Hi,
We need to set up a high-availability solution with Post-
greSQL. I've done some research and found several different
solutions for replication, but I'm somewhat confused about
which of them might be really suitable for us.
Currently we have a single PostgreSQL machine which runs
fine. It serves mostly databases for web CGI applications.
However, we need to extend it into high-availability set-up,
so the applications will continue to work if the PostgreSQL
server goes down.
We also currently have a small mysql cluster, which works
using master-slave replication with a separate load-balancer
in front. All updates are replicated from the master to the
slave. Normally, only the master is accesses by clients
(through the load balancer). If the balancer detects that
the master went out for lunch, it switches over to the
slave, so all accesses will go to the slave -- even if the
master returns, because it will be out-of-sync by now. The
system generates alert messages and requires manual
intervention: the slave will become the new master, all
changes will be synced back to the old master which becomes
the new slave.
Is there something for PostgreSQL wich works similarly, or
even better? The web applications perform SELECTs as well
as UPDATEs, so the slave must not be read-only in the event
of a fail-over.
I would like to migrate completely away from mysql, but it
requires that the high-availability problem is solved in a
reasonable way. A free (i.e. non-commercial) solution is
preferred, because the mysql setup was free, too, so there
is no budget available for this.
Any hints and pointers would be greatly appreciated!
Thankyou very much in advance!
Best regards
Oliver
--
Oliver Fromme, secnetix GmbH & Co KG, Oettingenstr. 2, 80538 München
Any opinions expressed in this message may be personal to the author
and may not necessarily reflect the opinions of secnetix in any way.
"Being really good at C++ is like being really good
at using rocks to sharpen sticks."
-- Thant Tessman
