My "hole" :+) is only to define that if you drop a table you loose the value between x and y and if you are using one unique sequence for all your schema you re not able to use these after.
A hole in sequence is not important but it's not very clean in database development.
For example table Person 1 to 5000 and Product 5000 to 10000, if you drop Person the next value of your sequence will be 10001 and you have a "hole" of none used value 1 to 5000.
It's only a method for Oracle and the way one sequence for a schema is not a good thing. Nothing in relation with Hibernate.
Christian have you seen my post about sequence and strange value? because it's a big problem and I don't understand why. I think it comes from Oracle 8 but I m not sure.
For Adaniz, you can use this, and create some sequences and reference them in your mapping files.
<id name="id"
type="java.lang.Integer"
column="ID">
<generator class="sequence">
<param name="sequence">WTOLD.SEQ</param>
</generator>
</id>
|