Le Fri, 19 Oct 2007 12:43:40 -0400, Sean Brown a écrit :
> We are setting up a new database system that is going to act as our
> central database. We are planning to have a Master/Slave setup,
> originally with one slave but there is a strong possibility of adding a
> second slave at our second site. We could run Mulit-Master, it just
> wasn't the original plan. (The original plan was MySQL log shipping). We
> do not want the possibility of loosing data in the event of loosing the
> master node so synchronous replication is preferable. We are using
> BLOBS, so the replication has to be able to handle that. The slave we
> want to have useable but it would be used for reporting only under
> normal usage.
Have a look at pgpool-I or pgpool-II in replication mode.
Pgpool proxies SQL requests, so that requests on one server are replayed
on the other. pgpool-I is limited to two nodes, whereas pgpool-II is not.
Latency is important for write performance. You can do failover, you just
need to detect by yourself when one node is dead.
There is also PITR recovery, but as it batches the transaction logs you
may lose a bit of data (up to a configured data size / duration).
In response to
pgsql-admin by date
|Next:||From: Ewald Geschwinde||Date: 2007-10-22 09:05:55|
|Subject: Re: Cluster/Replication|
|Previous:||From: Tino Schwarze||Date: 2007-10-22 08:17:20|
|Subject: Re: Postgres SQL with Suse Linux|