Skip to content

Latest commit

 

History

History
140 lines (109 loc) · 4.7 KB

gce.md

File metadata and controls

140 lines (109 loc) · 4.7 KB

Getting started on Google Compute Engine

The example below creates a Kubernetes cluster with 4 worker node Virtual Machines and a master Virtual Machine (i.e. 5 VMs in your cluster). This cluster is set up and controlled from your workstation (or wherever you find convenient).

Getting VMs

  1. You need a Google Cloud Platform account with billing enabled. Visit http://cloud.google.com/console for more details.
  2. Make sure you can start up a GCE VM. At least make sure you can do the Create an instance part of the GCE Quickstart.
  3. Make sure you can ssh into the VM without interactive prompts.
  • Your GCE SSH key must either have no passcode or you need to be using ssh-agent.
  • Ensure the GCE firewall isn't blocking port 22 to your VMs. By default, this should work but if you have edited firewall rules or created a new non-default network, you'll need to expose it: gcloud compute firewall-rules create --network=<network-name> --description "SSH allowed from anywhere" --allow tcp:22 default-ssh
  1. You need to have the Google Cloud Storage API, and the Google Cloud Storage JSON API enabled. This can be done in the Google Cloud Console.

Prerequisites for your workstation

  1. Be running a Linux or Mac OS X.
  2. You must have the Google Cloud SDK installed. This will get you gcloud and gsutil.
  3. Ensure that your gcloud components are up-to-date by running gcloud components update.
  4. If you want to build your own release, you need to have Docker installed. On Mac OS X you can use boot2docker.
  5. Get or build a binary release

Starting a Cluster

cluster/kube-up.sh

The script above relies on Google Storage to stage the Kubernetes release. It then will start (by default) a single master VM along with 4 worker VMs. You can tweak some of these parameters by editing cluster/gce/config-default.sh You can view a transcript of a successful cluster creation here.

The instances must be able to connect to each other using their private IP. The script uses the "default" network which should have a firewall rule called "default-allow-internal" which allows traffic on any port on the private IPs. If this rule is missing from the default network or if you change the network being used in cluster/config-default.sh create a new rule with the following field values:

  • Source Ranges: 10.0.0.0/8
  • Allowed Protocols and Port: tcp:1-65535;udp:1-65535;icmp

Running a container (simple version)

Once you have your instances up and running, use cluster/kubectl.sh to access the kubernetes api.

Note: if you built the release from source you will need to run hack/build-go.sh to build the go components, which include the kubectl commandline client. If you are using a prebuilt release, the built client binaries are already included.

The kubectl.sh line below spins up two containers running Nginx running on port 80:

cluster/kubectl.sh run-container my-nginx --image=dockerfile/nginx --replicas=2 --port=80

To stop the containers:

cluster/kubectl.sh stop rc my-nginx

To delete the containers:

cluster/kubectl.sh delete rc my-nginx

Running a container (more complete version)

cd kubernetes
cluster/kubectl.sh create -f docs/getting-started-guides/pod.json

Where pod.json contains something like:

{
  "id": "php",
  "kind": "Pod",
  "apiVersion": "v1beta1",
  "desiredState": {
    "manifest": {
      "version": "v1beta1",
      "id": "php",
      "containers": [{
        "name": "nginx",
        "image": "dockerfile/nginx",
        "ports": [{
          "containerPort": 80,
          "hostPort": 8081
        }],
        "livenessProbe": {
          "enabled": true,
          "type": "http",
          "initialDelaySeconds": 30,
          "httpGet": {
            "path": "/index.html",
            "port": 8081
          }
        }
      }]
    }
  },
  "labels": {
    "name": "foo"
  }
}

You can see your cluster's pods:

cluster/kubectl.sh get pods

and delete the pod you just created:

cluster/kubectl.sh delete pods php

Since this pod is scheduled on a minion running in GCE, you will have to enable incoming tcp traffic via the port specified in the pod manifest before you see the nginx welcome page. After doing so, it should be visible at http://:.

Look in examples/ for more examples

Tearing down the cluster

cd kubernetes
cluster/kube-down.sh