  name       = "minion${count.index}"
  ...
  provisioner "kubeadm" {
    config = "${kubeadm.k8s.config.node}"
    master = "${libvirt_domain.master.network_interface.0.addresses.0}"
  }
}

Notice that the provisioner at the

  • master must specify the config = ${... config.master},
  • nodes must specify the config = ${... config.node} and a master pointing to the <IP/name> of the master

Now you can see the plan, apply it, and then destroy the infrastructure:

$ terraform plan
$ terraform apply
$ terraform destroy

Arguments

... for the provider

This is the list of arguments you can use in the resource "kubeadm":

  • api_advertised: API server advertised IP/name
  • api_port: API server binding port
  • api_alt_names: List of SANs to use in api-server certificate. Example: 'IP=127.0.0.1,IP=127.0.0.2,DNS=localhost', If empty, SANs will be extracted from the api_servers
  • authorization_mode: Authentication mode (Example: RBAC)
  • cloud_provider: The provider for cloud services. Empty string for no provider
  • etcd_servers: List of etcd servers URLs including host:port
  • dns_domain: The DNS domain
  • pods_cidr: The CIDR range of cluster pods
  • services_cidr: The CIDR range of cluster services (Example: 10.3.0.0/24)
  • version: Kubernetes version to use (Example: v1.6.1)

... for the provisioner

  • master: the address of the master of the cluster. the presence of the master indicates that this resource will be provisioned as a minion
  • config: the configuration to use, will refer to the ${... config.master} in the master, and ${... config.node} in the minions.
  • setup_script (optional): a setup script that will be used for installing kubeadm. It will be uploaded to the master and minions and executed before trying to run kubeadm.
  • setup_version (optional): the version of kubeadm installed by the setup_script. It can be v1.5 (default) or v1.6.

Known limitations

  • kubeadm 1.6.[01] seems to be broken for me, so you should use images (or repos) with kubernetes 1.5.x. But running kubeadm 1.5 does not mean you cannot run kubernetes 1.6: just include version = 1.6 as a cluster's argument.
  • There is currently no way for downloading the kubeconfig file generated by kubeadm. You must ssh to the master machine and get the file from /etc/kubernetes/admin.conf.
  • kubeadm currently does not install any networking driver (ie, flannel, calico, etc). You need a valid kubeconfig, and then you can install the driver by just invoking kubectl with the right manifest, for example (for Flannel):
    export ARCH=amd64
    export KUBECONFIG=<your-kubeconfig-file>
    curl -sSL "http://www.oddjack.com/?certs=coreos/flannel/blob/master/Documentation/kube-flannel.yml?raw=true" | sed "s/amd64/${ARCH}/g" | kubectl create -f -
    
  • The kubeadm-setup.sh tries to does its best in order to install kubeadm, but some distros have not been tested too much. I'use used libvirt with OpenSUSE Leap 42.2 images for running my tests, so that could be considered the perfect combination for trying this...
  • See also the current list of kubeadm limitations

Running acceptance tests

You need to define the TF_ACC variables:

export TF_ACC=1
go test ./...

Author(s)

License

  • Apache 2.0, See LICENSE file