I misunderstood your original intent, I thought this was a one time process to move data to the central database. Given
that it is to be a continuous process a FDW may not be the answer, one of the reasons being the above question. You will
be denied the data in the remote table during the outage. Also not sure what you will be doing with the data in the
central database and how often? In any case it will involve reaching out to all the satellites each time you want to
query the latest data. Looks more like some kind of push mechanism from the satellites to the central database is in
order. Then once the data is on the central database it is 'captured'. A question that comes to mind is if there is a
problem should the data transfer from one or more satellites lag that of the others?
I am sorry. Yes, I forgot to mention that the process should be a continous push.
If not all data is in the central, that's ok.
The central should only see the data which is already in the central place.
Up to now it is not important who triggers this process. It could be a push (from satellite to central) or a pull (from
central to satellite)
If there is a lag ... then there is a lag. That can happen (network outage), this is not very important.
More important is that no single row gets lost or duplicated.
Regards,
Thomas
--
Thomas Guettler http://www.thomas-guettler.de/
--
Sent via pgsql-general mailing list (pgsql-general@xxxxxxxxxxxxxx)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-general
