Using Clone Pools to Implement Off-Host Processing

In this scenario, snapshots are taken of a set of five volumes, vol1 through vol5, in the data storage pool within the hrdg disk group. These snapshots are then made available to a different host so that a backup can be made of the data.

1. First create a clone pool, snappool1, to hold the snapshots:
     # vxpool -g hrdg create snappool1

   ---

     Note    As snappool1 is the second storage pool to be added to the disk group after the data pool, it is automatically created as a clone pool.

   ---

   The disk group can be split and clone pool taken to a new disk group. This new disk group should be deported from this host and imported on the host where the data from the volumes could be backed up.

2. Create five concat volumes in the clone pool to act as snapshots for the volumes in the data pool:
     # vxvoladm -g hrdg -p snappool1 make snapvol1 10g
     # vxvoladm -g hrdg -p snappool1 make snapvol2 1g
     # vxvoladm -g hrdg -p snappool1 make snapvol3 50g
     # vxvoladm -g hrdg -p snappool1 make snapvol4 25g
     # vxvoladm -g hrdg -p snappool1 make snapvol5 20g

   ---

     Note    You do not need to run the vxsnap prepare command. Volumes created in a clone pool are automatically allocated a data change object (DCO).

   ---

3. Using the newly created volumes, take full-sized instant snapshots of the corresponding five volumes in the data pool:
   # vxsnap -g hrdg make src=vol1/snap=snapvol1/sync=yes \
   src=vol2/snap=snapvol2/sync=yes \
   src=vol3/snap=snapvol3/sync=yes \
   src=vol4/snap=snapvol4/sync=yes \
   src=vol5/snap=snapvol5/sync=yes
4. After synchronization of the data in the snapshot volumes is complete, split off a new disk group, snapdg, from the hrdg disk group:
     # vxdg split hrdg snapdg snappool1

   The snapdg disk group contains the snapshot volumes from the snappool1 clone pool.

5. Deport the newly created disk group, snapdg:
     # vxdg deport snapdg

   This disk group can now be imported on a different host.