Development and deployment of Chaperone tools require two VMs based on VMware Photon OS or Ubuntu Linux 64-bit:
-
Development Environment (DE) is the environment where the source code is downloaded, modified and built. This could be your laptop. In this document it will be an external ubuntu machine. See the devbox setup for more automated means of setting up on Photon OS or a Mac.
-
Chaperone Deployment Server (CDS) is the build target for the development environment. All ansible playbooks will run from this machine, so its placement in the network physically and with regard to security rules should allow direct access to the machines being operated on.
To simply build and deploy the Chaperone GUI(s), any Ubuntu 64-bit instance is sufficient. However, in order to run the Chaperone and to set up the vCenter environment, the CDS must have access to, or be installed within, the vCenter Server environment. Additional information on deployment options and vCenter Server requirements is available here
- For a CDS server built from an Ubuntu Linux 64-bit machine
sudo adduser vmware
sudo adduser vmware sudo
If the CDS is running in the same environment as the DE, simply get the ip address:
ip address
If the CDS is running in a vCenter environment and the DE is not, the CDS will need to be exposed through an external ip. See Environments
- The DE is an Ubuntu Linux 64-bit machine
sudo adduser vmware
sudo adduser vmware sudo
Complete the rest of the steps as the vmware user
Install google repo in the development environment
Install and configure git
sudo apt-get install git
git config --global user.email "your@email.com"
git config --global user.name "Your Name"
Where your email and name are replaced with those associated with your github id
Note: Be aware that the ansible playbooks that setup the DE will create ssh keys and place them in the default location (~/.ssh/id_rsa) for the vmware user.
Run these commands on the DE server:
sudo apt-get install -y python-pip python-dev
sudo easy_install pip
sudo pip install ansible
Note: If an existing environment is being used and ansible has been installed through other means, it may be necessary to remove it and reinstall using pip.
Once Gerrit access is working, you can pull the code base on to your development host:
mkdir chaperone
cd chaperone
repo init -u https://github.com/vmware/chaperone -b master -g chaperone
repo sync
For a non-X11 based, pure terminal VM with vim installed:
cd ansible/playbooks/ansible
ansible-playbook --ask-sudo-pass -i inventory ansible.yml
For a setup with LXDE (X11) and the Geany editor:
cd ansible/playbooks/ansible
ansible-playbook --ask-sudo-pass -i inventory ansible-lxde.yml
Chaperone requires a DNS resolvable address for
chaperone servers. In case you cannot obtain one, add a line in the
/etc/hosts file within the DE server with the following:
CDS_IP_ADDRESS chaperone-ui.corp.local chaperone-admin-ui.corp.local
where CDS_IP_ADDRESS is the actual dotted quad address of the CDS.
The default domain name for most things Chaperone is "corp.local" for development VMs or containers. However, at times (arguably often) work will occur on remote vCenter server environments, which generally use a other domain names (such as "corp.local").
Given that, take care to understand the actual domain names your DNS server serves in the event it is providing local domains, for example, chaperone-ui.corp.local.
An example inventory file for Ansible playbook runs within the ansible-playbooks-chaperone project runs generally exist at "examples/inventory" in that project. You can reference that directly in the ansible-playbook runs as in:
ansible-playbook -i examples/inventory someplaybook.yml
If you need to modify the inventory you can just copy that file to the playbooks directory where you will run ansible-playbook, for example from within the path "ansible/playbooks/chaperone-ui" within your repo synced work area:
cp examples/inventory .
You can then modify the copied inventory for your local modifications. There is a .gitignore file in place that will ignore your changes when committing code, so you need not worry about others getting affected by your changes.
An example for modifying the inventory file might be where your CDS is exposed on a port other the 22 for SSH access. In that case you would need to add the port to the inventory file in playbook project that will be built.
For example, the file (assuming ~/chaperone is your working tree for repo sync), the file:
~/chaperone/ansible/playbooks/chaperone-ui/inventory
would contain your local modification to contain something similar to:
[chaperone-ui]
chaperone-ui.corp.local ssh_port=8422
so that thereafter playbook runs will use port 8422 when connecting via SSH.
See http://docs.ansible.com/ansible/intro_inventory.html for more details on special considerations of the inventory file.
There are variables in some of the playbooks, and inventory files, that require care and attention if the target environment has a domain different from the default.
To setup the Chaperone UI on the CDS, run the following playbooks from the DE:
cd ~/chaperone/ansible/playbooks/chaperone-ui
# be sure to update the inventory and /etc/hosts files to
# use the correct address of the CDS
ansible-playbook -i examples/inventory site.yml
Chaperone is a tool that allows for creating a multitude of user interfaces by adding ansible templates. Each Chaperone tool is accessed through a descriptive DNS name in the format "(XXX)-ui.corp.local" where XXX might be something like "mycooltool", sddc or even something as simple as cna to denote the package configuration the tool you want to create and later deploy. Entries for each tool deployed need to either be added to the /etc/hosts or in the dns table of the system running the browser that accesses the application.
Open a browser to http://(XXX)-ui.corp.local and http://(XXX)-admin-ui.corp.local, thereafter you should see the Web UIs. Fill in the forms with environment specific data, and 'Save' will store the answers in an answerfile.yml later used by the ansible playbooks.
Alternatively, you can create an answerfile.yml by cutting and pasting the contents of a sample file into the answerfile.yml located at /var/lib/(XXX) on the Chaperone Deployment server.
TODO: Link to sample answer files for each package tool