LVS
lvs-users
Google
 
Web LinuxVirtualServer.org

help with Director failover

To: lvs-users@xxxxxxxxxxxxxxxxxxxxxx
Subject: help with Director failover
From: colin hostert <chostert@xxxxxxxxx>
Date: Tue, 17 Jan 2006 16:11:48 -0500
Hello

ipvsadm v1.24
RHEL 4 2.6.9-22.0.1.EL

**NOTE**
I have already posted this to the Red Hat list. However it looks quite dead
so hopefully someone here can help me out.

I have a LVS cluster ( 2 Dir, 3 real servers) set up and working, but am
having problems with director failover. Everything works great during the
failover process except for one thing. When the backup director becomes
active, all settings transfer over except for the default gateway(dfgw). Not
only dose the correct dfgw not transfer over, but it erases the boxes curent
one, so there is no longer a dfgw. The dfgw is defined in the
/etc/sysconfig/network file.

Why doesn't the dfgw transfer over, or at least how can I make it so the box
keeps its assigned dfgw? Am i missing a line the the conf file (see below),
or is it something else?

 [cluster@RHC02 ha]# vi lvs.cf
serial_no = 278
primary = xxx.xxx.xxx.100
service = lvs
backup_active = 1
backup = xxx.xxx.xxx.101
heartbeat = 1
heartbeat_port = 539
keepalive = 15
deadtime = 25
network = nat
nat_router = 192.168.0.1 eth0
nat_nmask = 255.255.255.0
debug_level = NONE
virtual http {
active = 1
address = 150.176.247.100 eth1
vip_nmask = 255.255.255.0
port = 80
send = "GET / HTTP/1.0\r\n\r\n"
expect = "HTTP"
use_regex = 0
load_monitor = none
scheduler = wlc
protocol = tcp
timeout = 6
reentry = 15
quiesce_server = 0
server RHC03 {
address = 192.168.0.21
active = 1
weight = 1
}
server RHC01 {
address = 192.168.0.20
active = 1
weight = 1
}
}

-----------
Thanks !
-----------

<Prev in Thread] Current Thread [Next in Thread>
  • help with Director failover, colin hostert <=