On 10.05.2007 06:30, David Wall wrote:
On Wed, 2007-05-09 at 14:40 -0700, David Wall wrote:
Is there a "preferred" replication system for PG 8 db users?
Obviously, we're looking for robustness, ease of
operations/installation, low latency and efficient with system and
network resources, with an active open source community being preferred.
Jeff Davis wrote:
http://www.postgresql.org/docs/8.2/static/high-availability.html
Thanks. I've seen the options and was hoping for grunt-level
realities. Many projects seem to have fallen by the wayside over time.
My first impression was towards a Slony-I type solution, but I need
large objects and would prefer schema updates to be automatic. I was
hoping to hear back on any pitfalls or preferences or "how I'd do it if
I could do it again" type stories. We mostly need it for disaster
recovery since we're looking to improve upon our current nightly
backup/syncs in which we pg_dump the database, SCP it to the backup,
then pg_restore on the backup. It's possible WAL copying will do it,
too, but don't know if people find this workable or not.
Replicate the whole block device, PostgreSQL sits on.
--
Regards,
Hannes Dorbath
---------------------------(end of broadcast)---------------------------
TIP 9: In versions below 8.0, the planner will ignore your desire to
choose an index scan if your joining column's datatypes do not
match
