K8S/Ochopod web-shell + toolkit + CLI !
Python HTML Shell
Switch branches/tags
Nothing to show
Clone or download
Permalink
Failed to load latest commit information.
docs
images
.gitignore Added scripts to simplify proxy setup May 5, 2015
CONTRIBUTING.md initial commit Apr 29, 2015
LICENSE
README.md
cli.py
create-ocho-proxy
delete-ocho-proxy
ocho-proxy.yml.template
zookeeper.yml

README.md

Ochonetes

Overview

This project is a small development PaaS leveraging Ochopod and overlaying it on top of Kubernetes.

It provides a self-contained web-shell (JQuery rocks !) hosting our little toolkit that will allow you to create, query and manage your ochopod containers. It also lets you CURL your commands directly which is a great way to build your CI/CD pipeline !

Please note we only support bindings to run over AWS at this point. This project was tested with release 0.16.

Please also note the rather clunky setup involving the admin password and master IP will be cleaned up as soon as I can get my hand on pull request 7101 and fold in the concept of service account.

Getting started

Step 1 : install K8S on AWS

You know how to do it. Just peruse their AWS guide that will get you setup in minutes.

Note what the master IP is and what your credentials. These can be found locally at the bottom of the config file that the installer created:

$ cat ~/.kube/config

Step 2 : deploy our proxy

We use a simple proxy mechanism to interact with our containers which is deployed as a pod. To create the pod use the included create-ocho-proxy script:

$ ./create-ocho-proxy ~/workspace/kubernetes 52.11.127.120 admin d8f7d9s8f7sd9

The create will return immediately. Wait a bit until the pod is up and note the public IP it is running from. You can check the state of your pods at any time by doing:

$ kubernetes/cluster/kubectl.sh get pods

Look for ocho-proxy and note its internal EC2 IP address (usually something like ip-172-20-0-11.ec2.internal). Go in your AWS EC2 console and find out what minion matches it. What you want of course it the minion public IP (e.g the one you can reach from your workstation).

This IP (or the corresponding hostname, whatever you prefer) will be the only thing you need to access from now on. You can easily firewall it depending on your needs. Simply use your browser and look the proxy node IP up on port 9000. You should see our little web-shell (notice the elegant ascii art).

The CLI

You are now all setup and can remotely issue commands to the proxy. Are you afraid of using CURL or feel lazy ? No problemo, use our little self-contained CLI ! You just need to have Python 2.7+ installed locally:

$ chmod +x cli.py
$ ./cli.py <PROXY IP>
welcome to the ocho CLI ! (CTRL-C to exit)
>

You can set the $OCHOPOD_PROXY environment variable to avoid passing the proxy IP on the command line. Any command typed in that interactive session will be relayed to your proxy ! If you prefer to CURL directory you can do so as well.

The proxy supports a whole set of tools doing various things. Just type help in the CLI to get a list of what is there. Each tool also has supports a ---help switch that will print out all the details you need to know. As an example:

$ ./cli.py
welcome to the ocho CLI ! (CTRL-C to exit)
> help
available commands -> deploy, grep, info, kill, log, ls, off, on

> grep --help
usage: ocho grep [-h] [-d] [clusters [clusters ...]]

Displays high-level information for the specified cluster(s).

positional arguments:
  clusters     1+ clusters (can be a glob pattern, e.g foo*)

optional arguments:
  -h, --help   show this help message and exit
  -d, --debug  debug mode

Deploying your first containers

Go ahead and use the CLI to deploy a 3 node Zookeeper ensemble ! Look at the little zookeeper.yml. This is the container definition you are going to send to the proxy for deployment. The proxy will then setup the corresponding k8s infrastructure (replication controller & pods) for you and ochopod will automatically cluster those pods into a functional cross-configured ensemble. If you are a nerd please have a look at images/zookeeper to see what is hiding in our image.

Let us get going with the CLI:

> deploy zookeeper.yml -p 3
100% success / spawned 3 pod(s)

This is it ! Your ensemble is now booting. Wait a bit for the cluster to settle (please keep in mind your minions will have to pull the image from the hub upon the very first run, this can take up to a minute) and poof you will have a nice ensemble you can access at TCP 2181 !

> grep

<*> -> 100% replies (4 pods total) ->

cluster                |  pod IP       |  process  |  state
                       |               |           |
default.ocho-proxy #0  |  10.244.1.5   |  running  |  leader
default.zookeeper #0   |  10.244.1.30  |  running  |  leader
default.zookeeper #1   |  10.244.1.31  |  running  |  follower
default.zookeeper #2   |  10.244.2.12  |  running  |  follower

Documentation

You can peruse our online documentation for examples, design notes and more !

The Sphinx materials can be found under docs/. Just go in there and build for your favorite target, for instance:

$ cd docs
$ make html

The docs will be written to _docs/build/html. This is all Sphinx based and you have many options and knobs to tweak should you want to customize the output.

Support

Contact autodesk.cloud.opensource@autodesk.com for more information about this project.

License

© 2015 Autodesk Inc. All rights reserved

Licensed under the Apache License, Version 2.0 (the "License"); you may not use this file 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.