It is likely then that one or more cache settings need to be changed. although there is one more thing that could be a cause of a memory leak - Objects (not sure which version of 4D you are using). I have not worked with these yet, but I can see where they could be a possible memory leak.
Other than that - cache settings and/or cache flushing. if the flush rate (time between flushes is too long) when the disk cash *is* flushed it could take some time, and as I remember this is a locking action (i.e. it stops other activities until it completes). --- Are you *sure* that it is the query that is slow? could it be that you are loading progressively more data, and so the process looks like it is slowing down. i.e. - you do a query - get some result records - do something with them, BUT forget to set a flag that the action has already occurred. Later - you do the same query - get the original result records + new records - do something with them, and still forget to set a flag that the action has already occurred. repeat, again and again with the results getting larger each time. Then the action which might have acted on 10 records to start, is now acting on thousands - appearing - to slow down. On Mon, 7 Nov 2016 16:28:51 +0100, stardata.info wrote: > Hi Chip, > > At the end of the process that is slow after some time, i already do > the your suggestions. > > Thanks > /Ferdinando/ > > Il 07/11/2016 16:02, 4d_tech-requ...@lists.4d.com ha scritto: >> Message: 12 >> Date: Mon, 7 Nov 2016 10:02:35 -0500 >> From: Chip Scheide<4d_o...@pghrepository.org> >> To: 4D iNug Technical<4d_tech@lists.4d.com> >> Subject: RE: Progressive slow >> Message-ID:<20161107100235041589.9243f...@pghrepository.org> >> Content-Type: text/plain; charset=us-ascii >> >> Hi Ferdindao, >> >> I do not know what your process does. - but - >> It does sound like there is an issue with memory. >> Others have responded with ideas about changing cache settings, I would >> follow there suggestions, too. >> >> As I suggested, below, it is possible to create a memory leak - a >> situation where 4D (your application) requests memory form 4D/system, >> does stuff, finishes but does NOT return all of the requested memory to >> 4D/System. This is called a memory leak. >> >> On a Mac, Safari and Firefox (windows too?) both have issues like this. >> ex: open one of these programs memory usage is say 100 megs >> You open 10, or 30, or 50 tabs >> close all the tabs >> memory usage might be 250 megs >> >> repeat this enough times and your system slows down, or the application >> crashes. >> >> for 4D, as I said below, you need to make sure at the end of each >> process (as it quits) the following items are cleared/emptied/removed >> from memory; this may not be exhaustive, and if not others i'm sure >> will chime in:) >> (it is often best to clear these items as you finish using them, rather >> then waiting until the end) >> - sets >> - selections >> - arrays (locals are supposed to be cleared for you) >> - Lists >> - hierarchal lists (be sure to clear lowest level to highest) >> - named selections >> - and turn off Event, and Err call methods >> >> if you have done the cache changes others suggest, and insure that all >> of the above items are cleared, I would expect your slow downs to go >> away. > > ********************************************************************** > 4D Internet Users Group (4D iNUG) > FAQ: http://lists.4d.com/faqnug.html > Archive: http://lists.4d.com/archives.html > Options: http://lists.4d.com/mailman/options/4d_tech > Unsub: mailto:4d_tech-unsubscr...@lists.4d.com > ********************************************************************** ********************************************************************** 4D Internet Users Group (4D iNUG) FAQ: http://lists.4d.com/faqnug.html Archive: http://lists.4d.com/archives.html Options: http://lists.4d.com/mailman/options/4d_tech Unsub: mailto:4d_tech-unsubscr...@lists.4d.com **********************************************************************
