Fixed issues

An application ingestion and replication or database ingestion and replication job's status displayed in the monitoring interface might show a transitional state such as "Stopping" or "Running" even after the underlying run has finished, instead of reflecting the correct final status such as "Stopped" or "Completed." The discrepancy is caused by a race condition in which a concurrent metrics update overwrites the final status written by the status update path.

Calls to the DBMI REST API v2 endpoints /dbmi/public/api/v2/task/create and /dbmi/public/api/v2/task/details might return HTTP 500 errors with the message "Agent with id ... does not exist," even though the runtime environment is present and running.

When using the Data Ingestion and Replication Command-Line Interface (CLI) to retrieve per-table metrics for a table whose name is a prefix substring of another table name in the same job, and when rename rules cause the alphabetical sort order of source table names to differ from that of target table names, the CLI might return "Subtask name does not match either source table or target table name" even though the specified table exists and has valid metrics.

You do not have the option to configure task-level Java Virtual Memory (JVM) for tasks running in serverless runtime environments.

For application ingestion and replication and database ingestion and replication combined load jobs, markers intended for initial load and backlog processing might incorrectly trigger the incremental load processing with no recovery checkpoints written, which might lead to duplicate data on restart.

Data Ingestion and Replication components contain outdated jose4j release 0.9.4 inherited from a third party driver.

When an application ingestion and replication job or a database ingestion and replication job restarts after a failure, data for a table that was already sent to the target might be resent while recovering data for other tables not yet sent to the target.

For organizations that use the shared secure agent group with multiple agents from parent organization, same unload tasks within a database ingestion and replication or application ingestion and replication might be are assigned to more than one agent, resulting in duplicate unloads for the same table.

Application ingestion and replication combined load jobs and database ingestion and replication combined load jobs that use custom properties related to the load transition processing might ignore those custom properties and instead apply only the default values. The following error might be issued:

Adding the customPrimaryKeys source custom property to an existing incremental load or combined initial and incremental load task might result in a primary key change not allowed error. Following a resync, new DML changes are not replicated to the target, and a restart of the task shows the primary key change not allowed error for the resynced table.

When you view details of an application ingestion and replication job or a database ingestion and replication job from Monitor, the overall job status might be missing from the job details page.

Data Ingestion and Replication components contain vulnerable Simba JDBC driver.