5.3. Usability

5.3.1. [Critical] Locator Reachability DataReader could not receive assertions from more than four DomainParticipants

A misconfiguration in the Locator Reachability DataReader set the maximum number of instances to 4. As a result of this configuration, the Locator Reachability DataReader could not receive assertions from more than four DomainParticipants. Now, the Locator Reachability DataReader can receive assertions from as many DomainParticipants as configured in the DomainParticipantResourceLimits.remote_participant_allocation QoS setting.

[RTI Issue ID CORE-15703]

5.3.2. [Minor] DDS_RTPS_AUTO_ID_FROM_MAC failed if MAC address had all zeros for first 4 bytes or last 4 bytes

If you set the DomainParticipantQos wire_protocol.rtps_auto_id_kind to DDS_RTPS_AUTO_ID_FROM_MAC, and the hardware MAC address of your machine had four zeros at the start or the end, then DomainParticipant creation failed with this error:

PRESParticipant_new:!create facade

For example, a MAC address of 56,6f,c0,9e,0,0,0,0 or 0,0,0,0,56,6f,c0,9e incorrectly caused DomainParticipant creation to fail.

Now, DomainParticipant creation succeeds regardless of the value of the MAC address.

[RTI Issue ID CORE-15655]

5.3.3. [Minor] Inconsistent QoS error when enabling Monitoring Library 2.0 with a default profile (is_default_qos = TRUE) *

Enabling Monitoring Library 2.0 by setting <monitoring>/<enable> to TRUE in <participant_factory_qos> could result in unexpected inconsistent QoS policy errors. These errors prevented Monitoring Library 2.0 from being enabled.

This issue only occurred when the default profile—where is_default_qos was set to TRUE—contained QoS settings that were incompatible with the default BuiltinQosLib::Generic.Monitoring2 profile, which is used to create the Monitoring Library 2.0 entities.

[RTI Issue ID MONITOR-728]

5.3.4. [Minor] Lack of allowed multicast interfaces did not cause Connext to ignore multicast addresses

If none of your allowed interfaces supported multicast, then Connext failed to adapt to this situation and incorrectly did the following by default:

Outgoing participant discovery announcements were still sent to the 239.255.0.1 multicast address.
The 239.255.0.1 locator was still advertised in the outgoing participant discovery announcements.
multicast_resend_threshold was still taking effect. Repair traffic (for both metadata and user data) was still attempted and failed to be sent over multicast.

Warnings such as

NDDS_Transport_UDPv4_SocketFactory_create_receive_socket:No interface found enabled for multicast

and

send failed: locator: udpv4://239.255.0.1:7400

were still being logged.

[RTI Issue ID CORE-14825]

* This bug did not affect you if you are upgrading from 6.1.x or earlier.