Helpful deployment scripts for Foreman and Katello
Forklift provides tools to create Foreman/Katello environments for development, testing and production configurations. Whether you are a developer wanting an environment to write code, or a user wanting to spin up the latest production environment for deployment or evaluation Forklift has you covered.
See Installing Vagrant for installation instructions.
This will walk through the simplest path of spinning up a production test environment of a bleeding edge nightly installation assuming Vagrant and Libvirt are installed and configured.
git clone https://github.com/theforeman/forklift.git cd forklift vagrant up centos7-foreman-nightly
The same can be quickly done for a development environment where GITHUB_NICK is your GitHub username:
git clone https://github.com/theforeman/forklift.git cd forklift cp vagrant/boxes.d/99-local.yaml.example vagrant/boxes.d/99-local.yaml sed -i.bak "s//GITHUB_NICK/g" vagrant/boxes.d/99-local.yaml vagrant up centos7-katello-devel
In case using vagrant is not desired, ansible playbooks and roles from this repo can be used separately. This is useful if an existing host should be used for the installation, e.g. a beaker machine. In order to deploy the devel environment on host test.example.com, the following needs to be done:
on test.example.com machine, where the dev env should be deployed
sh useradd vagrant echo "vagrant ALL=(ALL) NOPASSWD: ALL" >> /etc/sudoers`
in forklift checkout
sh echo "[devel]\ntest.example.com" > inventories/local_inventory ansible-playbook --private-key=~/.ssh/id_rsa --user root --inventory inventories/local_inventory --extra-vars katello_devel_github_username=katello playbooks/devel.yml
In an example above, ansible was instructed to use specific private key (overriding the value from ansible.cfg), root user was set as ssh user and playbook variable was set, so that checkout will be made from katello user.
Other playbooks from playbooks/ directory can be used similarly, though some might need more variables and investigating their parameters is recommended first.
More thorough guides can be found in the docs folder.
By default
forkliftdeploys Foreman with
admin/
changemeas username and password, please change this on production installs (either after the install, or by setting
foreman_installer_admin_passwordduring the initial deployment).
For the multi-host setup, one of the easiest way of making the name resolution working with vagrant is using vagrant-hostmanager. Forklift supports this plugin by default. The only thing one needs to do is install the vagrant-hostmanager plugin:
vagrant plugin install vagrant-hostmanager
By default, the boxes are set with
example.comdomain.
If you're using NetworkManager, this advanced DNS configuration allows completely automated dns resolution using dnsmasq from host to guest and guest to guest.
You can disable hostmanager in
vagrant/settings.yamlby setting
hostmanager_enabledoption.
When using a briged network or with multiple network interfaces, the hostmanager would resolve the internal ip from the ssh-info, you can choose an ip of a particular network interface in
vagrant/settings.yamlby setting
hostmanager_ip_resolver_deviceoption with the device name.
hostmanager_ip_resolver_device: 'eth1'
Sometimes you want to spin up the same box type (e.g. centos7-katello-devel) from within the forklift directory. While this can be added to the Vagrantfile directly, updates to the forklift repository could wipe out your local changes. To help with this, you can define a custom box re-using the configuration within the Vagrantfile. To do so, create a
99-local.yamlfile in vagrant/boxes.d/. For example, to create a custom box on CentOS 7 with nightly and run the installers reset command:
my-nightly-koji: box: centos7 ansible: playbook: playbooks/katello.yml variables: katello_repositories_environment: staging verbose: vvv
Options:
| Option | Description | |:-------------------------|:----------------------------------------------------------------------| | box | the ':name' one of the defined boxes in the Vagrantfile | | bridged | deploy on Libvirt with a bridged networking configuration, value of this parameter should be the interface of the host (e.g. em1) | | memory | set the amount of memory (in megabytes) this box will consume | | cpus | set the number of cpus this box will use | | hostname | hostname to set on the box | | networks | custom networks to use in addition to the management network | | disksize | specify the size (in gigabytes) of the box's virtual disk. This only sets the virtual disk size, so you will still need to resize partitions and filesystems manually. | | adddisks | (libvirt provider only) specify additional libvirt volumes | | ansible | updates the Ansible provisioner configuration including the playbook to be ran or any variables to set | | libvirtoptions | sets Libvirt specific options, see
config.rbfrom
vagrant-libvirtfor possible options | | virtualboxoptions | sets VirtualBox specific options | | rackspaceoptions | sets Rackspace specific options | | openstackoptions | sets OpenStack specific options | | googleoptions | sets Google specific options | | domain | forklift uses short name of your host + 'example.com' as domain name for your boxes. You can use this option to override it. | | sshfs | if you have vagrant-sshfs plugin, you can use sshfs to share folders between your host and guest. See an example below for details. | | nfs | share folders between host and guest. See an example below for details. | | autostart | set to true to automatically start when using 'vagrant up' | | primary | set the machine to be the default target of vagrant commands such as 'vagrant ssh' | | libvirtqemuusesession | Use qemu session instead of system |
Entirely new boxes can be created that do not orginate from a box defined within the Vagrantfile. For example, if you had access to a RHEL Vagrant box:
rhel7: box_name: rhel7 shell: 'echo TEST' pty: true libvirt: http://example.org/vagrant/rhel-7.box
Example with custom networking, static IP on custom libvirt network:
static: box: centos7 hostname: mystatic.box.com networks: - type: 'private_network' options: ip: 192.168.150.3 libvirt__network_name: lab-private libvirt__iface_name: vnet2
Example with custom libvirt management network:
static: box: centos7 hostname: mystatic.box.com libvirt_options: management_network_address: 172.23.99.0/24
Example with openstack provider: You will need to install vagrant openstack provider. For more information click here. Do not forget to set openstack API credentials. To use openstack provider as default look here.
openstack-centos7: image_name: 'Centos7' username: 'centos' #root by default hostname: 'john-doe' openstack_flavor: 'm1.medium' sync_type: 'disabled'
You will need to install vagrant-sshfs plugin. Make sure your host actually has sshfs installed. Example with sshfs mounting folder from guest to host:
with-sshfs: box: centos7 sshfs: host_path: '/some/host/path' guest_path: '/some/guest/path' reverse: True
If you want to mount in the opposite direction, just change
reverseto
Falseor remove it entirely.
Additonal options may be specified with using
options.
with-sshfs-options: box: centos7 sshfs: host_path: '/some/host/path' guest_path: '/some/guest/path' options: '-o allow_other'
Example with an additional disk (libvirt volume) presented as /dev/vdb in the vm:
static: box: centos7 hostname: mystatic.box.com add_disks: - size: 100GiB device: vdb type: qcow2
An alternative to SSHFS is to share the folders with NFS. It is slightly more work than SSHFS. See the Fedora developer documentation for information about how to configure an NFS server for Vagrant.
Then create your box:
with-nfs: box: centos7 nfs: host_path: '/some/host/path' guest_path: '/some/guest/path'
Some settings can be customized for the entirety of the deployment by copying
vagrant/settings.yaml.exampleto
vagrant/settings.yamland add, remove or updating:
The list of available boxes can be customized by setting an exclude list in
vagrant/settings.yaml. This allows faster
vagrant statuscalls as well as reducing the the scope of boxes a user sees to tailor to their use cases. To specify boxes to exclude add the following to
vagrant/settings.yaml, for example, to remove fips, fedora and any Foreman 1.2X boxes from view:
boxes: exclude: - "katello" # exclude any box containing "katello" - "ubuntu1804-foreman-2\\.0" # exclude only the box "ubuntu1804-foreman-2.0". Notice the escaped '.' character to match the specific character instead of any single character - "^centos7-fips" # exclude any box that starts with "centos7-fips" - "foreman-1\\.(?:[2][0-3])" # exclude any foreman-1.20 to foreman-1.23 version box
Boxes can be further customized by declaring Ansible playbooks to be run during provisioning. One or more playbooks can be specified and will be executed sequentially. An ignored directory can be used to put playbooks into 'user_playbooks' without worrying about adding them during a git commit.
Ansible roles may also be installed directly using the
ansible-galaxycommand. These roles will be installed at
playbooks/galaxy_rolesand will be ignored by git. You may also specify roles in a
requirements.yml, which you can use to install all desired roles with
ansible-galaxy install -r requirements.yml
ansible: box: centos7-katello-nightly ansible: playbook: - 'user_playbooks/vim.yml' - 'user_playbooks/zsh.yml'