BUG #1879: ODBC Connection Dies

The following bug has been logged online:

Bug reference: 1879
Logged by: Paul O'Connor
Email address: paul(dot)oconnor(at)complinet(dot)com
PostgreSQL version: 7.4.2
Operating system: Redhat: kernel 2.4
Description: ODBC Connection Dies
Details:

The ODBC Connection dies, displaying various frontend messages (frontend is
MSAccess 2000/2003). Have tried almost everything, but found nothing.

here are some of the things I've tried:

Compile frontend on local machine that has problems

Check Version is being changed on update (frontend specific)

Check MDAC Type and change in frontend

Check Application Error Log in Windows

Check timeout with individual connection (through odbc)

Check ODBC Settings

Check Driver version for ODBC-PostrgeSQL

Check Windows Version

Check Office Version

Check updates to Jet Engine

Update MDAC

Check for corruption in frontend

Check using MDB for more explicit error messages (no result)

Check for postgres DB Corruption (vacuum db)

Check ActiveX Versions and if present

Check connection time (run overnight on working machine)

Possible indication towards table - test it (no problems, and errors were
not all here anyway)

Check for fields with caps (possible act of PEBKAC)

Check User data (for unexpected or wrong data)

Logging ODBC trace

Compare data structure with last known working

Added indexes to tables that don't have them

Vacuum tables again

None of these have given my anything useful. The problem is happening on
20(ish) machines, and there is no defined pattern as to which machines are
not gtting the errors (although this could just be a matter of usage). The
errors appear after random intervals (5mins/10mins/30mins/etc.) and the
application frontend has a connection sustaining module (that definitely
works).

if anyone could help, I'd be very much obliged.

Thank you.