Christopher Kings-Lynne wrote:
> Hi guys,
>
> Just out of interest, if someone was going to pay you to hack on Postgres
> for 6 months, what would you like to code for 7.4?
Replication. Replication. Replication. Replication. Replication.
Replication. Replication. Replication. Replication. Replication.
Replication. Replication. Replication. Replication. Replication.
Well, jokes apart, I think this is one of the most needed features to
me. Currently I'm using strange voodoo to replicate some tables on other
machines in order to spread load and resilency. Compared to what I am
doing now a good master to slave replication would be heaven.
I understand that a good replication is painful but in my experience, if
you start by integrating some rude, experimental implementation in the
mainstream PostgreSQL the rest will come by itself.
For example, RI was something I wouldn't consider "production level" in
7.2, but was a start, now in 7.3 is much much better, probably complete
in the most important parts.
Other wishes (not as important as the replication issue) are:
- Better granularity of security and access control, like in mysql.
- Ability to reset the state of an open backend, including aborting open
transaction to allow for better connection pooling and reusing, maybe
giving the client the ability to switch between users...
Bye!
-- Daniele Orlandi Planet Srl