[ https://issues.jboss.org/browse/WFLY-8954?page=com.atlassian.jira.plugin.... ] Nuno Godinho de Matos edited comment on WFLY-8954 at 9/1/17 9:08 AM: --------------------------------------------------------------------- No. The issue not solved! I have a system test where this issue can still be reproduced. I can no longer reproduce it using the sample application, but we do have a scenario in a system test where after refreshing the entity we get different data than what we have in the server session cache. Not good... Actually, good. The fix is most likely working perfectly. The issue I am looking at is a incorrectly named event class, that was leading for my dirty "state" check to be run against a the wrong entity field. However, the refresh action is being redundant. That is a very good indication. I believe my impression above is nothing but a false negative. We are still in the clear for now... Changes are so far looking good. I have taken the dirty code that was specific to wildfly onobsever bug from our code, and I will see if the CI system tests are fine all accross. NOTE: - I will be opening a new wildfly JIRA related to the unregister connection problem mentioned on a wildfly thread. {panel} java.lang.IllegalStateException: IJ000152: Trying to return an unknown connection: org.jboss.jca.adapters.jdbc.jdk7.WrappedConnectionJDK7@40a9ad99 {panel} This issue is not addressed by this fix. Thanks. was (Author: nuno.godinhomatos): No. The issue not solved! I have a system test where this issue can still be reproduced. I can no longer reproduce it using the sample application, but we do have a scenario in a system test where after refreshing the entity we get different data than what we have in the server session cache. Not good... Actually, good. The fix is most likely working perfectly. The issue I am looking at is a incorrectly named event class, that was leading for my dirty "state" check to be run against a the wrong entity field. However, the refresh action is being redundant. That is a very good indication. I believe my impression above is nothing but a false negative. We are still in the clear for now... -- This message was sent by Atlassian JIRA (v7.2.3#72005)