Received: from bastion.solipsys.com (bastion.solipsys.com [209.117.190.2]) by hub.org (8.9.3/8.9.3) with ESMTP id PAA34020 for ; Tue, 24 Aug 1999 15:44:03 -0400 (EDT) (envelope-from byronn@solipsys.com) Received: from solipsys.com ([209.117.190.166]) by bastion.solipsys.com (8.9.1a/8.9.1) with ESMTP id PAA27788; Tue, 24 Aug 1999 15:42:56 -0400 (EDT) Message-ID: <37C2F57E.DE133D21@solipsys.com> Date: Tue, 24 Aug 1999 15:41:50 -0400 From: Byron Nikolaidis X-Mailer: Mozilla 4.51 [en] (WinNT; I) X-Accept-Language: en MIME-Version: 1.0 To: "G. Anthony Reina" , interfaces@postgreSQL.org Subject: Re: [INTERFACES] psqlODBC/MS Access "Out of memory error (#1)" References: <37C2E032.A6199F3C@nsi.edu> <37C2EA1E.D966AC76@solipsys.com> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Transfer-Encoding: 7bit Byron Nikolaidis wrote: > "G. Anthony Reina" wrote: > > > I just figured out how to get my ODBC connection running between the > > Postgres 6.5.1 database on my RH Linux 6.0 box and my Windows 95 box. I > > am using Microsoft Access 97 on the Windows 95 box. I have been able to > > import some of the smaller tables using "Import from External Data > > Source" and then specifying the postODBC driver. It seems that some of > > my larger tables can't be loaded in because of the error: > > > > Error while executing the query: > > Out of memory while reading tuples (#1). > > > > The error seems pretty self-explanatory. However, does it mean that I > > (a) don't have enough system RAM to hold the table, (b) need more hard > > disk space to hold the file, or (c) none of the above? > > > > Try turning off the declare/fetch driver option. With this option on, the > driver is trying to store every single tuple in memory. With the option > off, it will only store 100 by default, or the number you enter for the > cache size. Since the import is accessing the tuples in a forward-only > fashion, there is no need to use the declare/fetch. > > Byron > > ************ Oops, geez, what I meant to say was, turn *ON* the declare/fetch option!!!! With it *OFF*, the driver is trying to store every single tuple in memory. With the option *ON", the driver will only store 100 tuples by default, or the number you enter for the cache size. Since the import is accessing the tuples in a forward-only fashion, there is no need for random-access to tuples (so the declare/fetch *should* be used). Sorry for any confusion. I think my mind is still on vacation. Byron