Hello Community,
I'm new to this list.
I am a user of the Postgres ODBC driver, mainly under MS Access and
Matlab. We deal in very large tabular datasets. (My largest table is
currently 72 million records, and it will grow several times that size
before the year is out.
The driver fails after several million record inserts (under access). Once
failed (with nothing more helpful tan an "ODBC call failed" message box),
it will process no further requests. (I believe it will also fail after
receiving that much data, as well a sending.)
(I would think of a memory leak of some sort, but in my 27 pro years of
programming, I have never dealt with the inner workings of an ODBC driver,
and I don't know how to get driver error information. This is also my
first open source experience, so please forgive my naivete.)
I would like to figure out how to fix this problem. I'm willing to help
debug, but I need help getting started. How do I go about trying to
understand this problem?
One thing that might be helpful: Is there a programmatic way to get Access
to close and unload, then reopen the ODBC driver without actually quitting
Access? I notice that when I start Access "fresh" I can pump another
several million records through before failing. If I can make my programs
unload then reload the driver every few million records, it might make a
good workaround for my problem. Any clues out there?
With thanks,
Robin
--------------------------------------------------
Robin Weber
Data Manager and Application Developer for
Marine Mammals and Birds
Tagging of Pacific Pelagics Project
University of California Santa Cruz
Long Marine Lab - Center for Ocean Health
100 Shaffer Road
Santa Cruz, CA 95060-5730
(831) 459-2691
