Decreasing the number of volumes in a copy pool

If you must remove a volume from a storage group that is defined to a copy pool, you must take special care to avoid data loss, which might occur if you request a recovery of the copy pool from which the volume was defined. For this reason, during the recovery of a copy pool, DFSMShsm fails the recovery of any individual volumes that existed within the copy pool at the time of the backup but that are not a part of the copy pool at the time of recovery. You may still recover these volumes individually.

Example: If you have two copy pools (COPYPOOLA and COPYPOOLB) and each contains a single storage group, SGA and SGB respectively. A daily fast replication backup is made of each copy pool. If storage group (SGA) contains an empty volume (named VOLA) that is moved to storage group (SGB) for space needs, after the volume (VOLA) is moved, data for copy pool (COPYPOOLB) is allocated on it. The potential for data loss arises from the fact that there may be a fast replication backup version for copy pool (COPYPOOLA) that contains the volume (VOLA). This backup version represents the volume (VOLA) in its empty state. If that fast replication backup version for copy pool (COPYPOOLA) is recovered, then the recovery of copy pool (COPYPOOLA) would recover the volume (VOLA) to its empty state and overlay any new data that had been allocated on it since it was moved to storage group (SGB). To prevent this scenario from happening and others similar to it, during the recovery of a copy pool, DFSMShsm fails individual volumes that are no longer defined to the copy pool that is recovering.

You should perform the following steps when you need to recover a backup version of a copy pool that contains a volume that was removed from the copy pool:

Recover the copy pool. Each volume that was removed from the copy pool since the time the backup version was created is failed with the ARC1811E message.
For each failed volume, determine if the volume should be recovered as a part of the copy pool. If so, perform the remaining steps for those volumes that need to be recovered.
Determine whether you need to preserve any of the existing data on the volume. If so, you must move the data off of this volume and onto another volume. If you do not do so, the recovery of the volume overwrites all existing data.
Enter the following volume recovery command to recover an individual volume:
```
FRRECOV TOVOLUME(volser) FROMCOPYPOOL(cpname) VERSION(vernum)
```
Note: FROMCOPYPOOL and VERSION are optional parameters
Determine whether the volume should be put back into the copy pool. If so, put it back.
Note: The volume should be returned to the copy pool if another backup version will be created after the recovery and that backup version should include the data that was recovered to the volume and any subsequent changes to that volume.