On 20:56, Sat 28 Jun 08, Juan Miscaro wrote:
> Hi, using OpenBSD (4.3 current), what is the recommended way to set up
> application level failover?  Let's say I have content that is
> constantly edited by several internal users (like a webserver dishing
> up text and images) and is available to the internet.  If I want to
> have failover that content needs to synchronize real-time with a
> second machine.  Apart from setting up the network part (with relayd I
> think), how is the data synchronization done?  For a more complex
> case, how is failover achieved for IMAP services?

What we do is have some sort of central storage.
We use netapp for that (and on another setup we have a HA NFS server
based on linux with drbd to do the sync of the data between the two NFS
servers)

A bunch of boxen is mounting this storage using NFS and running the
services.
A pair of OpenBSD firewalls with relayd is doing the
loadbalancing/failover stuff.

-- 

Michiel van Baak
[EMAIL PROTECTED]
http://michiel.vanbaak.eu
GnuPG key: http://pgp.mit.edu:11371/pks/lookup?op=get&search=0x71C946BD

"Why is it drug addicts and computer aficionados are both called users?"

Reply via email to