Quick install config file

2023-06-12 14:18:53 -07:00 · 2023-06-12 14:18:53 -07:00 · 60aa3d624e
--- a/docs/tutorials/quick_install.md
+++ b/docs/tutorials/quick_install.md
@ -55,7 +55,7 @@ Duration: 50 minutes
 In this exercise, you will learn how to deploy **Az-hop** utilizing a Linux machine.

 ### Task 1: Set up WSL2
-
+#
 To install **Az-hop** we require to have a Linux machine that could support the toolchain to use during the installation, it is preferred to use Ubuntu 20.04+. So, if you are running a Windows desktop you should use WSL2 with an Ubuntu 20.04 environment.

 Run the following commands to install WSL2 (you will only need to do this once on your local device) in your Windows computer.
--- a/examples/qinstall-config.yml
+++ b/examples/qinstall-config.yml
@ -0,0 +1,264 @@
+---
+# yaml-language-server: $schema=config.schema.json
+#
+# name of the cluster
+project_name: az-hop
+# azure location name as returned by the command : az account list-locations -o table
+location: westus2
+# Name of the resource group to create all resources
+resource_group: azhop-qinstall
+# If using an existing resource group set to true. Default is false
+# When using an existing resource group make sure the location match the one of the existing resource group
+use_existing_rg: false
+# If set to true, will disable telemetry for azhop. See https://azure.github.io/az-hop/deploy/telemetry.html.
+#optout_telemetry: true
+
+tags:
+  env: dev
+  project: azhop-quick-install
+# Define an Azure Netapp Files (ANF) account, single pool and volume
+# If not present, assume that there is an existing NFS share for the users home directory
+anf:
+  create: true
+  # Size of the ANF pool and unique volume (min: 4TB, max: 100TB)
+  homefs_size_tb: 4
+  # Service level of the ANF volume, can be: Standard, Premium, Ultra
+  homefs_service_level: Standard
+  # dual protocol
+  dual_protocol: false # true to enable SMB support. false by default
+  # If alerting is enabled, this value will be used to determine when to trigger alerts 
+  alert_threshold: 80 # alert when ANF volume reaches this threshold
+
+# For small deployments you can use Azure Files instead of ANF for the home directory
+azurefiles:
+  create: false
+  size_gb: 1024
+
+# These mounts will be listed in the Files menu of the OnDemand portal and automatically mounted on all compute nodes and remote desktop nodes
+mounts:
+  # mount settings for the user home directory
+  home: # This home name can't be changed
+    type: anf # anf or azurefiles, default to anf. One of the two should be defined in order to mount the home directory
+    mountpoint: /anfhome # /sharedhome for example
+    server: '{{anf_home_ip}}' # Specify an existing NFS server name or IP, when using the ANF built in use '{{anf_home_ip}}'
+    export: '{{anf_home_path}}' # Specify an existing NFS export directory, when using the ANF built in use '{{anf_home_path}}'
+    options: '{{anf_home_opts}}' # Specify the mount options. Default to rw,hard,rsize=262144,wsize=262144,vers=3,tcp,_netdev
+
+# name of the admin account
+admin_user: hpcadmin
+# List of identities (object ids) to grant read access to az-hop key vault (optional)
+# key_vault_readers:
+# Network
+network:
+  # Create Network and Application Security Rules, true by default, false when using an existing VNET if not specified
+  create_nsg: true
+  vnet:
+    name: hpcvnet # Optional - default to hpcvnet
+    #id: # If a vnet id is set then no network will be created and the provided vnet will be used
+    address_space: "10.0.0.0/23" 
+    # When using an existing VNET, only the subnet names will be used and not the adress_prefixes
+    subnets: # all subnets are optionals
+    # name values can be used to rename the default to specific names, address_prefixes to change the IP ranges to be used
+    # All values below are the default values
+      frontend:
+        name: frontend
+        address_prefixes: "10.0.0.0/29"
+        create: true # create the subnet if true. default to true when not specified, default to false if using an existing VNET when not specified
+      admin:
+        name: admin
+        address_prefixes: "10.0.0.16/28"
+        create: true
+      netapp:
+        name: netapp
+        address_prefixes: "10.0.0.32/28"
+        create: true
+      ad:
+        name: ad
+        address_prefixes: "10.0.0.8/29"
+        create: true
+      compute:
+        name: compute
+        address_prefixes: "10.0.1.0/24"
+        create: true
+  # Specify the Application Security Groups mapping if already existing
+
+# When working in a locked down network, uncomment and fill out this section
+locked_down_network:
+  enforce: false
+#   grant_access_from: [a.b.c.d] # Array of CIDR to grant access from, see https://docs.microsoft.com/en-us/azure/storage/common/storage-network-security?tabs=azure-portal#grant-access-from-an-internet-ip-range
+  public_ip: true # Enable public IP creation for Jumpbox, OnDemand and create images. Default to true
+
+# Base image configuration. Can be either an image reference or an image_id from the image registry or a custom managed image
+linux_base_image: "OpenLogic:CentOS:7_9-gen2:latest" # publisher:offer:sku:version or image_id
+# linux image plan if required, format is publisher:product:name
+#linux_base_plan:
+windows_base_image: "MicrosoftWindowsServer:WindowsServer:2019-Datacenter-smalldisk:latest" # publisher:offer:sku:version or image_id
+
+# Jumpbox VM configuration, only needed when deploying thru a public IP and without a configured deployer VM
+jumpbox:
+  vm_size: Standard_B2ms
+  # SSH port under which the jumpbox SSH server listens on the public IP. Default to 22
+  # Change this to, e.g., 2222, if security policies (like "zero trust") in your tenant automatically block access to port 22 from the internet
+  #ssh_port: 2222
+# Active directory VM configuration
+ad:
+  vm_size: Standard_B2ms
+  hybrid_benefit: false # Enable hybrid benefit for AD, default to false
+  high_availability: false # Build AD in High Availability mode (2 Domain Controlers) - default to false
+# On demand VM configuration
+ondemand:
+  vm_size: Standard_D4s_v5
+  #fqdn: azhop.foo.com # When provided it will be used for the certificate server name
+  generate_certificate: true # Generate an SSL certificate for the OnDemand portal. Default to true
+# Grafana VM configuration
+grafana:
+  vm_size: Standard_B2ms
+# Guacamole VM configuration
+guacamole:
+  vm_size: Standard_B2ms
+# Scheduler VM configuration
+scheduler:
+  vm_size: Standard_B2ms
+# CycleCloud VM configuration
+cyclecloud:
+  vm_size: Standard_B2ms
+  # version: 8.3.0-3062 # to specify a specific version, see https://packages.microsoft.com/yumrepos/cyclecloud/
+
+# List of users to be created on this environment
+users:
+  - { name: clusteradmin, uid: 10001, groups: [5001, 5002] }
+  - { name: hpcuser, uid: 10002 }
+
+usergroups:
+# These groups can’t be changed
+  - name: Domain Users # All users will be added to this one by default
+    gid: 5000
+  - name: az-hop-admins
+    gid: 5001
+    description: "For users with azhop admin privileges"
+  - name: az-hop-localadmins
+    gid: 5002
+    description: "For users with sudo right or local admin right on nodes"
+
+# scheduler to be installed and configured (openpbs, slurm)
+queue_manager: openpbs
+
+# Specific SLURM configuration
+slurm:
+  # Enable SLURM accounting, this will create a SLURM accounting database in a managed MariaDB server instance
+  accounting_enabled: false
+  # SLURM version to install. Currently supported: only 20.11.9 and 22.05.3.
+  # Other versions can be installed by building from source (See build_rpms setting in the slurmserver role)
+  slurm_version: 20.11.9
+  # Name of the SLURM cluster for accounting (optional, default to 'slurm')
+  # WARNING: changing this value on a running cluster will cause slurmctld to fail to start. This is a
+  # safety check to prevent accounting errors. To override, remove /var/spool/slurmd/clustername
+  cluster_name: slurm_azhop
+
+enroot:
+  enroot_version: 3.4.1
+
+# Authentication configuration for accessing the az-hop portal
+# Default is basic authentication. For oidc authentication you have to specify the following values
+# The OIDCClient secret need to be stored as a secret named <oidc-client-id>-password in the keyvault used by az-hop
+authentication:
+  httpd_auth: basic # oidc or basic
+
+image_gallery:
+  create: true # Create the shared image gallery to store custom images
+
+# Autoscale default settings for all queues, can be overriden on each queue depending on the VM type if needed
+autoscale:
+  idle_timeout: 1800 # Idle time in seconds before shutting down VMs - default to 1800 like in CycleCloud
+
+# List of queues (node arrays in Cycle) to be defined
+# don't use queue names longer than 8 characters in order to leave space for node suffix, as hostnames are limited to 15 chars due to domain join and NETBIOS constraints.
+queues:
+  - name: execute # name of the Cycle Cloud node array
+    # Azure VM Instance type
+    vm_size: Standard_F2s_v2
+    # maximum number of cores that can be instanciated
+    max_core_count: 1024
+    # Use the pre-built azhop image from the marketplace
+    image: azhpc:azhop-compute:centos-7_9:latest
+    # Set to true if AccelNet need to be enabled. false is the default value
+    EnableAcceleratedNetworking: false
+    # spot instance support. Default is false
+    spot: false
+    # Set to false to disable creation of placement groups (for SLURM only). Default is true
+    ColocateNodes: false
+    # Specific idle time in seconds before shutting down VMs, make sure it's lower than autoscale.idle_timeout
+    idle_timeout: 300
+    # Set the max number of vm's in a VMSS; requires additional limit raise through support ticket for >100; 
+    # 100 is default value; lower numbers will improve scaling for single node jobs or jobs with small number of nodes
+    MaxScaleSetSize: 100
+  - name: hc44rs
+    vm_size: Standard_HC44rs
+    max_core_count: 440
+    image: azhpc:azhop-compute:centos-7_9:latest
+    spot: true
+    EnableAcceleratedNetworking: true
+  - name: hb120v2
+    vm_size: Standard_HB120rs_v2
+    max_core_count: 1200
+    image: azhpc:azhop-compute:centos-7_9:latest
+    spot: true
+    EnableAcceleratedNetworking: true
+  - name: hb120v3
+    vm_size: Standard_HB120rs_v3
+    max_core_count: 1200
+    image: azhpc:azhop-compute:centos-7_9:latest
+    spot: true
+    EnableAcceleratedNetworking: true
+    # Queue dedicated to GPU remote viz nodes. This name is fixed and can't be changed
+  - name: viz3d
+    vm_size: Standard_NV12s_v3
+    max_core_count: 48
+    # Use the pre-built azhop image from the marketplace
+    image: azhpc:azhop-desktop:centos-7_9:latest
+    # Use this image ID when building your own custom images
+    #image: /subscriptions/{{subscription_id}}/resourceGroups/{{resource_group}}/providers/Microsoft.Compute/galleries/{{sig_name}}/images/azhop-centos79-desktop3d/latest
+    ColocateNodes: false
+    spot: false
+    EnableAcceleratedNetworking: true
+    max_hours: 12 # Maximum session duration
+    min_hours: 1 # Minimum session duration - 0 is infinite
+    # Queue dedicated to share GPU remote viz nodes. This name is fixed and can't be changed
+  - name: largeviz3d
+    vm_size: Standard_NV48s_v3
+    max_core_count: 96
+    image: azhpc:azhop-desktop:centos-7_9:latest
+    ColocateNodes: false
+    EnableAcceleratedNetworking: true
+    spot: false
+    max_hours: 12
+    min_hours: 1
+    # Queue dedicated to non GPU remote viz nodes. This name is fixed and can't be changed
+  - name: viz
+    vm_size: Standard_D8s_v5
+    max_core_count: 200
+    image: azhpc:azhop-desktop:centos-7_9:latest
+    ColocateNodes: false
+    spot: false
+    EnableAcceleratedNetworking: true
+    max_hours: 12
+    min_hours: 1
+
+# Application settings
+applications:
+  bc_codeserver:
+    enabled: true
+  bc_jupyter:
+    enabled: true
+  bc_amlsdk:
+    enabled: false
+  bc_rstudio:
+    enabled: false
+  bc_ansys_workbench:
+    enabled: false
+  bc_vmd:
+    enabled: false
+  bc_paraview:
+    enabled: true
+  bc_vizer:
+    enabled: false