[ http://jira.jboss.com/jira/browse/JGRP-643?page=comments#action_12391192 ] Bela Ban commented on JGRP-643: ------------------------------- Why is this an issue ? The static method is called only *once* with init=true, by the ProtocolStack. All other calls are *NOT* initializing it. So there is no concurrent invocation of this method with init=true -- This message is automatically generated by JIRA. - If you think it was sent incorrectly contact one of the administrators: http://jira.jboss.com/jira/secure/Administrators.jspa - For more information on JIRA, see: http://www.atlassian.com/software/jira

[ http://jira.jboss.com/jira/browse/JGRP-643?page=comments#action_12391326 ] Michael Newcomb commented on JGRP-643: -------------------------------------- ClassConfigurator.getInstance(true) is called indirectly from the JChannel constructor. If you create 2 or more JChannels in seperate threads this method can be called in an unsafe manner. This does indeed occur because I got a "magic key x (...) is already in map..." exception which indicates 2 init calls got through. Interestingly enough, it occurs only on my dual core computer ;). Could you make the creation and assignment of 'instance' a class-level static initializer block? That way there would be no need for synchronization and no chance of multiple calls? -- This message is automatically generated by JIRA. - If you think it was sent incorrectly contact one of the administrators: http://jira.jboss.com/jira/secure/Administrators.jspa - For more information on JIRA, see: http://www.atlassian.com/software/jira

[ http://jira.jboss.com/jira/browse/JGRP-643?page=all ] Bela Ban resolved JGRP-643. --------------------------- Resolution: Done -- This message is automatically generated by JIRA. - If you think it was sent incorrectly contact one of the administrators: http://jira.jboss.com/jira/secure/Administrators.jspa - For more information on JIRA, see: http://www.atlassian.com/software/jira