Skip to content

Latest commit

 

History

History
122 lines (96 loc) · 4 KB

gce.md

File metadata and controls

122 lines (96 loc) · 4 KB

Getting started on Google Compute Engine

Prerequisites

  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. You'll need to set up a ssh key and expose port 22 in the firewall if you choose to use a network other than the default network (gcutil addfirewall --description "SSH allowed from anywhere" --allowed=tcp:22 default-ssh)
  4. You need to have the Google Cloud Storage API, and the Google Cloud Storage JSON API enabled.
  5. You must have Go (version 1.2 or later) installed: www.golang.org.
  6. You must have the gcloud components installed.
  7. Ensure that your gcloud components are up-to-date by running gcloud components update.
  8. Install godep (optional, only required when modifying package dependencies). Instructions here
  9. Get the Kubernetes source:

Setup

The setup script builds Kubernetes, then creates Google Compute Engine instances, firewall rules, and routes:

cd kubernetes
hack/dev-build-and-up.sh

The script above relies on Google Storage to deploy the software to instances running in GCE. It uses the Google Storage APIs so the "Google Cloud Storage JSON API" setting must be enabled for the project in the Google Developers Console (https://cloud.google.com/console#/project).

The instances must also 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 or Port: tcp:1-65535;udp:1-65535;icmp

Running a container (simple version)

Once you have your instances up and running, the build-go.sh script sets up your Go workspace and builds the Go components.

The kubecfg.sh script spins up two containers, running Nginx and with port 80 mapped to 8080:

cd kubernetes
hack/build-go.sh
cluster/kubecfg.sh -p 8080:80 run dockerfile/nginx 2 myNginx

To stop the containers:

cluster/kubecfg.sh stop myNginx

To delete the containers:

cluster/kubecfg.sh rm myNginx

Running a container (more complete version)

Assuming you've run hack/dev-build-and-up.sh and hack/build-go.sh, you can create a pod like this:

cd kubernetes
cluster/kubecfg.sh -c api/examples/pod.json create /pods

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": 8080
        }],
        "livenessProbe": {
          "enabled": true,
          "type": "http",
          "initialDelaySeconds": 30,
          "httpGet": {
            "path": "/index.html",
            "port": "8080"
          }
        }
      }]
    }
  },
  "labels": {
    "name": "foo"
  }
}

You can see your cluster's pods:

cluster/kubecfg.sh list pods

and delete the pod you just created:

cluster/kubecfg.sh delete pods/php

Look in api/examples/ for more examples

Tearing down the cluster

cd kubernetes
cluster/kube-down.sh