cinder: restart services after upgrade

This patch is roughly an adaptation of
Ia6fc9011ee6f5461f40a1307b72709d769814a79 for cinder.

During an upgrade, cinder pins the version of RPC calls to the minimum
seen across all services. This ensures that old services do not receive
data they cannot handle. After the upgrade is complete, all cinder
services are supposed to be reloaded to cause them to check again the
RPC versions of services and use the new latest version which should now
be supported by all running services.

There is a second issue in that it takes some time for the upgraded
services to update the cinder services database table with their new
version. We need to wait until all cinder services have done this
before the restart is performed, otherwise the RPC version cap will
remain in place. There is currently no interface in cinder available for
checking these versions, so as a workaround we use a configurable
delay with a default duration of 30 seconds, as we do for nova.

This change restarts all cinder services after an upgrade, after a 30
second delay.

Closes-Bug: #1954932
Related-Bug: #1833069

Change-Id: I9164dc589386d2c2d4daf1bf84061b806ba9988d
(cherry picked from commit 80a32c3)

Author: markgoddard

ansible/roles/cinder/defaults/main.yml

@@ -201,6 +201,15 @@ cinder_logging_debug: "{{ openstack_logging_debug }}"
 
 openstack_cinder_auth: "{{ openstack_auth }}"
 
+# After upgrading cinder, services will have an RPC version cap in place.  We
+# need to restart all services in order to allow them to use the latest RPC
+# version. Ideally, there would be a way to check whether all cinder services
+# are using the latest version, but currently there is not. Instead, wait a
+# short time for all cinder services to update the version of their service in
+# the database.  This seems to take around 10 seconds, but the default is 30 to
+# allow room for slowness.
+cinder_rpc_version_startup_delay: 30
+
 
 ####################
 # Cinder

ansible/roles/cinder/handlers/main.yml

@@ -66,3 +66,23 @@
     healthcheck: "{{ service.healthcheck | default(omit) }}"
   when:
     - kolla_action != "config"
+
+# NOTE(mgoddard): After upgrading cinder, services will have an RPC version cap
+# in place.  We need to restart all services in order to allow them to use the
+# latest RPC version. Ideally, there would be a way to check whether all cinder
+# services are using the latest version, but currently there is not. Instead,
+# wait a short time for all cinder services to update the version of their
+# service in the database.  This seems to take around 10 seconds, but the
+# default is 30 to allow room for slowness.
+
+- name: Wait for cinder services to update service versions
+  pause:
+    seconds: "{{ cinder_rpc_version_startup_delay }}"
+  run_once: true
+  when:
+    - kolla_action == 'upgrade'
+  listen:
+    - Restart cinder-api container
+    - Restart cinder-scheduler container
+    - Restart cinder-volume container
+    - Restart cinder-backup container

ansible/roles/cinder/tasks/reload.yml

@@ -0,0 +1,10 @@
+---
+- name: Reload cinder services to remove RPC version pin
+  vars:
+    service: "{{ item.value }}"
+  become: true
+  kolla_docker:
+    action: "restart_container"
+    common_options: "{{ docker_common_options }}"
+    name: "{{ service.container_name }}"
+  with_dict: "{{ cinder_services | select_services_enabled_and_mapped_to_host }}"

ansible/roles/cinder/tasks/upgrade.yml

@@ -13,6 +13,8 @@
 - name: Flush handlers
   meta: flush_handlers
 
+- import_tasks: reload.yml
+
 - name: Running Cinder online schema migration
   vars:
     cinder_api: "{{ cinder_services['cinder-api'] }}"

releasenotes/notes/unpin-cinder-rpcs-8eb7e0858a91b9b8.yaml

@@ -0,0 +1,6 @@
+---
+fixes:
+  - |
+    Fixes an issue with Cinder upgrade where Cinder services would remain
+    pinned to the previous release's RPC & object versions. `LP#1954932
+    <https://bugs.launchpad.net/kolla-ansible/+bug/1954932>`__