|
> Hi.
>
> I now have a small cluster of RedHat 7.2 machines. I have two lvs load
> balancers running Piranha-0.6.x. They are serving a vip for two web servers.
>
> The problem is this: When I reboot the primary lvs the secondary seems to
> take over as it should. However after a few minutes, the (the primary one
> is back up) the secondary lvs stops funcitoning. I look at the processes
> running and it lists nanny for the two web servers, lvs and pulse (and
> piranha-gui) -- looks normal. From my browser, however, I get a 404
> error. When I do a ipvsadm -l on the secondary server, I active and
> inactive connections both register as 0. When I shutdown the pulse service
> on the secondary lvs machine, the first take it over again, and everything
> seems fine (so long as the primary remains up).
>
> When the first comes back up, is the second supposed give control back to
> it? Or does it take the job of lvs until it goes down (then the primary
> takes it back)? It is extremely frustrating. What good is clustering if
> the fail-over doesn't work?
add --forceactive to the pulse command for the primary server's init script.
This will force the primary to always try to be the active server.
# snip
case "$1" in
start)
echo -n "Starting pulse: "
daemon pulse --forceactive
RETVAL=$?
echo
[ $RETVAL -eq 0 ] && touch /var/lock/subsys/pulse
;;
# end snip
>
> I really appreciate your help.
>
> Chris
>
> Here is a copy of my lvs.cf:
>
> serial_no = 71
> primary = 10.100.100.2
> primary_private = 192.168.0.1
> service = lvs
> rsh_command = ssh
> backup_active = 1
> backup = 10.100.100.3
> backup_private = 192.168.0.2
> heartbeat = 1
> heartbeat_port = 539
> keepalive = 2
> deadtime = 5
> network = nat
> nat_router = 10.200.200.8 eth1:1
> nat_nmask = 255.255.255.0
> reservation_conflict_action = preempt
> debug_level = NONE
> virtual website {
> active = 1
> address = 10.100.100.8 eth0:1
> vip_nmask = 255.255.255.0
> port = 80
> send = "GET / HTTP/1.0\r\n\r\n"
> expect = "HTTP"
> load_monitor = ruptime
> scheduler = wlc
> protocol = tcp
> timeout = 6
> reentry = 15
> quiesce_server = 0
> server gluttony {
> address = 10.200.200.3
> active = 1
> weight = 50
> }
> server desire {
> address = 10.200.200.6
> active = 1
> weight = 50
> }
> }
>
>
> _______________________________________________
> LinuxVirtualServer.org mailing list - lvs-users@xxxxxxxxxxxxxxxxxxxxxx
> Send requests to lvs-users-request@xxxxxxxxxxxxxxxxxxxxxx
> or go to http://www.in-addr.de/mailman/listinfo/lvs-users
>
|
