Deepa Remesh wrote: >I have added procedure tests to xaSimplePositive.sql and attached it >with the mail. This patch is only for review. If it is okay, I can >combine it with my previous patch for derby-498 and attach it to the >JIRA issue. > >I ran xaSimplePositive.sql before and after applying my patch for >derby-498. In both cases, I did'nt get any error when the holdability >of a statement inside a procedure was set to HOLD_CURSORS_OVER_COMMIT. >Is this okay? > >Here is behaviour before and after the patch (It is the same for >global and local transactions) >------------------------------------- >before the patch: >------------------------------------- >Holdability of statement inside procedure | Result set holdability >default | HOLD_CURSORS_OVER_COMMIT >HOLD_CURSORS_OVER_COMMIT | HOLD_CURSORS_OVER_COMMIT >CLOSE_CURSORS_AT_COMMIT | HOLD_CURSORS_OVER_COMMIT > >------------------------------------- >after the patch: >------------------------------------- >Holdability of statement inside procedure | Result set holdability >default | CLOSE_CURSORS_AT_COMMIT >HOLD_CURSORS_OVER_COMMIT | HOLD_CURSORS_OVER_COMMIT >CLOSE_CURSORS_AT_COMMIT | CLOSE_CURSORS_AT_COMMIT > >Does this look okay? > >
Hi Deepa from the perspective of DERBY-498 this looks ok as long as the test runs ok with jdk131. But it seems wrong to me that you would be able to execute a statement with HOLD_CURSORS_OVER_COMMIT within an xa transaction. Dan or Mamta is this something for which we should file a bug? Kathey
