Hide Forgot
Date of First Response: 2007-12-13 10:29:53 project_key: SOA Reproducible: 100% 1. I took full IR8 and used dbinstall (https://svn.corp.jboss.com/repos/soa/trunk/qa/dbinstall/) for the 'all' configuration to use Oracle DB. 2. I unsecured jmx-service-invoker. 3. I copied the SOA-P to the second node. 4. I changed the server peer id on the both copies to be different. 5. I started the first node, it was successful. 6. I started the second node successfully as well. 7. I let it be for half an hour. 8. This appeared in the log files: Node 1: 2007-12-13 10:57:07,502 WARN [org.jgroups.blocks.ConnectionTable] peer closed connection, trying to re-send msg 2007-12-13 10:57:07,502 ERROR [org.jgroups.blocks.ConnectionTable] 2nd attempt to send data failed too Node 2: 2007-12-13 10:57:08,927 WARN [org.jgroups.blocks.ConnectionTable] peer closed connection, trying to re-send msg 2007-12-13 10:57:08,928 ERROR [org.jgroups.blocks.ConnectionTable] 2nd attempt to send data failed too It doesn't seem like there's anything related to the problems in the logs (no exceptions, no other JGroups related output...).
Link: Added: This issue incorporates JGRP-646
Phillip is going to recreate this and create a JBPAPP issue for it, and probably a JBMESSAGING bug.
I was unable to reproduce this in the performance lab. I used the latest EAP 4.3 build and mysql as the messaging persistence db. Martin, can you try to reproduce this in the QA lab?
I tried to reproduce it on our machines and it took 13 hours to happen. But it happened! I'll give it a try in the QA lab...
I tried to reproduce it in QA lab on machines cluster09 & 10 with MySQL database on perfdb01 and I was not able to reproduce the problem. Maybe it was a local network issue. I'm going to close the issue for now...