A method begins by a processing module of a dispersed storage network (DSN) determining that a set of storage units has less than a desired number of active storage units, where the DSN includes a plurality of storage units that randomly are active or inactive. The method continues with the processing module identifying another active storage unit of the storage units that is not currently part of the set of storage units and adding the other active storage unit to the set of storage units. For encoded data stored by the set of storage units, the method continues with the processing module increasing a pillar width number of a dispersed storage error encoding function, maintaining a decode threshold number of the dispersed storage error encoding function, creating new encoded data slices for the encoded data, and storing the new encoded data slices in the other active storage unit.