On December 14, 2004 05:00 am, Paul Slootman wrote:
> Hi,
> I'm experimenting a bit with slony, to get a "feel" for it.
> I have a basic replication between two hosts working fine (I first
> forgot about the sequences, which is pretty obvious once you do a
> move set and try to insert data :-).
>
> I just did a drop table of one table (it's not needed anymore by the
> application which uses the database), and then realised that perhaps I
> should have done a "set drop table" first. That doesn't work now, it
> says: "Slony-I: Table with id 1 not found".
>
> - Is there a way around this?

No, currently there is not a nice way around this.  To fix this you'll have to 
manualy delete rows from the _CLUSTER.* tables; most notably sl_table , 
sl_log_1 , sl_log_2  on all the nodes.

> - Would it be bad if I just left it as it was?

I'm not 100% sure if there are any ill effects by having a nonexistant table, 
I'm assuming not since data is pushed into the _CLUSTER.* tables.  but I 
personaly would try to clean them up.

> - Is there a way to list what tables have what IDs? I happened to know
>   that this table was added with ID 1, but I wonder how to find it out
>   if you don't know this...

No you can't not after it has been dropped.  If you want to find out the real 
relation name of a object you havn't yet deleted youcan use the following:

SELECT slt.tab_id, pgc.relname FROM pg_class pgc, "_cluster".sl_table slt 
WHERE pgc.oid = slt.reloid ......


On a side note, this will be easier in 1.1

>
>
> Thanks,
> Paul Slootman
> _______________________________________________
> Slony1-general mailing list
> Slony1-general at gborg.postgresql.org
> http://gborg.postgresql.org/mailman/listinfo/slony1-general

-- 
Darcy Buskermolen
Wavefire Technologies Corp.
ph: 250.717.0200
fx:  250.763.1759
http://www.wavefire.com