|
Hi Leandro
On Mon, 2008-12-15 at 18:22 +0000, Leandro Sales wrote:
> I have a basic configuration of LVS-NAT pointing to two real-servers
> (A and B). In case of a failure in one of the real servers (let's say
> A), does LVS take care of detecting this failure and always redirect
> requests to real-server B? Or should I have to use keepalived to
> provide this function?
LVS does not provide any health checking at all. There are many ways you
could do it, but the most popular are:
ldirectord
keepalived
They work in different ways to achieve the same end. Given the "basic"
nature of your setup, you could just as easily write your own scripts to
do this using a combination of (for example, for webservers)
shell/curl/ipvsadm or something similar.
If I were you I'd stick with one of the above pair, though - as you gain
complexity you'll pay for rolling your own (in most cases) :-)
Graeme
_______________________________________________
Please read the documentation before posting - it's available at:
http://www.linuxvirtualserver.org/
LinuxVirtualServer.org mailing list - lvs-users@xxxxxxxxxxxxxxxxxxxxxx
Send requests to lvs-users-request@xxxxxxxxxxxxxxxxxxxxxx
or go to http://lists.graemef.net/mailman/listinfo/lvs-users
|
