|
> I don't know much about Utramonkey, but since Horms
> is
> skiing in Australia and probably won't be reading
> this
> posting for a while, I'll have a go for what it's
> worth.
vacation?!?! the nerve! horms should be fired.
>
> These IPs are all x.x.100.x
>
> > I have the following IP setup:
> > VIP: 192.168.100.98
> > RIP: 192.168.100.96 (server1)
> > RIP: 192.168.100.97 (server2)
>
> these IPs are all x.x.0.x
>
> is this part of the problem?
crapolla. no. they are really 192.168.0.x IPs. when
i started the email i simply used 192.168.100.x to
explain, then forgot to change the IPs in the
"screenshot" ;-)
i actually just got it working - sorta. (I'll use
actual IPs this time)
i was using the example from ultramonkey.org and setup
/etc/sysconfig/network-scripts/ifcfg-lo:0 file which
still uses "ifconfig". i removed it, and setup
192.168.0.98 on lo using the "ip" utility instead of
"ifconfig" and a lo:0 alias (keepalived uses "ip" as
well). This worked, with both director-servers
serving content correctly, but then the IP got lost on
failover. As if eth0 wasn't sending ARP broadcasts,
the client's still had the VIP associated with the
original master director's MAC address, even though
the backup director now owned the IP.
-Ken
__________________________________
Do you Yahoo!?
Yahoo! SiteBuilder - Free, easy-to-use web site design software
http://sitebuilder.yahoo.com
|
