CONFIG-VARS.md

Valid Configuration Variables

Supported configuration variables are listed in the tables below. All variables can also be specified on the command line. Values specified on the command line will override values in configuration defaults files.

Table of Contents

• Valid Configuration Variables
  • Table of Contents
  • VMware vSphere/vCenter
    • Terraform terraform.tfvars file
      • General Items
      • vSphere
      • Systems
      • Kubernetes Cluster
      • Kubernetes Cluster Virtual IP Address
      • Kubernetes Load Balancer
      • Control Plane
      • Node Pools
      • Jump Server
      • NFS Server
      • PostgreSQL Server
  • Bare Metal
    • Ansible ansible-vars.yaml File
    • Labels/Taints
      • Labels
      • Taints
    • Ansible inventory file
  • Storage

VMware vSphere/vCenter

Terraform terraform.tfvars file

Terraform input variables can be set in the following ways:

• Individually, with the -var command line option.
• In variable definitions (.tfvars) files. We recommend using this method for most variables.
• As environment variables.

General Items

Name	Description	Type	Default	Notes
ansible_user	The user ID on your systems that Ansible uses to perform its tasks	string		Requires password-less sudo privileges.
ansible_password	The password for the user account on your systems that Ansible uses to perform its tasks	string
prefix	A prefix used in the names of all the resources created by this script	string
gateway	DNS gateway for vSphere/vCenter	string
netmask	Subnet mask for your network	number	16	The value must provide access from your machine's IP address to the gateway.

vSphere

Name	Description	Type	Default	Notes
vsphere_server	Name of the vSphere server	string
vsphere_datacenter	Name of the vSphere data center	string
vsphere_datastore	Name of the vSphere data store to use for the VMs	string
vsphere_resource_pool	Name of the vSphere resource pool to use for the VMs	string
vsphere_folder	Name of the vSphere folder to store the VMs	string
vsphere_template	Name of the VM template to clone to create VMs for the cluster	string
vsphere_network	Name of the network to to use for the VMs	string

Systems

Name	Description	Type	Default	Notes
system_ssh_keys_dir	Directory holding public keys to be used on each system	string	"~/.ssh"	These keys are applied to the operating system and root users of your machines.

Kubernetes Cluster

Name	Description	Type	Default	Notes
cluster_version	Kubernetes version	string	"1.29.7"	Valid values are listed here: SAS Viya platform Supported Kubernetes Versions.
cluster_cni	Kubernetes container network interface (CNI)	string	"calico"
cluster_cni_version	Kubernetes Container Network Interface (CNI) Version	string	"3.28.0"
cluster_cri	Kubernetes container runtime interface (CRI)	string	"containerd"
cluster_cri_version	Version of the CRI specifed by cluster_cri to be installed	string	"1.6.28"	Set as an empty string to use the latest upstream version from the Docker APT repository. Currently only containerd is supported, see the releases page for available versions
cluster_service_subnet	Kubernetes service subnet	string	"10.43.0.0/16"
cluster_pod_subnet	Kubernetes pod subnet	string	"10.42.0.0/16"
cluster_domain	Cluster domain suffix for DNS	string

Kubernetes Cluster Virtual IP Address

Name	Description	Type	Default	Notes
cluster_vip_version	kube-vip version	string	"0.7.1"	Currently kube-vip is the only supported Kubernetes virtual IP address. The minimum supported version is 0.5.7.
cluster_vip_ip	kube-vip IP address	string		IP address assigned to the FQDN value. You must access the cluster via the FQDN value supplied.
cluster_vip_fqdn	kube-vip DNS	string		FQDN used in the creation of the kubeconfig file, which is used to access the cluster.

Kubernetes Load Balancer

Name	Description	Type	Default	Notes
cluster_lb_type	Load balancer used in the cluster	string	"kube_vip"	Valid values: kube_vip, metallb. If you will be using SingleStore in your cluster it is required that you set this value to metallb.
cluster_lb_addresses	IP addresses used by the load balancer	list	null	Values change depending on the load balancer that is selected. This link provides more information about kube-vip load balancer addresses. This link provides more information about MetalLB load balancer addresses.

Control Plane

Name	Description	Type	Default	Notes
control_plane_ssh_key_name	Name for SSH key generated for the control plane	string	"cp_ssh"

Node Pools

Node pools are maps of objects. They represent information about each pool type, its physical hardware, and their labels and taints. Each node pool requires the following variables:

Name	Description	Type	Default	Notes
count	Number of nodes	number		Setting this variable creates nodes with dynamic IP addresses that are assigned from your network. It cannot be used with the ip_addresses field.
cpus	Number of CPU cores	number
memory	Memory in MB	number
os_disk	Size of operating system disk in GB	number		Operating system root disk.
misc_disk	Size of extra disks in GB	number		Miscellaneous disks that are used for the local-storage storage class. At this time, these disk are empty partitions created and attached to your VM once created. These disks are used for the local-storage storage class created for those applications that need local vs networked storage to run proficiently. Requirements and information on this can be found here
ip_addresses	List of static IP addresses to be used in creating control_plane nodes	list(string)		Setting this variable creates nodes with static IP addresses assigned from this list. It cannot be used if the count field is used.
node_taints		list(string)
node_labels		map(string)

There is no default type for SAS Viya platform node pools. However, examples that are based on SAS recommendations are listed in the example files. Below is a sample of a basic cluster node_pools definition that you could use in your terraform.tfvars file.

NOTE: These node pools are required for the node_pools variable:

• control_plane
• system

Sample node_pool entry:

# Cluster Node Pools config
#
#   Your node pools must contain at least 3 or more nodes.
#   The required node types are:
#
#   * control_plane - Having an odd number 3/5/7... ensures
#                     HA while using kube-vip
#   * system        - System node pool to run misc pods, etc
#   * cas           - CAS Nodes
#   * <node type>   - Any number of node types with unique names.
#                     These are typically: compute, stateful, and
#                     stateless. 
#
node_pools = {
  #
  # REQUIRED NODE TYPE - DO NOT REMOVE and DO NOT CHANGE THE NAME
  #                      Other variables may be altered
  control_plane = {
    cpus        = 2
    memory      = 4096
    os_disk     = 100
    ip_addresses = [
      "192.168.7.21",
      "192.168.7.22",
      "192.168.7.23",
    ]
    node_taints = []
    node_labels = {}
  },
  #
  # REQUIRED NODE TYPE - DO NOT REMOVE and DO NOT CHANGE THE NAME
  #                      Other variables may be altered
  system = {
    count       = 1
    cpus        = 8
    memory      = 16384
    os_disk     = 100
    node_taints = []
    node_labels = {
      "kubernetes.azure.com/mode" = "system" # REQUIRED LABEL - DO NOT REMOVE
    }
  },
  cas = {
    count       = 3
    cpus        = 16
    memory      = 131072
    os_disk     = 350
    misc_disks = [
      150,
      150,
    ]    
    node_taints = ["workload.sas.com/class=cas:NoSchedule"]
    node_labels = {
      "workload.sas.com/class" = "cas"
    }
  },
  compute = {
    count       = 1
    cpus        = 16
    memory      = 131072
    os_disk     = 100
    node_taints = ["workload.sas.com/class=compute:NoSchedule"]
    node_labels = {
      "workload.sas.com/class"        = "compute"
      "launcher.sas.com/prepullImage" = "sas-programming-environment"
    }
  },
  stateful = {
    count       = 1
    cpus        = 8
    memory      = 32768
    os_disk     = 100
    node_taints = ["workload.sas.com/class=stateful:NoSchedule"]
    node_labels = {
      "workload.sas.com/class" = "stateful"
    }
  },
  stateless = {
    count       = 2
    cpus        = 8
    memory      = 32768
    os_disk     = 100
    node_taints = ["workload.sas.com/class=stateless:NoSchedule"]
    node_labels = {
      "workload.sas.com/class" = "stateless"
    }
  },
  singlestore = {
    cpus    = 16
    memory  = 131072
    os_disk = 100
    misc_disks = [
      250,
      250,
      500,
      500,
    ]
    count       = 3
    node_taints = ["workload.sas.com/class=singlestore:NoSchedule"]
    node_labels = {
      "workload.sas.com/class" = "singlestore"
    }
  },
}

Jump Server

Name	Description	Type	Default	Notes
create_jump	Creation flag	bool	false
jump_num_cpu	# of CPU cores	number	4
jump_memory	Memory in MB	number	8092
jump_disk_size	Size of disk in GB	number	100
jump_ip	Static IP address for jump server	string

Sample:

# Jump server
create_jump    = true # Creation flag
jump_num_cpu   = 4    # 4 CPU cores
jump_memory    = 8092 # 8 GB
jump_disk_size = 100  # 100 GB
jump_ip        = ""   # Assigned values for static IP addresses

NFS Server

Name	Description	Type	Default	Notes
create_nfs	Creation flag	bool	false
nfs_num_cpu	# of CPUs	number	4
nfs_memory	Memory in MB	number	16384
nfs_disk_size	Size of disk in GB	number	400
nfs_ip	Static IP for NFS server	string

Sample:

# NFS server
create_nfs    = true  # Creation flag
nfs_num_cpu   = 8     # 8 CPUs
nfs_memory    = 16384 # 16 GB
nfs_disk_size = 500   # 500 GB
nfs_ip        = ""    # Assigned values for static IP addresses

PostgreSQL Servers

When setting up external database servers, you must provide information about those servers in the postgres_servers variable block. Each entry in the variable block represents a single database server.

This code only configures database servers. No databases are created during the infrastructure setup.

The variable has the following format:

postgres_servers = {
  default = {},
  ...
}

NOTE: The default = {} element is always required when creating external databases. This is the system's default database server.

Each server element, like foo = {}, can contain none, some, or all of the parameters listed. When naming the server element, it must follow a valid naming scheme, the name must consist of lower case alphanumeric characters or '-', and must start and end with an alphanumeric character.

Name	Description	Type	Default	Notes
server_num_cpu	# of CPUs	number	4
server_memory	Memory in MB	number	16385
server_disk_size	Size of disk in GB	number	128
server_ip	Static IP address for PostgreSQL server	string		This is a required field.
server_version	The version of the PostgreSQL server	string	"15"	Refer to the SAS Viya platform System Requirements for the supported versions of PostgreSQL for the SAS Viya platform.
server_ssl	Enable/disable SSL	string	"off"
server_ssl_cert_file	Path to the PostgreSQL SSL certificate file	string	""	If server_ssl is enabled and this variable is not defined, the system default SSL certificate is used.
server_ssl_key_file	Path to the PostgreSQL SSL key file	string	""	If server_ssl is enabled and this variable is not defined, the system default SSL key is used.
administrator_login	Admin user	string	"postgres"
administrator_password	Admin password	string	"my$up3rS3cretPassw0rd"
postgres_system_settings	Configure PostgreSQL system settings	list(object({}))	[{ name = "max_prepared_transactions", value = "1024" }, { name = "max_connections", value = "1024" }]	Currently max_prepared_transactions and max_connections are supported configuration values

NOTES:

1. If you enable server_ssl without defining either server_ssl_cert_file or server_ssl_key_file, the system's default SSL certificate and key are used instead. By default, on Ubuntu systems we create a copy of those files and name them ssl-cert-sas-${PG_HOST}.pem and ssl-cert-sas-${PG_HOST}.key.
   • The Ansible tasks that are performed include copying the certificate and key from the PostgreSQL VM into your local workspace directory.
2. If you are planning to use the viya4-deployment repository to perform a SAS Viya platform deployment where you have full-stack TLS configured, make sure that the V4_CFG_TLS_TRUSTED_CA_CERTS variable in the viya4-deployment ansible-vars.yaml file points to a directory that contains the server_ssl_cert_file.

Multiple SAS offerings require a second PostgreSQL instance referred to as SAS Common Data Store, or CDS PostgreSQL. For more information, see Common Customizations. A list of SAS offerings that require CDS PostgreSQL is provided in SAS Common Data Store Requirements. To create and configure an external CDS PostgreSQL instance in addition to the external platform PostgreSQL instance named default, specify cds-postgres as a second PostgreSQL instance, as shown in the example below.

Here is an example of the postgres_servers variable with the default server entry overriding only the administrator_password parameter and the cds-postgres entry overriding all the parameters:

postgres_servers = {
  default = {
    administrator_password       = "D0ntL00kTh1sWay"
    server_ip                    = "10.10.10.10"     # Assigned values for static IPs
  },
  cds-postgres = {
    server_num_cpu         = 4                       # 8 CPUs
    server_memory          = 16384                   # 16 GB
    server_disk_size       = 128                     # 256 GB
    server_ip              = "10.10.10.11"           # Assigned values for static IPs
    server_version         = 15                      # PostgreSQL version
    server_ssl             = "on"                    # SSL flag
    server_ssl_cert_file   = "./ssl_cert.pem"        # Path to the PostgreSQL SSL certificate file
    server_ssl_key_file    = "./ssl_cert.key"        # Path to the PostgreSQL SSL key file
    administrator_login    = "postgres"              # PostgreSQL admin user - CANNOT BE CHANGED
    administrator_password = "D0ntL00kTh1sWay"       # PostgreSQL admin user password
  }
}

Bare Metal

Ansible ansible-vars.yaml File

The following variables are used to describe the machine targets for the SAS Viya platform deployment.

Name	Description	Type	Default	Notes
ansible_user	The user ID on your systems that Ansible uses to perform its tasks	string		Requires password-less sudo privileges.
ansible_password	The password for the user account on your systems that Ansible uses to perform its tasks	string
vm_os	Operating system that is running on your machines	string	"ubuntu"	All machines in your cluster must have the same operating system.
vm_arch	Machine architecture	string	"amd64"
enable_cgroup_v2	Enable cgroup_v2 on your machines	bool	true
system_ssh_keys_dir	Directory location of public keys to be used on each system	string	"~/.ssh"
prefix	A prefix used in the names of all the resources created by this script	string
deployment_type	Type of deployment to be performed	string	"bare_metal"	Specify bare_metal or vsphere.
kubernetes_cluster_name	Cluster name	string	"{{ prefix }}-oss"	This item is auto-filled. ONLY change the prefix value described previously.
kubernetes_version	Kubernetes version	string	"1.29.7"	Valid values are listed here: Kubernetes Releases.
kubernetes_upgrade_allowed		bool	true	NOTE: Not currently used.
kubernetes_arch		string	"{{ vm_arch }}"	This item is auto-filled. ONLY change the vm_arch value described previously.
kubernetes_cni	Kubernetes Container Network Interface (CNI)	string	"calico"
kubernetes_cni_version	Kubernetes Container Network Interface (CNI) Version	string	"3.28.0"
kubernetes_cri	Kubernetes Container Runtime Interface (CRI)	string	"containerd"
kubernetes_cri_version	Version of the CRI specifed by kubernetes_cri to be installed	string	"1.6.28"	Set as an empty string to use the latest upstream version from the Docker APT repository. Currently only containerd is supported, see the releases page for available versions
kubernetes_service_subnet	Kubernetes service subnet	string	"10.43.0.0/16"
kubernetes_pod_subnet	Kubernetes pod subnet	string	"10.42.0.0/16"
kubernetes_vip_version	kube-vip version	string	"0.7.1"
kubernetes_vip_ip	kube-vip IP address	string
kubernetes_vip_fqdn	kube-vip DNS	string
kubernetes_loadbalancer	Load balancer provider	string	"kube_vip"	Choices are kube_vip or metallb.
kubernetes_loadbalancer_addresses	Load balancer IP addresses	string	[]	Values depend on the load balancer that is selected. This link provides more information about kube-vip load balancer addresses. This link provides more information about MetalLB load balancer addresses.
node_labels	Labels applied to nodes in your cluster	map(list(string))		See Labels/Taints below for more information.
node_taints	Taints applied to nodes in your cluster	map(list(string))		See Labels/Taints below for more information.
control_plane_ssh_key_name	Name for generated control plane SSH key	string	"cp_ssh"
jump_ip	Dynamic or static IP address that is assigned to your jump server	string
nfs_ip	Dynamic or static IP address that is assigned to your NFS server	string

NOTES:

• For bare metal systems in order to leverage the local-storage storage class created by these scripts you need to have empty partitions attached to your machines. These disks are used for the local-storage storage class created for those applications that need local vs networked storage to run proficiently. If they are not present this storage class cannot be used. Be sure to alter your manifest files to take advantage of this new storage class where its needed.

Labels/Taints

Labels and taints are applied to each node that matches the label or taint key name. The following examples outline these names and the labels/taints associated with those nodes.

Labels

To label your machines as specific nodes, add the following items to your ansible-vars.yaml file:

node_labels:
  cas:
    - workload.sas.com/class=cas
  compute:
    - launcher.sas.com/prepullImage=sas-programming-environment
    - workload.sas.com/class=compute
  stateful:
    - workload.sas.com/class=stateful
  stateless:
    - workload.sas.com/class=stateless
  singlestore:
    - workload.sas.com/class=singlestore
  system:
    - kubernetes.azure.com/mode=system

NOTE: The label on the system node pool is required if you DO NOT want SAS software to run on your system node(s).

Taints

To taint your machines as specific nodes, add the following items to your ansible-vars.yaml file:

node_taints:
  cas:
    - workload.sas.com/class=cas:NoSchedule
  compute:
    - workload.sas.com/class=compute:NoSchedule
  stateful:
    - workload.sas.com/class=stateful:NoSchedule
  stateless:
    - workload.sas.com/class=stateless:NoSchedule
  singlestore:
    - workload.sas.com/class=singlestore:NoSchedule

Ansible inventory file

The inventory file represents the machines that you will be using in your Kubernetes deployment of the SAS Viya platform. An example and information about this file can be found here.

The following variables are used to describe the machine targets for the SAS Viya platform deployment.

Note: If this file was generated as part of Terraform infrastructure creation through VMware vSphere/vCenter, these fields will be prepopulated with values you set in your terraform.tfvars file. For bare-metal, if you already have your machines, and are going to be populating the inventory file yourself, you can start with the sample-inventory file in this project to help you get started.

Name	Description	Type	Notes
postgres_server_name	Name of the PostgreSQL server	string
postgres_server_version	The version of the PostgreSQL server	string	Refer to the SAS Viya Platform Administration Guide for the supported versions of PostgreSQL for the SAS Viya platform.
postgres_server_ssl	Enable/disable SSL	string	Specify off or on
postgres_server_ssl_cert_file	Path to the PostgreSQL SSL certificate file	string	If postgres_server_ssl is enabled and this variable is not defined, the system default SSL certificate is used.
postgres_server_ssl_key_file	Path to the PostgreSQL SSL key file	string	If postgres_server_ssl is enabled and this variable is not defined, the system default SSL key is used.
postgres_administrator_login	PostgreSQL admin user	string
postgres_administrator_password	PostgreSQL admin password	string
postgres_system_setting_max_prepared_transactions	Allows you to configure the max_prepared_transactions setting for your PostgreSQL Server	string	If not defined in your inventory file, the value 1024 will automatically be configured as per the SAS Viya Platform PG tuning requirements
postgres_system_setting_max_connections	Allows you to configure the max_connections setting for your PostgreSQL Server	string	If not defined in your inventory file, the value 1024 will automatically be configured as per the SAS Viya Platform PG tuning requirements

Storage

An NFS server is set up by default. This is a required machine that is used as backing storage for the default storage class that is created. Information on setting up that machine is provided here.