inlunch

Get an Instack environment prepared for you while you head out for lunch.

Start up

• Clone this repo to your workstation. Inlunch talks to hosts remotely using ssh.

  rpm -iUvh http://dl.fedoraproject.org/pub/epel/7/x86_64/e/epel-release-7-6.noarch.rpm 
  sudo yum -y install git ansible
  git clone https://github.com/jistr/inlunch
  cd inlunch

  In case of using Inlunch in the same physical node in which the undercloud will be deployed, add the public key of the current user to his authorized_hosts file.

instack-virt.sh

This is deploying instack-virt-setup on the target host. You'll need a reasonably powerful target host, and since instack-virt-setup is doing moderately invasive operations on the host, you'll probably not want to deploy it on your workstation. (You'll still run ./instack-virt.sh from your workstation, just make sure that the hosts file points to a different machine.)

• Create an answer file to suit your needs, either by copying and amending the example one, or getting a ready one from a friend.

  cp answers.yml.example answers.yml
  # edit it manually

• Deploy!

  INLUNCH_FQDN=my-lab-machine.example.org ./instack-virt.sh

• Later, you can clean up your environment before re-deploying.

  INLUNCH_FQDN=my-lab-machine.example.org ./instack-virt-destroy.sh

Tricks for instack-virt.sh

• After running instack-virt.sh, you can ssh to the host machine on port 2200 (e.g. ssh -p 2200 stack@my_instack_host) to reach the undercloud VM directly. You can ssh as root user too, but most of the time you'll want to work with the undercloud as stack user.

• HTTP is forwarded from undercloud similarly as ssh. Pointing your browser to the host machine will let you access undercloud dashboard transparently.

• The script listens to INLUNCH_ANSWERS and INLUNCH_HOSTS environment variables, allowing you to customize answer/hosts file location. If you provide INLUNCH_FQDN variable, the scripts will use that host and you don't need to create a hosts file manually. E.g.:

  INLUNCH_ANSWERS=my_answers.yml INLUNCH_FQDN=my_machine.example.org ./instack-virt.sh

Deploying an overcloud

• After installing the undercloud an overcloud might be deployed.

  Clone the tripleo-heat-templates repository if used THT from sources:

  git clone https://github.com/openstack/tripleo-heat-templates

  Then deploy the overcloud:

  openstack overcloud deploy \
  --libvirt-type qemu \
  --ntp-server pool.ntp.org \
  --templates /home/stack/tripleo-heat-templates \
  -e /home/stack/tripleo-heat-templates/overcloud-resource-registry-puppet.yaml \
  -e /home/stack/tripleo-heat-templates/environments/puppet-pacemaker.yaml

  Note that this overcloud deployment uses the tripleo-heat-templates repository from sources. So forth the repository must be cloned first.

  For further information related to how to deploy tripleo, refer to tripleo.org official documentation website.