The open questions of the study, Q1 to Q14. The summary table lists them, and each question has a section that gives
why it matters and its status.
Note
Key answers
Open questions Q1–Q14: all are NOT AVAILABLE, except Q14, which is answered from the source
CODE, the control-flow part of Q6, also answered from the source CODE, and Q10, whose 2023.1
form is partly answered: F2 no longer applies, and the ansible-core 2.13 side is
not in these sources (Questions). Q1 asks whether running VMs survive a nova_libvirt
restart or recreate.
These questions cannot be settled by reading the sources, except Q14, which is answered in its section by reading the
kolla image sources. For Q6, the 2023.1 plays and the ansible-core 2.14.18 code answer whether later batches proceed;
the state of the failed node and the availability risk stay NOT AVAILABLE. Q10 was asked for the Ansible versions
Ussuri allows (2.8 and 2.9); 2023.1 allows ansible-core 2.13 and 2.14, and only 2.14.18 is in these sources, so its
2023.1 form is answered only in part. The others are NOT AVAILABLE.
Control flow answered from the sourceCODE. In 2023.1 the handlers only put the hosts into the mariadb_restart and mariadb_start groups, and separate plays with serial: "33%" restart them; neither these plays nor any other MariaDB play sets any_errors_fatal or max_fail_percentage (roles/mariadb/handlers/main.yml:56-73, mariadb.yml:14-40). A node that fails its port or WSREP wait (roles/mariadb/tasks/restart_services.yml:18-51) marks only its own host failed. When that is every host of the batch (always so with three nodes, one host per batch), ansible-core ends the playbook and the later batches are not restarted; when other hosts of the batch succeed, the next batch still restarts its nodes (ansible-core 2.14.18 lib/ansible/executor/playbook_executor.py:204-208, lib/ansible/utils/helpers.py:25-32; R9, MariaDB, fact sheet B §1.5). NOT AVAILABLE Whether the failed node is then still joining, whether that can leave a single Synced node acting as donor, and so how large the quorum and availability risk is
Ussuri allows 2.8. 2023.1 no longer runs Ansible 2.8 or 2.9: kolla-ansible accepts ansible-core 2.13 to 2.14 (tools/kolla-ansible:10-11), so the question becomes whether 2.13 and 2.14 behave the same
Status
Answered in partCODE. F2 no longer applies in 2023.1, whatever the Ansible version: common is its own play tagged common (site.yml:86-97), so -t common does not depend on tag inheritance through the dynamic include_tasks (roles/common/tasks/main.yml:2). F1 holds in ansible-core 2.14.18: one changed loop item notifies the handlers of every executed item (lib/ansible/plugins/strategy/__init__.py:614-621, lib/ansible/executor/task_executor.py:804-805); the tag inheritance of dynamic includes is in lib/ansible/playbook/task_include.py:131-150 and lib/ansible/playbook/block.py:297-368 (Tags §2.1, §2.4). The Ussuri comparison (Ansible 2.8.20 against 2.9.27: same code for both findings) does not carry over. NOT AVAILABLE Whether ansible-core 2.13 has the same code: only 2.14.18 is in these sources
Answered from the sourceCODE, see Kolla images §4. Nova's bootstrap containers run with KOLLA_UPGRADE: api_db sync + db sync --local_cell (nova-api, whose bootstrap nova.conf points [database] at cell0), db sync --local_cell (nova-conductor), so Nova's online data migrations never run outside upgrade; the images' KOLLA_BOOTSTRAP branches are unreachable. Placement's KOLLA_BOOTSTRAP runs db sync + online_data_migrations on every deploy/reconfigure. New in 2023.1: nova-api's KOLLA_UPGRADE_CHECK runs nova-status upgrade check, upgrade only. Glance adds db_load_metadefs. Heat creates heat_user_domain/heat_domain_admin only if the domain is absent, so a changed heat_domain_admin_password is never pushed to Keystone. neutron-server's KOLLA_BOOTSTRAP migrates neutron and every enabled subproject (networking-sfc, neutron-dynamic-routing, neutron-vpnaas, vmware-nsx), and the rolling upgrade migrates the same list, so the Ussuri gap on -dynamic-routing is closed (roles/neutron/defaults/main.yml:656-664,705,711). RabbitMQ writes the Erlang cookie only in its bootstrap container (kolla docker/rabbitmq/rabbitmq/extend_start.sh:7-18). Agent, OVS and OVN images have no one-shot branch