- nodeset: name: nested-virt-multinode nodes: - name: controller label: nested-virt-ubuntu-jammy # NOTE(artom) We can't name the node 'compute' because that seems to # take precedence over the 'compute' group in playbooks, so things we # want to run on all hosts in the 'compute' group would only run on the # subnode. - name: compute-host label: nested-virt-ubuntu-jammy groups: # Node where tests are executed and test results collected - name: tempest nodes: - controller # Nodes running the compute service - name: compute nodes: - controller - compute-host # Nodes that are not the controller - name: subnode nodes: - compute-host # Switch node for multinode networking setup - name: switch nodes: - controller # Peer nodes for multinode networking setup - name: peers nodes: - compute-host - job: name: whitebox-devstack-multinode parent: tempest-multinode-full-py3 nodeset: nested-virt-multinode description: | Runs the entire test suite on single-NUMA, non-SMT, nested virt VMs. required-projects: - openstack/whitebox-tempest-plugin - openstack/barbican - openstack/barbican-tempest-plugin pre-run: playbooks/whitebox/pre.yaml irrelevant-files: - ^test-requirements.txt$ vars: tox_envlist: all tempest_test_regex: ^whitebox_tempest_plugin\. # NOTE(jparker) in order for guest to boot via UEFI, the host will need the # open source implementation of UEFI for VMs via the OVMF package. In # addition to test vTPM hosts need swtpm as well extra_packages: ovmf,swtpm-tools tempest_exclude_regex: ^whitebox_tempest_plugin\.api\.compute\.test_hugepages compute_node_template_name: whitebox-devstack-multinode.yaml.j2 devstack_localrc: MAX_COMPUTE_NODES: 2 NOVA_SERVICE_REPORT_INTERVAL: 10 LIBVIRT_TYPE: kvm TEMPEST_PLUGINS: /opt/stack/whitebox-tempest-plugin /opt/stack/barbican-tempest-plugin WHITEBOX_PRIVKEY_PATH: /home/tempest/.ssh/id_rsa WHITEBOX_CPU_MODEL: Nehalem WHITEBOX_CPU_MODEL_EXTRA_FLAGS: vme,+ssse3,-mmx WHITEBOX_CPU_TOPOLOGY: "0: [0,1,2,3,4,5,6,7]" WHITEBOX_DEDICATED_CPUS_PER_NUMA: 3 devstack_local_conf: test-config: $TEMPEST_CONFIG: compute-feature-enabled: volume_backed_live_migration: true stable_compute_uuid_supported: true cpu_power_management: true auth: tempest_roles: creator post-config: $NOVA_CONF: key_manager: backend: barbican compute: cpu_dedicated_set: '1-3' cpu_shared_set: '4,5' max_disk_devices_to_attach: '7' libvirt: cpu_mode: custom cpu_models: Nehalem cpu_model_extra_flags: vme,+ssse3,-mmx cpu_power_management: True virt_type: kvm rx_queue_size: 1024 swtpm_enabled: True swtpm_user: swtpm swtpm_group: swtpm group-vars: subnode: devstack_localrc: LIBVIRT_TYPE: kvm NOVA_SERVICE_REPORT_INTERVAL: 10 devstack_local_conf: post-config: $NOVA_CONF: key_manager: backend: barbican compute: cpu_dedicated_set: '4-6' cpu_shared_set: '2,3' max_disk_devices_to_attach: '7' libvirt: cpu_mode: custom cpu_models: Nehalem cpu_model_extra_flags: vme,+ssse3,-mmx cpu_power_management: True virt_type: kvm rx_queue_size: 1024 swtpm_enabled: True swtpm_user: swtpm swtpm_group: swtpm tempest: devstack_plugins: barbican: https://opendev.org/openstack/barbican.git whitebox-tempest-plugin: https://opendev.org/openstack/whitebox-tempest-plugin.git - job: name: whitebox-devstack-multinode-hugepages parent: whitebox-devstack-multinode description: | Runs the hugepages tests on a deployment that has set up hugepages on the hosts. vars: tempest_test_regex: ^whitebox_tempest_plugin\.api\.compute\.test_hugepages # NOTE(artom) The parent job's exclude regex excludes the hugepages # tests, so we need to overwrite it here with a regex that matches # *nothing*. tempest_exclude_regex: $^ num_2M_pages: 512 num_1G_pages: 1 - job: name: whitebox-devstack-ceph-multinode parent: devstack-plugin-ceph-multinode-tempest-py3 description: | Runs test_rbd_direct_download test on single-NUMA, non-SMT, nested virt VMs with ceph plugin enabled. The job uses the default qcow2 file based imagebackend for ephemeral storage. It also enables the direct download of images via rbd into the local imagecache for Nova. voting: true required-projects: openstack/whitebox-tempest-plugin pre-run: playbooks/whitebox/pre.yaml irrelevant-files: - ^whitebox_tempest_plugin/api/compute/(?!test_rbd_direct_download).*py$ vars: tox_envlist: all compute_node_template_name: whitebox-devstack-ceph-multinode.yaml.j2 devstack_plugins: whitebox-tempest-plugin: https://opendev.org/openstack/whitebox-tempest-plugin.git tempest_test_regex: '^whitebox_tempest_plugin.api.compute.test_rbd_direct_download' devstack_localrc: COMPUTE_FEATURE_RBD_DOWNLOAD: True TEMPEST_PLUGINS: /opt/stack/whitebox-tempest-plugin WHITEBOX_PRIVKEY_PATH: /home/tempest/.ssh/id_rsa devstack_local_conf: post-config: $NOVA_CPU_CONF: glance: enable_rbd_download: True rbd_user: glance rbd_ceph_conf: /etc/ceph/ceph.conf rbd_pool: images libvirt: images_type: default group-vars: subnode: devstack_local_conf: post-config: $NOVA_CPU_CONF: glance: enable_rbd_download: True rbd_user: glance rbd_ceph_conf: /etc/ceph/ceph.conf rbd_pool: images libvirt: images_type: default - project: templates: - publish-to-pypi check: jobs: - whitebox-devstack-multinode - whitebox-devstack-ceph-multinode - openstack-tox-pep8 gate: jobs: - whitebox-devstack-multinode - whitebox-devstack-ceph-multinode - openstack-tox-pep8 experimental: jobs: - whitebox-devstack-multinode-hugepages