[prev in list] [next in list] [prev in thread] [next in thread] 

List:       linux-ha
Subject:    Re: [Linux-HA] [Newbie Question] Can one create a "bi-directional"
From:       David Lang <david.lang () digitalinsight ! com>
Date:       2004-01-29 8:31:44
Message-ID: Pine.LNX.4.58.0401290027560.12731 () dlang ! diginsite ! com
[Download RAW message or body]

On Tue, 28 Oct 2003, Owen McShane wrote:

> > On Tue, Oct 28, 2003 at 10:48:18AM +0200, Wayne de Villiers wrote:
> > > Hi Alan
> > > I was thinking of permanently configuring both servers to handle all 8
> > > web sites' requests.
> > > 
> > > All I then need to do is have the relevant server take on the dead
> > > server's IP address.  Apache will already be listening on all
> > > interfaces.  Do you think that's do-able?
> > 
> > Yes, just ommit the apache bit from the haresoures lines below.
> > Personally I would recommend this, there doesn't seem any point in
> > (re)starting apache if you don't need to.
> 
> Correct me if I'm wrong, but Apache will have to be (re)started for it to bind to \
> the new ips, unless you have all ips on both machines, and then have a shared ip \
> which fails over, and the ips are routed to this address.

sorry for the (very) late reply, I am finally catching up with old e-mail
and this doesn't seem to have been answered.

if apache is configured to bind to specific IP addresses then you will
need to change it's config and restart it (it won't even start if the IP
addresses aren't available)

however if you configure it to bind to * (the default) then it will
receive any packets that arrive on the machine (assuming it's the right
port) so just moving the IP addresses is enough.

I agree with Horms that there is no gain from restarting apache.

in fact I will go further and say that in general the entries in
haresources should be pruned to the minimum that is nessasary to change
when the machine becomes active, anything that can be running all the time
should be as it makes the failover go faster

David Lang

-- 
"Debugging is twice as hard as writing the code in the first place.
Therefore, if you write the code as cleverly as possible, you are,
by definition, not smart enough to debug it." - Brian W. Kernighan
_______________________________________________
Linux-HA mailing list
Linux-HA@lists.linux-ha.org
http://lists.linux-ha.org/mailman/listinfo/linux-ha


[prev in list] [next in list] [prev in thread] [next in thread]