On Thu, Feb 21, 2008 at 9:20 AM, Terry Lee Tucker <terry@xxxxxxxxxxxxx> wrote:
> Greetings:
>
> We have been working diligently toward integrating Slony into our production
> databases. We've been having trouble with various tables, although being
> replicated perfectly in the initial replication stage, afterwards, getting
> out of sync.
>
> I have finally figured out what the problem is. We have a Perl process that
> continually updates certain columns across all databases. That Perl process
> calls a function we have written called disable_triggers which updates
> pg_class, setting reltriggers to 0 for the given table, and then later, after
> the work is complete, resetting reltriggers to the original value.
> Unfortunately, during this process, the Slony trigger is disabled as well
> which is causing our problem.
Disabling all triggers is not something you do on a live, running
database with users accessing and possibly changing it, it's something
you do to a database during maintenance when no one else is connected.
You'll have to go with the solution you talked about, i.e. disabling
individual triggers by name, etc...
---------------------------(end of broadcast)---------------------------
TIP 9: In versions below 8.0, the planner will ignore your desire to
choose an index scan if your joining column's datatypes do not
match
