I am using SchemaSpy 5.0.0 with DB2 9.7 Fix Pack 5 and the udbt4 connection profile. When a table with a self-referencing foreign key is encountered, the program churns out new instances of Table and ForeignKeyConstraint until it runs out of space and dies with a StackOverflowError. This condition is easy to reproduce with DB2, since the db2sampl utility generates a pre-populated database called SAMPLE, which contains a handful of tables with self-referencing foreign keys.
Am I correct in thinking that the foreign keys are obtained via JDBC metadata, thereby preventing me from writing a SQL query I could substitute for the FK lookup?
Log in to post a comment.