Forums

An issue has been identified with hot-fixes 6.3.12.020 and 7.0.09.023, which were the topic of Support Bulletin #28. This new issue is also present in 7.0.10, but does not exist in any other releases. These hotfixes have now been withdrawn.

Details of Identified Issue
Secondary databases (including RPS) cannot successfully replay a database reorganisation, if the primary database server was restarted before the reorg completed on the Primary database.

If, for example, the primary database server is restarted while the reorg transition is pending, this issue can occur and will cause any attached secondary databases to become unusable and require re-cloning.

This issue only affects Secondary databases including RPS Nodes, however it does not affect Primary databases.

Actions

• Any system where either hotfix is currently installed should be reverted back to 6.3.12.016 or 7.0.09.022 respectively.
• Users of 7.0.10 systems should be aware that if a reorg must be performed on a 7.0.10 SDS system, then it is essential that this is done without the Primary database server restarting. This could be done in two ways:
  1. Online - ensure the Primary database server is not intentionally restarted during the reorg.
  2. Offline - ensure the complete database reorg is performed in one execution of jadloadb.exe. i.e. The initiateTransition parameter should be true.

• If the above conditions cannot be satisfied, it is likely that any attached Secondary databases will need to be re-cloned.

Further mitigations such as additional hot-fixes will be advised as soon as practical.

If you have any questions, please contact Jade Support.