That won't help, as the cursor would still
be held open in the pl/sql cursor cache -
despite the explicit close.
It's also more efficient to use the implicit
cursor in pl/sql for a single row fetch in
the user's version of Oracle.
Regards
Jonathan Lewis
http://www.jlcomp.demon.co.uk
The
It might be held in the cursor cache, it may even be
held in session cursors cache but it will not be counted
as an open cursor. My suggestion had diagnostic purpose only.
The problem is, probably, with the tool which explicitly closes
cursors too frequently and insufficiently sized shared pool
Run your test case, and check the contents
of v$open_cursor. Unless my memory has
got it backwards,
the pl/sql cursor cache is counted towards
max_open_cursors, but the cursors that have
been held open by the 'dirty tricks department'
are closed as required if the limit is
What tool are you using? HAve you considered putting select from the
sequence in an explicit cursor, open it, fetch it and close it again?
What Have in mind is something like this:
declare
cursor csr is
select sai.nextval from dual;
num integer :=0;
ind integer :=10;
begin
while (ind=0) loop
open
thanks mladen.
will give this a shot...again thanks a bunch
sai
--- Mladen Gogala [EMAIL PROTECTED] wrote:
What tool are you using? HAve you considered putting
select from the
sequence in an explicit cursor, open it, fetch it
and close it again?
What Have in mind is something like this:
thanks mlade..i will surely give this a shot.
can you please tell me whether a sequence creates such
issues. as mentioned earlier, the developers claim
that no code has changed. im am not able to give any
kind of reason for this though the trace shows this
statement being called more than 350
