"Benezra, Eric " <[EMAIL PROTECTED]> writes: > Hi everyone, > > I am very close to having replication work but I am running into a problem > now. Both my configuration and replication scripts run without error. When I > run my > configuration script, the _newschema is created in both the master and slave > databases with all the slony default tables. The replication set is also > created on both > machines except there is one difference. On the master, under replication > set--->nodes, both the master and slave contain 1 path and 1 listen. On the > slave machine, > the master node doesn't contain any paths or listens and there are 0 > subscriptions and the table is not replicated over. All connection > information is correct. > > I was wondering if there is a problem in my scripts that is preventing the > subscription (and table replication) to the slave node. Here are my two > scripts. ANY help at > all would be greatly appreciated!! Thanks!!
This sounds a lot like what happens when you submit slonik scripts to configure the cluster, but then haven't started up slon processes to do the replication work. Normally, if you neglect to start the slons, the two nodes will get initialized with the Slony-I cluster's namespace, and you'll see (generally speaking) that each node has half of the "STORE PATH" configuration in table sl_path, where the other half is on some other node. Start up slons and the STORE_PATH event should propagate from node to node, thereby populating sl_path more completely, and SUBSCRIBE_SET should be able to run. Alternatively, you should be able to report what errors were in the slon logs... But I think you haven't got slon processes running... -- "cbbrowne","@","cbbrowne.com" http://linuxfinances.info/info/rdbms.html "My definition of a free society is a society where it is safe to be unpopular." -- Adlai Stevenson _______________________________________________ Slony1-general mailing list [email protected] http://gborg.postgresql.org/mailman/listinfo/slony1-general
