RE: CURSOR_SHARING = FORCE ques.

<SPAN

class=774340322-02072001>You're probably referring to this bug, Ivan, I haven't seen a fix via MetaLink.
<SPAN

class=774340322-02072001>Somewhere I read a recommendation of cursor_sharing=exact, but can't pinpoint that
advice.
 
BUG 984132Description: If
automatic replacement of literals with bind variables is enabled by using setting CURSOR_SHARING = FORCE (either in as an initialization parameter or by setting it in an ALTER SYSTEM or ALTER SESSION statement), then statements with literals in a cursor expression can core dump. There
is no workaround for thisproblem, other than disabling replacement of literals. Eg: "selectcursor(select 'a' from dual) from po" may

dump                        

 
<FONT face=Tahoma

size=2>-----Original Message-----From: Ivan_Rivera_at_doh.state.fl.us [mailto:Ivan_Rivera_at_doh.state.fl.us]Sent: Monday, July 02, 2001 5:33 PMTo: Multiple recipients of list ORACLE-LSubject: CURSOR_SHARING = FORCE ques.
Gurus....
Well the boss is now getting on my back for me to upgrade my OCP to 8i.  I know, I know, I wasn't going to, please don't flame me, but work is insisting that I do.  So now I'm studying the upgrade book and it's sparked up a couple of questions regarding cursor_sharing = force. 1) I've heard a bunch of negative things (like bugs) regarding using this feature, was this just with 8.1.6 or is it still buggy? 2) Anyone out there using it?  Does it give that much better performance gains?  It seems to me like it's not worth implementing.
3)Can the guys playing with 9i out there verify if this is buggy in 9i?
Thanks to all. Ivan Rivera Received on Mon Jul 02 2001 - 17:28:24 CDT