Hi,
Le mercredi 16 avril 2008, Rob Collins a écrit :
> To clarify the requirement: much like you, we're not looking for
> synchronous replication, which would be too slow. The branches and central
> server can be different for about 5 to 10 minutes. But the branches need to
> be able to function independently if the network or central goes down.
Exactly my need, with the addition that central edited tables and branches
local edited tables are disjoint sets, which makes the point for master/slave
replication choice.
I'm experiencing 1.5s and 4.8s average lag on my production setup, with
respectively 20tps and 200tps (insert/update/delete only).
> Londiste looks interesting, though the documentation seems a bit sparse. Is
> it really that simple to set up, or is there a fair amount of trial and
> error in the setup process?
I had the same question some time ago when I wanted to choose between londiste
and other solutions, and came to write up a part of the current
documentation.
In my experience, londiste is really that easy to set up, the only trial and
errors I've had where either bad reading of the doc (forgot to launch pgq
daemon, for example) or bad psycopg version usage.
In my case, but several people on skytools mailing-list have different
experience, I had to stick with psycopg 1 and avoid 2.0.5 (debian stable
version). YMMV.
I want to add skytools-users@pgfoundry.org mailing list is pretty responsive,
don't hesitate asking there :)
Regards,
--
dim