Appendix E: Container networking¶

#
# Overview
# ========
#
# This file contains the configuration for OpenStack Ansible Deployment
# (OSA) core services. Optional service configuration resides in the
# conf.d directory.
#
# You can customize the options in this file and copy it to
# /etc/openstack_deploy/openstack_user_config.yml or create a new
# file containing only necessary options for your environment
# before deployment.
#
# OSA implements PyYAML to parse YAML files and therefore supports structure
# and formatting options that augment traditional YAML. For example, aliases
# or references. For more information on PyYAML, see the documentation at
#
# http://pyyaml.org/wiki/PyYAMLDocumentation
#
# Configuration reference
# =======================
#
# Level: cidr_networks (required)
# Contains an arbitrary list of networks for the deployment. For each network,
# the inventory generator uses the IP address range to create a pool of IP
# addresses for network interfaces inside containers. A deployment requires
# at least one network for management.
#
#   Option: <value> (required, string)
#   Name of network and IP address range in CIDR notation. This IP address
#   range coincides with the IP address range of the bridge for this network
#   on the target host.
#
# Example:
#
# Define networks for a typical deployment.
#
#   - Management network on 172.29.236.0/22. Control plane for infrastructure
#     services, OpenStack APIs, and horizon.
#   - Tunnel network on 172.29.240.0/22. Data plane for project (tenant) VXLAN
#     networks.
#   - Storage network on 172.29.244.0/22. Data plane for storage services such
#     as cinder and swift.
#
# cidr_networks:
#   container: 172.29.236.0/22
#   tunnel: 172.29.240.0/22
#   storage: 172.29.244.0/22
#
# Example:
#
# Define additional service network on 172.29.248.0/22 for deployment in a
# Rackspace data center.
#
#   snet: 172.29.248.0/22
#
# --------
#
# Level: used_ips (optional)
# For each network in the 'cidr_networks' level, specify a list of IP addresses
# or a range of IP addresses that the inventory generator should exclude from
# the pools of IP addresses for network interfaces inside containers. To use a
# range, specify the lower and upper IP addresses (inclusive) with a comma
# separator.
#
# Example:
#
# The management network includes a router (gateway) on 172.29.236.1 and
# DNS servers on 172.29.236.11-12. The deployment includes seven target
# servers on 172.29.236.101-103, 172.29.236.111, 172.29.236.121, and
# 172.29.236.131. However, the inventory generator automatically excludes
# these IP addresses. The deployment host itself isn't automatically
# excluded. Network policy at this particular example organization
# also reserves 231-254 in the last octet at the high end of the range for
# network device management.
#
# used_ips:
#   - 172.29.236.1
#   - "172.29.236.100,172.29.236.200"
#   - "172.29.240.100,172.29.240.200"
#   - "172.29.244.100,172.29.244.200"
#
# --------
#
# Level: global_overrides (required)
# Contains global options that require customization for a deployment. For
# example, load balancer virtual IP addresses (VIP). This level also provides
# a mechanism to override other options defined in the playbook structure.
#
#   Option: internal_lb_vip_address (required, string)
#   Load balancer VIP for the following items:
#
#   - Local package repository
#   - Galera SQL database cluster
#   - Administrative and internal API endpoints for all OpenStack services
#   - Glance registry
#   - Nova compute source of images
#   - Cinder source of images
#   - Instance metadata
#
#   Option: external_lb_vip_address (required, string)
#   Load balancer VIP for the following items:
#
#   - Public API endpoints for all OpenStack services
#   - Horizon
#
#   Option: management_bridge (required, string)
#   Name of management network bridge on target hosts. Typically 'br-mgmt'.
#
#   Option: tunnel_bridge (optional, string)
#   Name of tunnel network bridge on target hosts. Typically 'br-vxlan'.
#
#   Level: provider_networks (required)
#   List of container and bare metal networks on target hosts.
#
#     Level: network (required)
#     Defines a container or bare metal network. Create a level for each
#     network.
#
#       Option: type (required, string)
#       Type of network. Networks other than those for neutron such as
#       management and storage typically use 'raw'. Neutron networks use
#       'flat', 'vlan', or 'vxlan'. Coincides with ML2 plug-in configuration
#       options.
#
#       Option: container_bridge (required, string)
#       Name of unique bridge on target hosts to use for this network. Typical
#       values include 'br-mgmt', 'br-storage', 'br-vlan', 'br-vxlan', etc.
#
#       Option: container_interface (required, string)
#       Name of unique interface in containers to use for this network.
#       Typical values include 'eth1', 'eth2', etc.
#       NOTE: Container interface is different from host interfaces.
#
#       Option: container_type (required, string)
#       Name of mechanism that connects interfaces in containers to the bridge
#       on target hosts for this network. Typically 'veth'.
#
#       Option: host_bind_override (optional, string)
#       Name of the physical network interface on the same L2 network being
#       used with the br-vlan device. This host_bind_override should only
#       be set for the ' container_bridge: "br-vlan" '.
#       This interface is optional but highly recommended for vlan based
#       OpenStack networking.
#       If no additional network interface is available, a deployer can create
#       a veth pair, and plug it into the the br-vlan bridge to provide
#       this interface. An example could be found in the aio_interfaces.cfg
#       file.
#
#       Option: container_mtu (optional, string)
#       Sets the MTU within LXC for a given network type.
#
#       Option: ip_from_q (optional, string)
#       Name of network in 'cidr_networks' level to use for IP address pool. Only
#       valid for 'raw' and 'vxlan' types.
#
#       Option: is_container_address (required, boolean)
#       If true, the load balancer uses this IP address to access services
#       in the container. Only valid for networks with 'ip_from_q' option.
#
#       Option: is_ssh_address (required, boolean)
#       If true, Ansible uses this IP address to access the container via SSH.
#       Only valid for networks with 'ip_from_q' option.
#
#       Option: group_binds (required, string)
#       List of one or more Ansible groups that contain this
#       network. For more information, see the env.d YAML files.
#
#       Option: net_name (optional, string)
#       Name of network for 'flat' or 'vlan' types. Only valid for these
#       types. Coincides with ML2 plug-in configuration options.
#
#       Option: range (optional, string)
#       For 'vxlan' type neutron networks, range of VXLAN network identifiers
#       (VNI). For 'vlan' type neutron networks, range of VLAN tags. Supports
#       more than one range of VLANs on a particular network. Coincides with
#       ML2 plug-in configuration options.
#
#       Option: static_routes (optional, list)
#       List of additional routes to give to the container interface.
#       Each item is composed of cidr and gateway. The items will be
#       translated into the container network interfaces configuration
#       as a `post-up ip route add <cidr> via <gateway> || true`.
#
#       Option: gateway (optional, string)
#       String containing the IP of the default gateway used by the
#       container. Generally not needed: the containers will have
#       their default gateway set with dnsmasq, poitining to the host
#       which does natting for container connectivity.
#
#   Example:
#
#   Define a typical network architecture:
#
#   - Network of type 'raw' that uses the 'br-mgmt' bridge and 'management'
#     IP address pool. Maps to the 'eth1' device in containers. Applies to all
#     containers and bare metal hosts. Both the load balancer and Ansible
#     use this network to access containers and services.
#   - Network of type 'raw' that uses the 'br-storage' bridge and 'storage'
#     IP address pool. Maps to the 'eth2' device in containers. Applies to
#     nova compute and all storage service containers. Optionally applies to
#     to the swift proxy service.
#   - Network of type 'vxlan' that contains neutron VXLAN tenant networks
#     1 to 1000 and uses 'br-vxlan' bridge on target hosts. Maps to the
#     'eth10' device in containers. Applies to all neutron agent containers
#     and neutron agents on bare metal hosts.
#   - Network of type 'vlan' that contains neutron VLAN networks 101 to 200
#     and 301 to 400 and uses the 'br-vlan' bridge on target hosts. Maps to
#     the 'eth11' device in containers. Applies to all neutron agent
#     containers and neutron agents on bare metal hosts.
#   - Network of type 'flat' that contains one neutron flat network and uses
#     the 'br-vlan' bridge on target hosts. Maps to the 'eth12' device in
#     containers. Applies to all neutron agent containers and neutron agents
#     on bare metal hosts.
#
#   Note: A pair of 'vlan' and 'flat' networks can use the same bridge because
#   one only handles tagged frames and the other only handles untagged frames
#   (the native VLAN in some parlance). However, additional 'vlan' or 'flat'
#   networks require additional bridges.
#
#   provider_networks:
#     - network:
#         group_binds:
#           - all_containers
#           - hosts
#         type: "raw"
#         container_bridge: "br-mgmt"
#         container_interface: "eth1"
#         container_type: "veth"
#         ip_from_q: "container"
#         is_container_address: true
#         is_ssh_address: true
#     - network:
#         group_binds:
#           - glance_api
#           - cinder_api
#           - cinder_volume
#           - nova_compute
#           # Uncomment the next line if using swift with a storage network.
#           # - swift_proxy
#         type: "raw"
#         container_bridge: "br-storage"
#         container_type: "veth"
#         container_interface: "eth2"
#         container_mtu: "9000"
#         ip_from_q: "storage"
#     - network:
#         group_binds:
#           - neutron_linuxbridge_agent
#         container_bridge: "br-vxlan"
#         container_type: "veth"
#         container_interface: "eth10"
#         container_mtu: "9000"
#         ip_from_q: "tunnel"
#         type: "vxlan"
#         range: "1:1000"
#         net_name: "vxlan"
#     - network:
#         group_binds:
#           - neutron_linuxbridge_agent
#         container_bridge: "br-vlan"
#         container_type: "veth"
#         container_interface: "eth11"
#         type: "vlan"
#         range: "101:200,301:400"
#         net_name: "vlan"
#     - network:
#         group_binds:
#           - neutron_linuxbridge_agent
#         container_bridge: "br-vlan"
#         container_type: "veth"
#         container_interface: "eth12"
#         host_bind_override: "eth12"
#         type: "flat"
#         net_name: "flat"
#
# --------
#
# Level: shared-infra_hosts (required)
# List of target hosts on which to deploy shared infrastructure services
# including the Galera SQL database cluster, RabbitMQ, and Memcached. Recommend
# three minimum target hosts for these services.
#
#   Level: <value> (required, string)
#   Hostname of a target host.
#
#     Option: ip (required, string)
#     IP address of this target host, typically the IP address assigned to
#     the management bridge.
#
# Example:
#
# Define three shared infrastructure hosts:
#
# shared-infra_hosts:
#   infra1:
#     ip: 172.29.236.101
#   infra2:
#     ip: 172.29.236.102
#   infra3:
#     ip: 172.29.236.103
#
# --------
#
# Level: repo-infra_hosts (required)
# List of target hosts on which to deploy the package repository. Recommend
# minimum three target hosts for this service. Typically contains the same
# target hosts as the 'shared-infra_hosts' level.
#
#   Level: <value> (required, string)
#   Hostname of a target host.
#
#     Option: ip (required, string)
#     IP address of this target host, typically the IP address assigned to
#     the management bridge.
#
# Example:
#
# Define three package repository hosts:
#
# repo-infra_hosts:
#   infra1:
#     ip: 172.29.236.101
#   infra2:
#     ip: 172.29.236.102
#   infra3:
#     ip: 172.29.236.103
#
# --------
#
# Level: os-infra_hosts (required)
# List of target hosts on which to deploy the glance API, nova API, heat API,
# and horizon. Recommend three minimum target hosts for these services.
# Typically contains the same target hosts as 'shared-infra_hosts' level.
#
#   Level: <value> (required, string)
#   Hostname of a target host.
#
#     Option: ip (required, string)
#     IP address of this target host, typically the IP address assigned to
#     the management bridge.
#
# Example:
#
# Define three OpenStack infrastructure hosts:
#
# os-infra_hosts:
#   infra1:
#     ip: 172.29.236.101
#   infra2:
#     ip: 172.29.236.102
#   infra3:
#     ip: 172.29.236.103
#
# --------
#
# Level: identity_hosts (required)
# List of target hosts on which to deploy the keystone service. Recommend
# three minimum target hosts for this service. Typically contains the same
# target hosts as the 'shared-infra_hosts' level.
#
#   Level: <value> (required, string)
#   Hostname of a target host.
#
#     Option: ip (required, string)
#     IP address of this target host, typically the IP address assigned to
#     the management bridge.
#
# Example:
#
# Define three OpenStack identity hosts:
#
# identity_hosts:
#   infra1:
#     ip: 172.29.236.101
#   infra2:
#     ip: 172.29.236.102
#   infra3:
#     ip: 172.29.236.103
#
# --------
#
# Level: network_hosts (required)
# List of target hosts on which to deploy neutron services. Recommend three
# minimum target hosts for this service. Typically contains the same target
# hosts as the 'shared-infra_hosts' level.
#
#   Level: <value> (required, string)
#   Hostname of a target host.
#
#     Option: ip (required, string)
#     IP address of this target host, typically the IP address assigned to
#     the management bridge.
#
# Example:
#
# Define three OpenStack network hosts:
#
# network_hosts:
#   infra1:
#     ip: 172.29.236.101
#   infra2:
#     ip: 172.29.236.102
#   infra3:
#     ip: 172.29.236.103
#
# --------
#
# Level: compute_hosts (optional)
# List of target hosts on which to deploy the nova compute service. Recommend
# one minimum target host for this service. Typically contains target hosts
# that do not reside in other levels.
#
#   Level: <value> (required, string)
#   Hostname of a target host.
#
#     Option: ip (required, string)
#     IP address of this target host, typically the IP address assigned to
#     the management bridge.
#
# Example:
#
# Define an OpenStack compute host:
#
# compute_hosts:
#   compute1:
#     ip: 172.29.236.121
#
# --------
#
# Level: ironic-compute_hosts (optional)
# List of target hosts on which to deploy the nova compute service for Ironic.
# Recommend one minimum target host for this service. Typically contains target
# hosts that do not reside in other levels.
#
#   Level: <value> (required, string)
#   Hostname of a target host.
#
#     Option: ip (required, string)
#     IP address of this target host, typically the IP address assigned to
#     the management bridge.
#
# Example:
#
# Define an OpenStack compute host:
#
# ironic-compute_hosts:
#   ironic-infra1:
#     ip: 172.29.236.121
#
# --------
#
# Level: storage-infra_hosts (required)
# List of target hosts on which to deploy the cinder API. Recommend three
# minimum target hosts for this service. Typically contains the same target
# hosts as the 'shared-infra_hosts' level.
#
#   Level: <value> (required, string)
#   Hostname of a target host.
#
#     Option: ip (required, string)
#     IP address of this target host, typically the IP address assigned to
#     the management bridge.
#
# Example:
#
# Define three OpenStack storage infrastructure hosts:
#
# storage-infra_hosts:
#   infra1:
#     ip: 172.29.236.101
#   infra2:
#     ip: 172.29.236.102
#   infra3:
#     ip: 172.29.236.103
#
# --------
#
# Level: storage_hosts (required)
# List of target hosts on which to deploy the cinder volume service. Recommend
# one minimum target host for this service. Typically contains target hosts
# that do not reside in other levels.
#
#   Level: <value> (required, string)
#   Hostname of a target host.
#
#     Option: ip (required, string)
#     IP address of this target host, typically the IP address assigned to
#     the management bridge.
#
#     Level: container_vars (required)
#     Contains storage options for this target host.
#
#       Option: cinder_storage_availability_zone (optional, string)
#       Cinder availability zone.
#
#       Option: cinder_default_availability_zone (optional, string)
#       If the deployment contains more than one cinder availability zone,
#       specify a default availability zone.
#
#       Level: cinder_backends (required)
#       Contains cinder backends.
#
#         Option: limit_container_types (optional, string)
#         Container name string in which to apply these options. Typically
#         any container with 'cinder_volume' in the name.
#
#         Level: <value> (required, string)
#         Arbitrary name of the backend. Each backend contains one or more
#         options for the particular backend driver. The template for the
#         cinder.conf file can generate configuration for any backend
#         providing that it includes the necessary driver options.
#
#           Option: volume_backend_name (required, string)
#           Name of backend, arbitrary.
#
#           The following options apply to the LVM backend driver:
#
#           Option: volume_driver (required, string)
#           Name of volume driver, typically
#           'cinder.volume.drivers.lvm.LVMVolumeDriver'.
#
#           Option: volume_group (required, string)
#           Name of LVM volume group, typically 'cinder-volumes'.
#
#           The following options apply to the NFS backend driver:
#
#           Option: volume_driver (required, string)
#           Name of volume driver,
#           'cinder.volume.drivers.nfs.NfsDriver'.
#           NB. When using NFS driver you may want to adjust your
#           env.d/cinder.yml file to run cinder-volumes in containers.
#
#           Option: nfs_shares_config (optional, string)
#           File containing list of NFS shares available to cinder, typically
#           '/etc/cinder/nfs_shares'.
#
#           Option: nfs_mount_point_base (optional, string)
#           Location in which to mount NFS shares, typically
#           '$state_path/mnt'.
#
#           Option: nfs_mount_options (optional, string)
#           Mount options used for the NFS mount points.
#
#           Option: shares (required)
#           List of shares to populate the 'nfs_shares_config' file. Each share
#           uses the following format:
#           - { ip: "{{ ip_nfs_server }}", share "/vol/cinder" }
#
#           The following options apply to the NetApp backend driver:
#
#           Option: volume_driver (required, string)
#           Name of volume driver,
#           'cinder.volume.drivers.netapp.common.NetAppDriver'.
#           NB. When using NetApp drivers you may want to adjust your
#           env.d/cinder.yml file to run cinder-volumes in containers.
#
#           Option: netapp_storage_family (required, string)
#           Access method, typically 'ontap_7mode' or 'ontap_cluster'.
#
#           Option: netapp_storage_protocol (required, string)
#           Transport method, typically 'scsi' or 'nfs'. NFS transport also
#           requires the 'nfs_shares_config' option.
#
#           Option: nfs_shares_config (required, string)
#           For NFS transport, name of the file containing shares. Typically
#           '/etc/cinder/nfs_shares'.
#
#           Option: netapp_server_hostname (required, string)
#           NetApp server hostname.
#
#           Option: netapp_server_port (required, integer)
#           NetApp server port, typically 80 or 443.
#
#           Option: netapp_login (required, string)
#           NetApp server username.
#
#           Option: netapp_password (required, string)
#           NetApp server password.
#
# Example:
#
# Define an OpenStack storage host:
#
# storage_hosts:
#   lvm-storage1:
#     ip: 172.29.236.131
#
# Example:
#
# Use the LVM iSCSI backend in availability zone 'cinderAZ_1':
#
#     container_vars:
#       cinder_storage_availability_zone: cinderAZ_1
#       cinder_default_availability_zone: cinderAZ_1
#       cinder_backends:
#         lvm:
#           volume_backend_name: LVM_iSCSI
#           volume_driver: cinder.volume.drivers.lvm.LVMVolumeDriver
#           volume_group: cinder-volumes
#           iscsi_ip_address: "{{ cinder_storage_address }}"
#         limit_container_types: cinder_volume
#
# Example:
#
# Use the NetApp iSCSI backend via Data ONTAP 7-mode in availability zone
# 'cinderAZ_2':
#
#     container_vars:
#       cinder_storage_availability_zone: cinderAZ_2
#       cinder_default_availability_zone: cinderAZ_1
#       cinder_backends:
#         netapp:
#           volume_backend_name: NETAPP_iSCSI
#           volume_driver: cinder.volume.drivers.netapp.common.NetAppDriver
#           netapp_storage_family: ontap_7mode
#           netapp_storage_protocol: iscsi
#           netapp_server_hostname: hostname
#           netapp_server_port: 443
#           netapp_login: username
#           netapp_password: password
#
#
# Example:
#
# Use the ceph RBD backend in availability zone 'cinderAZ_3':
#
#     container_vars:
#       cinder_storage_availability_zone: cinderAZ_3
#       cinder_default_availability_zone: cinderAZ_1
#       cinder_backends:
#         limit_container_types: cinder_volume
#         volumes_hdd:
#           volume_driver: cinder.volume.drivers.rbd.RBDDriver
#           rbd_pool: volumes_hdd
#           rbd_ceph_conf: /etc/ceph/ceph.conf
#           rbd_flatten_volume_from_snapshot: 'false'
#           rbd_max_clone_depth: 5
#           rbd_store_chunk_size: 4
#           rados_connect_timeout: -1
#           volume_backend_name: volumes_hdd
#           rbd_user: "{{ cinder_ceph_client }}"
#           rbd_secret_uuid: "{{ cinder_ceph_client_uuid }}"
#
#
# --------
#
# Level: log_hosts (required)
# List of target hosts on which to deploy logging services. Recommend
# one minimum target host for this service.
#
#   Level: <value> (required, string)
#   Hostname of a target host.
#
#     Option: ip (required, string)
#     IP address of this target host, typically the IP address assigned to
#     the management bridge.
#
# Example:
#
# Define a logging host:
#
# log_hosts:
#   log1:
#     ip: 172.29.236.171
#
# --------
#
# Level: haproxy_hosts (optional)
# List of target hosts on which to deploy HAProxy. Recommend at least one
# target host for this service if hardware load balancers are not being
# used.
#
#   Level: <value> (required, string)
#   Hostname of a target host.
#
#     Option: ip (required, string)
#     IP address of this target host, typically the IP address assigned to
#     the management bridge.
#
#
# Example:
#
# Define a virtual load balancer (HAProxy):
#
# While HAProxy can be used as a virtual load balancer, it is recommended to use
# a physical load balancer in a production environment.
#
# haproxy_hosts:
#   lb1:
#     ip: 172.29.236.100
#   lb2:
#     ip: 172.29.236.101
#
# In case of the above scenario(multiple hosts),HAProxy can be deployed in a
# highly-available manner by installing keepalived.
#
# To make keepalived work, edit at least the following variables
# in ``user_variables.yml``:
#  haproxy_keepalived_external_vip_cidr: 192.168.0.4/25
#  haproxy_keepalived_internal_vip_cidr: 172.29.236.54/16
#  haproxy_keepalived_external_interface: br-flat
#  haproxy_keepalived_internal_interface: br-mgmt
#
# To always deploy (or upgrade to) the latest stable version of keepalived.
# Edit the ``/etc/openstack_deploy/user_variables.yml``:
#   keepalived_use_latest_stable: True
#
# The HAProxy playbook reads the ``vars/configs/keepalived_haproxy.yml``
# variable file and provides content to the keepalived role for
# keepalived master and backup nodes.
#
# Keepalived pings a public IP address to check its status. The default
# address is ``193.0.14.129``. To change this default,
# set the ``keepalived_ping_address`` variable in the
# ``user_variables.yml`` file.
#
# You can define additional variables to adapt keepalived to your
# deployment. Refer to the ``user_variables.yml`` file for
# more information. Optionally, you can use your own variable file.
# For example:
#   haproxy_keepalived_vars_file: /path/to/myvariablefile.yml
#