Hi.

I don't now if my database is too complex for Slony, or if I'm doing something 
wrong, anyway, I cannot replicate my database. 

Could you tell me if Foreign keys, Constraints and Triggers can exist in 
replicated database? If yes - can they also exist in slave?

First of all, I have a problem (already reported on this list and not 
answered) that if I prepare for replication database containing data - after 
starting of replication (as described in "Time To Replicate" in README)  I 
receive a lot of errors:

WARN   remoteWorkerThread_1: data copy for set 1 failed - sleep 60 seconds
ERROR  remoteWorkerThread_1: "delete from public.agents; copy public.agents 
from stdin; " 
ERROR:  Slony-I: Table agents is replicated and cannot be modified on a 
subscriber node

Second: what should I do with tables containing primary key  based on multiple 
columns and what without primary key at all? Is 'table add key' enough for 
them?

I have an idea to first create database schema (on master and slave, but just 
tables, primary keys and views), then prepare it for slony, then import data 
into master and finally create foreign keys, triggers, constraints etc. Will 
it work, or not, or it depends on my database schema?

TIA,

Richard.
-- 
"First they ignore you. Then they laugh at you. Then they
fight you. Then you win." - Mohandas Gandhi.