Staggered upgrade procedure

doc/source/operations/upgrading-openstack.rst

@@ -1063,6 +1063,12 @@ This will block the upgrade, but may be overridden by setting
 ``etc/kayobe/kolla/globals.yml`` or
 ``etc/kayobe/environments/<env>/kolla/globals.yml``.
 
+Depending on the networking architecture of your cloud, the steps used
+to upgrade the containerised services will differ.
+
+OVN
+^^^
+
 To upgrade the containerised control plane services:
 
 .. code-block:: console
@@ -1076,6 +1082,65 @@ scope of the upgrade:
 
    kayobe overcloud service upgrade --tags config --kolla-tags keystone
 
+OVS (w/ Dedicated network nodes)
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+You should first stop the Octavia health manager to prevent alerts during
+the service upgrade.
+
+.. code-block:: console
+
+   kayobe overcloud host command run --command "docker stop octavia_health_manager" --limit controllers --become
+
+Upgrade the control plane services
+
+.. code-block:: console
+
+   kayobe overcloud serivce upgrade --kolla-limit controllers
+
+To ensure L3 reliability during the upgrade, we will need to manually drain
+the network nodes of all agents, and upgrade the nodes sequentially.
+
+Kolla credentials will need to be activated before running the neutron-namespace-drain
+role.
+
+.. code-block:: console
+
+   source $KOLLA_CONFIG_PATH/public-openrc.sh
+
+You should substitute <network0> with the first network node to be drained, To set
+the node for maintenance and begin draining the agents:
+
+.. code-block:: console
+
+   kayobe playbook run $KAYOBE_CONFIG_PATH/ansible/neutron-l3-drain.yml -e neutron_drain_host=<network0> -e maintenance=true -e drain_dhcp_agents=true
+
+You can monitor the L3/DHCP agents being drained from the node by running:
+
+.. code-block:: console
+   
+   ssh -t <network0> watch ip netns ls
+
+Once all agents have been drained, you can upgrade the containerised services
+on the network node.
+
+.. code-block:: console
+
+   kayobe overcloud service upgrade --kolla-limit <network0>
+
+Following the service upgrade, the agents can be restored on the node by disabling maintenance:
+
+.. code-block:: console
+
+   kayobe playbook run $KAYOBE_CONFIG_PATH/ansible/neutron-l3-drain.yml -e neutron_drain_host=<network0> -e maintenance=false -e drain_dhcp_agents=true
+
+The above steps should be repeated for the remaining network nodes, once all network nodes have been upgraded
+the remaining containerised services can be upgraded:
+
+.. code-block:: console
+
+   kayobe overcloud service upgrade --kolla-tags common,nova,prometheus,openvswitch,neutron --skip-prechecks  -kl controllers,compute --limit controllers,compute
+
 Updating the Octavia Amphora Image
 ----------------------------------
 

etc/kayobe/ansible/neutron-l3-drain.yml

@@ -0,0 +1,23 @@
+---
+- name: Drain neutron of l3 agents and dhcp agents
+  hosts: localhost
+  gather_facts: true
+  tags:
+    - neutron-l3-drain
+  vars:
+    maintenance: false
+    drain_ctrl1: false
+    drain_ctrl2: false
+    drain_ctrl3: false
+  tasks:
+    - import_role:
+        name: neutron-namespace-drain
+        tasks_from: main.yml
+      when: drain_ctrl1 | bool or drain_ctrl2 | bool or drain_ctrl3 | bool or neutron_drain_host is defined
+    - name: "Print Info"
+      debug:
+        msg:
+          - "{{ neutron_drain_host }} is read for maintenance"
+          - "rerun this play book with -e maintenance=false to re-add"
+          - "routers"
+      when: maintenance | bool

etc/kayobe/ansible/roles/neutron-namespace-drain/defaults/main.yml

@@ -0,0 +1,8 @@
+---
+neutron_drain_venv: "{{ virtualenv_path }}/openstack"
+neutron_drain_host: "{% if drain_ctrl1 | bool %}{{ groups['controllers'][0] }}{% elif drain_ctrl2 | bool %}{{ groups['controllers'][1] }}{% elif drain_ctrl3 | bool %}{{ groups['controllers'][2] }}{% endif %}"
+neutron_drain_venv_python: "{{ 'python' ~ ansible_facts.python.version.major ~ '.' ~ ansible_facts.python.version.minor }}"
+drain_ctrl1: false
+drain_ctrl2: false
+drain_ctrl3: false
+drain_dhcp_agents: false

etc/kayobe/ansible/roles/neutron-namespace-drain/tasks/add-dhcp.yml

@@ -0,0 +1,51 @@
+---
+- name: Query source SRC_DHCP_ID
+  command: >
+    {{ neutron_drain_venv }}/bin/openstack
+    network agent list --host {{ neutron_drain_host }}
+    --agent-type dhcp -f value -c ID
+  register: SRC_DHCP_ID
+  environment: "{{ openstack_auth_env }}"
+  delegate_to: "{{ neutron_drain_host }}"
+  vars:
+    ansible_host: "{{ hostvars[neutron_drain_host].ansible_host }}"
+
+- name: Set fact containing SRC_DHCP_ID
+  set_fact:
+    DHCP_SRC_ID: "{{ SRC_DHCP_ID.stdout }}"
+
+- name: Enable DHCP agent
+  command: >
+    {{ neutron_drain_venv }}/bin/openstack
+    network agent set "{{ DHCP_SRC_ID }}" --enable
+  environment: "{{ openstack_auth_env }}"
+  delegate_to: "{{ neutron_drain_host }}"
+  vars:
+    ansible_host: "{{ hostvars[neutron_drain_host].ansible_host }}"
+
+- name: Get Network IDs
+  command: >
+    {{ neutron_drain_venv }}/bin/openstack
+    network list -f value -c ID
+  register: network_ids
+  environment: "{{ openstack_auth_env }}"
+  delegate_to: "{{ neutron_drain_host }}"
+  vars:
+    ansible_host: "{{ hostvars[neutron_drain_host].ansible_host }}"
+
+- name: Set Network IDs
+  set_fact:
+    NETWORK_IDS: "{{ network_ids.stdout_lines }}"
+
+- name: Add DHCP agent
+  command: >
+    {{ neutron_drain_venv }}/bin/openstack
+    network agent add network {{ DHCP_SRC_ID }} {{ item }} --dhcp
+  environment: "{{ openstack_auth_env }}"
+  delegate_to: "{{ neutron_drain_host }}"
+  vars:
+    ansible_host: "{{ hostvars[neutron_drain_host].ansible_host }}"
+  with_items:
+    - "{{ NETWORK_IDS }}"
+  loop_control:
+    pause: 10 

etc/kayobe/ansible/roles/neutron-namespace-drain/tasks/add-new-l3.yml

@@ -0,0 +1,43 @@
+---
+- name: Query L3_IDs
+  command: >
+    {{ neutron_drain_venv }}/bin/openstack
+    network agent list --agent-type l3 -f value -c ID
+  register: L3_ID
+  environment: "{{ openstack_auth_env }}"
+  delegate_to: "{{ neutron_drain_host }}"
+  vars:
+    ansible_host: "{{ hostvars[neutron_drain_host].ansible_host }}"
+
+- name: Set fact containing SRC_L3_IDs
+  set_fact:
+    L3_IDS: "{{ L3_ID.stdout_lines }}"
+
+- name: Get agents for each router
+  command: >
+    {{ neutron_drain_venv }}/bin/openstack
+    network agent list --router {{ router_id }} --agent-type l3 -f value -c ID
+  environment: "{{ openstack_auth_env }}"
+  delegate_to: "{{ neutron_drain_host }}"
+  failed_when: false
+  register: ROUTER_L3_IDS
+  vars:
+    ansible_host: "{{ hostvars[neutron_drain_host].ansible_host }}"
+  loop: "{{ ROUTER_IDS }}"
+  loop_control:
+    loop_var: router_id
+
+- name: Add agent to router
+  command: >
+    {{ neutron_drain_venv }}/bin/openstack
+    network agent add router --l3 {{ L3_ADD }} {{ item.router_id }}
+  loop: "{{ ROUTER_L3_IDS.results }}"
+  loop_control:
+    label: "{{ item.router_id }}"
+  environment: "{{ openstack_auth_env }}"
+  delegate_to: "{{ neutron_drain_host }}"
+  vars:
+    ansible_host: "{{ hostvars[neutron_drain_host].ansible_host }}"
+    L3_ADD_DIFF: "{{ L3_IDS | difference([L3_SRC_ID]) | difference(item.stdout_lines) }}"
+    L3_ADD: "{{ L3_ADD_DIFF[:1] | first }}"
+  when: L3_ADD_DIFF | length > 0

etc/kayobe/ansible/roles/neutron-namespace-drain/tasks/drain-dhcp.yml

@@ -0,0 +1,87 @@
+---
+- name: Query source SRC_DHCP_ID
+  command: >
+    {{ neutron_drain_venv }}/bin/openstack
+    network agent list --host {{ neutron_drain_host }}
+    --agent-type dhcp -f value -c ID
+  register: SRC_DHCP_ID
+  environment: "{{ openstack_auth_env }}"
+  delegate_to: "{{ neutron_drain_host }}"
+  vars:
+    ansible_host: "{{ hostvars[neutron_drain_host].ansible_host }}"
+
+- name: Set fact containing SRC_DHCP_ID
+  set_fact:
+    DHCP_SRC_ID: "{{ SRC_DHCP_ID.stdout }}"
+
+- name: Get DHCP agent network IDs
+  command: >
+    {{ neutron_drain_venv }}/bin/openstack
+     network list --agent {{ DHCP_SRC_ID }} -f value -c ID
+  register: dhcp_agent_ids
+  environment: "{{ openstack_auth_env }}"
+  delegate_to: "{{ neutron_drain_host }}"
+  vars:
+    ansible_host: "{{ hostvars[neutron_drain_host].ansible_host }}"
+
+- name: Set DHCP agent network IDs
+  set_fact:
+    DHCP_AGENT_IDS: "{{ dhcp_agent_ids.stdout_lines }}"
+
+- name: Remove DHCP agent
+  command: >
+    {{ neutron_drain_venv }}/bin/openstack
+    network agent remove network {{ DHCP_SRC_ID }} {{ item }} --dhcp
+  environment: "{{ openstack_auth_env }}"
+  delegate_to: "{{ neutron_drain_host }}"
+  vars:
+    ansible_host: "{{ hostvars[neutron_drain_host].ansible_host }}"
+  with_items:
+    - "{{ DHCP_AGENT_IDS }}"
+
+- name: Wait for no more dhcp agents to be attached to the host
+  block:
+    - name: Retry count
+      set_fact:
+        retry_count: "{{ 0 if retry_count is undefined or retry_count == 'reset' else retry_count | int + 1 }}"
+        max_retries: 20
+
+    - name: Verify dhcp agents exist
+      command: >
+        {{ neutron_drain_venv }}/bin/openstack
+        network list --agent {{ DHCP_SRC_ID }} -f value -c ID
+      environment: "{{ openstack_auth_env }}"
+      delegate_to: "{{ neutron_drain_host }}"
+      vars:
+        ansible_host: "{{ hostvars[neutron_drain_host].ansible_host }}"
+      register: agent_status
+
+    - name: Fail if DHCP agent still attached
+      fail:
+        msg: a DHCP agent is still attached to "{{ L3_SRC_ID }}"
+      when: agent_status.stdout | length > 0
+
+    - name: Reset retry count after success
+      set_fact:
+        retry_count: reset
+  rescue:
+    - fail:
+        msg: |
+          Maximum retries waiting for DHCP agents to be detached reached
+      when: retry_count | int == max_retries
+
+    - name: Reset retry counter if max retries reached (exit loop)
+      set_fact:
+        retry_count: reset
+      failed_when: retry_count == 'reset'
+      when: retry_count | int >= max_retries | int
+
+- name: Disable DHCP agent
+  command: >
+    {{ neutron_drain_venv }}/bin/openstack
+    network agent set "{{ DHCP_SRC_ID }}" --disable
+  environment: "{{ openstack_auth_env }}"
+  delegate_to: "{{ neutron_drain_host }}"
+  vars:
+    ansible_host: "{{ hostvars[neutron_drain_host].ansible_host }}"
+  when: agent_status.stdout | length == 0

etc/kayobe/ansible/roles/neutron-namespace-drain/tasks/drain-l3.yml

@@ -0,0 +1,90 @@
+---
+- name: Query source SRC_L3_ID
+  command: >
+    {{ neutron_drain_venv }}/bin/openstack
+    network agent list --host {{ neutron_drain_host }}
+    --agent-type l3 -f value -c ID
+  register: SRC_L3_ID
+  environment: "{{ openstack_auth_env }}"
+  delegate_to: "{{ neutron_drain_host }}"
+  vars:
+    ansible_host: "{{ hostvars[neutron_drain_host].ansible_host }}"
+
+- name: Set fact containing SRC_L3_ID
+  set_fact:
+    L3_SRC_ID: "{{ SRC_L3_ID.stdout }}"
+
+- name: Get Router IDs
+  command: >
+    {{ neutron_drain_venv }}/bin/openstack
+    router list --agent {{ L3_SRC_ID }} -f value -c ID
+  register: router_ids
+  environment: "{{ openstack_auth_env }}"
+  delegate_to: "{{ neutron_drain_host }}"
+  vars:
+    ansible_host: "{{ hostvars[neutron_drain_host].ansible_host }}"
+
+- name: Set Router IDs
+  set_fact:
+    ROUTER_IDS: "{{ router_ids.stdout_lines }}"
+
+- name: Add agents to router
+  include_tasks: add-new-l3.yml
+  when: ROUTER_IDS | length > 0
+
+- name: Remove router
+  command: >
+    {{ neutron_drain_venv }}/bin/openstack
+    network agent remove router {{ L3_SRC_ID }} {{ item }} --l3
+  environment: "{{ openstack_auth_env }}"
+  delegate_to: "{{ neutron_drain_host }}"
+  vars:
+    ansible_host: "{{ hostvars[neutron_drain_host].ansible_host }}"
+  with_items:
+    - "{{ ROUTER_IDS }}"
+
+- name: Wait for no more routers to be attached to the host
+  block:
+    - name: Retry count
+      set_fact:
+        retry_count: "{{ 0 if retry_count is undefined or retry_count == 'reset' else retry_count | int + 1 }}"
+        max_retries: 20
+
+    - name: Verify routers exist
+      command: >
+        {{ neutron_drain_venv }}/bin/openstack router list --agent {{ L3_SRC_ID }} -f value -c ID
+      environment: "{{ openstack_auth_env }}"
+      delegate_to: "{{ neutron_drain_host }}"
+      register: agent_status
+      vars:
+        ansible_host: "{{ hostvars[neutron_drain_host].ansible_host }}"
+
+    - name: Fail if routers still attached
+      fail:
+        msg: a Router is still attached to agent "{{ L3_SRC_ID }}"
+      when: agent_status.stdout | length > 0
+
+    - name: Reset retry count after success
+      set_fact:
+        retry_count: reset
+  rescue:
+    - fail:
+        msg: |
+          Maximum retries waiting for routers to be detached reached
+      when: retry_count | int == max_retries
+
+    - name: Reset retry counter if max retries reached (exit loop)
+      set_fact:
+        retry_count: reset
+      failed_when: retry_count == 'reset'
+      when: retry_count | int >= max_retries | int
+
+- name: Disable L3 agent
+  command: >
+    {{ neutron_drain_venv }}/bin/openstack
+    network agent set "{{ L3_SRC_ID }}" --disable
+  environment: "{{ openstack_auth_env }}"
+  delegate_to: "{{ neutron_drain_host }}"
+  when: agent_status.stdout | length == 0
+  vars:
+    ansible_host: "{{ hostvars[neutron_drain_host].ansible_host }}"