Merge pull request #619 from stackhpc/upstream/2023.1-2024-09-23

Synchronise 2023.1 with upstream

ansible/gather-facts.yml

@@ -13,6 +13,7 @@
         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
@@ -53,8 +54,11 @@
       delegate_facts: True
       delegate_to: "{{ item }}"
       with_items: "{{ delegate_hosts }}"
-      # We gathered facts for all hosts in the batch during the first play.
       when:
+        # We gathered facts for all hosts in the batch during the first play.
+        # Ensure that we don't try again if they failed.
+        - 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

ansible/module_utils/kolla_systemd_worker.py

@@ -23,6 +23,7 @@
 [Unit]
 Description=docker ${service_name}
 After=docker.service
+Wants=docker.service
 StartLimitIntervalSec=${restart_timeout}
 
 StartLimitBurst=${restart_retries}

ansible/roles/octavia/templates/octavia-interface.service.j2

@@ -7,6 +7,10 @@ After=docker.service
 Type=oneshot
 User=root
 Group=root
+Restart=on-failure
+{% if octavia_interface_wait_timeout is defined %}
+TimeoutStartSec={{ octavia_interface_wait_timeout }}
+{% endif %}
 RemainAfterExit=true
 ExecStartPre=/sbin/ip link set dev {{ octavia_network_interface }} address {{ port_info.port.mac_address }}
 ExecStart=/sbin/dhclient -v {{ octavia_network_interface }} -cf /etc/dhcp/octavia-dhclient.conf

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

@@ -184,6 +184,9 @@ therefore RabbitMQ version 3.13).
 
    This command should be run from the Antelope release.
 
+   Note that this command is NOT idempotent. See "RabbitMQ versions" below for
+   an alternative approach.
+
 .. code-block:: console
 
    kolla-ansible rabbitmq-upgrade 3.12

doc/source/reference/networking/octavia.rst

@@ -437,6 +437,24 @@ Add ``octavia_network_type`` to ``globals.yml`` and set the value to ``tenant``
 
 Next，follow the deployment instructions as normal.
 
+Failure handling
+----------------
+
+On large deployments, where neutron-openvswitch-agent sync could takes
+more then 5 minutes, you can get an error on octavia-interface.service
+systemd unit, because it can't wait either o-hm0 interface is already
+attached to br-int, or octavia management VxLAN is already configured
+on that host. In this case you have to add ``octavia_interface_wait_timeout``
+to ``globals.yml`` and set the value to new timeout in seconds
+
+.. code-block:: yaml
+
+   octavia_interface_wait_timeout: 1800
+
+On deployments with up to 2500 network ports per network node sync process
+could take up to 30mins. But you have to consider this value according
+to your deployment size.
+
 OVN provider
 ============
 

releasenotes/notes/fact-gathering-double-failure-4662ef2f54d58291.yaml

@@ -0,0 +1,6 @@
+---
+fixes:
+  - |
+    Fixes an issue during fact gathering when using the ``--limit`` argument
+    where a host that fails to gather facts could cause another host to fail
+    during delegated fact gathering.

releasenotes/notes/fix-octavia-interface-timeout-5e87ea2501d5ab3c.yaml

@@ -0,0 +1,10 @@
+---
+fixes:
+  - |
+    Fixes 2067036.
+    Added ``octavia_interface_wait_timeout`` to control
+    octavia-interface.service timeout to be able wait
+    openvswitch agent sync has been finished and
+    octavia-lb-net is reachable from the host.
+    Also set restart policy for this unit to on-failure
+    `LP#2067036 <https://launchpad.net/bugs/2067036>`__