The next_random configuration to the Session.open_cursor method configures the cursor to return a pseudo-random record from a row-store object (the next_random configuration is not supported on other types of objects).

Applications should use the Cursor.next method to retrieve records from the object, most other cursor methods are not supported. For example, it's not possible to update using a cursor configured for random retrieval.

By default, each returned record is pseudo-randomly selected from the underlying object as a whole. That can lead to skewed results when the underlying tree structure is unbalanced or records are not uniformly distributed. In such cases, the next_random_sample_size configuration can also be specified. Setting next_random_sample_size configures the number of samples the application expects to take using the cursor. A cursor configured using next_random_sample_size divides the object into next_random_sample_size pieces, and each subsequent retrieval returns a record from the next one of those pieces.

For example, setting next_random_sample_percent to 10 would cause the cursor to sequentially return records from each tenth part of the object. Setting next_random_sample_percent to 1000 would cause the cursor to sequentially return records from each .1% of the object.