Prime Network Services Controller automatically selects the data store to use for the InterCloud Extender instance.

• If you did not enable high availability, navigate to and select the ESXi host to use for the InterCloud Extender instance.

• If you enabled high availability, do one of the following:

  • To use the same ESXi host as the primary InterCloud Extender, in the Secondary area, check the Same as Primary check box.

  • To use an ESXi host other than the primary InterCloud Extender, in the Secondary area, navigate to and select the ESXi host to use for the secondary InterCloud Extender instance.

• If you did not check the Use Marketplace ICS check box in the Configure InterCloud Link screen, select the template that you want to use, then click Next. The template version must match the version of the InterCloud extender image that you selected in the InterCloud Extender screen.
• If you checked the Use Marketplace ICS check box in the Configure InterCloud Link screen, select the required image from Amazon Marketplace as follows, then click Next:
  1. Click Refresh Marketplace to ensure that the latest information is displayed.

     Note	The Refresh Marketplace button is available when selecting templates from Amazon Marketplace.

  2. Select the required InterCloud Switch template with the number of VMs that you want to purchase. The template version must match the version of the InterCloud Extender image that you selected in the InterCloud Extender screen.

1. Review the configuration to ensure that it is correct.
2. Check the Deploy check box to create the InterCloud link when you click Finish. Uncheck the Deploy check box to create the InterCloud link later.
3. Click Finish.

The InterCloud Link Update Wizard is displayed.

The screens that are displayed in the wizard depend on the images that you select. For example, if you select to update the InterCloud extender image, the screen for the InterCloud switch image is not displayed.

1. Click Refresh to ensure that the latest information is displayed.

   The table is refreshed with the available InterCloud Switch templates, both locally and on Amazon Marketplace.

2. Select the required InterCloud Switch template, then click Next.

# vmware-mount -f vmdk-image /mount/point

where vmdk-image is the VMDK filename and /mount/point is the desired directory.

# losetup /dev/loopn /mnt/vmdk/file

where loopn is the loop device and file is the name of the flat file.

1. Enter the fdisk command to view the disk partitions as shown in the following example for loop device /dev/loop0 :

   # fdisk -l /dev/loop0
   
   Device          Boot     Start     End     Blocks     Id     System
   /dev/loop0p1     *       1         64      512000     83     Linux
   /dev/loop0p2             64        784     5778432    8e     Linux LVM
   

2. Create a device for each partition on the disk image by entering the kpartx command, as follows:

   # kpartx -a /dev/loop0
   # ls  /dev/mapper/
   control   loop0p1   loop0p2
   

   This command creates device files for all physical partitions on the disk file and maps the logical volumes that reside in the partition.

3. (Optional) Enter the pvscan command to identify the volume groups that are present on the disk image. The command with example output resembles the following:

   # pvscan
   
   PV /dev/mapper/loop0p2 VG vg_mowgli lvm2 [5.51 GiB / 0  free]
   

   In this example, the second partition (/dev/mapper/loop0p2) contains the volume group vg_mowgli.

4. (Optional) View the logical volumes by entering the lvscan command as shown in the following example:

   # lvscan
   inactive      '/dev/vg_mowgli/lv_root'[1.85 GiB] inherit
   inactive      '/dev/vg_mowgli/lv_swap'[3.66 GiB] inherit
   

1. Activate the logical volume by using the lvchange command so that it can be mounted:

   # lvchange -ay /dev/vg_mowgli/lv_root

2. Mount the logical volume on a directory on the host system (/mnt/fs in this example):

   # mount /dev/vg_mowgli/lv_root /mnt/fs/

3. List the contents of the logical volume to confirm that this is the root of the filesystem:

   # ls /mnt/fs/
   bin boot cgroup dev etc home lib lib64 lost+found media mnt opt proc root sbin selinux 
   srv sys tmp usr var
   

4. Obtain the filesystem mount points by entering the cat command:

   # cat /mnt/fs/etc/fstab
   
   /dev/mapper/vg_mowgli-lv_root   /        ext4    defaults    1 1
   UUID=44cb64be-62bc-4297-9a8d-beb3493a2362   /boot    ext4     defaults     1 2
   /dev/mapper/vg_mowgli-lv_swap swap       swap    defaults    0 0 
   

   In this example:

   • /dev/mapper/vg_mowgli-lv_root mounts at /.

   • /dev/mapper/vg_mowgli-lv_swap is the swap partition.

   • The partition with the UUID 44cb64be-62bc-4297-9a8d-beb3493a2362 mounts at /boot.

5. Obtain the UUID and LABEL of a partition by entering the blkid command on the partition device file:

   # blkid /dev/mapper/loop0p1
   
   /dev/mapper/loop0p1:UUID="44cb64be-62bc-4297-9a8d-beb3493a2362"
   TYPE="ext4"
   

6. Recreate the filesystem by using the information gained from the fstab command:

   # mount /dev/mapper/loop0p1 /mnt/fs/boot/

   The Linux file system is successfully recreated with all partitions mounted as in fstab. That is, the file system present on vmdk has been recreated inside the /mnt/fs directory on the host system.

1. Verify that the operating system and version are supported by Prime Network Services Controller by checking the contents of the file etc/redhat-release:

   # cat /mnt/fs/edt/redhat-release
   Red Hat Enterprise Linux Server release 6.2 (Santiago)

2. Review the contents of grub.conf to determine the version of Linux kernel present on the filesystem that is set to boot by default. The following example displays the contents of a grub.conf file and provides the following information:

   • The value of the parameter default identifies the default entry (0).

   • A single kernel is present in the disk image and is set to boot by default.

   • The version of the kernel that is set to boot by default is displayed in the kernel entry. In this example, the version is 2.6.32-220.el6.x86_64.

   # grub.conf generated by anaconda
   # 
   # Note that you do not have to rerun grub after making changes to this file
   # NOTICE: You have a /boot partition. This means that 
   #     all kernel and initrd paths are relative to /boot/, eg.
   #     root (hd0,0)
   #     kernel /vmlinuz-version ro root=/dev/mapper/vg-mowgli-lv_root
   #     initrd /initrd-[generic-]version.img
   # boot
   default=0
   timeout=5
   splashimage-(hd0,0)/grub/splash.xmp.gz
   hiddenmenu
   title Red Hat Enterprise Linux (2.6.32-220.el6.x86_64)
         root (hd0,0)
         kernel /vmlinuz-2.6.32.220.el6.x86_64 ro root=/dev/mapper/vg_mowgli-lv_root
         initrd /initramfs-2.6.32-220.el6.x86_64.img

3. In the file /mnt/fs/boot/config-kernel-version, verify that the following flags are set to y:

   • CONFIG_PARAVIRT_GUEST=y

   • CONFIG_XEN=y

   • CONFIG_PARAVIRT=y

   • CONFIG_NETFILTER=y

1. Create a new disk by using the qemu-img command:

   qemu-img create -f raw myami.img 6G
   Formatting 'myami.img', fmt=raw size=6442450944

2. Attach the new image file to a loop device and format it so that it is the same as the source VMDK image filesystem:

   # losetup /dev/loop1 myami.img
   # mkfs.ext4 /dev/loop1

3. Label the new image, as in the following example:

   # e4label /dev/loop1 _/
   # blkid /dev/loop1
   /dev/loop1: LABEL="_/"  UUID="9cf14199-a0ff-4501-bb2f-9a7bc020b1e2"
   TYPE="ext4"
   

4. Mount the new image file:

   # mkdir /mnt/amifs
   # mount /dev/loop1 /mnt/amifs

# cp -ar /mnt/fs/* /mnt/amifs/

1. Modify /mnt/amifs/etc/fstab so that it reflects the partitioning on the new image. For example:

   LABEL=_/ ext4 defaults 1 1
   devpts /dev/pts devpts gid=5,mode=620 0 0 
   sysfs /sys  sysfs  defaults  0 0 
   proc  /proc  proc  defaults  0 0 

2. Create a udev rule to change the first Ethernet interface to csc0 by creating the file /mnt/amifs/etc/udev/rules.d/70-persistent-net.rules and adding the following content:

   SUBSYSTEM=="net", DRIVERS=="vif", ATTRS{nodename}=="device/vif/0", NAME="csc0"

3. Create an interface file named /mnt/p1/etc/sysconfig/network-scripts/ifcfg-csc0 and add the following content:

   ONBOOT=yes
   DEVICE=csc0
   BOOTPROTO=dhcp

4. Enable networking by editing the file /mnt/amifs/etc/sysconfig/network and setting NETWORKING to yes.
5. Edit grub.conf so that it looks similar to the following:

   default=0
   timeout=2
   title Red Hat Enterprise Linux (2.6.32-71.el6.x86_64)
          root (hd0)
          kernel    /boot/vmlinuz-2.6.32-71.el6.x86_64  ro  root=LABEL=_/  selinux=0  console=hvc0

1. Create the directory /mnt/amifs/opt/.cisco and copy the driver RPM into this directory:

   mkdir -p /mnt/amifs/opt/.cisco
   cp csw.1.1.2.rhel6_2.x86-64.rpm /mnt/amifs/opt/.cisco

2. Create the files version_cur and version_gold in /opt/.cisco and place the driver version in each file. You can obtain the driver version from the name of the file. For example, a file with the name csw.1.1.2.rhel6_2.x86_64 has the driver version 1.1.2.

   cat /mnt/amifs/opt/.cisco/version_cur
   1.1.2
   cat /mnt/amifs/opt/.cisco/version_gold
   1.1.2

3. For each ifcfg-ethn file in /mnt/amifs/sysconfig/network-scripts/, create an entry in interface.conf using the syntax interface-interface-number,interface-name,random-mac-address where:

   • interface-number is the number assigned to the interface, starting with 1.

   • interface-name is the name assigned to the interface.

   • random-mac-address is a MAC address.

   The following is an example interface.conf file for two interfaces:

   cat /mnt/amifs/opt/.cisco/interface.conf
   
   interface-1,ether0,00:0f:f7:dd:8a:37
   interface-2,ether1,00:0f:f7:34:40:ae

# cp csw /mnt/amifs/etc/init.d/
# chroot /mnt/amifs/ chkconfig --level 34 csw on

# cp getkeys /mnt/amifs/etc/init.d/
# chroot /mnt/amifs/ chkconfig --34 getkeys on

//Unmount the disk image file
# umount /mnt/amifs/

//Detach the loop device
# losetup -d dev/loop0

//Unmount partitions
# umount /mnt/fs/boot
# umount /mnt/fs

//Deactivate all logical volumes present
# lvchange -an /dev/vg-mowgli/lv_root

//Delete device mappings
# kpartx -d /dev/loop0

//Detach the loop device
# losetup -d /dev/loop0

//Unmount the VMDK file 
# vmware-mount -d /mnt/vmdk