Provisioning di modelli LBaaS di Avi Load Balancer

L'amministratore del cloud può sfruttare le risorse di VMware Avi Load Balancer per creare l'infrastruttura. È inoltre possibile utilizzare gli helper di allocazione per specificare la logica di allocazione per le risorse di Avi Load Balancer.

Prima di iniziare

Prima di iniziare con uno qualsiasi dei modelli di esempio, è necessario creare l'account cloud di Avi Load Balancer in Automation Assembler, creare un progetto e aggiungere la zona cloud. Per istruzioni, vedere Creazione di un account cloud di VMware Avi Load Balancer.

Modello LBaaS semplice

Questo modello di esempio mostra come creare un servizio virtuale, un pool e risorse IP virtuali con input minimi.

Tutte le risorse di Avi Load Balancer utilizzate in questo esempio fanno riferimento a costrutti di Avi Load Balancer che puntano al controller di Avi Load Balancer.

Nessuna macchina o server di cui viene eseguito il provisioning in questo modello. Si ottiene l'elenco dei server, che possono essere aggiornati continuamente.

Per impostazione predefinita, il modello è un pool vuoto, ma il modulo e l'input sono configurati in modo da poter immettere IP reali e viene creato un pool normale.

formatVersion: 1
name: ALB - LBaaS Demo - Simple
version: 1
inputs:
  port:
    type: integer
    title: Port
    description: Traffic sent to servers will use this destination server port unless overridden by the server's specific port attribute.
    default: 80
  servers:
    type: array
    title: Servers
    description: The pool directs load balanced traffic to this list of destination servers. The servers can be configured by IP address, name, network or via IP Address Group. Leaving 0.0.0.0 will result in an empty pool being created.
    items:
      type: string
    minItems: 0
    default:
      - 0.0.0.0
resources:
  VIP:
    type: Idem.AVILB.APPLICATIONS.VS_VIP
    properties:
      name: vip-${uuid()}
      account: avi-account
      description: Managed by Aria Automation
      vip:
        - vip_id: 0
          auto_allocate_ip: true
          ipam_network_subnet:
            network_ref: VMNetwork-PortGroup
  VirtualService:
    type: Idem.AVILB.APPLICATIONS.VIRTUAL_SERVICE
    properties:
      name: vs-${uuid()}
      account: avi-account
      description: Managed by Aria Automation
      services:
        - port: ${input.port}
      pool_ref: ${resource.Pool.name}
      vsvip_ref: ${resource.VIP.name}
  Pool:
    type: Idem.AVILB.APPLICATIONS.POOL
    properties:
      name: pool-${uuid()}
      account: avi-account
      description: Managed by Aria Automation
      default_server_port: ${input.port}
      servers: '${input.servers[0] == "0.0.0.0" ? null :map_by(input.servers, address => {"ip": {"addr": address, "type" : "V4"}})}'

Modello LBaaS avanzato

Questo modello di esempio mostra come è possibile creare un servizio virtuale, un pool e risorse IP virtuali connesse a una rete esistente insieme a un helper di allocazione della zona cloud. Il pool è inoltre configurato con il monitoraggio dell'integrità HTTP del sistema.

formatVersion: 1
name: ALB - LBaaS Demo
version: 1
inputs:
  env:
    type: string
    title: Environment
    description: Select Dev or Prod ALB environment
    default: env:dev
    oneOf:
      - title: Dev ALB
        const: env:dev
      - title: Prod ALB
        const: env:prod
  port:
    type: integer
    title: Port
    description: Input the server ip's to add to the pool.
    default: 80
  servers:
    type: array
    title: Servers
    description: Input the server ip's to add to the pool. Leaving 0.0.0.0 will result in an empty pool being created.
    items:
      type: string
    minItems: 0
    default:
      - 0.0.0.0
resources:
  Allocations_CloudZone_1:
    type: Allocations.CloudZone
    properties:
      accountType: avilb
      constraints:
        - tag: ${input.env}
  VIP:
    type: Idem.AVILB.APPLICATIONS.VS_VIP
    properties:
      name: vip-${uuid()}
      account: ${resource.Allocations_CloudZone_1.selectedCloudAccount.name}
      description: Managed by Aria Automation
      vip:
        - vip_id: 0
          auto_allocate_ip: true
          ipam_network_subnet:
            network_ref: ${resource.Cloud_vSphere_Network_1.resourceName}
  VirtualService:
    type: Idem.AVILB.APPLICATIONS.VIRTUAL_SERVICE
    properties:
      name: vs-${uuid()}
      account: ${resource.Allocations_CloudZone_1.selectedCloudAccount.name}
      description: Managed by Aria Automation
      services:
        - port: ${input.port}
      pool_ref: ${resource.Pool.name}
      vsvip_ref: ${resource.VIP.name}
  Pool:
    type: Idem.AVILB.APPLICATIONS.POOL
    properties:
      name: pool-${uuid()}
      account: ${resource.Allocations_CloudZone_1.selectedCloudAccount.name}
      description: Managed by Aria Automation
      default_server_port: ${input.port}
      servers: '${input.servers[0] == "0.0.0.0" ? null :map_by(input.servers, address => {"ip": {"addr": address, "type" : "V4"}})}'
      health_monitor_refs:
        - System-HTTP
  Cloud_vSphere_Network_1:
    type: Cloud.vSphere.Network
    properties:
      networkType: existing
      constraints:
        - tag: net:vm

Modello di server Web semplice

Questo modello di esempio mostra come creare un servizio virtuale, un pool, un IP virtuale e due risorse delle macchine virtuali del server Web connesse a una rete esistente. Il pool è configurato per monitorare la porta 80 utilizzando l'algoritmo Round Robin.

formatVersion: 1
name: ALB - Web Servers
version: 1
inputs:
  env:
    type: string
    title: Environment
    description: Select Dev or Prod ALB environment
    default: env:dev
    oneOf:
      - title: Dev ALB
        const: env:dev
      - title: Prod ALB
        const: env:prod
  clusterSize:
    type: string
    enum:
      - small
      - medium
      - large
    default: small
    title: Web Server Cluster Size
    description: Web Server Cluster Size. Small creates one web server and no ALB. Medium creates 2 web servers and a ALB. Large creates 4 web servers and a ALB.
  username:
    type: string
    title: Username
    default: demouser
  password:
    type: string
    title: Password
    encrypted: false
    default: VMware1!
  port:
    type: integer
    title: Port
    default: 80
  health_monitor:
    type: array
    items:
      type: string
    title: Health Monitors
    default:
      - System-HTTP
    minItems: 1
    maxItems: 10
  lb_algorithm:
    type: string
    title: Load Balancer Algorithm
    description: The load balancing algorithm will pick a server within the pool's list of available servers.
    readOnly: false
    default: LB_ALGORITHM_LEAST_CONNECTIONS
    oneOf:
      - title: Least Connections
        const: LB_ALGORITHM_LEAST_CONNECTIONS
      - title: Round Robin
        const: LB_ALGORITHM_ROUND_ROBIN
      - title: Fastest Response
        const: LB_ALGORITHM_FASTEST_RESPONSE
      - title: Consistent Hash
        const: LB_ALGORITHM_CONSISTENT_HASH
      - title: Least Load
        const: LB_ALGORITHM_LEAST_LOAD
      - title: Fewest Servers
        const: LB_ALGORITHM_FEWEST_SERVERS
      - title: Random
        const: LB_ALGORITHM_RANDOM
      - title: Fewest Tasks
        const: LB_ALGORITHM_FEWEST_TASKS
      - title: Core Affinity
        const: LB_ALGORITHM_CORE_AFFINITY
resources:
  ALB_HEALTH_MONITOR:
    type: Idem.AVILB.PROFILES.HEALTH_MONITOR
    properties:
      name: web-health-${uuid()}
      description: Managed by Aria Assembler
      type: ${input.health_monitor}
      account: ${resource.Allocations_CloudZone_1.selectedCloudAccount.name}
  Allocations_CloudZone_1:
    type: Allocations.CloudZone
    properties:
      accountType: avilb
      constraints:
        - tag: ${input.env}
  ALB_APPLICATION_PERSISTENCE_PROFILE:
    type: Idem.AVILB.PROFILES.APPLICATION_PERSISTENCE_PROFILE
    properties:
      name: apache-appprofile-${uuid()}
      account: ${resource.Allocations_CloudZone_1.selectedCloudAccount.name}
      description: Managed by Aria Automation
      persistence_type: PERSISTENCE_TYPE_CLIENT_IP_ADDRESS
      server_hm_down_recovery: HM_DOWN_PICK_NEW_SERVER
      ip_persistence_profile:
        ip_mask: 24
        ip_persistent_timeout: 20
  SecurityGroup:
    type: Cloud.SecurityGroup
    properties:
      constraints:
        - tag: ${input.env}
      securityGroupType: existing
  VIP:
    type: Idem.AVILB.APPLICATIONS.VS_VIP
    properties:
      name: web-vip-${uuid()}
      description: Managed by Aria Automation
      count: ${input.clusterSize == "small" ? 0:1}
      account: ${resource.Allocations_CloudZone_1.selectedCloudAccount.name}
      vip:
        - vip_id: 0
          auto_allocate_ip: true
          ipam_network_subnet:
            network_ref: ${resource.Cloud_vSphere_Network_1.resourceName}
  VirtualService:
    type: Idem.AVILB.APPLICATIONS.VIRTUAL_SERVICE
    properties:
      name: web-vs-${uuid()}
      description: Managed by Aria Assembler
      count: ${input.clusterSize == "small" ? 0:1}
      account: ${resource.Allocations_CloudZone_1.selectedCloudAccount.name}
      cloud_type: CLOUD_VCENTER
      services:
        - port: ${input.port}
      vsvip_ref: ${resource.VIP[0].name}
      pool_ref: ${resource.Pool[0].name}
  Pool:
    type: Idem.AVILB.APPLICATIONS.POOL
    properties:
      name: web-pool-${uuid()}
      description: Managed by Aria Automation
      count: ${input.clusterSize == "small" ? 0:1}
      account: ${resource.Allocations_CloudZone_1.selectedCloudAccount.name}
      default_server_port: ${input.port}
      health_monitor_refs:
        - ${resource.ALB_HEALTH_MONITOR.name}
      lb_algorithm: ${input.lb_algorithm}
      application_persistence_profile_ref: ${resource.ALB_APPLICATION_PERSISTENCE_PROFILE.name}
      servers: '${map_by(resource.Web_Server[*].address, address => {"ip": {"addr": address, "type" : "V4"}})}'
  Web_Server:
    type: Cloud.vSphere.Machine
    properties:
      name: web
      count: 2
      flavor: small
      imageRef: https://cloud-images.ubuntu.com/releases/18.04/release/ubuntu-18.04-server-cloudimg-amd64.ova
      cloudConfig: |
        packages:
          - apache2
          - open-vm-tools

        runcmd:
          ### update the top of the web page to the web servers resource name
          - sed -i 's/Apache2 Ubuntu Default Page/${self.resourceName}/g' /var/www/html/index.html
          
          ### Restart services
          - systemctl reload apache2
          - systemctl restart apache2
          
          ### Log completion
          - echo 'Cloud-init is done!' >> /tmp/finished.txt
      networks:
        - network: ${resource.Cloud_vSphere_Network_1.id}
          securityGroups:
            - ${resource.SecurityGroup.id}
  Cloud_vSphere_Network_1:
    type: Cloud.vSphere.Network
    properties:
      networkType: existing
      constraints:
        - tag: net:vm

Modello server Web avanzato

Questo modello mostra come è possibile creare un servizio virtuale, un pool e risorse IP virtuali connesse a una rete esistente insieme a un helper di allocazione della zona cloud. Inoltre, nel pool vengono creati e configurati un monitoraggio dell'integrità e un profilo di persistenza dell'applicazione. Infine, i gruppi di sicurezza vengono enumerati in base al tag e i server Web delle macchine virtuali vengono aggiunti a tale gruppo.

formatVersion: 1
name: ALB - Web Servers
version: 1
inputs:
  env:
    type: string
    title: Environment
    description: Select Dev or Prod ALB environment
    default: env:dev
    oneOf:
      - title: Dev ALB
        const: env:dev
      - title: Prod ALB
        const: env:prod
  clusterSize:
    type: string
    enum:
      - small
      - medium
      - large
    default: small
    title: Web Server Cluster Size
    description: Web Server Cluster Size. Small creates one web server and no ALB. Medium creates 2 web servers and a ALB. Large creates 4 web servers and a ALB.
  username:
    type: string
    title: Username
    default: demouser
  password:
    type: string
    title: Password
    encrypted: false
    default: VMware1!
  port:
    type: integer
    title: Port
    default: 80
  health_monitor:
    type: array
    items:
      type: string
    title: Health Monitors
    default:
      - System-HTTP
    minItems: 1
    maxItems: 10
  lb_algorithm:
    type: string
    title: Load Balancer Algorithm
    description: The load balancing algorithm will pick a server within the pool's list of available servers.
    readOnly: false
    default: LB_ALGORITHM_LEAST_CONNECTIONS
    oneOf:
      - title: Least Connections
        const: LB_ALGORITHM_LEAST_CONNECTIONS
      - title: Round Robin
        const: LB_ALGORITHM_ROUND_ROBIN
      - title: Fastest Response
        const: LB_ALGORITHM_FASTEST_RESPONSE
      - title: Consistent Hash
        const: LB_ALGORITHM_CONSISTENT_HASH
      - title: Least Load
        const: LB_ALGORITHM_LEAST_LOAD
      - title: Fewest Servers
        const: LB_ALGORITHM_FEWEST_SERVERS
      - title: Random
        const: LB_ALGORITHM_RANDOM
      - title: Fewest Tasks
        const: LB_ALGORITHM_FEWEST_TASKS
      - title: Core Affinity
        const: LB_ALGORITHM_CORE_AFFINITY
resources:
  ALB_HEALTH_MONITOR:
    type: Idem.AVILB.PROFILES.HEALTH_MONITOR
    properties:
      name: web-health-${uuid()}
      description: Managed by Aria Assembler
      type: ${input.health_monitor}
      account: ${resource.Allocations_CloudZone_1.selectedCloudAccount.name}
  Allocations_CloudZone_1:
    type: Allocations.CloudZone
    properties:
      accountType: avilb
      constraints:
        - tag: ${input.env}
  ALB_APPLICATION_PERSISTENCE_PROFILE:
    type: Idem.AVILB.PROFILES.APPLICATION_PERSISTENCE_PROFILE
    properties:
      name: apache-appprofile-${uuid()}
      account: ${resource.Allocations_CloudZone_1.selectedCloudAccount.name}
      description: Managed by Aria Automation
      persistence_type: PERSISTENCE_TYPE_CLIENT_IP_ADDRESS
      server_hm_down_recovery: HM_DOWN_PICK_NEW_SERVER
      ip_persistence_profile:
        ip_mask: 24
        ip_persistent_timeout: 20
  SecurityGroup:
    type: Cloud.SecurityGroup
    properties:
      constraints:
        - tag: ${input.env}
      securityGroupType: existing
  VIP:
    type: Idem.AVILB.APPLICATIONS.VS_VIP
    properties:
      name: web-vip-${uuid()}
      description: Managed by Aria Automation
      count: ${input.clusterSize == "small" ? 0:1}
      account: ${resource.Allocations_CloudZone_1.selectedCloudAccount.name}
      vip:
        - vip_id: 0
          auto_allocate_ip: true
          ipam_network_subnet:
            network_ref: ${resource.Cloud_vSphere_Network_1.resourceName}
  VirtualService:
    type: Idem.AVILB.APPLICATIONS.VIRTUAL_SERVICE
    properties:
      name: web-vs-${uuid()}
      description: Managed by Aria Assembler
      count: ${input.clusterSize == "small" ? 0:1}
      account: ${resource.Allocations_CloudZone_1.selectedCloudAccount.name}
      cloud_type: CLOUD_VCENTER
      services:
        - port: ${input.port}
      vsvip_ref: ${resource.VIP[0].name}
      pool_ref: ${resource.Pool[0].name}
  Pool:
    type: Idem.AVILB.APPLICATIONS.POOL
    properties:
      name: web-pool-${uuid()}
      description: Managed by Aria Automation
      count: ${input.clusterSize == "small" ? 0:1}
      account: ${resource.Allocations_CloudZone_1.selectedCloudAccount.name}
      default_server_port: ${input.port}
      health_monitor_refs:
        - ${resource.ALB_HEALTH_MONITOR.name}
      lb_algorithm: ${input.lb_algorithm}
      application_persistence_profile_ref: ${resource.ALB_APPLICATION_PERSISTENCE_PROFILE.name}
      servers: '${map_by(resource.Web_Server[*].address, address => {"ip": {"addr": address, "type" : "V4"}})}'
  Web_Server:
    type: Cloud.vSphere.Machine
    properties:
      name: web
      count: 2
      flavor: small
      imageRef: https://cloud-images.ubuntu.com/releases/18.04/release/ubuntu-18.04-server-cloudimg-amd64.ova
      cloudConfig: |
        packages:
          - apache2
          - open-vm-tools

        runcmd:
          ### update the top of the web page to the web servers resource name
          - sed -i 's/Apache2 Ubuntu Default Page/${self.resourceName}/g' /var/www/html/index.html
          
          ### Restart services
          - systemctl reload apache2
          - systemctl restart apache2
          
          ### Log completion
          - echo 'Cloud-init is done!' >> /tmp/finished.txt
      networks:
        - network: ${resource.Cloud_vSphere_Network_1.id}
          securityGroups:
            - ${resource.SecurityGroup.id}
  Cloud_vSphere_Network_1:
    type: Cloud.vSphere.Network
    properties:
      networkType: existing
      constraints:
        - tag: net:vm