LVS
lvs-users
Google
 
Web LinuxVirtualServer.org

Redundant Load balanced cluster,

To: "LinuxVirtualServer.org users mailing list." <lvs-users@xxxxxxxxxxxxxxxxxxxxxx>
Subject: Redundant Load balanced cluster,
From: Jan Klopper <janklopper@xxxxxxxxx>
Date: Tue, 15 Feb 2005 14:51:10 +0100
I just bought 5 new servers to replace/extend my old 2 webservers.

I'd like to create a cluster out of them, and i had the following setup
in mind.

Two machines working as routers/proxy's/cache;s (is this needed when
using lvs?)
These would be connected with a hearthbeat serial cable to do redundancy.

The rest of the machines would be apache's (and if possible, the two
routers would also double as apache servers)
There would be one dedicated Mysql server. (a 6th machine)
There would be one NFS file server for the webroots, backups/logs. (7th
server)

The file server would also be a mysql slave (trough binary logs) to
always have a very recent backup of the mysql database. (not intended
for queries)

Would this be a good setup?
The 2 old servers now in service could have their websites migrated to
the cluster, en than ater on join the cluster.
Internal lan will be 100mbits, internet on both the routers also 100mbit.
Hearthbeats between routers and between mysqls will be serial.

The application which i going to run on this is a really database heavy
php application, caching with squids won't do much good. And i need
loads of apache/mysql power to crunch the numbers.

I think this setup would do, do you guys see any weak points?

I have a little bit of a problem with the single NFS server, and i hear
people say "coda" to me everytime, would that be a good sollution? Or
will it be good enough to rsync the webroots to the apache's since that
will also increase speed?

Thanks for any advice. (im still wondering why an application developer
would need to build his own clusters :( )

<Prev in Thread] Current Thread [Next in Thread>