[Pgcluster-general] Hanging through Perl

[fixed]

Hi,

Well we made the transition from MySQL to PostgreSQL and we are running
CyberCluster which appears to basically be PGCluster (am I correct)?

Cybercluster is here: http://www.postgresql.at/english/start_e.html

We are running PostgreSQL 8.2.4 and have Replication just running on one
server (for now). Things seem to be running fine. Here is the config:

<Cluster_Server_Info>
    <Host_Name>   db00.ourhost.com  </Host_Name>
    <Port>        5432                </Port>
    <Recovery_Port>       7001        </Recovery_Port>
</Cluster_Server_Info>

<Host_Name> db00.ourhost.com	</Host_Name>
<Replication_Port>		8001		</Replication_Port>
<Recovery_Port>			8101		</Recovery_Port>
<RLOG_Port>				8301		</RLOG_Port>
<Response_Mode>			normal		</Response_Mode>
<Use_Replication_Log>	no			</Use_Replication_Log>
<Replication_Timeout>	1min		</Replication_Timeout>
<LifeCheck_Timeout>		3s
</LifeCheck_Timeout>
<LifeCheck_Interval>	15s			</LifeCheck_Interval>

I've noticed that when we run a Perl script which uses DBI, a simple
"SELECT count(*) FROM table" causes a hang. Nothing happens. In the
process list is a BIND process. However, when we run the same SQL
statement through the command line interface (psql), things come up in
less than a second. If I stop the pgreplicate process, the query from the
Perl script is run immediately.

Is there something odd with DBI that doesn't play well with pgreplicate? I
have searched the lists and found little on this.  Any input is
appreciated.

Thank you

Ogden