We are getting an "Out of memory while reading tuples" error while
iterating through a large record set. Our environment can be roughly
described as MS Access with linked tables connecting to a PG backend
through ODBC. The ODBC Driver version is PostgreSQL Unicode 8.02.05.00.
We understand that the "Use Declare/Fetch" setting would probably
circumvent this problem. However that setting has posed challenges for
us in the past. Those challenges were significant enough that we'd
deliberately chosen to run our application with that setting off.
So before diving back into figuring workarounds for those challenges, we
are wondering if there is any way of increasing the memory overhead of
the ODBC connection. We'll go further out on a limb and ask, is this
simply a bug in the ODBC driver? When this occurs it appears as if there
is plenty of physical memory available on the machine. And while I'm not
sure if the ODBC driver's memory allocation would appear under the
MSACCESS.EXE process, MS Access often doesn't appear to allocate more
than 512MB for itself.
I have a mylog prepared already should that be of benefit to you all.
You all have been very helpful to us in the past, so thanks again for
your time and energy...
-Garner
