Обсуждение: How do clients failover in hot standby/SR?

We are looking at a replication solution aimed at high availability.

So we want to use PostgreSQL 9's streaming replication/hot standby. But I seem to be missing a very basic piece of
information:suppose the primary is host1 and the secondary is host2. Suppose that when host1 fails host2 detects that
andcreates the trigger file that causes the secondary to act as primary. 

How do all clients, which have connection strings aimed at host1 know to fail over and use host2?

Is there a good Internet resource for reading on this?

Thank you,
Herouth

Herouth Maoz <herouth@unicell.co.il> wrote:

> We are looking at a replication solution aimed at high availability.
>
> So we want to use PostgreSQL 9's streaming replication/hot standby.
> But I seem to be missing a very basic piece of information: suppose
> the primary is host1 and the secondary is host2. Suppose that when
> host1 fails host2 detects that and creates the trigger file that
> causes the secondary to act as primary.
>
> How do all clients, which have connection strings aimed at host1 know
> to fail over and use host2?

You can, for instance, use pgpool as connection-pooler. pgpool can
detect a failed node, can create the trigger-file and connects clients
now to the other server.

>
> Is there a good Internet resource for reading on this?

google -> pgpool, for instance. There are other solutions, heartbeat for
instance (with flying service-ip's).


Andreas
--
Really, I'm not out to destroy Microsoft. That will just be a completely
unintentional side effect.                              (Linus Torvalds)
"If I was god, I would recompile penguin with --enable-fly."   (unknown)
Kaufbach, Saxony, Germany, Europe.              N 51.05082°, E 13.56889°