RE: AUD$ TIMESTAMP and LOGOFF_TIME are equal
Date: Thu, 31 Dec 2009 08:27:45 -0600
Message-ID: <C970F08BBE1E164AA8063E01502A71CF01647868_at_WIN02.hotsos.com>
No what I see is not normal in our experience. We have collected like data from many other systems and have only occasionally seen these two being equal. This time around 99% of the data has them equal.
I saw that note too but it didn't seem to apply to our situation.
Thanks all for your input.
- Ric
-----Original Message-----
From: Yong Huang [mailto:yong321_at_yahoo.com]
Sent: Wednesday, December 30, 2009 5:14 PM
To: Dave.Herring_at_acxiom.com; Ric Van Dyke
Cc: oracle-l_at_freelists.org
Subject: RE: AUD$ TIMESTAMP and LOGOFF_TIME are equal
> Is there any consistency of the audit rows where TIMESTAMP =
LOGOFF_TIME?
I remember reading on MOS that said what Dave said earlier "if the
session
ends properly, TIMESTAMP and LOGOFF_TIME should be the same." At logoff
time, the timestamp is updated to that time, unless the action is
'LOGOFF
BY CLEANUP', i.e. the session terminates abnormally.
I didn't find that note, but found Note:271615.1 that says "From 10g onwards the time shown in TIMESTAMP and LOGOFF_TIME columns are in different time zones". I thought that would create a difference in query
result. But no. TIMESTAMP of the dba_audit_* views in 10g and up is already converted to the local timezone, even though the ntimestamp# column of aud$ is in UTC. Ric, you're talking about the columns of dba_audit_*, not aud$. What you see is normal.
Yong Huang
-- http://www.freelists.org/webpage/oracle-lReceived on Thu Dec 31 2009 - 08:27:45 CST