JBoss 3.2.6 does this as well. If you play around with clustered indexes you can make it go away, however JBoss does access the DB in a non-optimal way (at least from SQL Servers perspective). I've been meaning to fix this (or at least supply a working XML file for SQL server) but dont have the time right now. Dont know if its documented but if you are using 3.2.6 or 4.0, or 4.0.1 (and probably 3.2.5) I highly sugest not using SQL server for JMS persistance.
The longer term answer is to profile and look at how JBoss creates, selects and deletes message with regards to tx_id's. We've seen jboss delete all messages with a tx_id while its still trying to insert them. However thats from SQL's perspective and since we moved to a different JMS setup we no longer see these issues and only saw them while under load. Funny thing is you can reproduce this with even the singleton JMS if you pump thousands of small messages through. View the original post : http://www.jboss.org/index.html?module=bb&op=viewtopic&p=3860430#3860430 Reply to the post : http://www.jboss.org/index.html?module=bb&op=posting&mode=reply&p=3860430 ------------------------------------------------------- The SF.Net email is sponsored by: Beat the post-holiday blues Get a FREE limited edition SourceForge.net t-shirt from ThinkGeek. It's fun and FREE -- well, almost....http://www.thinkgeek.com/sfshirt _______________________________________________ JBoss-user mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/jboss-user
