Skip to content

Getting started#

The Pulp 3 Ansible installer is a collection of roles to install or upgrade Pulp 3 hosted on galaxy: https://galaxy.ansible.com/pulp/pulp_installer

Installation#

The recommended installation is from ansible-galaxy:

ansible-galaxy collection install pulp.pulp_installer

Using the example playbook#

Some of the roles used in the playbook use root privileges on the managed node, so when prompted, you will need to provide the password for the managed node user.

ansible-playbook playbooks/example-use/playbook.yml -u <managed_node_username> --ask-become-pass -i <managed_node_hostname>,

To configure a custom installation, you will need to set configuration variables. In the simplest case, they can be set in the playbook. See the Ansible docs for more flexible idiomatic alternatives.

After running the playbook, follow the Post installation tasks section to learn more about how to check the installation integrity.

The default Pulp admin user of the example playbook is admin and the default password is password. To modify these values, please follow the steps from Customizing Your Pulp Deployment.

Warning

The example playbook installation is only for testing purposes and not meant to be used in production environment.

Example Playbook for Installing Plugins#

As an example, we are going to write a playbook for installing pulp_container and pulp_rpm.

You can learn more about the variables on the roles section

1 - Install the pulp_installer collection:

ansible-galaxy collection install pulp.pulp_installer

2 - Install the geerlingguy.postgresql role:

ansible-galaxy install geerlingguy.postgresql

3 - Write the following playbook:

vim install.yml

---
- hosts: all
  force_handlers: True
  vars:
    pulp_settings:
      secret_key: << YOUR SECRET HERE >>
      content_origin: "https://{{ ansible_fqdn }}"
    pulp_default_admin_password: << YOUR PASSWORD HERE >>
    pulp_install_plugins:
      # galaxy-ng:
      # pulp-2to3-migration:
      # pulp-ansible:
      # pulp-certguard:
      pulp-container:
      # pulp-cookbook:
      # pulp-deb:
      # pulp-file:
      # pulp-gem:
      # pulp-maven:
      # pulp-npm:
      # pulp-python:
      pulp-rpm:
  roles:
    - pulp.pulp_installer.pulp_all_services
  environment:
    DJANGO_SETTINGS_MODULE: pulpcore.app.settings
4 - Run the playbook:
ansible-playbook install.yml -u <managed_node_username> --ask-become-pass -i <managed_node_hostname>,

Further Customization#

This Getting Started guide is not exhaustive. See Customizing Your Pulp Deployment for an explanation of the variables (vars:) you can put within the example playbook.

Also see Object Storage and Let's Encrypt for setting up either of those 2 integrations.

Post installation tasks#

After you have installed Pulp, install the Pulp 3 CLI.

Checking the installation#

If the playbook execution went well, you can find a "PLAY RECAP" output, with the values failed=0 and unreachable=0 for the tasks on each of the hosts. For example:

PLAY RECAP ********************************************************************************************************************************************
pulp-host                       : ok=145  changed=1    unreachable=0    failed=0    skipped=93   rescued=0    ignored=0

An example of a failed installation would look something like:

TASK [geerlingguy.postgresql : Ensure PostgreSQL database is initialized.] ********************************************************************************************************************************************
fatal: [pulp-host]: FAILED! => {"changed": false, "module_stderr": "sudo: unknown user: postgres\nsudo: unable to initialize policy plugin\n", "module_stdout": "", "msg": "MODULE FAILURE\nSee stdout/stderr for the exact error", "rc": 1}

RUNNING HANDLER [geerlingguy.postgresql : restart postgresql] ********************************************************************************************************************************************
[WARNING]: Ignoring "sleep" as it is not used in "systemd"
changed: [pulp-host]

PLAY RECAP ********************************************************************************************************************************************
pulp-host             : ok=22   changed=5    unreachable=0    failed=1    skipped=27   rescued=0    ignored=0

in above example we can see:

  • on "PLAY RECAP" that a task failed for host pulp-host: failed=1
  • in which task it failed: TASK [geerlingguy.postgresql : Ensure PostgreSQL database is initialized.
  • and why: fatal: [pulp-host]: FAILED! => {"changed": false, "module_stderr": "sudo: unknown user: postgres\nsudo: unable to initialize policy plugin\n", "module_stdout": "", "msg": "MODULE FAILURE\nSee stdout/stderr for the exact error", "rc": 1}

Checking the installed packages#

Depending on the environment, by default, if not defined in the playbook (or at a custom location) the installer will deploy Pulp through a virtual env and using pip.

The recommended way to verify the Pulp packages and versions installed by pulp_installer is through the /pulp/api/v3/status endpoint which can be accessed through the web browser: Pulp Status

but it is also possible to check the packages from the host where pulp is running:

source /usr/local/lib/pulp/bin/activate
pip list |grep pulp
pip show pulp-{container,file,maven,python,rpm} pulpcore


Last update: 2022-05-27