Hello:

We have implemented Slony since 7 months ago. Below are the configuration:

Server A:
Redhat 9, Postgres 8.3, Slony 1.2, Set 1 master, Set 2 Slave.

Server B:
Centos 5, Postgres 8.3, Slony 1.2, Set 1 Slave, Set 2 Master.

Things went very well for the beginning 6 months. Just about a month ago
we are starting to see data not being replicating for Set 2 (perhaps Set 1
as well just hadn't verified it yet). Investigation so far found Slon
daemons 1 and 2 to be running when replication hangs, and reviewing the
logfile had not yield any out of the ordinary messages. The fix so far is
to bounce both Slon daemons and the replication would pick up from where
it had left off.

So anyone has a suggestion on how we should troubleshoot this issue? To be
specify it is kinda important as Set 1 contains parts and on hand data for
real time inventory check, and Set 2 contains order information. I just
wanted to bounce this issue off this list before having to resolve to a
nightly bouncing of the Slon daemon.

Thanks ahead for your assistance!

Regards,

Tim