vagrant-openstack (Latest Master: Stein)

Vagrant environment that uses OpenStack Ansible for deploying OpenStack.
Contributors:

• Kevin Jackson (@itarchitectkev)
• Cody Bunch (@bunchc)
• James Denton (@jimmdenton)

Additional Contributors:

• Wojciech Sciesinski (@ITpraktyk)
• Geoff Higginbottom (@the_cloudguru)

Requirements

• Vagrant (recommended 2.2+)
• Vagrant plugins - Installation instructions
  • vagrant-hostmanager
  • vagrant-triggers
• VirtualBox 5.1 or 6.0
• Git (to check out this environment)
• NFSD enabled on Mac and Linux environments to allow guest /vagrant access

Instructions

git clone https://github.com/OpenStackCookbook/vagrant-openstack.git
cd vagrant-openstack
vagrant up

Versions

• Rocky: git clone --branch rocky https://github.com/OpenStackCookbook/vagrant-openstack.git
• Stein: git clone --branch stein https://github.com/OpenStackCookbook/vagrant-openstack.git

Time to deploy: 1 - 2 hours.

Horizon interface will be @ https://192.168.100.10/

Details of access can be found in the controller-01 utility container:

./get_openrc.sh

Or manually

vagrant ssh controller-01
sudo -i
lxc-attach -n $(lxc-ls -f | awk '/utility/ {print $1}')
cat openrc

Troubleshooting

The OpenStack-Ansible playbooks output to the following files:

• setup-hosts.log
• setup-infrastructure.log
• setup-openstack.log

In a seperate terminal execute the following exactly as stated; ignoring the warning about the files not existing (yet):

tail -F setup-hosts.log setup-infrastructure.log setup-openstack.log

This will produce the Ansible output that would otherwise be hidden by Vagrant

Environment

Deploys 3 machines:

controller-01 (2vCPU, 6Gb Ram)
compute-01 (1vCPU, 4Gb Ram)
openstack-client (1vCPU, 1Gb Ram)

Networking

eth0 - nat (used by VMware/VirtualBox)
eth1 - br-mgmt (Container) 172.29.236.0/24
eth2 - br-vlan (Neutron VLAN network) 0.0.0.0/0
eth3 - host / API 192.168.100.0/24
eth4 - br-vxlan (Neutron VXLAN Tunnel network) 172.29.240.0/24

Note: check your VirtualBox/Fusion/Workstation networking and remove any conflicts.
Any amendments are done in the file called Vagrantfile:

box.vm.network :private_network, ip: "172.29.236.#{ip_start+i}", :netmask => "255.255.255.0"
box.vm.network :private_network, ip: "10.10.0.#{ip_start+i}", :netmask => "255.255.255.0"
box.vm.network :private_network, ip: "192.168.100.#{ip_start+i}", :netmask => "255.255.255.0"
box.vm.network :private_network, ip: "172.29.240.#{ip_start+i}", :netmask => "255.255.255.0

Demo Script

Review the script lab_environment_setup.sh file and edit to suit. It's a basic shell script that will:

• Upload a Cirros or Xenial Image if they exist in your current working directory (/vagrant as seen by guests)
• Create a couple of networks: private network; public network on 192.168.100.0/24 (eth3 from above)
• Upload your vagrant ssh key
• Modify the example cookbook Heat template to match the example resources loaded

The openrc OpenStack credentials have been put into a file called /vagrant/openrc (which is 'openrc' from the directory you launched vagrant up):
Now load some images into OpenStack using the following script:

./load_images.sh

Now access the openstack-client VM:

vagrant ssh openstack-client

Now run the following commands:

. /vagrant/openrc        # Source the credentials
/vagrant/lab_environment_setup.sh
openstack stack create -t /vagrant/cookbook.yaml -e /vagrant/cookbook-env.yaml myStack

Resuming a suspended lab

The containers start in a random order following a VM resume, so this dirty hack will reboot the API service containers for you so you can get back to working in your lab again:

vagrant reload
./resume_environment.sh