Red Hat Training

A Red Hat training course is available for Red Hat OpenStack Platform

Chapter 10. Example: Configuring OVS-DPDK and SR-IOV with VXLAN tunnelling

This section describes how to deploy Compute nodes with both OVS-DPDK and SR-IOV interfaces. The cluster will be installed with ML2/OVS and VXLAN tunnelling.

Important

You must determine the best values for the OVS-DPDK parameters that you set in the network-environment.yaml file to optimize your OpenStack network for OVS-DPDK. See Deriving DPDK parameters with workflows for details.

10.1. Configuring roles

Configure a custom role by copying and editing the default roles_data.yaml file found in /usr/share/openstack-tripleo-heat-templates.

For the purposes of this example, the ComputeOvsDpdkSriov role is created. For information on creating roles in Red Hat OpenStack Platform, see Advanced Overcloud Customization. For details on the specific role used for this example, see roles_data.yaml.

10.2. Configuring OVS-DPDK parameters

Important

You must determine the best values for the OVS-DPDK parameters that you set in the network-environment.yaml file to optimize your OpenStack network for OVS-DPDK. See Network Functions Virtualization Planning and Configuration for details.

  1. Add the custom resources for OVS-DPDK under resource_registry:

      resource_registry:
        # Specify the relative/absolute path to the config files you want to use for override the default.
        OS::TripleO::ComputeOvsDpdkSriov::Net::SoftwareConfig:
          nic-configs/computeovsdpdksriov.yaml
        OS::TripleO::Controller::Net::SoftwareConfig:
          nic-configs/controller.yaml
  2. Under parameter_defaults, set the tunnel type and network type to vxlan.:

    NeutronTunnelTypes: 'vxlan'
    NeutronNetworkType: 'vxlan'
  3. Under parameters_defaults, set the bridge mapping:

    # The OVS logical->physical bridge mappings to use.
    NeutronBridgeMappings:
      - dpdk-mgmt:br-link0
  4. Under parameter_defaults, set the role-specific parameters for the ComputeOvsDpdkSriov role:

      ##########################
      # OVS DPDK configuration #
      # ########################
      ComputeOvsDpdkSriovParameters:
        KernelArgs: "default_hugepagesz=1GB hugepagesz=1G hugepages=32 iommu=pt intel_iommu=on isolcpus=2-19,22-39"
        TunedProfileName: "cpu-partitioning"
        IsolCpusList: "2-19,22-39"
        NovaVcpuPinSet: ['4-19,24-39']
        NovaReservedHostMemory: 2048
        OvsDpdkSocketMemory: "3072,1024"
        OvsDpdkMemoryChannels: "4"
        OvsDpdkCoreList: "0,20,1,21"
        OvsPmdCoreList: "2,22,3,23"
        NovaLibvirtRxQueueSize: 1024
        NovaLibvirtTxQueueSize: 1024
    Note

    To prevent failures during guest creation, assign at least one CPU with sibling thread on each NUMA node. In the example, the values for the OvsPmdCoreList parameter denote cores 2 and 22 from NUMA 0, and cores 3 and 23 from NUMA 1.

    Note

    Huge pages are consumed by virtual machines, as well as OVS-DPDK using the OvsDpdkSocketMemory parameter. To calculate the number of huge pages available to the virtual machine, subtract the OvsDpdkSocketMemory value from the boot parameter value. You must also add hw:mem_page_size=1GB to the flavor you associate with the DPDK instance.

    Note

    OvsDPDKCoreList and OvsDpdkMemoryChannels are required settings for this procedure and must be set correctly to prevent failures.

  5. Configure the role-specific parameters for SR-IOV:

      ##########################
      #  SR-IOV configuration  #
      ##########################
      NeutronMechanismDrivers: ['openvswitch','sriovnicswitch']
      NovaSchedulerDefaultFilters: ['RetryFilter','AvailabilityZoneFilter','RamFilter','ComputeFilter','ComputeCapabilitiesFilter','ImagePropertiesFilter','ServerGroupAntiAffinityFilter','ServerGroupAffinityFilter','PciPassthroughFilter','NUMATopologyFilter']
      NovaSchedulerAvailableFilters: ["nova.scheduler.filters.all_filters","nova.scheduler.filters.pci_passthrough_filter.PciPassthroughFilter"]
      NovaPCIPassthrough:
        - vendor_id: "8086"
          product_id: "1528"
          address: "0000:06:00.0"
          trusted: "true"
          physical_network: "sriov-1"
        - vendor_id: "8086"
          product_id: "1528"
          address: "0000:06:00.1"
          trusted: "true"
          physical_network: "sriov-2"

10.3. Configuring the Controller node

  1. Create the control plane Linux bond for an isolated network.

      - type: linux_bond
        name: bond_api
        bonding_options: "mode=active-backup"
        use_dhcp: false
        dns_servers:
          get_param: DnsServers
        members:
        - type: interface
          name: nic2
          primary: true
  2. Assign VLANs to this Linux bond.

      - type: vlan
        vlan_id:
          get_param: InternalApiNetworkVlanID
        device: bond_api
        addresses:
        - ip_netmask:
            get_param: InternalApiIpSubnet
    
      - type: vlan
        vlan_id:
          get_param: StorageNetworkVlanID
        device: bond_api
        addresses:
        - ip_netmask:
            get_param: StorageIpSubnet
    
      - type: vlan
        vlan_id:
          get_param: StorageMgmtNetworkVlanID
        device: bond_api
        addresses:
        - ip_netmask:
            get_param: StorageMgmtIpSubnet
    
      - type: vlan
        vlan_id:
          get_param: ExternalNetworkVlanID
        device: bond_api
        addresses:
        - ip_netmask:
            get_param: ExternalIpSubnet
        routes:
        - default: true
          next_hop:
            get_param: ExternalInterfaceDefaultRoute
  3. Create the OVS bridge for access to neutron-dhcp-agent and neutron-metadata-agent services.

      - type: ovs_bridge
        name: br-link0
        use_dhcp: false
        mtu: 9000
        members:
        - type: interface
          name: nic3
          mtu: 9000
        - type: vlan
          vlan_id:
            get_param: TenantNetworkVlanID
          mtu: 9000
          addresses:
          - ip_netmask:
              get_param: TenantIpSubnet

10.4. Configuring the Compute node for DPDK and SR-IOV

Create the computeovsdpdksriov.yaml file from the default compute.yaml file and make the following changes:

  1. Create the control plane Linux bond for an isolated network.

      - type: linux_bond
        name: bond_api
        bonding_options: "mode=active-backup"
        use_dhcp: false
        dns_servers:
          get_param: DnsServers
        members:
        - type: interface
          name: nic3
          primary: true
        - type: interface
          name: nic4
  2. Assign VLANs to this Linux bond.

      - type: vlan
        vlan_id:
          get_param: InternalApiNetworkVlanID
        device: bond_api
        addresses:
        - ip_netmask:
            get_param: InternalApiIpSubnet
    
      - type: vlan
        vlan_id:
          get_param: StorageNetworkVlanID
        device: bond_api
        addresses:
        - ip_netmask:
            get_param: StorageIpSubnet
  3. Set a bridge with a DPDK port to link to the controller.

      - type: ovs_user_bridge
        name: br-link0
        use_dhcp: false
        ovs_extra:
          - str_replace:
              template: set port br-link0 tag=_VLAN_TAG_
              params:
                _VLAN_TAG_:
                   get_param: TenantNetworkVlanID
        addresses:
          - ip_netmask:
              get_param: TenantIpSubnet
        members:
          - type: ovs_dpdk_bond
            name: dpdkbond0
            mtu: 9000
            rx_queue: 2
            members:
              - type: ovs_dpdk_port
                name: dpdk0
                members:
                  - type: interface
                    name: nic7
              - type: ovs_dpdk_port
                name: dpdk1
                members:
                  - type: interface
                    name: nic8
    Note

    To include multiple DPDK devices, repeat the type code section for each DPDK device you want to add.

    Note

    When using OVS-DPDK, all bridges on the same Compute node should be of type ovs_user_bridge. The director may accept the configuration, but Red Hat OpenStack Platform does not support mixing ovs_bridge and ovs_user_bridge on the same node.

10.5. Deploying the overcloud

Run the overcloud_deploy.sh script to deploy the overcloud.

#!/bin/bash

openstack overcloud deploy \
--templates \
-r /home/stack/ospd-13-vxlan-dpdk-sriov-ctlplane-dataplane-bonding-hybrid/roles_data.yaml \
-e /usr/share/openstack-tripleo-heat-templates/environments/network-isolation.yaml \
-e /usr/share/openstack-tripleo-heat-templates/environments/host-config-and-reboot.yaml \
-e /usr/share/openstack-tripleo-heat-templates/environments/neutron-ovs-dpdk.yaml \
-e /usr/share/openstack-tripleo-heat-templates/environments/services/neutron-sriov.yaml \
-e /home/stack/ospd-13-vxlan-dpdk-sriov-ctlplane-dataplane-bonding-hybrid/ovs-dpdk-permissions.yaml \
-e /home/stack/ospd-13-vxlan-dpdk-sriov-ctlplane-dataplane-bonding-hybrid/overcloud_images.yaml \
-e /home/stack/ospd-13-vxlan-dpdk-sriov-ctlplane-dataplane-bonding-hybrid/network-environment.yaml \
--log-file overcloud_install.log &> overcloud_install.log