Re: V$SESSION - SQL_ADDRESS for background processes

From: Martin Berger <martin.a.berger_at_gmail.com>
Date: Fri, 17 Apr 2009 20:29:48 +0200
Message-Id: <F977E35A-A5A5-4DED-BD7A-14D9767B0563_at_gmail.com>



Paul,

can you enable trace 10046 for ARC2 and try to reproduce your findings? I'm curious if you see any PARSE and EXEC rows for statements in question.

just an idea,
  Martin

> I'm doing a query to join V$SESSION with V$SQLAREA to get the
> current SQL for each running session. Nothing fancy, it's just a
> quick view to help diagnose an issue. But I see some odd results,
> with background processes (specifically, ARC2) running what looks
> like user SQL (SELECT /*+ FIRST_ROWS */ * FROM user_table_1,
> user_table_2). I'm tempted to discount these, on the assumption that
> background processes wouldn't be running user SQL, so presumably
> SQL_ADDRESS and SQL_HASH_VALUE are meaningless for such processes,
> but is that a fair assumption? And if not, then why is that SQL
> present?
>
> It's also interesting to note that EXECUTIONS in V$SQLAREA for that
> statement is 0. Presumably that implies that even though the
> statement was parsed, it has never been executed? What does that mean?
>
> Thanks,
> Paul
>
> Paul Moore
> Oracle DBA Support
> Atos Origin UK Ltd
> Daresbury Court
> Runcorn



--
http://www.freelists.org/webpage/oracle-l


  • application/pkcs7-signature attachment: smime.p7s
Received on Fri Apr 17 2009 - 13:29:48 CDT

Original text of this message