4.15. Instances

4.15.1. Instances transitioned due to instance state consistency did not respect propagate_dispose_of_unregistered_instances

By default, Connext does not support transitions between NOT_ALIVE instance states; however, this can be configured on the DataReader by setting propagate_dispose_of_unregistered_instances and/or propagate_unregister_of_disposed_instances in the DATA_READER_PROTOCOL QoS policy. Instances that were transitioned due to instance state consistency (i.e., instances that transitioned upon recovering liveliness with a previously matched DataWriter) were not abiding by this configuration and may have transitioned from NOT_ALIVE_NO_WRITERS to NOT_ALIVE_DISPOSED even though propagate_dispose_of_unregister_instances was false. This issue has been resolved.

[RTI Issue ID CORE-13477]

4.15.2. Instance purging based on source timestamp did not work

In 7.1.0, the source timestamp-based purge delay did not purge instances based on their source timestamp. Instead, it purged instances based on their sequence number. This problem has been resolved. Now, the dds.data_writer.history.source_timestamp_based_autopurge_instances_delay property works as intended.

[RTI Issue ID CORE-13911]