We found the problem. Our database (SQL Server) is a little flaky and likes to not answer from time to time. Just doesn't scale well. And by default, JTDS doesn't set a query timeout. We've added: ;loginTimeout=20 to the JTDS URL, to make sure the connection pool has a chance at refreshing iteself, and added: private static final int SQL_QUERY_TIMEOUT = 2; // by query | ... | callableStatement.setQueryTimeout(SQL_QUERY_TIMEOUT); to all of our DAOs for scheduled jobs. I made these changes last night, and as expected the jobs threw several query timeout exceptions, and the jobs restarted on the next scheduled interval. So - hopefully - this is the entire issue we're experiencing. We'll be able to confirm after a couple of days in production. Thanks! Larry View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=3958103#... Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&a...