Ram, Everyone else already answered your question and I agree with them. Kill the unix session first. And Richard gave a precise explanation of why things hang when they are killed incorrectly. Tom ________________________________ From: Ram Raman [mailto:veeeraman@xxxxxxxxx] Sent: Friday, January 30, 2009 11:45 AM To: Mercadante, Thomas F (LABOR) Cc: ORACLE-L Subject: Re: v$session question Thank you Tom. I would think that killing via oracle first is safer because oracle would roll back any work done. If done via OS first, will Oracle roll the transaction back and leave the database in a consistent state? If there is no entry in the V$transaction, then I think killing the process from the OS should be ok. Any corrections to this theory? On 1/30/09, Mercadante, Thomas F (LABOR) <Thomas.Mercadante@xxxxxxxxxxxxxxxxx> wrote: Ram, On Unix, you need to kill the unix session first before killing the Oracle session. Kill the spid value returned from this query on the unix prompt: kill -9 999999 Then kill the session within the database with your "alter kill" statement. select v$session.username, spid from v$session, v$process where v$session.username in ('yourusername') and v$session.paddr=v$process.addr order by last_call_et desc Hope this helps. Tom ________________________________ From: oracle-l-bounce@xxxxxxxxxxxxx [mailto:oracle-l-bounce@xxxxxxxxxxxxx] On Behalf Of Ram Raman Sent: Thursday, January 29, 2009 8:52 PM To: Paul Drake Cc: ORACLE-L Subject: Re: v$session question Thanks Paul. It is HP-UX 11i. Oracle is 10.202. I used the 'Alter system kill session' command from SQL PLus. orakill.exe? Is it for Windows? On Thu, Jan 29, 2009 at 5:31 PM, Ram Raman <veeeraman@xxxxxxxxx> wrote: Listers, When I kill a session and the status shows as "KILLED" in v$session, I assumed that the rollback is taking place. I am not sure what is happening here: