> first of all please excuse my English - I'm trying to improve, but
> ...
[quoted text clipped - 13 lines]
> that loc<i>1.mdb synchronizes with loc<j>2.mdb and loc<i>1.mdb
> synchronizes with loc<j>3.mdb indirectly ?
If the synchronizers can see each other, you can set up a
synchronization schedule in Replication Manager.
> Or does this topology don#t make any sense at all ?
Absolutely!
All sounds correct already, and all you need to do is initiate a
synch between two of the synchronizers to see if it works.
I'm assuming that when you look at the Replication Manager diagram
of the replica set, you see all three of your synchronizers? If not,
that means the synchronizers don't know about each other, so can't
synch with each other.
The easiest way to rectify that is to use an Empty Partial Replica
to transfer it's knowledge about the replication topology without
needing to also transfer data (this makes the MDB very small).
Michael Kaplan wrote an article explaining how to do this when
synchronizers have become invalidated:
http://trigeminal.com/usenet/usenet016.asp
and at the end, he cites my CEPR utility, an Access MDB that has a
dialog form that helps you create the empty partial replica for
sending to the remote site. THe direct link to the download is:
http://trigeminal.com/samples/CEPR.zip
I recently used it to initiate synchronization between two
previously non-connected sites, and it still works just great.

Signature
David W. Fenton http://www.bway.net/~dfenton
dfenton at bway dot net http://www.bway.net/~dfassoc
Uelsasser - 13 Aug 2005 08:15 GMT
Hi David,
thanks a lot - this works perfectly
Uli