Merge pull request #630 from stackhpc/upstream/2024.1-2024-10-10

Synchronise 2024.1 with upstream

ansible/gather-facts.yml

@@ -4,33 +4,25 @@
 # building their configurations.
 - name: Gather facts for all hosts
   hosts: all
-  any_errors_fatal: "{{ kolla_ansible_setup_any_errors_fatal | bool }}"
   max_fail_percentage: >-
     {{ gather_facts_max_fail_percentage |
        default(kolla_max_fail_percentage) |
        default(100) }}
   serial: '{{ kolla_serial|default("0") }}'
   gather_facts: false
   tasks:
+    - name: Group hosts to determine when using --limit
+      group_by:
+        key: "all_using_limit_{{ (ansible_play_batch | length) != (groups['all'] | length) }}"
+      changed_when: false
+
     - name: Gather facts
       setup:
         filter: "{{ kolla_ansible_setup_filter }}"
         gather_subset: "{{ kolla_ansible_setup_gather_subset }}"
       when:
         # Don't gather if fact caching is in use
         - not ansible_facts
-
-    - name: Gather package facts
-      package_facts:
-      when:
-        - "'packages' not in ansible_facts"
-        - kolla_action is defined
-        - kolla_action == "precheck"
-
-    - name: Group hosts to determine when using --limit
-      group_by:
-        key: "all_using_limit_{{ (ansible_play_batch | length) != (groups['all'] | length) }}"
-      changed_when: false
   tags: always
 
 # NOTE(pbourke): This case covers deploying subsets of hosts using --limit. The
@@ -41,7 +33,6 @@
 # the limit.
 - name: Gather facts for all hosts (if using --limit)
   hosts: all_using_limit_True
-  any_errors_fatal: "{{ kolla_ansible_setup_any_errors_fatal | bool }}"
   max_fail_percentage: >-
     {{ gather_facts_max_fail_percentage |
        default(kolla_max_fail_percentage) |
@@ -68,14 +59,4 @@
         - item not in groups["all_using_limit_True"]
         # Don't gather if fact caching is in use
         - not hostvars[item].ansible_facts
-
-    - name: Gather package facts
-      package_facts:
-      delegate_facts: True
-      delegate_to: "{{ item }}"
-      with_items: "{{ delegate_hosts }}"
-      when:
-        - "'packages' not in hostvars[item].ansible_facts"
-        - kolla_action is defined
-        - "kolla_action == 'precheck'"
   tags: always

ansible/group_vars/all.yml

@@ -36,14 +36,6 @@ kolla_ansible_setup_filter: "{{ omit }}"
 # By default, we do not provide a gather subset.
 kolla_ansible_setup_gather_subset: "{{ omit }}"
 
-# This variable is used as "any_errors_fatal" setting for the setup (gather
-# facts) plays.
-# This is useful for weeding out failing hosts early to avoid late failures
-# due to missing facts (especially cross-host).
-# Do note this still supports host fact caching and it will not affect
-# scenarios with all facts cached (as there is no task to fail).
-kolla_ansible_setup_any_errors_fatal: false
-
 ###################
 # Kolla options
 ###################

ansible/module_utils/kolla_podman_worker.py

@@ -177,7 +177,11 @@ def parse_volumes(self, volumes, mounts, filtered_volumes):
                 mounts.append(mount_item)
             else:
                 try:
-                    src, dest = item.split(':')
+                    mode = 'rw'
+                    if item.count(':') == 2:
+                        src, dest, mode = item.split(':')
+                    else:
+                        src, dest = item.split(':')
                 except ValueError:
                     self.module.fail_json(
                         msg="Wrong format of volume: {}".format(item),
@@ -192,7 +196,7 @@ def parse_volumes(self, volumes, mounts, filtered_volumes):
                 else:
                     filtered_volumes[src] = dict(
                         bind=dest,
-                        mode='rw'
+                        mode=mode
                     )
 
     def parse_dimensions(self, dimensions):

ansible/roles/ironic/defaults/main.yml

@@ -82,6 +82,7 @@ ironic_services:
     enabled: true
     cap_add:
       - NET_ADMIN
+      - NET_RAW
     image: "{{ ironic_dnsmasq_image_full }}"
     volumes: "{{ ironic_dnsmasq_default_volumes + ironic_dnsmasq_extra_volumes }}"
     dimensions: "{{ ironic_dnsmasq_dimensions }}"

ansible/roles/openvswitch/defaults/main.yml

@@ -103,5 +103,5 @@ openvswitch_ovs_vsctl_wrapper_enabled: false
 #############
 
 openvswitch_system_id: "{{ ansible_facts.hostname }}"
-openvswitch_hostname: "{{ ansible_facts.hostname }}"
+openvswitch_hostname: "{{ ansible_facts.fqdn }}"
 openvswitch_hw_offload: "no"

ansible/roles/rabbitmq/tasks/upgrade.yml

@@ -1,4 +1,6 @@
 ---
+- import_tasks: version-check.yml
+
 - include_tasks: remove-ha-all-policy.yml
   when:
     - not om_enable_rabbitmq_high_availability | bool

ansible/roles/rabbitmq/tasks/version-check.yml

@@ -0,0 +1,59 @@
+---
+- block:
+    - name: Get current RabbitMQ version
+      vars:
+        service_name: "rabbitmq"
+        service: "{{ rabbitmq_services[service_name] }}"
+      become: true
+      command: "{{ kolla_container_engine }} exec {{ service.container_name }} rabbitmqctl --version"
+      register: rabbitmq_version_current
+      changed_when: false
+
+    - name: Get new RabbitMQ version
+      become: true
+      vars:
+        rabbitmq_container: "{{ rabbitmq_services['rabbitmq'] }}"
+      kolla_container:
+        action: "start_container"
+        command: "rabbitmqctl --version"
+        detach: false
+        environment:
+          KOLLA_CONFIG_STRATEGY: "{{ config_strategy }}"
+        image: "{{ rabbitmq_container.image }}"
+        name: "rabbitmq_version_check"
+        restart_policy: oneshot
+        volumes: "{{ rabbitmq_default_volumes + rabbitmq_extra_volumes }}"
+      register: rabbitmq_version_new
+      failed_when: false
+      check_mode: false
+
+    # As an example, when the new RabbitMQ version is 3.13.6:
+    # new_major_version = 3
+    # new_minor_version = 13
+    # new_version = 3.13
+    # And if the current RabbitMQ version is 3.11.28:
+    # upgrade_version = 3.12
+    - name: Check if running RabbitMQ is at most one version behind
+      vars:
+        current_version_major: "{{ rabbitmq_version_current.stdout | regex_search('^[0-9]+') }}"
+        current_version_minor: "{{ rabbitmq_version_current.stdout | regex_search('(?<=.)[^.].') }}"
+        current_version: "{{ rabbitmq_version_current.stdout | regex_replace('.[^.]+$', '') }}"
+        new_version_major: "{{ rabbitmq_version_new.stdout | regex_search('^[0-9]+') }}"
+        new_version_minor: "{{ rabbitmq_version_new.stdout | regex_search('(?<=.)[^.].') }}"
+        new_version: "{{ rabbitmq_version_new.stdout | regex_replace('.[^.]+$', '') }}"
+        # Note: this assumes 3.13 will be the last release before 4.0.
+        upgrade_version: "{{ '4.0' if current_version == '3.13' else current_version_major + '.' + (current_version_minor | int + 1) | string }}"
+      assert:
+        that: (current_version_major == new_version_major and
+              new_version_minor | int - current_version_minor | int <= 1) or
+              (new_version | float == 4.0 and current_version | float == 3.13)
+        fail_msg: >
+          Looks like you're trying to run a skip-release upgrade!
+          RabbitMQ must be at most one version behind the target release version ({{ rabbitmq_version_new.stdout | trim }}) to run this upgrade.
+          You are running {{ rabbitmq_version_current.stdout }}.
+          Please first upgrade to {{ upgrade_version }} with the command ``kolla-ansible rabbitmq-upgrade {{ upgrade_version }}``.
+          See these docs for more details: https://docs.openstack.org/kolla-ansible/latest/reference/message-queues/rabbitmq.html#slurp
+
+  delegate_to: "{{ groups[role_rabbitmq_groups] | first }}"
+  run_once: true
+  tags: rabbitmq-version-check

doc/source/reference/message-queues/rabbitmq.rst

@@ -128,7 +128,8 @@ additional steps needed to migrate from transient to durable queues.
 .. warning::
 
    Since the default changed from non-HA to Quorum queues in Bobcat release,
-   following procedure is required to be carried out before an upgrade.
+   following procedure is required to be carried out before a SLURP upgrade to
+   Caracal.
 
 1. Stop all OpenStack services which use RabbitMQ, so that they will not
    attempt to recreate any queues yet.
@@ -199,7 +200,7 @@ you must override the image. if you want to use version 3.12 change
 
 .. code-block:: yaml
 
-   rabbitmq_image: "{{ docker_registry ~ '/' if docker_registry else '' }}{{ docker_namespace }}/rabbitmq-3.12"
+   rabbitmq_image: "{{ docker_registry ~ '/' if docker_registry else '' }}{{ docker_namespace }}/rabbitmq-3-12"
 
 You can then upgrade RabbitMQ with the usual command:
 

doc/source/user/operating-kolla.rst

@@ -176,6 +176,16 @@ issues:
 
 At a convenient time, the upgrade can now be run.
 
+SLURP extra preparations
+++++++++++++++++++++++++
+
+RabbitMQ has two major version releases per year but does not support jumping
+two versions in one upgrade. So if you want to perform a skip-level upgrade,
+you must first upgrade RabbitMQ to an intermediary version. Please see the
+`RabbitMQ SLURP section
+<https://docs.openstack.org/kolla-ansible/latest/reference/message-queues/rabbitmq.html#slurp>`__
+for details.
+
 Perform the Upgrade
 -------------------
 

etc/kolla/globals.yml

@@ -24,14 +24,6 @@
 # Dummy variable to allow Ansible to accept this file.
 workaround_ansible_issue_8743: yes
 
-# This variable is used as "any_errors_fatal" setting for the setup (gather
-# facts) plays.
-# This is useful for weeding out failing hosts early to avoid late failures
-# due to missing facts (especially cross-host).
-# Do note this still supports host fact caching and it will not affect
-# scenarios with all facts cached (as there is no task to fail).
-#kolla_ansible_setup_any_errors_fatal: false
-
 # This variable may be used to set the maximum failure percentage for all
 # plays. More fine-grained control is possible via per-service variables, e.g.
 # nova_max_fail_percentage. The default behaviour is to set a max fail
@@ -757,7 +749,7 @@ workaround_ansible_issue_8743: yes
 #enable_prometheus_mysqld_exporter: "{{ enable_mariadb | bool }}"
 #enable_prometheus_node_exporter: "{{ enable_prometheus | bool }}"
 #enable_prometheus_cadvisor: "{{ enable_prometheus | bool }}"
-#enable_prometheus_fluentd_integration: "{{ enable_prometheus | bool and enable fluentd | bool }}"
+#enable_prometheus_fluentd_integration: "{{ enable_prometheus | bool and enable_fluentd | bool }}"
 #enable_prometheus_memcached: "{{ enable_prometheus | bool }}"
 #enable_prometheus_alertmanager: "{{ enable_prometheus | bool }}"
 #enable_prometheus_alertmanager_external: "{{ enable_prometheus_alertmanager | bool }}"

releasenotes/notes/adds-net-raw-capability-to-ironic-dnsmasq-40f5894c5180b12d.yaml

@@ -0,0 +1,6 @@
+---
+fixes:
+  - |
+    Fixes an issue with ironic dnsmasq failing to start in deployments
+    using podman because it requires the NET_RAW capability. See `LP#2055282
+    <https://bugs.launchpad.net/kolla-ansible/+bug/2055282>`__.

releasenotes/notes/bug-2080552-714078144170a5bd.yaml

@@ -0,0 +1,8 @@
+---
+fixes:
+  - |
+    Fixes #2080552.
+    ``openvswitch`` role will now set ``external-ids:hostname`` to
+    ``{{ ansible_facts.fqdn }}`` instead of ``{{ ansible_facts.hostname }}``
+    due to Neutron using FQDN based hostnames in ``requested-chassis`` field.
+    `LP#2080552 <https://launchpad.net/bugs/2080552>`__

releasenotes/notes/fixes-podman-named-volume-issues-when-using-mode-36b9862321eb9e68.yaml

@@ -0,0 +1,6 @@
+---
+fixes:
+  - |
+    Fixes an issue, when using podman, with named volumes that use a mode
+    specifier. See `LP#2054834
+    <https://bugs.launchpad.net/kolla-ansible/+bug/2054834>`_ for more details.

releasenotes/notes/remove-any-errors-fatal-664fc2207074f0ef.yaml

@@ -0,0 +1,14 @@
+---
+upgrade:
+  - |
+    Support for failing execution early if fact collection fails on any of the
+    hosts by setting ``kolla_ansible_setup_any_errors_fatal`` to ``true`` has
+    been removed. This is due to Ansible's ``any_errors_fatal`` parameter not
+    being templated, resulting in the value always being interpreted as
+    ``true``, even though the default value of
+    ``kolla_ansible_setup_any_errors_fatal`` is ``false``.
+
+    Equivalent behaviour is possible by setting the maximum failure percentage
+    to 0. This may be done specifically for fact gathering using
+    ``gather_facts_max_fail_percentage`` or globally using
+    ``kolla_max_fail_percentage``.

tests/templates/globals-default.j2

@@ -5,10 +5,6 @@
 ansible_python_interpreter: /usr/bin/python3
 {% endif %}
 
-# NOTE(yoctozepto): In CI it makes sense to always try to fail
-# as early as possible.
-kolla_ansible_setup_any_errors_fatal: true
-
 kolla_base_distro: "{{ base_distro }}"
 network_interface: "{{ api_interface_name }}"
 network_address_family: "{{ address_family }}"
@@ -189,6 +185,7 @@ enable_octavia: "yes"
 octavia_provider_drivers: "ovn:OVN provider"
 octavia_provider_agents: "ovn"
 enable_redis: "yes"
+neutron_dns_domain: "example.org."
 {% endif %}
 
 {% if scenario == "prometheus-opensearch" %}