problems on INSERTS via ODBC?

From: "Moray McConnachie" <moray(dot)mcconnachie(at)computing-services(dot)oxford(dot)ac(dot)uk>
To: <pgsql-general(at)postgresql(dot)org>
Subject: problems on INSERTS via ODBC?
Date: 1999-11-25 18:44:54
Message-ID: 003e01bf3775$2a3faa60$760e01a3@oucs.ox.ac.uk
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-general

6.5.3 running on RedHat Linux 6.0. Front-end running on MS Access 2000
over ODBC on WinNT

I have recently been experiencing some worrying problems whereby an
"INSERT INTO..." query executed in Access succeeds to all appearances.
Opening the linked table or querying it from Access shows that the
records are present. However, on the server itself these records are
not there, nor are they there if you quit Access and enter it again.

I have debug logs to level 2 off the server, and the psql odbc driver
log. I'm trying to understand them to work out if it was my query
which was somehow causing the problem. I'm not very conscious of where
the responsibility is under ODBC connections to report problems, with
the server, or with the driver.

I think I have solved the problems by dropping all indices, rebuilding
them, and redesigning the problematic queries. (One of the functions
in one had an incorrectly specified parameter - however, Access did
not complain while building the insert, and the final insert looks
correct).

I am worried as to whether this might happen again, and worried that
it could happen at all. I don't know whether the problem is in the
client, the server, or the ODBC driver, but the trouble about this
kind of error is that it is the worst type - records apparently being
inserted correctly, but in fact not being so.

I attach a part of the server log - the inserts seem to be being done
on a one-by-one basis, and the partial log attached contains all
references to one of them. Could someone maybe take a look, and see if
what is recorded in these logs explains the behaviour? It seems
particularly odd, because the final insert appears to have nothing
wrong with it!

Yours,
Moray McConnachie

----------------------------------------------------------------------
----------------
Moray(dot)McConnachie(at)computing-services(dot)oxford(dot)ac(dot)uk

Attachment Content-Type Size
oneproc.txt text/plain 13.1 KB

Browse pgsql-general by date

  From Date Subject
Next Message Stephen Birch 1999-11-25 20:22:35 Re: [GENERAL] Referencial integrity - Re: Is PostgreSQL ready formission criti calapplications?
Previous Message Bruce Momjian 1999-11-25 18:28:28 Re: [GENERAL] Table names case sensitive?