[ http://issues.apache.org/jira/browse/JCR-623?page=comments#action_12454688 ] Michael Singer commented on JCR-623: ------------------------------------
We did some tests with 3 clusternodes. We had problems with the following use case: JBoss-4.0.5-GA with latest JCA and latest Jackrabbit. Only Jackrabbit is configured for clustering, JBoss is running standalone on each clusternode. All 3 clusternodes already have several jcr-nodes and an existing index-folder. 2 nodes are running, the third node is stopped. Added a jcr-node on clusternode A, the new jcr-node is immediately visible on clusternode B (after a browser refresh of course). Then I start clusternode C which boots up normaly. I then load my testapp. Consoleoutput shows, that all jcr-nodes are loaded except the one added on clusternode A. It just stopps working there, I don't get an exception and no response in the browser window. If I delete the index-folder in the local workspace folder everything works fine. So I suppose that item changes are not properly merged into existing index if a clusternode was down. I haven't had the time yet to dig into this. Maybe someone can confirm this behavior with another configuration? I will file a new jira issue if I know more about this... > Clustering > ---------- > > Key: JCR-623 > URL: http://issues.apache.org/jira/browse/JCR-623 > Project: Jackrabbit > Issue Type: New Feature > Components: core > Reporter: Dominique Pfister > Assigned To: Dominique Pfister > Fix For: 1.2 > > > Implement basic clustering, i.e. make two or more repositories available at > the same time, allowing them to stay in sync with changes applied to only > one of them. -- This message is automatically generated by JIRA. - If you think it was sent incorrectly contact one of the administrators: http://issues.apache.org/jira/secure/Administrators.jspa - For more information on JIRA, see: http://www.atlassian.com/software/jira
