RE: Memory Leak

> Yes that's what we were told. We did this but it had little effect. > If a cursor was declared and freed, then the resources were released, > but > subsequent declares and frees had no effect. > > I'm no longer working for the company but I have confirmed that the > reported bug is still unresolved. > > Informix were unable to duplicate the bug, but then again they were > attempting the > same thing on an HP box not a IBM J40 > > Regards > Rekaish > > > -----Original Message----- > From: Jonathan Leffler > [SMTP:jleffler@earthlink.net.compans.com] > Sent: Tuesday, January 26, 1999 4:43 AM > To: informix-list@iiug.org > Subject: Re: Memory Leak > > Rekaish Bhardwaj wrote: > > I've encountered this problem before. > > I've logged a call with informix who have so far been unable > to > > reproduce the problem. > > > > In my case we tracked the leak to cursors declared within our > 4ge's. > > After each cursor had been used, the free statements that we > used seem > > to stop working > > in V7,23 of Informix (running on AIX 4.1.5). > > > > Our work-a-round has so far been to ask users to log off every > 3 or 4 > > hours e.g. at lunchtimes or breaks. > > If anyone else has had this problem I'd be interested > > Are you freeing both the cursor and the statement? In 6.x and > later > versions of I4GL (and in 5.x and later versions of ESQL/C), you > need > to free both the prepared statements and the cursors declared > for them. > With the 4.x and earlier versions of I4GL (and ESQL/C), freeing > both > the cursor and the statement was an absolute no-no, because the > system > freed the same memory twice, rendering the dynamic memory > allocation > system wholly unreliable. > > -- > Jonathan Leffler (jleffler@informix.com, jleffler@earthlink.net) > Guardian of DBD::Informix v0.60 -- see http://www.perl.com/CPAN > #include <disclaimer.h>