[Pgcluster-general] Last chance to get everything working

[Rick Vernam]

On Thursday 24 May 2007 06:14:31 am Gábriel Ákos wrote:
> On Wed, 23 May 2007 22:01:51 -0500
>
> Rick Vernam <rickv at hobi.com> wrote:
> > Is there a trend here that jdbc->pgcluster does not work well?
>
> Yes I think so as well. I haven't got any reasonable answer of my
> question that a primary key is needed or not when using jdbc with
> replication. My statement that if there is a key then replication
> works, if there is no key, then it doesn't work was simply ignored.
> It is very sad for me as we should have a working cluster very soon and
> I'm in doubt that it will work with pgcluster.
Ahhh...  perhaps your mail 'Table requirements' refers to the requirements of 
JDBC, not pgcluster.
We have an app that connects to pgcluster via the pgODBC driver.
We had to put a few special keys in a couple tables to satisfy the odbc 
driver...
In some cases the client's communication with the odbc driver created queries 
that did not replicate reliably.  Specifically, when the odbc driver would 
create prepared statements.

>
> I will try to investigate the problem further down to the level of C
> code.
>
> Rgds,
> Akos
> _______________________________________________
> Pgcluster-general mailing list
> Pgcluster-general at pgfoundry.org
> http://pgfoundry.org/mailman/listinfo/pgcluster-general