ViPR 2.2 - Deploy the Compute Image Server Using the ViPR REST API

Table of Contents


ViPR system administrators can learn the ViPR configuration requirements for a deploying a compute image server for ViPR, how to deploy the ViPR compute image server, or create a custom compute image server, and how to configure a compute image server once it has been deployed for ViPR.

A compute image server is required by ViPR to deploy the compute images when you run a VCE Vblock™ System provisioning service, which performs operating system installation from ViPR.

For information about ViPR support for Vblock system, refer to: Understanding ViPR Support for VCE Vblock Systems.

Back to Top

ViPR network requirements for the compute image server

A network administrator must configure two networks before deploying the compute image server for ViPR.

Management Network

The management network is required for communication between ViPR, and the compute image server.

Private OS Install Network

The OS Install Network is a private network for operating system (OS) installation. The OS installation Network is used by ViPR during provisioning, for communication between the hosts, and the ViPR compute image server. Once the hosts, and ViPR compute image server are connected over the OS Install Network, the operating system installation is then performed over the OS Install Network. Once installation is complete, the OS Install Network is removed from the hosts.

The Private OS Install Network must be:

Back to Top

Deploy the compute image server

ViPR provides a compute image server OVF template that you can deploy, or you can create a custom compute image server, which adheres to the ViPR compute image server requirements.

Back to Top

Deploy the ViPR Compute Image Server OVF file

ViPR is provided with a compute image server OVF template that you can deploy as a VM.

Before you begin

  • You need access to the computer image server deployment file, OSImageServer.x86_64-, where xx is the compute image server build version number. You can get the file from the ViPR download page on
  • You need credentials to log in to vSphere for the vCenter Server where you are deploying the compute image server.
  • During deployment you will need to provide:
    • Management Network
    • OS Install Network
    • A fully-qualified hostname for the compute image server
    • IPv4 address for the management network interface
    • IPv4 address for the private OS install network interfaces
    • Netmasks and gateway addresses for both the Management Network
    • One or more DNS server IPv4 addresses
    • Search domain
    • Time zone of the compute image server


  1. Download the compute image server image from the ViPR product page to a temporary directory.
  2. Start the vSphere Client and log in to the vCenter Server on which you will be deploying the virtual appliance.
  3. From the File menu, select Deploy OVF Template.
  4. Browse to and select the ViPR compute image server file located in the temporary directory you created earlier.
  5. On the OVF Template Details page, review the details about the appliance.
  6. Accept the End User License Agreement.
  7. Specify a name and location for the appliance.
  8. Select the host or cluster on which to run the virtual appliance.
  9. If resource pools are configured, select one.
  10. If more than one datastore is attached to the ESX Server, select the datastore for your appliance.
  11. Select a disk format: Thick Provision Lazy Zeroed, Thick Provision Eager Zeroed, or Thin Provision.
  12. On the Network Mapping page, specify a destination network for the Management Network and for the private OS Install Network.
  13. Enter the values for the properties:
    Property Description
    Appliance fully qualified name FQDN of the image server host name.
    Management Network IP Address IPv4 address for the Management Network interface
    Management Network Netmask IPv4 netmask for the Management Network interface
    Management Network Gateway IPv4 address for the Management Network gateway
    Private OS Install Network IP address IPv4 address for the OS Install Network interface
    DNS Server(s) IPv4 addresses for one or more DNS servers
    Search Domain(s) One or more domains for directing searches.
    Time Zone Select the time zone where the image server resides.
  14. Power on the VM.
Back to Top

Requirements to create a custom compute image server

If you choose to create a custom compute image server for the ViPR compute images, the image server must be configured as follows:

  • Compute Image Server must run on Linux OS
  • Compute Image Server must have 2 vNICs
    • Management Network vNIC
    • OS Install Network vNIC

      OS Install vNIC netmask must be for example:

  • Compute Image Server must have DHCP server
    • DHCP server must be listening on the OS Install Network
    • DHCP response must contain "next-server" option with its own OS Install Network IP and "filename" option set to "/pxelinux.0"
    • Suggested DHCP version: Internet Systems Consortium DHCP Server 4.2 as demonstrated in the following example. Note the next-server, and filename.
      ddns-update-style none;
      ignore client-updates;
      subnet netmask {
          option subnet-mask;
          option time-offset      -18000; # Eastern Standard Time
      # --- DHCP pool configuration
          default-lease-time 3600;
          max-lease-time 7200;
      # --- TFTP/PXE configuration
          filename "/pxelinux.0";
      # listen on eth1 only
  • Compute Image Server must have TFTP server
    service tftp
            socket_type             = dgram
            protocol                = udp
            wait                    = yes
            user                    = root
            server                  = /usr/sbin/in.tftpd
            server_args             = -s /opt/tftpboot/ -vvvvvvv
            disable                 = no
            per_source              = 11
            cps                     = 100 2
            flags                   = IPv4
  • SSH acces
    • User account must have permissions to write to TFTPBOOT directory.
    • User account must have permissions to execute mount/umount commands
  • Python
  • Enough disk space to store multiple OS images - at least 50 GB is recommended
  • No firewall blocking standard SSH, DHCP, TFTP ports and HTTP on 44491 (or a custom port chosen for HTTP).
  • wget binary must be installed.

Back to Top

Configure the compute image server in ViPR

Once the compute image server has been installed, you must configure it in ViPR, before you can add the compute images to ViPR.

Before you begin

After you make changes to these properties by submitting a request using the REST API with these parameters in the payload, as shown below, you must initiate a reboot of the ViPR nodes.
Note Image
Rebooting the ViPR nodes may disrupt ViPR processes that are currently running.

In this example, the default values of the TFTPBOOT directory and the os install timeout will not be changed.


  1. Configure the compute image server using PUT /config/properties.
    PUT https://<ViPR_VIP>:4443/config/properties
    Content-Type: application/xml
    HTTP 202 Accepted
    Content-Type: application/xml
    <?xml version="1.0" encoding="UTF-8" standalone="yes"?>
         <link rel="self" href="/upgrade/cluster-state"/>
  2. Reboot the ViPR nodes.
  3. You can verify the changes using GET /config/properties.
Back to Top