[lvs-users] CentOS-5 VLS doesn't virtual httpd doesnt failover when httpd dies

[Mark]

> > I apologize, CentOS = RHEL, and the default monitoring 
> script is simply 
> > and http get:
> >
> >      send = "GET / HTTP/1.0\r\n\r\n"
> >      expect = "1.3.37"
> >      use_regex = 0
> >   
> OK. So you're using piranha. Can you send the full config ?
> 
> Using 1.3.37 as the expect string is, IMHO, a bad idea because you'll 
> have to alter your config once you update your realservers 
> and will have 
> problems if not all RS are running the same version. It think 
> the "real" 
> default was more something like "HTTP 200". Anyway.

Maybe this is already happening right now - it might actually be the reason why it does not fail over to the other server.
If the fallback server returns the wrong version then it was never added to the pool to begin with, so it would never be used to
fail back to.


MARK