Tibco JMS Queue Sender activity in Hung State

We are using Tibco 5.11 BW and EMS 8.0.0.9 on separate virtual windows servers 2008 R2 6.1

Periodically approximately once per 1-2 month random tibco BW process from random tibco BW package hang on "JMS Queue Sender" activity. We have around 80 Tibco BW packages and couple of hundred processes in each package. Most of the processes have really simple logic inside. The processes are using Tibco JMS transport.

When the problem happened It's not throwing any errors in the tra.log or application.log. I only see hanging threads with current activity "JMS Queue Sender" in Tibco administrator. After "JMS Queue Sender" activity start hanging all the process from the package start hanging too and eventually the package not responding at all. Package restart resolving the issue.

The "JMS Sender activity" is using TMP queue in "destinationQueue" . The services can be called from java using tibjms.jar-7.0.1.jar or from another tibco BW service using "JMS Queue Requestor".

Here is our typical service example that hang on "JMS Queue sender"

Similar issue described in tibco forum but no resolution https://community.tibco.com/questions/jms-queue-sender-hung-state

the "JMS Sender activity" hang is probably related to temporary queue that set in ReplayTo. As workaround we are changing Temporary Queue to Static queue that resolving the issue.

The question is what is causing "JMS Sender activity" hang in case of temporary queue in ReplayTo?

UPD: It may be related to

5.14.0: BW-17137 The JMS Receiver and the JMS Sender activities that used the same connection resource got into a dead lock situation while reconnecting to the EMS Server.

5.13.0: BW-16413 The JMS Receiver and JMS Sender that used the same connection resource got into a dead lock while reconnecting to the EMS Server.

Unfortunately I see only issue header in release notes for 5.14 BW . I haven't found the issues detail description in public

The problem was finally fixed by migration BW to version 5.14.