> I'm setting up my first Slony cluster (Slony 1.1.5), using the altperl
> tools in a simple master-slave setup with two machines.
>
> I've set up my slon_tools.conf, have duped the schema on the slave, have
> run slonik_init_cluster, have fired up slon daemons on each node, have
> run slonik_create_set, and then slonik_subscribe_set set1 node2.
>
> This all appears to be working as it should. In the slon logs I see
> Slony copying over all the sequences, and then truncating and copying
> each table's data in turn on the slave.
>
> However, once it's gone through the whole schema, it appears to be
> looping and doing the same process of copying the sequences, truncating
> and populating all the tables over and over again.
>
> As I understand it once it has done this initially there should be no
> more need to truncate.

You understand correctly...

> Any thoughts on why this might be happening? A pg_dump of this database
> is about 13G so it's somewhat sizeable.

The reason for this to keep repeating is that something broke along the way.

Was there some error somewhere in the process?  If a table or sequence was
missing, that could cause it.  If the COPY_SET event deadlocked against
some VACUUMs, that could cause it.

I think the answer to this is probably to Look To Thine Logs...