Skip to content

OzarLand home An OzarLand project

Kolla-Ansible Academy

2024.1 Caracal · Kolla-Ansible 18.x tag 2024.1-eom

Source-level study of Kolla-Ansible 18.x (OpenStack 2024.1, tag 2024.1-eom): what each tag really executes and what commands do to a running platform.

A diverse group of travellers studies a map on a wooden terrace above a bright island city and sea.
Lumo, the guide, waves hello while raising a glowing brass lantern.

Hi, I'm Lumo, your guide

New to Kolla-Ansible Day-2 operations? We'll take this one step at a time. The learning journey walks you through five stages, and every page tells you where you are and what to read next. (My name comes from lumen, Latin for light: like my lantern, the Academy sheds light on what the code really does.)

Start the learning journey

Start from what you need

  • I am about to run a command


    Operations: one page per Day-2 operation, with its commands, its impact classes and the risks it names.

  • A run did something unexpected


    Troubleshooting: where to start during an incident. A symptom index, the known bugs and limitations, the open questions, and what is not known.

  • I need the facts about one component


    Components: one page per OpenStack service and per infrastructure component, each ending with its service card.

  • I need to judge the risk of a change


    Risk and impact: the 30 rated risks (R25 does not apply to 2024.1), what a tag really executes, and why a small change creates a large workload.

  • I want to understand how a run works


    Concepts: repository architecture, deployment execution flow, compute scaling, and what the kolla images do at container start.

  • I need the detailed evidence


    Reference: findings, glossary, generated inventory, per-role fact sheets, Ansible listings, and the method and limits of the study.

Operator situations, from an incident to a failed VM deployment, are mapped to the pages to open first in Start from a situation.

New in 2024.1

What a cloud deployed by 2023.1 meets on its first 2024.1 run, each with the section that gives the evidence:

  • Restarts without an operator change. The systemd unit template changed, so the first 2024.1 check reports drift for a container whose unit 2023.1 wrote (F16); RabbitMQ quorum queues are on by default and prechecks fails while classic queues exist (F17). How to see both before the run: Predict and inspect a run §2.
  • Docker or Podman. The container module is now kolla_container, which drives Docker (the default) or Podman (Repository architecture §6).
  • Every play sets max_fail_percentage, default 100, which never stops a play early (Write config first, apply in controlled batches §1.3).

Frequently used procedures

The operation pages that Find it fast and Ten things to know before a production run point to most often:

Every operation page is in the Operations catalogue, and the Command quick lookup by tag has one row per command and tag.

Before any production run

Recently updated

Every change is in the commit history of the repository.

How to read this portal

Every statement carries one of two evidence labels, defined in Evidence labels:

  • CODE shown by the sources, with a path:line citation nearby.
  • NOT AVAILABLE the sources do not establish it.

An operation page shows its impact classes under its title. A class says that at least one command of the page has that impact, not that all do. The classes and the sections that support each one are in Operation impact classes:

  • read-only — the corpus says the command changes nothing.
  • writes-config — the command renders config and restarts no container; in 2024.1 it can pull the fluentd image.
  • pulls-images — the command downloads images; running containers do not change.
  • restarts-on-drift — the command can restart containers, only if drift is detected.
  • recreates-containers — the command recreates containers (stop, rm, create, start) when their image or parameters differ.
  • critical — the page covers an operation rated Critical in the risk register.
  • maintenance-window — the corpus asks to plan the operation as a maintenance.
  • recovery — the page is a backup or recovery procedure.

Find it fast lists typical operator questions and the section that answers each one. There are four more ways into the study: by tag (Tag index), by service (Components), by term (Glossary) and by what is not known (What is not known). This page adds no finding of its own: every entry restates what the linked section says, and the linked section is the reference.

Scope

The portal is a source-level study of Kolla-Ansible 2024.1-eom (openstack/kolla-ansible, tag 2024.1-eom, commit 890396f). All conclusions come from the 2024.1 implementation, not from older or newer releases (Method and limits of the study).

Component Version
OpenStack release 2024.1 (openstack_release: "2024.1", group_vars/all.yml:679)
Kolla-Ansible 18.x series, the 2024.1 series: tag 2024.1-eom, commit 890396f
Kolla (images) tag 2024.1-eom of openstack/kolla, commit ed7a36a
ansible-collection-kolla tag 2024.1-eom, commit 82a0f5c (requirements.yml pins stable/2024.1, requirements.yml:5)
Ansible ansible-core 2.16.20 (tools/kolla-ansible:10-11 accepts 2.15 to 2.16)
  • Source code only. The study rests only on reading source code: kolla-ansible 2024.1-eom, the kolla image repository 2024.1-eom, ansible-collection-kolla 2024.1-eom and the ansible-core 2.16.20 source. The only Ansible output used is its --list-tags/--list-tasks parse of the playbooks, which runs nothing against a host.
  • There are no timings. No run is used, so no wall time is known; work can only be counted from the code.
  • No container or VM behaviour is used as evidence. Every statement about data-plane or VM impact is NOT AVAILABLE. The Podman paths new in 2024.1 were read in the code only, and ansible-core 2.15 was not read.

The other limits are in Limits of this study.