anonymous wrote : No, sorry, I managed to reproduce the problem with one thread on DEBUG level. Here is the log: | Code: | | [2009-07-01 16:11:48,367] [Detection1] DEBUG JDBCCacheLoader - executing sql: INSERT INTO Cached_Eve | nts (fqn, node, parent) SELECT ?, ?, ? FROM Cached_Events_D WHERE NOT EXISTS (SELECT fqn FROM Cached | _Events WHERE fqn = ?) (/__JBossInternal__/RTDE/pattern-matching-threshold/663330655/_ID_/aaaa-uiq8a | n-fwm4m5h3-1-fwm4m8s0-z) | | | | | Still don't know how to fill in the question marks to run it... yep, I've just noticed that the cache loader was modified to create a new table postfixed with "_D", so Cached_Events_D is indeed created by jbosscache. This was a changed made for following JIRA JBCACHE-1410. Guess the query (which looks a bit odd to me) doesn't work on oracle. Is this happening from the start, when you firt try to add something to the cache, or only after a while? View the original post : http://www.jboss.org/index.html?module=bb&op=viewtopic&p=4241417#... Reply to the post : http://www.jboss.org/index.html?module=bb&op=posting&mode=reply&a...