Does anyone have a working keepalived.conf they'd like to share which has
at least two virtual_server blocks which refer to the same VIP?
Keepalived sets everything up fine for me as long as I only declare one
service per IP. As soon as I add another then one of two things happens:
The healthchecker fails or, the healthcheck succeeds but instead of adding
a new service, another realserver is added to the first virtual_server.
For instance, if I have:
virtual_server 206.114.148.54 22 {
delay_loop 10
lb_algo rr
lb_kind NAT
protocol TCP
real_server 10.0.0.100 22 {
TCP_CHECK { connect_timeout 3 }
}
}
virtual_server 206.114.148.54 80 {
delay_loop 10
lb_algo rr
lb_kind NAT
protocol TCP
real_server 10.0.0.100 80 {
TCP_CHECK { connect_timeout 3 }
}
}
Then I see in the log:
Apr 25 16:49:00 bbi-lvs1 Keepalived_healthcheckers: TCP connection to
[10.0.0.100:80] failed !!!
Apr 25 16:49:00 bbi-lvs1 Keepalived_healthcheckers: Removing service
[10.0.0.100:80] from VS [206.114.148.54:22]
I don't know why it fails first, since I can telnet to the port, or why it
is removing it from VS [206.114.148.54:22].
Thanks in advance for any good ideas.
--
mrb
|