Configuration of WORM-protected file volumes for event-based retention

Data that is stored in Write Once Read Many (WORM)-protected file volumes that are managed by IBM Spectrum® Protect for Data Retention and event-based retention, can result in excessive reclamation, which causes performance degradation of the server. The WORM-protected file volumes can be stored on NetApp SnapLock volumes or volumes in IBM Spectrum Scale immutable filesets.

If data is managed by event-based retention, IBM Spectrum Protect initially sets the retention period to the greater of the RETVER and RETMIN parameter values for the archive copy group. When the volume enters the reclamation period and data that remains on the volume is moved, the retention period for the target volume is set to the remaining retention period of the data, which is typically 0. The new volume then enters the reclamation period shortly after the volume receives the data, resulting in the reclamation of volumes that were just created. For example, if you set the RETVER value to 365 and the RETMIN value to 0, the retention period is set to 365 and the reclamation will start after 365 days and move the data to the target volume. The retention period of the target volume is set to 0 because the RETVER value has expired. Therefore, the target volume would be reclaimed again and again. You can avoid this situation by using the RETENTIONEXTENSION server option.

By specifying the RETENTIONEXTENSION option allows the server to set or extend the retention date of a SnapLock volume or volume in an IBM Spectrum Scale immutable fileset. You can specify a value in the range 30 - 9999 days. The default is 365 days. For more information, see RETENTIONEXTENSION.

When you select volumes in a storage pool for data retention servers for reclamation, the server verifies whether the volume is within the reclamation period:
  • If the volume is not within the reclamation period, no action is taken. The volume is not reclaimed, and the retention date is unchanged.
  • If the volume is within the reclamation period, the server verifies whether the percentage of reclaimable space on the volume is greater than the reclamation threshold of the storage pool or of the threshold percentage that is passed in on the THRESHOLD parameter of a RECLAIM STGPOOL command:
    • If the reclaimable space is greater than the threshold, the server reclaims the volume and sets the retention date of the target volume is set to the greater of these values:
      • The remaining retention time of the data plus 30 days for the reclamation period.
      • The RETENTIONEXTENSION value plus 30 days for the reclamation period.
    • If the reclaimable space is not greater than the threshold, the server resets the retention date of the volume by the amount that is specified in the RETENTIONEXTENSION option. The new retention period is calculated by adding the number of days that are specified to the current date.
In the following examples, the SnapLock volume, VolumeA, is in a storage pool whose reclamation threshold is set to 60%. The RETENTIONEXTENSION server option is set to 365 days. The retention period for VolumeA is in the reclamation period. The following situations show how retention is affected:
  • The reclaimable space on VolumeA is less than 60%. The retention date of VolumeA is extended by 365 days.
  • The reclaimable space on VolumeA is greater than 60%, and the remaining retention time of the data is more than 365 days. VolumeA is reclaimed, and the retention date of the target volume is set based on the remaining retention period of the data plus 30 days for the reclamation period.
  • The reclaimable space on VolumeA is greater than 60%, and the retention time of the data is less than 365 days. VolumeA is reclaimed, and its retention date is set to 365 days, the RETENTIONEXTENTION value, plus 30 days for the reclamation period.