ExtremeTransactions 5.0.86

180713: Refactor LogCloud to allow multiple instances that share the same DB schema.

You can now share the same DBMS for different microservices - so you don't need to instantiate multiple DB instances.

To upgrade, it is best to do a clean shutdown (in no-force mode) so you are sure that there are no pending transactions.

The following schema changes are required on your end:

Add a column OWNING_RECOVERY_DOMAIN_NAME to all tables

The default value should be whatever you have for com.atomikos.icatch.tm_unique_name in your microservice.

Table LOG_CLOUD_STARTUP_CONFIG: com.atomikos.icatch.tm_unique_name no longer a separate row

The property value com.atomikos.icatch.tm_unique_name is now implicit in part of the primary key columns. So you don't need to insert this as a separate row any more.

180766: Remoting recovery: improve logging if getOutcome fails

Made the log files and stack traces less confusing when resolution of a pending in-doubt transaction fails due to connectivity issues.

180765: NPE in remoting commit replay if coordinator / root not found

Fixed a NullPointerException with remoting recovery towards a remote participant that cannot find the transaction in question.

180060: Improve log warning on timeout before commit

When a transaction timed out we used to hint at different options / causes in the warning. This has been improved to be accurate now.

180154: Bug in LogCloud recovery: max_timeout in application node must equal that in LogCloud

Fixed a bug that could cause problems in recovery when multiple nodes compete for the recovery master role.

180730: Bug in HTTP recovery

Fixed a bug where some pending HTTP remote participants were not recovered correctly.

Only for customers. Not a customer yet?