Page History

...

• Configure the management node to check in with the database
  • Management Nodes -> Edit Management Node Information -> Add
    • Hostname: localhost
    • IP Address: 127.0.0.1
    • Owner: admin
    • State: available
    • Predictive Loading Module: Predictive Loading Level 0 Module
    • Check-in Interval: 5
    • Install Path: /install
    • End Node SSH Identitiy Key Files: /etc/vcl/vcl.key
    • SSH Port for this node: 22
    • Enable Image Library: no
• Add VM host computer:
  • Manage Computers -> Edit Computer Information -> Submit -> Add
    • Hosthame: localvmhost
    • IP Address: 192.168.0.1 (NOTE: there is currently a bug restricting the size of the field, so hopefully one octet has less than 3 digits.)
    • State: available
      Note: do not attempt to add the computer in the maintenance state because of the following issue:
      https://issues.apache.org/jira/browse/VCL-189
    • Owner: admin
    • Platform: i386
    • Schedule: VCL 24x7 
    • RAM: 1024
    • No Processors: 1
    • Processor Speed: 2000
    • Network Speed: 100 
    • Type: blade
    • Provisioning engine: xCAT 1.x
      Note: the Type and Provisioning engine values don't matter for the localvmhost computer in this test environment because vcld isn't provisioning it
    • Computer Groups: all vm host computers
• Add VM guest computer:
  • Manage Computers -> Edit Computer Information -> Submit -> Add
    • Hosthame: vmguest-1
    • IP Address: current public ip address used by Windows XP VM
    • State: available 
    • Owner: admin
    • Platform: i386
    • Schedule: VCL 24x7 
    • RAM: 1024
    • No Processors: 1
    • Processor Speed: 2000
    • Network Speed: 100 
    • Type: virtualmachine
    • Provisioning engine: VMWare Server Provisioning
    • Computer Groups: all vm guest computers
• Note the fix for the IP address bug is recorded here: https://issues.apache.org/jira/browse/VCL-193
• Configure the VM guest's MAC address in the computer table in the database (as there currently does not seem to be a way to do this through the UI):
  • eth0: 00:50:56:1a:01:01
  • eth1: 00:50:56:1a:01:02
• Configure the VM host profile:
  • Virtual Hosts -> VM Host Profiles tab -> VMware GSX standard -> configure Profile
    • Virtual Switch 0: VMnet1
    • Virtual Switch 1: VMnet0
• Change state of localvmhost to vmhostinuse (edit the database directly, do not use the Virtual Hosts utility on the website)
• Assign vm guests to localvmhost
• Create node in priviledge tree
  • Name: VM image access
  • Add resource groups:
    • all vm guest computers
    • all vm guest images

...

ddns-update-style none;
 shared-network vmnet1 {
         subnet 192.168.0.0 netmask 255.255.0.0 {
                 ignore unknown-clients;
                 option routers 192.168.0.1;                host vmguest-1 {
                         option host-name "vmguest-1";
                         hardware ethernet 00:50:56:1a:01:01;
                         fixed-address 192.168.1.1;
                         filename "/tftpboot/pxelinux.0";
                         option dhcp-server-identifier 192.168.0.1;
                         next-server 192.168.0.1;
                 }
      }  }
}

The DHCP daemon should only listen on the virtual private network (vmnet1) to avoid conflicts with other production VCL networks.  Configure the dhcpd service startup script to only listen on the vmnet1 interface:

start() {
    [ -x $dhcpd ] || return 5
    [ -f $conf ] || return 6

    pidofproc $prog >/dev/null 2>&1
    RETVAL=$?
    [ $RETVAL -eq 0 ] && return $RETVAL

    echo -n $"Starting $prog: "
    daemon $dhcpd vmnet1 $DHCPDARGS 2>/dev/null
    RETVAL=$?
    echo
    [ $RETVAL = 0 ] && touch $lockfile
    return $RETVAL
 }

Configure the dhcpd service to automatically start at runlevels 3-5:

...