|
iceam@xxxxxxxxxx wrote:
>
> I did the following:
>
> Director:
> eth0 = 10.1.1.1/24
> eth0:0 = 192.168.1.110/24
>
> Real1:
> eth0 = 10.1.1.2/24
>
> Real2:
> eth0 = 10.1.1.3/24
>
> I have used the following lvs_nat.conf file:
>
> LVS_TYPE=VS_NAT
> INITIAL_STATE=on
> VIP=eth0:0 extdirector.icepicc.co.uk 255.255.255.0
> 192.168.1.110
> DIRECTOR_INSIDEIP=eth0 director.server.net 10.1.1.1
> 255.255.255.0 10.1.1.0
> DIRECTOR_DEFAULT_GW=192.168.1.110
> SERVICE=t http wlc real1.server.net:80
> SERVER_NET_DEVICE=eth0
> SERVER_DEFAULT_GW=director.server.net
>
> If I don't put a gateway in for the director, the script
> rc.lvs_nat won't run.
sorry, my mistake. You must set a default_gw for the director.
The default_gw for the real-servers is set automatically for you
by the script in VS-NAT
> Plus I don't have fping, which means
> when I run rc.lvs_nat, I get errors about that. Replacing
> the FPING=/usr/.....fping line with:
>
> FPING="/bin/ping -c 3" makes it work.
You handled it. I'll put a note in about this somewhere.
Presumably old pings don't do this and fping must have been specially
written.
I have run the script
> on all of the machines. I can ping the director on both
> interfaces from the real servers, and I can ping the real servers from the
> director. When I try to connect to the director (on its VIP) using a client
> with IP 192.168.1.254,
> the director spews out errors like:
where does it spew these out?
which version of ipvs/kernel are you running?
> IP_VS: Schedule: port zero only supported in persistent
> services, check your ipvs configuration
>
> Is that any more helpful? I was a little non-descript last time.
please send the output of ipvsadm
> Oh, and Joe, all your posts are coming up twice.
hopefully you'll only get one this time.
Joe
--
Joseph Mack PhD, Senior Systems Engineer, Lockheed Martin
contractor to the National Environmental Supercomputer Center,
mailto:mack.joseph@xxxxxxx ph# 919-541-0007, RTP, NC, USA
|
