RE: [Pound Mailing List] Pound Failover

I have two boxes, set ip_nonlocal_bind, each one runs pound all the time, and corosync/pacemaker moves the ip between them.  So pound is always running but it only receives requests when the cluster makes the server active.

Joe

> -----Original Message-----
> From: James Bensley [mailto:[email protected]]
> Sent: Tuesday, March 19, 2013 7:08 PM
> To: [email protected]
> Subject: [Pound Mailing List] Pound Failover
> 
> Hi All,
> 
> How are people achieving fail-over with Pound, in the event the server
> becomes unavailable?
> 
> I am thinking of something like keepalived to use VRRP to fail-over to
> a second box if the first one fails (so they share the same public IP),
> 
> OR
> 
> IPVS, again to share the same public.
> 
> However, I haven't had a chance to lab up either scenario yet. Are
> people using either of the scenarios, which do they prefer? Or if you
> are doing something else, what are you doing, how is it working out for
> you?
> 
> I am primarily after active/passive fail-over, active/active is a
> luxury extra.
> 
> Many thanks,
> James.
> 
> --
> To unsubscribe send an email with subject unsubscribe to
> [email protected]
> Please contact [email protected] for questions.

--
To unsubscribe send an email with subject unsubscribe to [email protected]
Please contact [email protected] for questions.