Hello,

I'm maintaining an application that uses Slony-I. Users may very
occasionally stop replication, to build a backlog of updates, and
later restart replication and propagate that backlog all at once. It
isn't likely to be used more than every once in a long while. At the
moment, I simply stop the Slony-I windows service (i.e. all slon
daemons), and later restart it through win32 API calls. This has the
considerable disadvantage of requiring that the user be present on the
master - if the user is connecting to the database remotely, the Slony
service cannot be restarted. It also requires that the user have OS
administrative privileges.

My questions are:

1. Is stopping Slony-I in this manner sensible? I've had acceptable
results so far, but I'm aware that "Extended periods of downtime will
require to remove or deactivate the node in question in the
configuration", and that Slony-I is not suitable for  replicating
"offline nodes that only become available sporadic [sic] for
synchronization".

2. If it is sensible, is it possible to achieve the same result
through calls to "bare metal" slony functions?

Thanks,
Peter Geoghegan