SUR CETTE PAGE
Annexe A : Exemples de fichiers de configuration
Note:
AppFormix a été renommé Contrail Insights, mais il est toujours appelé appformix dans les fichiers de configuration YAML.
Exemple de fichier de configuration site.yml
# Copyright 2023 Juniper Networks, Inc. All rights reserved.
# Licensed under the Juniper Networks Script Software License (the "License").
# You may not use this script file except in compliance with the License, which is located at
# http://www.juniper.net/support/legal/scriptlicense/
# Unless required by applicable law or otherwise agreed to in writing by the parties,
# software distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
#
#
#
global:
# List of DNS nameservers
dns:
# Google Public DNS
- "8.8.8.8"
- "8.8.4.4"
# List of NTP time servers
ntp:
# public pool.ntp.org
- "0.pool.ntp.org"
- "1.pool.ntp.org"
- "2.pool.ntp.org"
- "3.pool.ntp.org"
# Timezone for all servers
timezone: 'America/Los_Angeles'
rhel:
# Contrail Cloud Activation Key
# These details are provided when you request an activation key from
# contrail cloud subscriptions <contrail_cloud_subscriptions@juniper.net>
#
satellite:
#SATELLITE_KEY should be defined in vault-data.yml file
#SATELLITE_ORG
organization: "ContrailCloud16"
#SATELLITE_FQDN
fqdn: contrail-cloud.juniper.net
# DNS domain information.
# Must be unique for every deployment to avoid name conflicts.
# Need not be a registered DNS domain.
domain: "my.unique.domain"
jumphost:
network:
# network used for provisioning (PXE booting) servers
provision:
# jumphost nic to be used for provisioning (PXE booting) servers
nic: eno1
control_hosts:
# Contains a list of label to disk mappings for roles
disk_mapping:
# the control host always uses the "baremetal" role
baremetal:
# Mapping of labels to disk devices. The label is assigned to the disk
# device so that the disk can be referenced by the alias in other
# configurations. for example /dev/disk/by-alias/<label>
# Each list element contains:
# label: label to assign
# name: disk device path (e.g. /dev/sdb)
# OR
# hctl: alternative notation for disk paths specifying SCSI address
# (Host, Channel, Target and Lun) The HCTL can be found with the
# lsscsi (or lspci) command or it can be found in introspection data
#
- label: spinning-0
name: /dev/sdb
- label: spinning-1
name: /dev/sdc
- label: spinning-2
name: /dev/sdd
- label: spinning-3
name: /dev/sde
- label: ssd-0
hctl: "0:2:3:0"
storage:
# Define a set of disk groups that can be referenced for VM virtual disk allocations
# These become virsh storage pools on the control host
# Each pool has:
# mountpoint: "/absolute/path/where/lvm/will/get/mounted"
# type: Either "dir" or "logical".
# "dir" does not create any new volumes
# it is useful if one large hardware raid is used as /
# "logical" is a LVM volume placed on the list of "disk".
# disk: List of disk devices to use for the pool
hdd_storage:
mountpoint: "/srv/hdd_storage"
type: logical
disk:
- "/dev/disk/by-alias/spinning-0"
- "/dev/disk/by-alias/spinning-1"
- "/dev/disk/by-alias/spinning-2"
- "/dev/disk/by-alias/spinning-3"
ssd_storage:
mountpoint: "/srv/ssd_storage"
type: logical
disk:
- "/dev/disk/by-alias/ssd-0"
#srv:
# mountpoint: "/srv"
# type: dir
vm:
# VM for Openstack Controller role
control:
disk:
# Root disk
vda:
# Virsh storage pool (see storage above)
pool: hdd_storage
# VMs for ContrailController role
contrail-k8s:
disk:
# Root disk
vda:
# Virsh storage pool (see storage above)
pool: hdd_storage
# VM for ContrailTsn role
contrail-tsn:
disk:
# Root disk
vda:
# Virsh storage pool (see storage above)
pool: hdd_storage
# VM for AppFormix controller role
appformix-controller:
disk:
# Root disk
vda:
# Virsh storage pool (see storage above)
pool: hdd_storage
compute_hosts:
sriov:
#enable sriov support
enabled: true
#enable sriov with dpdk
# Contrail vrouter mode:
# supported values are: dpdk or anything else means kernel vRouter
mode: dpdk
#Sriov NumVFs separated by comma
num_vf:
- "ens2f1:7"
#NovaPCIPassthrough settings
pci_passthrough:
- devname: "ens2f1"
physical_network: "sriov1"
root_disk:
# Define root disk for the listed ironic profiles.
# The default of "/dev/sda" will be used if there is no
# specific profile definition
#
# In case 'name' hint should be dropped in favor of other
# hints the block scalar can be used:
#
# ComputeKernel0Hw0: |
# vendor: VendorName
#
# which will overwrite default values.
# To keep key-value structure and not use 'name' hint, it can be set to the
# value which always evaluate to True, for example:
#
# ComputeKernel0Hw0:
# name: "s!=NonExistingDevice"
# vendor: VendorName
#
# For more details please check:
# https://docs.openstack.org/ironic/latest/install/advanced.html#specifying-the-disk-for-deployment-root-device-hints
#
ComputeKernel0Hw0:
name: "/dev/sda"
ComputeKernel0Hw1:
name: "/dev/sda"
ComputeKernel1Hw1:
name: "/dev/sda"
ComputeKernel1Hw0:
name: "/dev/sda"
ComputeDpdk0Hw2:
name: "/dev/sda"
ComputeDpdk1Hw3:
name: "/dev/sda"
ComputeSriov0Hw4:
name: "/dev/sda"
ComputeSriov1Hw5:
name: "/dev/sda"
resource:
minimal_disk:
# This value will be used as the local_gb size for the listed ironic profiles
# If not defined for a profile then the default will be used
ComputeKernel0Hw0: 50
ComputeKernel0Hw1: 50
ComputeKernel1Hw1: 50
ComputeKernel1Hw0: 50
ComputeDpdk0Hw2: 50
ComputeDpdk1Hw3: 50
ComputeSriov0Hw4: 50
ComputeSriov1Hw5: 50
storage_hosts:
root_disk:
# Define root disk for the listed ironic profiles.
# The default of "/dev/sda" will be used if there is no
# specific profile definition
CephStorage0Hw6:
name: "/dev/sda"
CephStorage1Hw7:
name: "/dev/sda"
undercloud:
nova:
# Nova flavor definitions for roles
flavor:
CephStorage0Hw6:
cpu: 1
memory: 4
disk: 40
ephemeral: 0
CephStorage1Hw7:
cpu: 1
memory: 4
disk: 40
ephemeral: 0
ComputeKernel0Hw0:
cpu: 8
memory: 24
disk: 40
ephemeral: 0
ComputeKernel0Hw1:
cpu: 8
memory: 24
disk: 40
ephemeral: 0
ComputeKernel1Hw1:
cpu: 8
memory: 24
disk: 40
ephemeral: 0
ComputeKernel1Hw0:
cpu: 8
memory: 24
disk: 40
ephemeral: 0
ComputeDpdk0Hw2:
cpu: 8
memory: 24
disk: 40
ephemeral: 0
ComputeDpdk1Hw3:
cpu: 8
memory: 24
disk: 40
ephemeral: 0
ComputeSriov0Hw4:
cpu: 8
memory: 24
disk: 40
ephemeral: 0
ComputeSriov1Hw5:
cpu: 8
memory: 24
disk: 40
ephemeral: 0
k8s:
external_vip_ip: 10.10.10.102
internal_api_vip_ip: 172.16.0.91
overcloud:
# Contains a list of label to disk mappings for roles.
# When Ceph Storage is disabled, compute-related roles (Compute* and
# ComputeDpdk* roles) will use any disks labeled with
# "ephemeral-<digits>" for local Nova ephemeral storage.
disk_mapping:
ComputeKernel:
# Mapping of labels to disk devices. The label is assigned to the disk
# device so that the disk can be referenced by the alias in other
# configurations. for example /dev/disk/by-alias/<label>
# Each list element contains:
# label: label to assign
# hctl: disk device path H:C:T:L (the path must exist). see lsscsi
- label: ephemeral-0
hctl: '5:0:0:0'
- label: ephemeral-1
hctl: '6:0:0:0'
- label: ephemeral-2
hctl: '7:0:0:0'
- label: ephemeral-3
hctl: '8:0:0:0'
ComputeKernel0Hw0:
# Mapping of labels to disk devices. The label is assigned to the disk
# device so that the disk can be referenced by the alias in other
# configurations. for example /dev/disk/by-alias/<label>
# Each list element contains:
# label: label to assign
# hctl: disk device path H:C:T:L (the path must exist). see lsscsi
- label: ephemeral-0
hctl: '5:0:0:0'
- label: ephemeral-1
hctl: '6:0:0:0'
- label: ephemeral-2
hctl: '7:0:0:0'
- label: ephemeral-3
hctl: '8:0:0:0'
ComputeKernel1Hw0:
# Mapping of labels to disk devices. The label is assigned to the disk
# device so that the disk can be referenced by the alias in other
# configurations. for example /dev/disk/by-alias/<label>
# Each list element contains:
# label: label to assign
# hctl: disk device path H:C:T:L (the path must exist). see lsscsi
- label: ephemeral-0
hctl: '5:0:0:0'
- label: ephemeral-1
hctl: '6:0:0:0'
- label: ephemeral-2
hctl: '7:0:0:0'
- label: ephemeral-3
hctl: '8:0:0:0'
ComputeKernel1Hw1:
# Mapping of labels to disk devices. The label is assigned to the disk
# device so that the disk can be referenced by the alias in other
# configurations. for example /dev/disk/by-alias/<label>
# Each list element contains:
# label: label to assign
# hctl: disk device path H:C:T:L (the path must exist). see lsscsi
- label: ephemeral-0
hctl: '5:0:0:0'
- label: ephemeral-1
hctl: '6:0:0:0'
- label: ephemeral-2
hctl: '7:0:0:0'
- label: ephemeral-3
hctl: '8:0:0:0'
ComputeKernel0Hw1:
# Mapping of labels to disk devices. The label is assigned to the disk
# device so that the disk can be referenced by the alias in other
# configurations. for example /dev/disk/by-alias/<label>
# Each list element contains:
# label: label to assign
# hctl: disk device path H:C:T:L (the path must exist). see lsscsi
- label: ephemeral-0
hctl: '5:0:0:0'
- label: ephemeral-1
hctl: '6:0:0:0'
- label: ephemeral-2
hctl: '7:0:0:0'
- label: ephemeral-3
hctl: '8:0:0:0'
ComputeDpdk:
# Mapping of labels to disk devices. The label is assigned to the disk
# device so that the disk can be referenced by the alias in other
# configurations. for example /dev/disk/by-alias/<label>
# Each list element contains:
# label: label to assign
# hctl: disk device path H:C:T:L (the path must exist). see lsscsi
- label: ephemeral-0
hctl: '5:0:0:0'
- label: ephemeral-1
hctl: '6:0:0:0'
- label: ephemeral-2
hctl: '7:0:0:0'
- label: ephemeral-3
hctl: '8:0:0:0'
ComputeDpdk0Hw2:
# Mapping of labels to disk devices. The label is assigned to the disk
# device so that the disk can be referenced by the alias in other
# configurations. for example /dev/disk/by-alias/<label>
# Each list element contains:
# label: label to assign
# hctl: disk device path H:C:T:L (the path must exist). see lsscsi
- label: ephemeral-0
hctl: '5:0:0:0'
- label: ephemeral-1
hctl: '6:0:0:0'
- label: ephemeral-2
hctl: '7:0:0:0'
- label: ephemeral-3
hctl: '8:0:0:0'
ComputeDpdk1Hw3:
# Mapping of labels to disk devices. The label is assigned to the disk
# device so that the disk can be referenced by the alias in other
# configurations. for example /dev/disk/by-alias/<label>
# Each list element contains:
# label: label to assign
# hctl: disk device path H:C:T:L (the path must exist). see lsscsi
- label: ephemeral-0
hctl: '5:0:0:0'
- label: ephemeral-1
hctl: '6:0:0:0'
- label: ephemeral-2
hctl: '7:0:0:0'
- label: ephemeral-3
hctl: '8:0:0:0'
ComputeSriov:
# Mapping of labels to disk devices. The label is assigned to the disk
# device so that the disk can be referenced by the alias in other
# configurations. for example /dev/disk/by-alias/<label>
# Each list element contains:
# label: label to assign
# hctl: disk device path H:C:T:L (the path must exist). see lsscsi
- label: ephemeral-0
hctl: '5:0:0:0'
- label: ephemeral-1
hctl: '6:0:0:0'
- label: ephemeral-2
hctl: '7:0:0:0'
- label: ephemeral-3
hctl: '8:0:0:0'
ComputeSriov0Hw4:
# Mapping of labels to disk devices. The label is assigned to the disk
# device so that the disk can be referenced by the alias in other
# configurations. for example /dev/disk/by-alias/<label>
# Each list element contains:
# label: label to assign
# hctl: disk device path H:C:T:L (the path must exist). see lsscsi
- label: ephemeral-0
hctl: '5:0:0:0'
- label: ephemeral-1
hctl: '6:0:0:0'
- label: ephemeral-2
hctl: '7:0:0:0'
- label: ephemeral-3
hctl: '8:0:0:0'
ComputeSriov1Hw5:
# Mapping of labels to disk devices. The label is assigned to the disk
# device so that the disk can be referenced by the alias in other
# configurations. for example /dev/disk/by-alias/<label>
# Each list element contains:
# label: label to assign
# hctl: disk device path H:C:T:L (the path must exist). see lsscsi
- label: ephemeral-0
hctl: '5:0:0:0'
- label: ephemeral-1
hctl: '6:0:0:0'
- label: ephemeral-2
hctl: '7:0:0:0'
- label: ephemeral-3
hctl: '8:0:0:0'
extra_config:
ComputeDpdkParameters:
TunedProfileName: "cpu-partitioning"
ContrailDpdkOptions: "--vr_flow_entries=2000000 --yield_option 0"
KernelArgs: "intel_iommu=on iommu=pt default_hugepagesz=1GB hugepagesz=1G hugepages=64 hugepagesz=2M hugepages=2048 isolcpus=2-9,22-29"
IsolCpusList: "2-9,22-29"
NovaVcpuPinSet: ['4-9','24-29']
ContrailSettings:
SERVICE_CORE_MASK: '0x1'
DPDK_CTRL_THREAD_MASK: '0x1'
ExtraSysctlSettings:
vm.nr_hugepages:
value: 64
vm.max_map_count:
value: 128960
ComputeDpdk0Hw2Parameters:
TunedProfileName: "cpu-partitioning"
ContrailDpdkOptions: "--vr_flow_entries=2000000 --yield_option 0"
KernelArgs: "intel_iommu=on iommu=pt default_hugepagesz=1GB hugepagesz=1G hugepages=64 hugepagesz=2M hugepages=2048 isolcpus=2-9,22-29"
IsolCpusList: "2-9,22-29"
NovaVcpuPinSet: ['4-9','24-29']
ContrailSettings:
SERVICE_CORE_MASK: '0x1'
DPDK_CTRL_THREAD_MASK: '0x1'
ExtraSysctlSettings:
vm.nr_hugepages:
value: 64
vm.max_map_count:
value: 128960
ComputeDpdk1Hw3Parameters:
TunedProfileName: "cpu-partitioning"
ContrailDpdkOptions: "--vr_flow_entries=2000000 --yield_option 0"
KernelArgs: "intel_iommu=on iommu=pt default_hugepagesz=1GB hugepagesz=1G hugepages=64 hugepagesz=2M hugepages=2048 isolcpus=2-9,22-29"
IsolCpusList: "2-9,22-29"
NovaVcpuPinSet: ['4-9','24-29']
ContrailSettings:
SERVICE_CORE_MASK: '0x1'
DPDK_CTRL_THREAD_MASK: '0x1'
ExtraSysctlSettings:
vm.nr_hugepages:
value: 64
vm.max_map_count:
value: 128960
ComputeKernelParameters:
ContrailVrouterHugepages1GB: 64
ContrailVrouterHugepages2MB: 2048
KernelArgs: "intel_iommu=on iommu=pt default_hugepagesz=1GB hugepagesz=1G hugepages=64 hugepagesz=2M hugepages=2048"
ExtraSysctlSettings:
vm.nr_hugepages:
value: 64
vm.max_map_count:
value: 128960
ComputeKernel0Hw0Parameters:
ContrailVrouterHugepages1GB: 64
ContrailVrouterHugepages2MB: 2048
KernelArgs: "intel_iommu=on iommu=pt default_hugepagesz=1GB hugepagesz=1G hugepages=64 hugepagesz=2M hugepages=2048"
ExtraSysctlSettings:
vm.nr_hugepages:
value: 64
vm.max_map_count:
value: 128960
ComputeKernel0Hw1Parameters:
ContrailVrouterHugepages1GB: 64
ContrailVrouterHugepages2MB: 2048
KernelArgs: "intel_iommu=on iommu=pt default_hugepagesz=1GB hugepagesz=1G hugepages=64 hugepagesz=2M hugepages=2048"
ExtraSysctlSettings:
vm.nr_hugepages:
value: 64
vm.max_map_count:
value: 128960
ComputeKernel1Hw0Parameters:
ContrailVrouterHugepages1GB: 64
ContrailVrouterHugepages2MB: 2048
KernelArgs: "intel_iommu=on iommu=pt default_hugepagesz=1GB hugepagesz=1G hugepages=64 hugepagesz=2M hugepages=2048"
ExtraSysctlSettings:
vm.nr_hugepages:
value: 64
vm.max_map_count:
value: 128960
ComputeKernel1Hw1Parameters:
ContrailVrouterHugepages1GB: 64
ContrailVrouterHugepages2MB: 2048
KernelArgs: "intel_iommu=on iommu=pt default_hugepagesz=1GB hugepagesz=1G hugepages=64 hugepagesz=2M hugepages=2048"
ExtraSysctlSettings:
vm.nr_hugepages:
value: 64
vm.max_map_count:
value: 128960
network:
# The external network is used for referencing the overcloud APIs from outside the infrastructure.
external:
# Network name used by TripleO Heat Templates
heat_name: External
# CIDR (IP/prefix) for the external network subnet
# Corresponds to the ExternalIpSubnet heat property
cidr: "10.84.36.64/28"
# Default route for the external network
# Corresponds to the ExternalInterfaceDefaultRoute heat property
gateway: "10.84.36.78"
# VLAN tag for the external network
# Corresponds to the ExternalNetworkVlanID heat property
vlan: 1350
# Floating virtual IP for the Openstack APIs on the external network
# Corresponds to the PublicVirtualFixedIPs heat property
vip: "10.84.36.77"
# DHCP pool for the external network
# Be sure that the range is large enough to accomodate all nodes in the external network
pool:
# Range start for the DHCP pool
start: "10.84.36.65"
# Range end for the DHCP pool
end: "10.84.36.75"
# MTU for external network
# Corresponds to the ExternalNetworkMtu heat property
mtu: 9000
# List of roles that can be on this network
role:
- Controller
- AppformixController
# The internal API network is used for control plane signalling and service API calls
internal_api:
# Network name used by TripleO Heat Templates
heat_name: InternalApi
# VLAN tag for the internal API network
# Corresponds to the InternalApiNetworkVlanID heat property
vlan: 1200
# CIDR (IP/prefix) for the internal api supernet network subnet
# Corresponds to the InternalApiSupernet heat property
# Supernet is used in spine/leaf configuration
# Supernet accommodate all related leaf networks, e.g. internal_api0 and internal_api1
# Supernet is used to create static routes between leafs
# Supernet is defined only for main network, not per leafs
supernet: "172.16.0.0/22"
# CIDR (IP/prefix) for the internal api network subnet
# Corresponds to the InternalApiIpSubnet heat property
cidr: "172.16.0.0/24"
# Default route for the internal api network
# Corresponds to the InternalApiInterfaceDefaultRoute heat property
gateway: 172.16.0.1
# MTU for internal api network
# Corresponds to the InternalApiNetworkMtu heat property
mtu: 9000
# DHCP pool for the internal api network
# Be sure that the range is large enough to accomodate all nodes in the internal api network
pool:
# Range start for the DHCP pool
start: 172.16.0.100
# Range end for the DHCP pool
end: 172.16.0.160
# Floating virtual IP for the Openstack APIs on the internal api network
# Corresponds to the InternalApiVirtualFixedIPs heat property
vip: 172.16.0.90
# List of roles that can be on this network
role:
- Controller
- ContrailTsn
- AppformixController
- ComputeKernel
- ComputeDpdk
- ComputeSriov
# Leaf 0 subnet of the internal_api network
internal_api0:
# Network name used by TripleO Heat Templates
heat_name: InternalApi0
# VLAN tag for the internal API 0 network
# Corresponds to the InternalApi0NetworkVlanID heat property
vlan: 1201
# CIDR (IP/prefix) for the internal api 0 network subnet
# Corresponds to the InternalApi0IpSubnet heat property
cidr: "172.16.1.0/24"
# Default route for the internal api 0 network
# Corresponds to the InternalApi0InterfaceDefaultRoute heat property
gateway: 172.16.1.1
# MTU for internal api 0 network
# Corresponds to the InternalApi0NetworkMtu heat property
mtu: 9000
# DHCP pool for the internal api 0 network
# Be sure that the range is large enough to accomodate all nodes in the internal api network
pool:
# Range start for the DHCP pool
start: 172.16.1.100
# Range end for the DHCP pool
end: 172.16.1.200
# List of roles that can be on this network
role:
- ComputeKernel0Hw0
- ComputeKernel0Hw1
- ComputeSriov0Hw4
- ComputeDpdk0Hw2
# Leaf 1 subnet of the internal_api network
internal_api1:
# Network name used by TripleO Heat Templates
heat_name: InternalApi1
# VLAN tag for the internal API 1 network
# Corresponds to the InternalApi1NetworkVlanID heat property
vlan: 1202
# CIDR (IP/prefix) for the internal api 1 network subnet
# Corresponds to the InternalApi1IpSubnet heat property
cidr: "172.16.2.0/24"
# Default route for the internal api 1 network
# Corresponds to the InternalApi1InterfaceDefaultRoute heat property
gateway: 172.16.2.1
# MTU for internal api 1 network
# Corresponds to the InternalApi1NetworkMtu heat property
mtu: 9000
# DHCP pool for the internal api 1 network
# Be sure that the range is large enough to accomodate all nodes in the internal api network
pool:
# Range start for the DHCP pool
start: 172.16.2.100
# Range end for the DHCP pool
end: 172.16.2.200
# List of roles that can be on this network
role:
- ComputeSriov1Hw5
- ComputeKernel1Hw0
- ComputeDpdk1Hw3
- ComputeKernel1Hw1
# The management network is defined for backwards-compatibility in RHOSP and is not
# used by default by any roles.
management:
# Network name used by TripleO Heat Templates
heat_name: Management
# VLAN tag for the management network
# Corresponds to the ManagementNetworkVlanID heat property
vlan: 1300
# CIDR (IP/prefix) for the network subnet
# Corresponds to the ManagementIpSubnet heat property
cidr: "192.168.1.0/24"
# MTU for the network
# Corresponds to the ManagementNetworkMtu heat property
mtu: 9000
# DHCP pool for the network
# Be sure that the range is large enough to accomodate all nodes in the network
pool:
# Range start for the DHCP pool
start: 192.168.1.100
# Range end for the DHCP pool
end: 192.168.1.200
# The storage network is used for Compute storage access
storage:
# Network name used by TripleO Heat Templates
heat_name: Storage
# VLAN tag for the storage network
# Corresponds to the StorageNetworkVlanID heat property
vlan: 1500
supernet: "172.16.16.0/22"
cidr: "172.16.16.0/24"
gateway: 172.16.16.1
mtu: 9000
pool:
start: 172.16.16.100
end: 172.16.16.200
# List of roles that can be on this network
role:
- Controller
- CephStorage
- ComputeKernel
- ComputeDpdk
- ComputeSriov
- ContrailTsn
# Leaf 0 subnet of the storage network
storage0:
# Network name used by TripleO Heat Templates
heat_name: Storage0
vlan: 1501
cidr: "172.16.17.0/24"
gateway: 172.16.17.1
mtu: 9000
pool:
start: 172.16.17.100
end: 172.16.17.200
# List of roles that can be on this network
role:
- CephStorage0Hw6
- ComputeKernel0Hw0
- ComputeKernel0Hw1
- ComputeDpdk0Hw2
- ComputeSriov0Hw4
# Leaf 1 subnet of the storage network
storage1:
# Network name used by TripleO Heat Templates
heat_name: Storage1
vlan: 1502
cidr: "172.16.18.0/24"
gateway: 172.16.18.1
mtu: 9000
pool:
start: 172.16.18.100
end: 172.16.18.200
# List of roles that can be on this network
role:
- ComputeSriov1Hw5
- ComputeKernel1Hw0
- ComputeDpdk1Hw3
- ComputeKernel1Hw1
- CephStorage1Hw7
# The storage management network is used for storage operations such as replication
storage_mgmt:
# Network name used by TripleO Heat Templates
heat_name: StorageMgmt
# VLAN tag for the storage management network
# Corresponds to the StorageMgmtNetworkVlanID heat property
vlan: 1450
supernet: "172.16.20.0/22"
cidr: "172.16.20.0/24"
gateway: 172.16.20.1
vip_enable: false
mtu: 9000
pool:
start: 172.16.20.100
end: 172.16.20.200
# List of roles that can be on this network
role:
- Controller
# Leaf 0 subnet of the storage_mgmt network
storage_mgmt0:
# Network name used by TripleO Heat Templates
heat_name: StorageMgmt0
vlan: 1451
cidr: "172.16.21.0/24"
gateway: 172.16.21.1
mtu: 9000
pool:
start: 172.16.21.100
end: 172.16.21.200
# List of roles that can be on this network
role:
- CephStorage0Hw6
# Leaf 1 subnet of the storage_mgmt network
storage_mgmt1:
# Network name used by TripleO Heat Templates
heat_name: StorageMgmt1
vlan: 1452
cidr: "172.16.22.0/24"
gateway: 172.16.22.1
mtu: 9000
pool:
start: 172.16.22.100
end: 172.16.22.200
# List of roles that can be on this network
role:
- CephStorage1Hw7
# The tenant network is used for tenant workload data
tenant:
# Network name used by TripleO Heat Templates
heat_name: Tenant
# VLAN tag for the tenant network
# Corresponds to the TenantNetworkVlanID heat property
vlan: 1250
supernet: "172.16.80.0/21"
cidr: "172.16.81.0/24"
gateway: 172.16.81.1
vrouter_gateway: 172.16.81.1
mtu: 9000
pool:
start: 172.16.81.100
end: 172.16.81.200
# List of roles that can be on this network
role:
- ContrailTsn
- ComputeKernel
- ComputeDpdk
- ComputeSriov
# Leaf 0 subnet of the tenant network
tenant0:
# Network name used by TripleO Heat Templates
heat_name: Tenant0
vlan: 1251
cidr: "172.16.82.0/24"
gateway: 172.16.82.1
vrouter_gateway: 172.16.82.1
mtu: 9000
pool:
start: 172.16.82.100
end: 172.16.82.200
# List of roles that can be on this network
role:
- ComputeKernel0Hw0
- ComputeKernel0Hw1
- ComputeDpdk0Hw2
- ComputeSriov0Hw4
# Leaf 1 subnet of the tenant network
tenant1:
# Network name used by TripleO Heat Templates
heat_name: Tenant1
vlan: 1252
cidr: "172.16.83.0/24"
gateway: 172.16.83.1
vrouter_gateway: 172.16.83.1
mtu: 9000
pool:
start: 172.16.83.100
end: 172.16.83.200
# List of roles that can be on this network
role:
- ComputeSriov1Hw5
- ComputeKernel1Hw0
- ComputeDpdk1Hw3
- ComputeKernel1Hw1
# Contrail sepcific settings
#contrail:
# aaa_mode: cloud-admin
# vrouter:
# contrail_settings:
# # Settings per profile.
# # Profile's contrail_settings replace default settings and should include
# # all keys and values which are intended to be exported on given role.
# # When leafs are used it implies per profile configuration as it defines
# # VROUTER_GATEWAY for profile by quering node's tenant network for
# # vrouter_gateway value.
# default:
# VROUTER_GATEWAY: 172.16.81.254
# BGP_ASN: 64512
# LACP_RATE: 1
# ComputeKernel1Hw0:
# LACP_RATE: 1
# Information used to generate the SSL certificates for the public Openstack service APIs
tls:
# SSL key size in KB for CA generation
ca_key_size: 2048
# SSL key size in KB for SSL of openstack external VIP
ssl_key_size: 2048
#countryName_default
country: "US"
#stateOrProvinceName_default
state: "CA"
#localityName_default
city: "Sunnyvale"
#organizationalUnitName_default
organization: "JNPR"
#commonName_default - this is typically the external VIP
common_name: "10.10.10.90"
ceph:
# Choice to enable Ceph storage in the overcloud.
# "true" means that Ceph will be deployed as the backed for Cinder and Glance services.
# "false" false means that Ceph will not be deployed.
enabled: true
# Ceph OSD disk configuration
osd:
# Update the Ceph crush map when OSDs are started
crush_update_on_start: true
# Ceph OSD disk assignments. The named disks will be exclusively used by Ceph for persistence.
# Lvm is a default scenario for ceph deployment with bluestore as a backend.
# When all named disks are the same type, spinning or solid state, all of them will be used
# as ceph osds. When disks with mixed types are defined spinning disks will be used as osds
# and on solid state disks ceph db will be created. For mixed types of disks the automatic pgp
# number calculation requires assigning key 'contents' with value 'db' to ssd disks.
# In below example disks sd[b-e] are spinning disks and sdf is solid state disk.
default:
disk:
'/dev/sdb':
'/dev/sdc':
'/dev/sdd':
'/dev/sde':
'/dev/sdf':
contents: db
CephStorage0Hw6:
disk:
'/dev/sdb':
'/dev/sdc':
'/dev/sdd':
'/dev/sde':
'/dev/sdf':
contents: db
CephStorage1Hw7:
disk:
'/dev/sdb':
'/dev/sdc':
'/dev/sdd':
'/dev/sde':
'/dev/sdf':
contents: db
# By default, pgp number is calculated by contrail cloud. If you want, you can give this parameter
# by yourself. Use the calculator on the website: https://ceph.com/pgcalc/. Calculator takes into
# account also pool utilization. Calculated pgp_num can be introduced in configuration as below.
# It's defined per used pool.
# pool:
# vms:
# pgp_num: 32
# rbd:
# pgp_num: 32
# images:
# pgp_num: 32
# volumes:
# pgp_num: 32
# backups:
# pgp_num: 32
#
# Rados Gateway when enabled, which is a default behaviour, creates it's own ceph pools
# not tracked by contrail cloud. Those pools can be predefined to better control
# their sizes. Below pools definitions are not an exhaustive, please consult with
# https://ceph.com/pgcalc/
# Pools should have enabled application according to their use.
# If not changed explicit, pools are created with 'rbd' application assigned.
# Available options are:
# - rbd for the Ceph Block Device
# - rgw for the Ceph Object Gateway
# - cephfs for the Ceph Filesystem
# or user defined value for custom application.
# More details can be found on
# https://access.redhat.com/documentation/en-us/red_hat_ceph_storage/3/html/storage_strategies_guide/pools-1#enable-application
# .rgw.root:
# pgp_num: 16
# enabled: true
# replica: 3
# application: rgw
# default.rgw.control:
# pgp_num: 16
# enabled: true
# replica: 3
# application: rgw
# default.rgw.meta:
# pgp_num: 16
# enabled: true
# replica: 3
# application: rgw
# default.rgw.log:
# pgp_num: 16
# enabled: true
# replica: 3
# application: rgw
# default.rgw.buckets.index:
# pgp_num: 16
# enabled: true
# replica: 3
# application: rgw
# default.rgw.buckets.data:
# pgp_num: 16
# enabled: true
# replica: 3
# application: rgw
# default.rgw.buckets.non-ec:
# pgp_num: 16
# enabled: true
# replica: 3
# application: rgw
appformix:
# Set to true if you have multiple control hosts which allows Apformix to run in HA mode
enable_ha: true
# Floating virtual IP for the Appformix APIs on the external network, used and required by HA mode.
vip: "10.10.10.101"
# Floating virtual IP for the Appformix APIs on the internal network, used and required by HA mode.
secondary_vip: "172.16.0.89"
keepalived:
# Set which interface will be used for vrrp
vrrp_interface: "enp2s0"
# Set which interface will be used for second vrrp
secondary_vrrp_interface: "vlan1200"
Exemple de fichier de configuration inventory.yml
# Copyright 2023 Juniper Networks, Inc. All rights reserved.
# Licensed under the Juniper Networks Script Software License (the "License").
# You may not use this script file except in compliance with the License, which is located at
# http://www.juniper.net/support/legal/scriptlicense/
# Unless required by applicable law or otherwise agreed to in writing by the parties,
# software distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
#
#
#
# Common values shared among group of nodes
ipmi_hardware1: &hardware1
pm_type: "ipmi"
pm_user: "{{ vault['inventory_nodes']['hardware1']['pm_user'] }}"
pm_password: "{{ vault['inventory_nodes']['hardware1']['pm_password'] }}"
capabilities: "boot_mode:uefi"
# List of baremetal server nodes that can be used for the deploying roles
# Each list item contains:
# name: logical name to assign this resource (string)
# pm_addr: IP address for resourceIPMI interface (string)
# pm_type: Ironic driver to interface with this resource (typically ipmi) (string)
# pm_user: IPMI user account (string)
# pm_password: IPMI account user password (string)
# capabilities: String of comma separated list of node capabilities.
# Capabilities 'profile' and 'boot_option' are managed
# by Contrail Cloud and will be omitted. (string)
# e.g capabilities: "boot_mode:uefi" set boot mode to uefi
#
# Some values common for nodes can be moved to dedicated section like ipmi_hardware1
# and be referred like this:
# <<: *hardware1
inventory_nodes:
- name: "control-host1"
pm_addr: "10.10.11.58"
<<: *hardware1
- name: "control-host2"
pm_addr: "10.10.11.59"
<<: *hardware1
- name: "control-host3"
pm_addr: "10.10.11.60"
<<: *hardware1
- name: "storage1"
pm_addr: "10.10.11.61"
<<: *hardware1
- name: "storage2"
pm_addr: "10.10.11.62"
<<: *hardware1
- name: "storage3"
pm_addr: "10.10.11.63"
<<: *hardware1
- name: "computedpdk1"
pm_addr: "10.10.11.64"
<<: *hardware1
- name: "computedpdk2"
pm_addr: "10.10.11.65"
<<: *hardware1
- name: "compute1"
pm_addr: "10.10.11.66"
<<: *hardware1
- name: "compute2"
pm_addr: "10.10.11.67"
<<: *hardware1
- name: "compute3"
pm_addr: "10.10.11.68"
<<: *hardware1
- name: "compute4"
pm_addr: "10.10.11.69"
<<: *hardware1
- name: "computesriov1"
pm_addr: "10.10.11.70"
<<: *hardware1
- name: "computesriov2"
pm_addr: "10.10.11.71"
<<: *hardware1
Exemple de fichier de configuration control-host-nodes.yml
# Copyright 2023 Juniper Networks, Inc. All rights reserved.
# Licensed under the Juniper Networks Script Software License (the "License").
# You may not use this script file except in compliance with the License, which is located at
# http://www.juniper.net/support/legal/scriptlicense/
# Unless required by applicable law or otherwise agreed to in writing by the parties,
# software distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
#
#
#
# List of nodes to use as control host role
# Each list item contains a set of variables which can be referenced
# with "{{ host.<variable> }}" in control_host_nodes_network_config below.
# Other ad-hoc variables can be added as needed.
# name: name of a node in the inventory (string)
# hostname: hostname to assign the node after it is imaged (string)
# control_ip_netmask: static CIDR address on Control Plane network.
# Choose a value outside the DHCP range. (string)
# dns_server1,dns_server2: dns server addresses (string)
# max_mtu: The largest MTU supported by an interface
#
control_host_nodes:
- name: "control-host1"
control_ip_netmask: "192.168.213.5/24"
max_mtu: 9216
- name: "control-host2"
control_ip_netmask: "192.168.213.6/24"
max_mtu: 9216
- name: "control-host3"
control_ip_netmask: "192.168.213.7/24"
max_mtu: 9216
# Template for network layout on all control host nodes
# This follows the syntax
# https://cloudinit.readthedocs.io/en/latest/topics/network-config-format-v1.html#network-config-v1
# or
# https://cloudinit.readthedocs.io/en/latest/topics/network-config-format-v2.html#network-config-v2
# variables from control_host_nodes can be refered with "{{ host.<variable> }}"
control_host_nodes_network_config:
version: 1
config:
- type: physical
name: eno1
subnets:
- type: dhcp
mtu: "{{ host.max_mtu }}"
nm_controlled: true
- type: physical
name: eno2
mtu: "{{ host.max_mtu }}"
nm_controlled: true
- type: physical
name: ens7f0
mtu: "{{ host.max_mtu }}"
nm_controlled: true
- type: physical
name: ens7f1
mtu: "{{ host.max_mtu }}"
nm_controlled: true
- type: bond
name: bond0
mtu: "{{ host.max_mtu }}"
nm_controlled: true
bond_interfaces:
- ens7f0
- ens7f1
params:
bond-mode: "802.3ad"
xmit_hash_policy: layer3+4
lacp_rate: fast
miimon: "100"
control_hosts:
# The mapping from control host interfaces to the control VM interfaces
# The first interface (enp1s0) must always be the Control Plane network to allow the VM to PXE boot
# VM interface names must be sequential with no gaps (e.g. enp1s0, enp2s0, enp3s0,...)
vm_interfaces:
- interface: enp1s0
physical_interface: eno1
- interface: enp2s0
physical_interface: eno2
- interface: enp3s0
physical_interface: bond0
Exemple de fichier de configuration overcloud-nics.yml
# Copyright 2023 Juniper Networks, Inc. All rights reserved.
# Licensed under the Juniper Networks Script Software License (the "License").
# You may not use this script file except in compliance with the License, which is located at
# http://www.juniper.net/support/legal/scriptlicense/
# Unless required by applicable law or otherwise agreed to in writing by the parties,
# software distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
#
#
#
Controller_network_config:
- type: interface
name: enp1s0
dns_servers:
get_param: DnsServers
use_dhcp: false
mtu:
get_param: ControlPlaneMtu
addresses:
- ip_netmask:
list_join:
- '/'
- - get_param: ControlPlaneIp
- get_param: ControlPlaneSubnetCidr
- type: vlan
device: enp1s0
vlan_id:
get_param: StorageNetworkVlanID
mtu:
get_param: StorageMtu
addresses:
- ip_netmask:
get_param: StorageIpSubnet
routes:
-
ip_netmask:
get_param: StorageSupernet
next_hop:
get_param: StorageInterfaceDefaultRoute
- type: vlan
device: enp1s0
vlan_id:
get_param: StorageMgmtNetworkVlanID
mtu:
get_param: StorageMgmtMtu
addresses:
- ip_netmask:
get_param: StorageMgmtIpSubnet
routes:
-
ip_netmask:
get_param: StorageMgmtSupernet
next_hop:
get_param: StorageMgmtInterfaceDefaultRoute
- type: vlan
device: enp1s0
vlan_id:
get_param: InternalApiNetworkVlanID
mtu:
get_param: InternalApiMtu
addresses:
- ip_netmask:
get_param: InternalApiIpSubnet
routes:
-
ip_netmask:
get_param: InternalApiSupernet
next_hop:
get_param: InternalApiInterfaceDefaultRoute
- type: interface
name: enp2s0
mtu:
get_param: ExternalMtu
addresses:
- ip_netmask:
get_param: ExternalIpSubnet
routes:
-
default: True
next_hop:
get_param: ExternalInterfaceDefaultRoute
- type: interface
name: enp3s0
use_dhcp: false
AppformixController_network_config:
- type: interface
name: enp1s0
dns_servers:
get_param: DnsServers
use_dhcp: false
mtu:
get_param: ControlPlaneMtu
addresses:
- ip_netmask:
list_join:
- '/'
- - get_param: ControlPlaneIp
- get_param: ControlPlaneSubnetCidr
- type: vlan
device: enp1s0
vlan_id:
get_param: InternalApiNetworkVlanID
mtu:
get_param: InternalApiMtu
addresses:
- ip_netmask:
get_param: InternalApiIpSubnet
routes:
-
ip_netmask:
get_param: InternalApiSupernet
next_hop:
get_param: InternalApiInterfaceDefaultRoute
- type: interface
name: enp2s0
mtu:
get_param: ExternalMtu
addresses:
- ip_netmask:
get_param: ExternalIpSubnet
routes:
-
default: True
next_hop:
get_param: ExternalInterfaceDefaultRoute
- type: interface
name: enp3s0
use_dhcp: false
ContrailTsn_network_config:
- type: interface
name: enp1s0
dns_servers:
get_param: DnsServers
mtu:
get_param: ControlPlaneMtu
addresses:
- ip_netmask:
list_join:
- '/'
- - get_param: ControlPlaneIp
- get_param: ControlPlaneSubnetCidr
use_dhcp: false
routes:
-
default: True
next_hop:
get_param: ControlPlaneDefaultRoute
- type: vlan
device: enp1s0
vlan_id:
get_param: InternalApiNetworkVlanID
mtu:
get_param: InternalApiMtu
addresses:
- ip_netmask:
get_param: InternalApiIpSubnet
routes:
-
ip_netmask:
get_param: InternalApiSupernet
next_hop:
get_param: InternalApiInterfaceDefaultRoute
- type: interface
name: enp2s0
use_dhcp: false
- type: interface
name: enp3s0
use_dhcp: false
mtu:
get_param: TenantMtu
- type: vlan
device: enp3s0
vlan_id:
get_param: TenantNetworkVlanID
mtu:
get_param: TenantMtu
use_dhcp: false
- type: contrail_vrouter
name: vhost0
members:
-
type: interface
name:
str_replace:
template: vlanVLANID
params:
VLANID: {get_param: TenantNetworkVlanID}
use_dhcp: false
mtu:
get_param: TenantMtu
addresses:
- ip_netmask:
get_param: TenantIpSubnet
routes:
-
ip_netmask:
get_param: TenantSupernet
next_hop:
get_param: TenantInterfaceDefaultRoute
ComputeKernel0Hw1_network_config:
- type: interface
name: nic1
dns_servers:
get_param: DnsServers
use_dhcp: false
mtu:
get_param: ControlPlaneMtu
addresses:
- ip_netmask:
list_join:
- '/'
- - get_param: ControlPlaneIp
- get_param: ControlPlaneSubnetCidr
routes:
-
default: True
next_hop:
get_param: ControlPlaneDefaultRoute
- type: vlan
device: nic1
vlan_id:
get_param: Storage0NetworkVlanID
mtu:
get_param: Storage0NetworkMtu
addresses:
- ip_netmask:
get_param: Storage0IpSubnet
routes:
-
ip_netmask:
get_param: StorageSupernet
next_hop:
get_param: Storage0InterfaceDefaultRoute
- type: vlan
device: nic1
vlan_id:
get_param: InternalApi0NetworkVlanID
mtu:
get_param: InternalApi0NetworkMtu
addresses:
- ip_netmask:
get_param: InternalApi0IpSubnet
routes:
-
ip_netmask:
get_param: InternalApiSupernet
next_hop:
get_param: InternalApi0InterfaceDefaultRoute
- type: interface
name: nic2
use_dhcp: false
- type: linux_bond
name: bond0
use_dhcp: false
bonding_options: "mode=802.3ad xmit_hash_policy=layer3+4 lacp_rate=fast updelay=1000 miimon=100"
mtu:
get_param: Tenant0NetworkMtu
members:
- type: interface
name: nic3
primary: true
mtu:
get_param: Tenant0NetworkMtu
- type: interface
name: nic4
mtu:
get_param: Tenant0NetworkMtu
- type: vlan
vlan_id:
get_param: Tenant0NetworkVlanID
device: bond0
- type: contrail_vrouter
name: vhost0
use_dhcp: false
members:
-
type: interface
name:
str_replace:
template: vlanVLANID
params:
VLANID: {get_param: Tenant0NetworkVlanID}
use_dhcp: false
addresses:
- ip_netmask:
get_param: Tenant0IpSubnet
mtu:
get_param: Tenant0NetworkMtu
routes:
-
ip_netmask:
get_param: TenantSupernet
next_hop:
get_param: Tenant0InterfaceDefaultRoute
ComputeKernel0Hw0_network_config:
- type: interface
name: nic1
dns_servers:
get_param: DnsServers
use_dhcp: false
mtu:
get_param: ControlPlaneMtu
addresses:
- ip_netmask:
list_join:
- '/'
- - get_param: ControlPlaneIp
- get_param: ControlPlaneSubnetCidr
routes:
-
default: True
next_hop:
get_param: ControlPlaneDefaultRoute
- type: vlan
device: nic1
vlan_id:
get_param: Storage0NetworkVlanID
mtu:
get_param: Storage0NetworkMtu
addresses:
- ip_netmask:
get_param: Storage0IpSubnet
routes:
-
ip_netmask:
get_param: StorageSupernet
next_hop:
get_param: Storage0InterfaceDefaultRoute
- type: vlan
device: nic1
vlan_id:
get_param: InternalApi0NetworkVlanID
mtu:
get_param: InternalApi0NetworkMtu
addresses:
- ip_netmask:
get_param: InternalApi0IpSubnet
routes:
-
ip_netmask:
get_param: InternalApiSupernet
next_hop:
get_param: InternalApi0InterfaceDefaultRoute
- type: interface
name: nic2
use_dhcp: false
- type: linux_bond
name: bond0
use_dhcp: false
bonding_options: "mode=802.3ad xmit_hash_policy=layer3+4 lacp_rate=fast updelay=1000 miimon=100"
mtu:
get_param: Tenant0NetworkMtu
members:
- type: interface
name: nic3
primary: true
mtu:
get_param: Tenant0NetworkMtu
- type: interface
name: nic4
mtu:
get_param: Tenant0NetworkMtu
- type: vlan
vlan_id:
get_param: Tenant0NetworkVlanID
device: bond0
- type: contrail_vrouter
name: vhost0
use_dhcp: false
members:
-
type: interface
name:
str_replace:
template: vlanVLANID
params:
VLANID: {get_param: Tenant0NetworkVlanID}
use_dhcp: false
addresses:
- ip_netmask:
get_param: Tenant0IpSubnet
mtu:
get_param: Tenant0NetworkMtu
routes:
-
ip_netmask:
get_param: TenantSupernet
next_hop:
get_param: Tenant0InterfaceDefaultRoute
ComputeKernel1Hw0_network_config:
- type: interface
name: nic1
dns_servers:
get_param: DnsServers
use_dhcp: false
mtu:
get_param: ControlPlaneMtu
addresses:
- ip_netmask:
list_join:
- '/'
- - get_param: ControlPlaneIp
- get_param: ControlPlaneSubnetCidr
routes:
-
default: True
next_hop:
get_param: ControlPlaneDefaultRoute
- type: vlan
device: nic1
vlan_id:
get_param: Storage1NetworkVlanID
mtu:
get_param: Storage1NetworkMtu
addresses:
- ip_netmask:
get_param: Storage1IpSubnet
routes:
-
ip_netmask:
get_param: StorageSupernet
next_hop:
get_param: Storage1InterfaceDefaultRoute
- type: vlan
device: nic1
vlan_id:
get_param: InternalApi1NetworkVlanID
mtu:
get_param: InternalApi1NetworkMtu
addresses:
- ip_netmask:
get_param: InternalApi1IpSubnet
routes:
-
ip_netmask:
get_param: InternalApiSupernet
next_hop:
get_param: InternalApi1InterfaceDefaultRoute
- type: interface
name: nic2
use_dhcp: false
- type: linux_bond
name: bond0
use_dhcp: false
bonding_options: "mode=802.3ad xmit_hash_policy=layer3+4 lacp_rate=fast updelay=1000 miimon=100"
mtu:
get_param: Tenant1NetworkMtu
members:
- type: interface
name: nic3
primary: true
mtu:
get_param: Tenant1NetworkMtu
- type: interface
name: nic4
mtu:
get_param: Tenant1NetworkMtu
- type: vlan
vlan_id:
get_param: Tenant1NetworkVlanID
device: bond0
- type: contrail_vrouter
name: vhost0
use_dhcp: false
members:
-
type: interface
name:
str_replace:
template: vlanVLANID
params:
VLANID: {get_param: Tenant1NetworkVlanID}
use_dhcp: false
addresses:
- ip_netmask:
get_param: Tenant1IpSubnet
mtu:
get_param: Tenant1NetworkMtu
routes:
-
ip_netmask:
get_param: TenantSupernet
next_hop:
get_param: Tenant1InterfaceDefaultRoute
ComputeKernel1Hw1_network_config:
- type: interface
name: nic1
dns_servers:
get_param: DnsServers
use_dhcp: false
mtu:
get_param: ControlPlaneMtu
addresses:
- ip_netmask:
list_join:
- '/'
- - get_param: ControlPlaneIp
- get_param: ControlPlaneSubnetCidr
routes:
-
default: True
next_hop:
get_param: ControlPlaneDefaultRoute
- type: vlan
device: nic1
vlan_id:
get_param: Storage1NetworkVlanID
mtu:
get_param: Storage1NetworkMtu
addresses:
- ip_netmask:
get_param: Storage1IpSubnet
routes:
-
ip_netmask:
get_param: StorageSupernet
next_hop:
get_param: Storage1InterfaceDefaultRoute
- type: vlan
device: nic1
vlan_id:
get_param: InternalApi1NetworkVlanID
mtu:
get_param: InternalApi1NetworkMtu
addresses:
- ip_netmask:
get_param: InternalApi1IpSubnet
routes:
-
ip_netmask:
get_param: InternalApiSupernet
next_hop:
get_param: InternalApi1InterfaceDefaultRoute
- type: interface
name: nic2
use_dhcp: false
- type: linux_bond
name: bond0
use_dhcp: false
bonding_options: "mode=802.3ad xmit_hash_policy=layer3+4 lacp_rate=fast updelay=1000 miimon=100"
mtu:
get_param: Tenant1NetworkMtu
members:
- type: interface
name: nic3
primary: true
mtu:
get_param: Tenant1NetworkMtu
- type: interface
name: nic4
mtu:
get_param: Tenant1NetworkMtu
- type: vlan
vlan_id:
get_param: Tenant1NetworkVlanID
device: bond0
- type: contrail_vrouter
name: vhost0
use_dhcp: false
members:
-
type: interface
name:
str_replace:
template: vlanVLANID
params:
VLANID: {get_param: Tenant1NetworkVlanID}
use_dhcp: false
addresses:
- ip_netmask:
get_param: Tenant1IpSubnet
mtu:
get_param: Tenant1NetworkMtu
routes:
-
ip_netmask:
get_param: TenantSupernet
next_hop:
get_param: Tenant1InterfaceDefaultRoute
ComputeSriov0Hw4_network_config:
- type: interface
name: nic1
dns_servers:
get_param: DnsServers
use_dhcp: false
mtu:
get_param: ControlPlaneMtu
addresses:
- ip_netmask:
list_join:
- '/'
- - get_param: ControlPlaneIp
- get_param: ControlPlaneSubnetCidr
routes:
-
default: True
next_hop:
get_param: ControlPlaneDefaultRoute
- type: vlan
device: nic1
vlan_id:
get_param: Storage0NetworkVlanID
mtu:
get_param: Storage0NetworkMtu
addresses:
- ip_netmask:
get_param: Storage0IpSubnet
routes:
-
ip_netmask:
get_param: StorageSupernet
next_hop:
get_param: Storage0InterfaceDefaultRoute
- type: vlan
device: nic1
vlan_id:
get_param: InternalApi0NetworkVlanID
mtu:
get_param: InternalApi0NetworkMtu
addresses:
- ip_netmask:
get_param: InternalApi0IpSubnet
routes:
-
ip_netmask:
get_param: InternalApiSupernet
next_hop:
get_param: InternalApi0InterfaceDefaultRoute
- type: interface
name: nic2
use_dhcp: false
- type: linux_bond
name: bond0
use_dhcp: false
bonding_options: "mode=802.3ad xmit_hash_policy=layer3+4 lacp_rate=fast updelay=1000 miimon=100"
mtu:
get_param: Tenant0NetworkMtu
members:
- type: interface
name: nic3
primary: true
mtu:
get_param: Tenant0NetworkMtu
- type: interface
name: nic4
mtu:
get_param: Tenant0NetworkMtu
- type: vlan
vlan_id:
get_param: Tenant0NetworkVlanID
device: bond0
- type: contrail_vrouter
name: vhost0
use_dhcp: false
members:
-
type: interface
name:
str_replace:
template: vlanVLANID
params:
VLANID: {get_param: Tenant0NetworkVlanID}
use_dhcp: false
addresses:
- ip_netmask:
get_param: Tenant0IpSubnet
mtu:
get_param: Tenant0NetworkMtu
routes:
-
ip_netmask:
get_param: TenantSupernet
next_hop:
get_param: Tenant0InterfaceDefaultRoute
ComputeSriov1Hw5_network_config:
- type: interface
name: nic1
dns_servers:
get_param: DnsServers
use_dhcp: false
mtu:
get_param: ControlPlaneMtu
addresses:
- ip_netmask:
list_join:
- '/'
- - get_param: ControlPlaneIp
- get_param: ControlPlaneSubnetCidr
routes:
-
default: True
next_hop:
get_param: ControlPlaneDefaultRoute
- type: vlan
device: nic1
vlan_id:
get_param: Storage1NetworkVlanID
mtu:
get_param: Storage1NetworkMtu
addresses:
- ip_netmask:
get_param: Storage1IpSubnet
routes:
-
ip_netmask:
get_param: StorageSupernet
next_hop:
get_param: Storage1InterfaceDefaultRoute
- type: vlan
device: nic1
vlan_id:
get_param: InternalApi1NetworkVlanID
mtu:
get_param: InternalApi1NetworkMtu
addresses:
- ip_netmask:
get_param: InternalApi1IpSubnet
routes:
-
ip_netmask:
get_param: InternalApiSupernet
next_hop:
get_param: InternalApi1InterfaceDefaultRoute
- type: interface
name: nic2
use_dhcp: false
- type: linux_bond
name: bond0
use_dhcp: false
bonding_options: "mode=802.3ad xmit_hash_policy=layer3+4 lacp_rate=fast updelay=1000 miimon=100"
mtu:
get_param: Tenant1NetworkMtu
members:
- type: interface
name: nic3
primary: true
mtu:
get_param: Tenant1NetworkMtu
- type: interface
name: nic4
mtu:
get_param: Tenant1NetworkMtu
- type: vlan
vlan_id:
get_param: Tenant1NetworkVlanID
device: bond0
- type: contrail_vrouter
name: vhost0
use_dhcp: false
members:
-
type: interface
name:
str_replace:
template: vlanVLANID
params:
VLANID: {get_param: Tenant1NetworkVlanID}
use_dhcp: false
addresses:
- ip_netmask:
get_param: Tenant1IpSubnet
mtu:
get_param: Tenant1NetworkMtu
routes:
-
ip_netmask:
get_param: TenantSupernet
next_hop:
get_param: Tenant1InterfaceDefaultRoute
ComputeDpdk0Hw2_network_config:
- type: interface
name: nic1
dns_servers:
get_param: DnsServers
use_dhcp: false
mtu:
get_param: ControlPlaneMtu
addresses:
- ip_netmask:
list_join:
- '/'
- - get_param: ControlPlaneIp
- get_param: ControlPlaneSubnetCidr
routes:
-
default: True
next_hop:
get_param: ControlPlaneDefaultRoute
- type: vlan
device: nic1
vlan_id:
get_param: Storage0NetworkVlanID
mtu:
get_param: Storage0NetworkMtu
addresses:
- ip_netmask:
get_param: Storage0IpSubnet
routes:
-
ip_netmask:
get_param: StorageSupernet
next_hop:
get_param: Storage0InterfaceDefaultRoute
- type: vlan
device: nic1
vlan_id:
get_param: InternalApi0NetworkVlanID
mtu:
get_param: InternalApi0NetworkMtu
addresses:
- ip_netmask:
get_param: InternalApi0IpSubnet
routes:
-
ip_netmask:
get_param: InternalApiSupernet
next_hop:
get_param: InternalApi0InterfaceDefaultRoute
- type: interface
name: nic2
use_dhcp: false
- type: contrail_vrouter_dpdk
name: vhost0
vlan_id:
get_param: Tenant0NetworkVlanID
driver: "{{ overcloud['contrail']['vrouter']['dpdk']['driver'] }}"
bond_mode: 4
bond_policy: layer2+3
cpu_list: 1,2
members:
- type: interface
name: nic3
- type: interface
name: nic4
addresses:
- ip_netmask:
get_param: Tenant0IpSubnet
mtu:
get_param: Tenant0NetworkMtu
routes:
-
ip_netmask:
get_param: TenantSupernet
next_hop:
get_param: Tenant0InterfaceDefaultRoute
ComputeDpdk1Hw3_network_config:
- type: interface
name: nic1
dns_servers:
get_param: DnsServers
use_dhcp: false
mtu:
get_param: ControlPlaneMtu
addresses:
- ip_netmask:
list_join:
- '/'
- - get_param: ControlPlaneIp
- get_param: ControlPlaneSubnetCidr
routes:
-
default: True
next_hop:
get_param: ControlPlaneDefaultRoute
- type: vlan
device: nic1
vlan_id:
get_param: Storage1NetworkVlanID
mtu:
get_param: Storage1NetworkMtu
addresses:
- ip_netmask:
get_param: Storage1IpSubnet
routes:
-
ip_netmask:
get_param: StorageSupernet
next_hop:
get_param: Storage1InterfaceDefaultRoute
- type: vlan
device: nic1
vlan_id:
get_param: InternalApi1NetworkVlanID
mtu:
get_param: InternalApi1NetworkMtu
addresses:
- ip_netmask:
get_param: InternalApi1IpSubnet
routes:
-
ip_netmask:
get_param: InternalApiSupernet
next_hop:
get_param: InternalApi1InterfaceDefaultRoute
- type: interface
name: nic2
use_dhcp: false
- type: contrail_vrouter_dpdk
name: vhost0
vlan_id:
get_param: Tenant1NetworkVlanID
driver: "{{ overcloud['contrail']['vrouter']['dpdk']['driver'] }}"
bond_mode: 4
bond_policy: layer2+3
cpu_list: 1,2
members:
- type: interface
name: nic3
- type: interface
name: nic4
addresses:
- ip_netmask:
get_param: Tenant1IpSubnet
mtu:
get_param: Tenant1NetworkMtu
routes:
-
ip_netmask:
get_param: TenantSupernet
next_hop:
get_param: Tenant1InterfaceDefaultRoute
CephStorage0Hw6_network_config:
- type: interface
name: nic1
dns_servers:
get_param: DnsServers
use_dhcp: false
mtu:
get_param: ControlPlaneMtu
addresses:
- ip_netmask:
list_join:
- '/'
- - get_param: ControlPlaneIp
- get_param: ControlPlaneSubnetCidr
routes:
-
default: True
next_hop:
get_param: ControlPlaneDefaultRoute
- type: vlan
device: nic1
vlan_id:
get_param: Storage0NetworkVlanID
mtu:
get_param: Storage0NetworkMtu
addresses:
- ip_netmask:
get_param: Storage0IpSubnet
routes:
-
ip_netmask:
get_param: StorageSupernet
next_hop:
get_param: Storage0InterfaceDefaultRoute
- type: vlan
device: nic1
vlan_id:
get_param: StorageMgmt0NetworkVlanID
mtu:
get_param: StorageMgmt0NetworkMtu
addresses:
- ip_netmask:
get_param: StorageMgmt0IpSubnet
routes:
-
ip_netmask:
get_param: StorageMgmtSupernet
next_hop:
get_param: StorageMgmt0InterfaceDefaultRoute
- type: interface
name: nic2
use_dhcp: false
- type: interface
name: nic3
use_dhcp: false
- type: interface
name: nic4
use_dhcp: false
CephStorage1Hw7_network_config:
- type: interface
name: nic1
dns_servers:
get_param: DnsServers
use_dhcp: false
mtu:
get_param: ControlPlaneMtu
addresses:
- ip_netmask:
list_join:
- '/'
- - get_param: ControlPlaneIp
- get_param: ControlPlaneSubnetCidr
routes:
-
default: True
next_hop:
get_param: ControlPlaneDefaultRoute
- type: vlan
device: nic1
vlan_id:
get_param: Storage1NetworkVlanID
mtu:
get_param: Storage1NetworkMtu
addresses:
- ip_netmask:
get_param: Storage1IpSubnet
routes:
-
ip_netmask:
get_param: StorageSupernet
next_hop:
get_param: Storage1InterfaceDefaultRoute
- type: vlan
device: nic1
vlan_id:
get_param: StorageMgmt1NetworkVlanID
mtu:
get_param: StorageMgmt1NetworkMtu
addresses:
- ip_netmask:
get_param: StorageMgmt1IpSubnet
routes:
-
ip_netmask:
get_param: StorageMgmtSupernet
next_hop:
get_param: StorageMgmt1InterfaceDefaultRoute
- type: interface
name: nic2
use_dhcp: false
- type: interface
name: nic3
use_dhcp: false
- type: interface
name: nic4
use_dhcp: false
Exemple de fichier de configuration compute-nodes.yml
# Copyright 2023 Juniper Networks, Inc. All rights reserved.
# Licensed under the Juniper Networks Script Software License (the "License").
# You may not use this script file except in compliance with the License, which is located at
# http://www.juniper.net/support/legal/scriptlicense/
# Unless required by applicable law or otherwise agreed to in writing by the parties,
# software distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
#
#
#
# Each list item contains:
# name: name of a node in the inventory (string)
# profile: name of hardware profile, group of servers (optional, string)
# leaf: leaf name (optional, string)
# List of nodes to use as compute role using Contrail DPDK vRouter
compute_nodes_dpdk:
- name: computedpdk1
leaf: '0'
profile: hw2
- name: computedpdk2
leaf: '1'
profile: hw3
# List of nodes to use as compute role using Sriov
compute_nodes_sriov:
- name: computesriov1
leaf: '0'
profile: hw4
- name: computesriov2
leaf: '1'
profile: hw5
# List of nodes to use as compute role using Contrail kernel vRouter
compute_nodes_kernel:
- name: compute1
leaf: '0'
profile: hw0
- name: compute2
leaf: '0'
profile: hw1
- name: compute3
leaf: '1'
profile: hw1
- name: compute4
leaf: '1'
profile: hw0
Exemple de fichier de configuration storage-nodes.yml
# Copyright 2023 Juniper Networks, Inc. All rights reserved.
# Licensed under the Juniper Networks Script Software License (the "License").
# You may not use this script file except in compliance with the License, which is located at
# http://www.juniper.net/support/legal/scriptlicense/
# Unless required by applicable law or otherwise agreed to in writing by the parties,
# software distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
#
#
#
# List of nodes to use as storage host role
# List item contains:
# name: name of a node in the inventory (string)
# profile: name of hardware profile, group of servers (optional, string)
# leaf: leaf name (optional, string)
storage_nodes:
- name: storage1
leaf: '0'
profile: hw6
- name: storage2
leaf: '0'
profile: hw6
- name: storage3
leaf: '1'
profile: hw7
Exemple de fichier s-host-nodes.yml configuration de K8
# Copyright 2023 Juniper Networks, Inc. All rights reserved.
# Licensed under the Juniper Networks Script Software License (the "License").
# You may not use this script file except in compliance with the License, which is located at
# http://www.juniper.net/support/legal/scriptlicense/
# Unless required by applicable law or otherwise agreed to in writing by the parties,
# software distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
#
#
# +-------------------+ +--------------+ +-----------------+
# |Undercloud/Jumphost| |OpenStack Ctrl| |OpenStack Compute|
# +-------------------+ +--------------+ +-----------------+
# |Provision & deploy | |Neutron | |vrouter |
# +-+-----------------+ |Heat | | |
# | |Keystone | | |
# | ++-------------+ | |
# | | ++--+-------------+
# | | | |
# | | | |
# | | | +----+--------------------------------------tenant_ip_netmask------------+
# | | | |
# | | | |
# | +------------+------+----------------------+-------------+---------internalapi_ip_netmask-------+
# | | | | |
# | | | | |
# | +-------------------------------------------------------------+ +-----------+ +-----------+
# | | | | | | | | | | |
# +--control_ip_netmask----+ +--------+--+ +-------+--+ +-------+--+ +-------+--+ | | | | |
# | |Config | |Control | |Command | |K8s api | | | | | |
# | +-----------+ +----------+ +----------+ +-+--------+ | | | | |
# | |Analytics | |WebUI | | | | | | |
# | +-----------+ +-----+----+ | | | | | |
# | | | | | | | |
# | +--+---------+--+ | | | | |
# | K8s host | VIP & PROXY | | | K8s host | | K8s host |
# +-----------------------------------+---------------+---------+ +-----------+ +-----------+
# | |
# | |
# +---+---------+---------------external_ip_netmask----------+
#
# Contrail control plane deployed on Kubernetes.
#
# Every node defined in k8s_host_nodes list should have following keys defined:
# * name: hostname without dns domain. Later it will be concatenated with
# global['domain'] value from site.yml.
#
# * hypervisor: Name of control host defined in control-host-nodes.yml where the
# k8s nodes VM will be hosted.
#
# * control_ip_netmask: IP address in CIDR notation in control network.
# This network will be used to provision, deploy and access
# node.
#
# * internalapi_ip_netmask: IP address in CIDR notation in internal api network.
# This network will be used for communication between
# OpenStack components like Neutron and Heat and
# Contrail services. Contrail vrouter will use this
# network to provision itself. K8s components utilises
# this network for internal communication.
#
# * tenant_ip_netmask: IP address in CIDR notation in tenant network.
# Through this network vrouter will communicate with
# Contrail control.
#
# * external_ip_netmask: IP address in CIDR notation in external network.
# This network will be used to setup external VIP
# managed by keepalived. Haproxy will be configured
# to expose services like Command, Contrail web ui and
# k8s api through this VIP.
#
# * extra keys: nodes can define variables used later in network configuration
# like dns servers, default gw etc.
common: &common
dns_server1: "8.8.4.4"
dns_server2: "8.8.8.8"
gw: "192.2.0.254"
k8s_host_nodes:
- name: k8s-contrail1
hypervisor: controler1
control_ip_netmask: "192.168.213.21/24"
internalapi_ip_netmask: "172.16.0.21/24"
tenant_ip_netmask: "172.17.131.21/24"
external_ip_netmask: "192.2.0.230/25"
<<: *common
- name: k8s-contrail2
hypervisor: controler2
control_ip_netmask: "192.168.213.22/24"
internalapi_ip_netmask: "172.16.0.22/24"
tenant_ip_netmask: "172.17.131.22/24"
external_ip_netmask: "192.2.0.231/25"
<<: *common
- name: k8s-contrail3
hypervisor: controler3
control_ip_netmask: "192.168.213.23/24"
internalapi_ip_netmask: "172.16.0.23/24"
tenant_ip_netmask: "172.17.131.23/24"
external_ip_netmask: "192.2.0.232/25"
<<: *common
# Template for network layout on all kubernetes cluster nodes
# This follows the syntax
# https://cloudinit.readthedocs.io/en/latest/topics/network-config-format-v1.html#network-config-v1
# or
# https://cloudinit.readthedocs.io/en/latest/topics/network-config-format-v2.html#network-config-v2
# variables from k8s_host_nodes can be refered with "{{ host.<variable> }}"
k8s_host_nodes_network_config:
version: 1
config:
- type: physical
name: enp1s0
subnets:
- type: static
address: "{{ host.control_ip_netmask }}"
- type: vlan
name: "enp1s0.{{ overcloud.network.internal_api.vlan }}"
vlan_link: enp1s0
vlan_id: "{{ overcloud.network.internal_api.vlan }}"
subnets:
- type: static
address: "{{ host.internalapi_ip_netmask }}"
routes:
- gateway: "{{ overcloud.network.internal_api.gateway }}"
network: "{{ overcloud.network.internal_api.supernet | ipaddr('network') }}"
netmask: "{{ overcloud.network.internal_api.supernet | ipaddr('netmask') }}"
- type: physical
name: enp2s0
subnets:
- type: static
address: "{{ host.external_ip_netmask }}"
gateway: "{{ host.gw }}"
dns_nameservers:
- "{{ host.dns_server1 }}"
- "{{ host.dns_server2 }}"
- type: physical
name: enp3s0
- type: vlan
name: "enp3s0.{{ overcloud.network.tenant.vlan }}"
vlan_link: enp3s0
vlan_id: "{{ overcloud.network.tenant.vlan }}"
subnets:
- type: static
address: "{{ host.tenant_ip_netmask }}"
routes:
- gateway: "{{ overcloud.network.tenant.gateway }}"
network: "{{ overcloud.network.tenant.supernet | ipaddr('network') }}"
netmask: "{{ overcloud.network.tenant.supernet | ipaddr('netmask') }}"
Exemple de fichier de configuration vault-data.yml
# This config structure can be used to hold information that needs to be encrypted for privacy
# If there is a password stored in /var/lib/contrail_cloud/config/.vault_password then it will be used
# Otherwise the password can be entered interactively
#
# This file can be edited with the "ansible-vault edit" command
# This file can be re-encrypted with a new password with the "ansible-vault rekey" command
vault:
global:
rhel:
# Contrail Cloud Activation Key
satellite:
#SATELLITE_KEY
key: "PUT_YOUR_KEY_HERE"
# User account used for all Contrail Cloud automation
# This account will be created on:
# - jumphost
# - control hosts
# - all overcloud roles
# - appformix controllers
service_user:
# Account Name
name: "contrail"
# Account Password
password: "c0ntrail123"
# Passphrase used to encrypt ssh key of service user.
# If not defined ssh private key will not be encrypted.
# ssh_key_passphrase: "c0ntrail123"
rhvm:
vm:
# rhvm user name
user: "contrail"
# password for the rhvm vm user
password: "c0ntrail123"
# root password for the rhvm VM
root_password: "c0ntrail123"
# webui admin password
admin_password: "c0ntrail123"
# Passphrase used to encrypt ssh key of rhvm user.
# If not defined ssh private key will not be encrypted.
# ssh_key_passphrase: "c0ntrail123"
vnc:
# VNC console password for the rhvm VM
password: "contrail123"
undercloud:
#Administrator password - default is randomly generated
#admin_password: "c0ntrail123"
vm:
# undercloud user name
user: "stack"
# password for the undercloud vm user
password: "contrail123"
# root password for the undercloud VM
root_password: "contrail123"
# Passphrase used to encrypt ssh key of undercloud user.
# If not defined ssh private key will not be encrypted.
# ssh_key_passphrase: "c0ntrail123"
vnc:
# VNC console password for the undercloud VM
password: "contrail123"
overcloud:
#Administrator password
admin_password: "c0ntrail123"
# Root password used for local login to overcloud nodes through console
# root_password: "contrail123"
contrail:
rabbitmq:
# contrail rabbitmq user name
user: "contrail_rabbitmq"
# contrail rabbitmq user password
password: "c0ntrail123"
registry:
local_instance:
openstack:
username: "contrail"
password: "c0ntrail123"
contrail:
username: "contrail"
password: "c0ntrail123"
contrail-insights:
username: "contrail"
password: "c0ntrail123"
control_hosts:
vm:
vnc:
# VNC console password for all control VMs
password: "contrail123"
appformix:
mysql:
# Approfmix MySQL user account
user: "appformix"
# Approfmix MySQL user password
password: "c0ntrail123"
rabbitmq:
# Approfmix RabbitMQ user account
user: "appformix"
# Approfmix RabbitMQ user password
password: "c0ntrail123"
# Credentials used to connect external ceph cluster
#ceph_external:
# client_key: "CLIENT_KEY"
# client_user: "openstack"
# List of inventory hardware types that can hold hardware-specific properties
# You can create similar configutations to allow reference from inventory-nodes.yml
inventory_nodes:
# A sample configuration for a hardware type
hardware1:
# IPMI user account for Ironic inevntory resources
pm_user: "ADMIN"
# IPMI user password for Ironic inevntory resources
pm_password: "ADMIN"
# A sample configuration for a hardware type
hardware2:
# IPMI user account for Ironic inevntory resource
pm_user: "admin"
# IPMI user password for Ironic inevntory resource
pm_password: "admin"
# User defined sensitive data can be stored under 'other' key.
# Schema validation will only check if key,value format is used.
#other:
# mykey: myvalue