Universal Messaging 10.15 | Resolved Issues | Release 10.5
 
Release 10.5
*NUM-10743 (MQTT QoS 2 connections are not downgraded to QoS 1)
Universal Messaging doesn’t support MQTT QoS 2 and will automatically downgrade client connections to QoS 1.
This issue is resolved.
*NUM-11606 (Universal Messaging status shown as ERROR in CCE)
Command Central displays an ERROR run-time status for a clustered Universal Messaging server when you replace interfaces on one or more of the realm servers in the cluster from outside Command Central.
When you monitor a Universal Messaging realm server that is part of a cluster in Command Central, and you add a new interface and remove the previously active interface on one or more of the realm servers in the cluster from outside Command Central (for example, in Enterprise Manager), Command Central displays an ERROR status for the monitored realm and also displays offline cluster members until Platform Manager is restarted.
The issue is resolved.
*NUM-11965 (Universal Messaging cluster formation may take too long to reform)
Logic for purging events from a multi-file store is now optimized.
Previously, events were read twice from disk although this was not really needed.
The issue is resolved.
*NUM-11971 (NSPS connection from DotNet to UM)
C# clients were not able to connect to NSPS interfaces using SASL authentication.
The issue is resolved.
*NUM-12285 (Selectors starting with % sign are not parsed correctly)
An error was thrown when parsing a selector which starts with the string "%_". This was an unhandled case in the code, which is now corrected and this scenario is working properly.
The issue is resolved.
*NUM-12300 (Batch consumption for serial durable consumers does not use the allowed batch size)
Serial consumers did not correctly consume messages in batches even when there was a large backlog of events. For example, if the batch size was 50, a serial consumer could receive only 2 or 3 messages in a single batch instead of all 50.
The issue is resolved.