We will keep monitoring but at the surface it seems that this bug has in fact manifested itself in 9.2.0.8. After restarting the listener, minimizing the log level and adding the subscription parameter to the listener.ora the load is as expected for a listener process. Thanx to Hemant for the heads up and to all who contributed. This is the main reason that oracle-l is the resource it has become. Thanx again Mike Haddon On 10/17/2011 9:33 PM, Chitale, Hemant Krishnarao wrote: > On 10gR1 and 10gR2 this was a known issue. > See references to Bug#4518443 at support.oracle.com > > For example > Intermittent TNS Listener Hang, New Child Listener Process Forked [ID > 340091.1] > 10g Listener: High CPU Utilization - Listener May Hang [ID 284602.1] > > I wonder if it manifested in 9.2.0.8 as well. > > > Hemant K Chitale > > -- //www.freelists.org/webpage/oracle-l