Thank you for your answer.
Sorry, i forgot to tell the list. I circumvented the problem by  
uninstalling and re-initializing and subscribing the slave. Seems to  
work fine.


Am 27.03.2007 um 02:09 schrieb Andrew Hammond:

> On 3/16/07, Tilman Baumann <tilman.baumann at collax.com> wrote:
>> Hi,
>>
>> in a inattentive moment i addes tome table to the replacation without
>> creating the tables on the slave fist.
>> I used the slony helpers in pgadmin3.
>> I added the new tables to a new replication set. After that i  
>> merged the
>> new set with my previous one. And than it struk me that i missed a  
>> step.
>> I guess if i had not merged the sets. I could remove the slave  
>> note from
>> that set, create the tables and add ist again.
>
> Possibly, but you'd probably have to do some surgery to get rid of the
> subscribe event (and probably subsequence SYNCs) that will never
> actually succeed. At which point you may as well drop the cluster and
> start over again.
>
>> Can anyone please give me a direction what i could do to bring the
>> synchronisation in 'sync' again?
>> Settiung the slave up new would be a ok option. The main thing is  
>> that
>> it syncs all tables after that. :)
>
> Well, it's gonna be hard for that to succeed since you'll never be
> able to get past the subscribe event.
>
>> Are there some tricks to tell slony to have a look for not yet synced
>> tables?
>
> No, slony is not that user-friendly yet. I'm sure that there'd be some
> interest in figuring out how to make it more user-friendly if you had
> a well thought out, well researched proposal along those lines. And
> some willingness to contribute code to the effort.
>
> Andrew