[ https://issues.jboss.org/browse/JGRP-2152?page=com.atlassian.jira.plugin.... ] Richard Achmatowicz edited comment on JGRP-2152 at 1/12/17 2:50 PM: -------------------------------------------------------------------- I added an ASYM_ENCRYT layer to the server configuration for the clustering testsuite and turned on TRACE logging for the ASYM_ENCRYPT protocol. Here is the stack: {noformat} <stack name="tcp"> <transport type="TCP" socket-binding="jgroups-tcp"/> <protocol type="MPING" socket-binding="jgroups-mping"> <property name="ip_ttl">0</property> </protocol> <protocol type="MERGE3"/> <protocol type="FD_SOCK" socket-binding="jgroups-tcp-fd"/> <protocol type="FD"/> <protocol type="VERIFY_SUSPECT"/> <protocol type="ASYM_ENCRYPT"> <property name="encrypt_entire_message">true</property> <property name="asym_keylength">512</property> <property name="asym_algorithm">RSA</property> </protocol> <protocol type="pbcast.NAKACK2"/> <protocol type="UNICAST3"/> <protocol type="pbcast.STABLE"/> <protocol type="pbcast.GMS"/> <protocol type="MFC"/> <protocol type="FRAG2"/> </stack> {noformat} The clustering tests pass with the old config and fail with the new config. One such test, CdiFailoverTestCase, fails with the same error message as in this issue. In this test, two servers, node-0 and node-1, are started. Then node-1 is stopped and then restarted, followed by node-0 which is stopped and then restarted. Looking at the logs, encryption seems to be progressing normally until the time at which node-1 is restarted. Then messages start getting enqueued and the cipher is reported as being null. I am attaching the file containing the test case output, which includes logs from both servers. So this might be a JGroups issue. was (Author: rachmato): I added an ASYM_ENCRYT layer to the server configuration for the clustering testsuite and turned on TRACE logging for the ASYM_ENCRYPT protocol. Here is the stack: {noformat} <stack name="tcp"> <transport type="TCP" socket-binding="jgroups-tcp"/> <protocol type="MPING" socket-binding="jgroups-mping"> <property name="ip_ttl">0</property> </protocol> <protocol type="MERGE3"/> <protocol type="FD_SOCK" socket-binding="jgroups-tcp-fd"/> <protocol type="FD"/> <protocol type="VERIFY_SUSPECT"/> <protocol type="ASYM_ENCRYPT"> <property name="encrypt_entire_message">true</property> <property name="asym_keylength">512</property> <property name="asym_algorithm">RSA</property> </protocol> <protocol type="pbcast.NAKACK2"/> <protocol type="UNICAST3"/> <protocol type="pbcast.STABLE"/> <protocol type="pbcast.GMS"/> <protocol type="MFC"/> <protocol type="FRAG2"/> </stack> {noformat} The clustering tests pass with the old config and fail with the new config. One such test, CdiFailoverTestCase, fails with the same error message as in this issue. In this test, two servers, node-0 and node-1, are started. Then node-1 is stopped and then restarted, followed by node-0 which is stopped and then restarted. Looking at the logs, encryption seems to be progressing normally until the time at which node-1 is restarted. Then messages start getting enqueued and the cipher is reported as being null. I am attaching the files containing the test case output, which includes logs from both servers. So this might be a JGroups issue. -- This message was sent by Atlassian JIRA (v7.2.3#72005)