For the sake of history, and to make it absolutely clear, in addition to this fix one may need to set DBDictionary.schemaCase as appropriate for the use case. E.g.:
<property name="openjpa.jdbc.DBDictionary" value="schemaCase=upper"/>
As another example, take this Column definition:
and assume we use a different case when referred as a join column:
Given the referencedColumnName is a different case than the actual column, OpenJPA will not find the column. To resolve this, one must have the fix of this JIRA, as well as setting the DBDictionary.schemaCase.