Merge stackhpc/2023.1 into stackhpc/2024.1

Author: markgoddard

.automation

@@ -137,6 +137,17 @@ jobs:
         run: ssh-keygen -f id_rsa -N ''
         working-directory: ${{ github.workspace }}/terraform/aio
 
+      # TODO: Remove the following step in Antelope.
+      # NOTE: In Ansible 2.10 and lower the synchronize module used in the
+      # ansible/diagnostics.yml playbook does not respect SSH connection
+      # variables. This may result in Permission Denied issues if using an SSH
+      # key that is not in ~/.ssh.
+      - name: Copy SSH keypair to .ssh/
+        run: |
+          install -d ~/.ssh -m 700 &&
+          cp id_rsa* ~/.ssh/
+        working-directory: ${{ github.workspace }}/terraform/aio
+
       - name: Generate clouds.yaml
         run: |
           cat << EOF > clouds.yaml

.github/workflows/stackhpc-all-in-one.yml

@@ -0,0 +1,71 @@
+---
+# This workflow provides a workflow_dispatch (manual) trigger to deploy a
+# multi-node test cluster.
+
+name: Multinode
+'on':
+  workflow_dispatch:
+    # NOTE: workflow_dispatch is limited to 10 inputs.
+    inputs:
+      multinode_name:
+        description: Multinode cluster name
+        type: string
+        required: true
+      os_distribution:
+        description: Host OS distribution
+        type: choice
+        default: rocky
+        options:
+          - rocky
+          - ubuntu
+      neutron_plugin:
+        description: Neutron ML2 plugin
+        type: choice
+        default: ovn
+        options:
+          - ovn
+          - ovs
+      upgrade:
+        description: Whether to perform an upgrade
+        type: boolean
+        default: false
+      break_on:
+        description: When to break execution for manual interaction
+        type: choice
+        default: never
+        options:
+          - always
+          - failure
+          - never
+          - success
+      break_duration:
+        description: How long to break execution for (minutes)
+        type: number
+        default: 60
+      ssh_key:
+        description: SSH public key to authorise on Ansible control host
+        type: string
+      terraform_kayobe_multinode_version:
+        description: terraform-kayobe-multinode version
+        type: string
+        default: main
+jobs:
+  multinode:
+    name: Multinode
+    uses: stackhpc/stackhpc-openstack-gh-workflows/.github/workflows/[email protected]
+    with:
+      multinode_name: ${{ inputs.multinode_name }}
+      os_distribution: ${{ inputs.os_distribution }}
+      os_release: ${{ inputs.os_distribution == 'rocky' && '9' || 'jammy' }}
+      ssh_username: ${{ inputs.os_distribution == 'rocky' && 'cloud-user' || 'ubuntu' }}
+      neutron_plugin: ${{ inputs.neutron_plugin }}
+      upgrade: ${{ inputs.upgrade }}
+      break_on: ${{ inputs.break_on }}
+      # Workaround loss of number type using fromJSON: https://github.com/orgs/community/discussions/67182
+      break_duration: ${{ fromJSON(inputs.break_duration) }}
+      ssh_key: ${{ inputs.ssh_key }}
+      stackhpc_kayobe_config_version: ${{ github.ref_name }}
+      # NOTE(upgrade): Reference the PREVIOUS release here.
+      stackhpc_kayobe_config_previous_version: stackhpc/zed
+      terraform_kayobe_multinode_version: ${{ inputs.terraform_kayobe_multinode_version }}
+    secrets: inherit

.github/workflows/stackhpc-multinode.yml

@@ -74,7 +74,7 @@ on the overcloud hosts:
 .. code-block:: console
 
     (kayobe) [stack@node ~]$ cd etc/kayobe
-    (kayobe) [stack@node kayobe]$ kayobe playbook run ansible/smartmontools.yml
+    (kayobe) [stack@node kayobe]$ kayobe playbook run ansible/smartmon-tools.yml
 
 SMART reporting should now be enabled along with a Prometheus alert for
 unhealthy disks and a Grafana dashboard called ``Hardware Overview``.

doc/source/configuration/monitoring.rst

@@ -21,9 +21,9 @@ node_types:
     volumes:
       # There is a minimum disk space capacity requirement of 4GiB when using Ironic Python Agent:
       # https://github.com/openstack/ironic-python-agent/blob/master/ironic_python_agent/utils.py#L290
-      - capacity: 10GiB
+      - capacity: 15GiB
       # Ceph volume
-      - capacity: 10GiB
+      - capacity: 20GiB
     physical_networks:
       - provision-net
       - cloud-net
@@ -34,7 +34,7 @@ node_types:
     volumes:
       # There is a minimum disk space capacity requirement of 4GiB when using Ironic Python Agent:
       # https://github.com/openstack/ironic-python-agent/blob/master/ironic_python_agent/utils.py#L290
-      - capacity: 10GiB
+      - capacity: 15GiB
     physical_networks:
       - provision-net
       - cloud-net

etc/kayobe/environments/aufn-ceph/tenks.yml

@@ -344,7 +344,7 @@ kolla_build_blocks:
   magnum_base_footer: |
     RUN curl https://raw.githubusercontent.com/helm/helm/main/scripts/get-helm-3 | head -n -1 | bash
     {% raw %}
-    {% set magnum_capi_packages = ['magnum-capi-helm==1.0.0'] %}
+    {% set magnum_capi_packages = ['magnum-capi-helm==1.1.0'] %}
     RUN {{ macros.install_pip(magnum_capi_packages | customizable("pip_packages")) }}
     {% endraw %}
   prometheus_alertmanager_repository_version: | # 2023.1 kolla has 0.24.0

etc/kayobe/kolla.yml

@@ -0,0 +1,6 @@
+---
+features:
+  - |
+    ``magnum-capi-helm`` driver has been updated to 1.1.0.
+    Please see `magnum-capi-helm release notes <https://docs.openstack.org/releasenotes/magnum-capi-helm/unreleased.html#relnotes-1-1-0>`_
+    for changes.