[TriLUG] clustering or server mirroring

[Matt Pusateri]

On Tue, April 19, 2005 8:58 am, David McDowell said:
> One in the same?  Here's my idea.  I'd like to use CentOS 4 if
> possible to do this.  I would like to have my webserver mirrored on
> another machine so that if one goes down, the site continues to run.
> If I change a config on one machine, the config should change on the
> mirrored machine.  Is this running a cluster or is this some other
> kind of setup?  Basically I have some time at work to play.  Any good
> resources for this kind of information?  Basically I want 2 servers to
> be identical mirrors of one another so that if one of the 2 goes down,
> I'm still online.  And, if I repair the broken one, it can resync
> itself so that the mirror of the 2 machines is identical again.
> Suggestions, links, etc?
>
> thanks,
> David McD


First off unless I am mis-understanding clustering, it is not what you
want.  To me clustering is taking a group of machines and grouping
them together so the can take advantage of a piece of software that
knows how to disperse the processing load across each of the cluster
members.  I think mirrored servers are more what your looking for.

So what about this, you have 1 web server that you consider the
primary web server. You make all changes to Server A and rysnc all
changes to server B. You then set both servers to have same names in
DNS so DNS can do a simple round robin load balancing for you. Now
when server A goes down, all you have to do is figure out how to
manage the down box/dns.   Some of that depends on how down server A
is.  How much of this that becomes automated depends on your script
fu, and how critical web services are.  For me I don't think I would
need to automate more than the web data.  This of course assumes that
your web data is fairly static, other wise both web servers would have
to point to a central DB, which then becomes the single point of
failure unless you do failover for that as well.  If your not running
Nagios.  I would heavily suggest you look into it.  It's one of the
best tools on my network.  I have it monitor all servers and critial
services and sends a email to my phone if things go down.  I still
have to implement a failover for nagios not being able to send the
email due to the t-1 being down.  But that has been a scenario I have
had to face yet :).  In any case most of the time nagios tells me of
the failure before people do and I can work on restoring service.

Note: I wrote this real quick, their are some obvious holes!  And you
need to evaluate what the specific needs are to determine how
intricate your setup would need to be.  I throw these simple
suggestions out just to get everyone's creative thoughts flowing.

Matt