iI followed the tutorial "Installing A Web, Email And MySQL Database Cluster (Mirror) On Debian 5.0 With ISPConfig 3" and it all seems to have completed correctly. So the theory is that I can place web sites in the first server and they will replicate (glusterfs) to the other server, right? My question is how does the "cluster" work? What drives the traffic to one server or the other? If www.domain.com points to the first ip, then what happens to load balance or to redirect to the second server if the first goes down? How or where is this set up or is it already? The tutorial seemed to end abruptly. It would have been nice to have a "Try this to see how it works" section. Any insight?
If you have a failover-IP address that you can switch between the master and the mirror (e.g. automatically with heartbeat/keepalived/etc. or manually, e.g. from your hoster's control panel), you can achieve high-availability because if the master fails, the mirror can take over.
These tutorials should give you the idea: http://www.howtoforge.com/setting-u...lancer-with-haproxy-heartbeat-on-debian-lenny http://www.howtoforge.com/setting-u...ancer-with-haproxy-keepalived-on-debian-lenny Some hosters (e.g. Hetzner: http://www.hetzner.de/en/hosting/produkte_rootserver/flexipack/ ) give you a failover IP that you can configure over a web interface.
