[ https://issues.jboss.org/browse/JGRP-1797?page=com.atlassian.jira.plugin.... ] Richard Achmatowicz commented on JGRP-1797: ------------------------------------------- I reproduced this error on my local machine by artificially bumping up the number of messages sent from 1000 to 3000 while keeping the wait timeout the same. With some additional debugging thrown in, I get results like this: {noformat} log4j:WARN No appenders could be found for logger (org.jgroups.conf.ClassConfigurator). log4j:WARN Please initialize the log4j system properly. C: received msg #1500 from C A: received msg #1500 from C B: received msg #1500 from C B: received msg #1500 from B C: received msg #1500 from B A: received msg #1500 from B C: received msg #3000 from C Channel C: all messages sent in 718 ms A: received msg #3000 from C B: received msg #3000 from C Channel B: all messages sent in 758 ms B: received msg #3000 from B A: received msg #3000 from B C: received msg #3000 from B java.lang.AssertionError: Incorrect number of messages received by the receiver thread: A: 5950 messages,B: 5886 messages,C: 5994 messages Expected :true Actual :false {noformat} This indicated that there may be some possible interference in updating the message counts in the receivers. Changing the receiver method to synchronized fixes the problem. -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators For more information on JIRA, see: http://www.atlassian.com/software/jira