On Tuesday September 21 2004 10:33, Jan Wieck wrote:
>
> The other way around ... on the subscribers (slaves), the subscribed
> tables have their triggers and rules parked on their PK index OID ...
> which is ugly and eggebeh and pfui, but it's the way it is for now. So
> simply dropping the replication schema will leave the slave database in
> an inconsistent state.

Attached is a complete example script of what I'm doing start to finish for 
a simple DB with one table and one replicated insert.  You should be able 
to change the shell variables at the top and run it to see what I do.  
Again, I don't see any replication fingerprints whatsoever left over on the 
slave, at least in the dump, after dropping the replication schema.  Still 
curious why am I not seeing what you describe?

Ed
-------------- next part --------------
A non-text attachment was scrubbed...
Name: example.sh
Type: application/x-shellscript
Size: 2749 bytes
Desc: not available
Url : http://gborg.postgresql.org/pipermail/slony1-general/attachments/20040921/d22a4a5f/example-0001.bin