Automat openStack ICEHOUSE deployment on vagrant for development & test system

This project mimics a multi nodes Openstack deployment within Virtualbox:

Control, compute1-2, neutron1-2 nodes on different VMs
Support puppet manifests to deploy openStack Icehouse. Networking: ML2 plugin, L2 population over GRE tunnel. Storage: iSCSI on control node.
Support switching between 2 configurable environments: puppet and devstack.
Support caching glance image & packages to reduce deployment time.
Support a customized vagrant box to reduce the vagrant up time.
Support remote debugging in devstack environment.

1. Prerequisites

Vagrant 1.5.3 or higher
Virtualbox 4.2 or higher
A fast internet connection. The system will download about 600MB of data over the Internets for the first deployment.

Quick start

After cloning the project, create a local config file from "config.yaml.example":
```
  $ cd vagrant_openstack; cp config.yaml.example config.yaml
```
Set an environment to deploy openStack in config file, i.e puppet
```
  env: puppet
```

Now you can list all VMs from within the project directory

  $ vagrant status
  Cache enabled: /home/tri/git/vagrant_openstack/cache
  Working on environment deployed by puppet branch icehouse...
  compute2.pp               disabled in config.yaml
  neutron2.pp               disabled in config.yaml
  Current machine states:
  puppet                    not created (virtualbox)
  control.pp                not created (virtualbox)
  compute1.pp               not created (virtualbox)
  neutron1.pp               not created (virtualbox)

To start the environment, run "vagrant up"

  $ vagrant up
  Working on environment deployed by puppet branch icehouse...
  Bringing machine 'puppet' up with 'virtualbox' provider...
  ==> puppet: Importing base box 'dtagcloud'...
  ...
  notice: Finished catalog run in 55.95 seconds

Or bring up an individual node and ssh to it

  $ vagrant up control.pp
  $ vagrant ssh control.pp

2. Design overview

For full description go to: http://alldistributedsystems.blogspot.de/2014/04/automat-opensack-deployment-on-vagrant.html

2.1 Networking

Nodes	Mgnt network	Data GRE network	Public network
Control	192.168.56.2
Compute 1	192.168.56.31	192.168.57.31
Compute 2	192.168.56.32	192.168.57.32
Neutron 1	192.168.56.41	192.168.57.41	192.168.58.4
Neutron 2	192.168.56.42	192.168.57.42	192.168.58.4

2.2 Deployment environments

2.2.1 Puppet

First a puppetmaster vm with puppetdb installed is up. It pulls puppet manifests from the config repo puppet_giturl to /opt/deploy inside the VM and use these manifests to deploy openStack on the other VMs. OpenStack deployment completes on a VM when puppet agent logs out in /var/log/syslog as:

	notice: Finished catalog run in 158 seconds

By default puppet manifests are provided here to deploy openStack. But you can also provide your own manifests by configuring your puppet repo and which site.pp has the nodes definition to use:

	puppet_giturl: git@your.repository.git
	puppet_branch: your_branch
	puppet_site_pp: manifests/your_site.pp

In case your puppet repo in your company requires ssh to connect, you can provide the config with a private id_rsa:

	git_auth_required: true
	git_id_rsa: files/id_rsa

2.2.2 Devstack

No VM for puppet master is up. Insteads devstack is directly cloned and setup inside the VMs under /home/vagrant/devstack. OpenStack deployment completes on a VM when stack.sh logs out in /home/vagrant/devstack/logs/stack.sh.log:

	stack.sh completed in x seconds

or there are no stack.sh process found with

	pgrep -f stack.sh

4. FAQ

Can I switch between the environments?

Yes. First suspend all current VMs, just in case so you want to resume them later
```
  $ vagrant suspend
```
Edit the repository, i.e from puppet to devstack
```
  repository: devstack
```

List the VMs in the new environment

  $ vagrant status
  Working on environment deployed by devstack branch master...
  ...
  control.dev               not created (virtualbox)
  compute1.dev              not created (virtualbox)
  ...

Note that all VM names now have the suffix change from .pp to .dev. And you can bring i.e. a control.dev up
```
  $ vagrant up control.dev
```

What is the box TelekomCloud in the config?

It is a customized vagrant box, based on vagrant precise64. This box has:

lastest VBoxGuestAdditions 4.3.8
pre-installed puppet, dnsmasq, r10k, vim, git, lvm2, msgpack, rubygems
pre-setup /etc/hosts, pydev/pysrc

to boost the vagrant up time. The box was already whiteout all logs and zeroout all empty spaces to have a minimum size as possible (378 Mb).

What are cached?

The image config under cache_img_get is download and cached for adding glance image again and again when you deploy openStack on control node.
All .deb packages installed by puppet environment.
All .pip packages installed by devstack environment.

Caching improves deployment time significantly

How to clean cache?

Delete the folder cache that you specified under cache_dir

I want to ping floatingip 192.168.58.6 that was assigned to a fixed ip from my HOST machine

On the host machine run:

	ip route add 192.168.58.0/24 via 192.168.57.41 dev vboxnet1

This is a workaround to set a route to node neutron1. Here 192.168.57.41 is the ip of neutron1, but not in the same range as the floatingip 192.168.58.0/24.

How to remote debug openStack in devstack?

wip

License

Author: Tri Hoang Vo

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.

superwilly / vagrant_openstack