[Pgcluster-general] Duplicate inserts, cluster node no recovering

[Raúl Alexis Betancor Santana]

On Monday 04 June 2007 11:42:33 Karsten Nielsen wrote:
> On Mon, 2007-06-04 at 08:22 +0000, Raúl Alexis Betancor Santana wrote:
> > El Monday 04 June 2007 07:44:25 Karsten Nielsen escribió:
> > > Hi,
> > >
> > > I have access to 64bit dual opteron system that is my setup.
> > >
> > > If I do an insert through my lb the insert is executed on both
> > > clusterdb's and that is the point og the cluster is it not?
> >
> > Yes it runs that way, but inserts are duplicated on the db's nodes
>
> Sorry but I cannot replicte that error on my debian etch amd64 running
> pgcluster 1.7.0.rc7

I'll try with 1.7.0rc7 this evening, the error I reported was with 1.7.0rc5 .. 
so i'll try with the new patch

Have you modified the init.d files to run with the -R parameter to recover 
from a dissaster? .. I ask you, because you told you use Debian like me and 
I'm trying to figure a way to "auto recover" a db node from a fail out.

-- 
Raúl Alexis Betancor Santana
Dimensión Virtual S.L.