-1

I configured two lvs keepalived servers. The first is master server, ip 10.1.88.76, iptables closed. The second is backup server, ip 10.1.88.77, iptables closed. the network connection between two servers is ok (verified by ping)

But, the problem is both servers are in MASTER state. What I expect is one in Master and the other in BACKUP mode.

What should I do to settle the problem? Thank you

The configuration in 76 server.

! Configuration File for keepalived

global_defs {
   notification_email {
    [email protected]
    [email protected]
        [email protected]
   }
   notification_email_from [email protected]
   smtp_server 192.168.200.1
   smtp_connect_timeout 30
   router_id LVS_DEVEL
}

vrrp_instance VI_1 {
    state MASTER
    interface eth0
    virtual_router_id 100
    priority 100
    advert_int 1
    authentication {
        auth_type PASS
        auth_pass 2013
    }
    virtual_ipaddress {
        10.1.88.81
    }
}

virtual_server 10.1.88.81 23 {
    delay_loop 6
    lb_algo rr
    lb_kind DR
    nat_mask 255.255.255.0
    persistence_timeout 50
    protocol TCP

    real_server 10.1.88.78 23 {
        weight 1
        TCP_CHECK {
            connect_timeout 3
            connect_port 23
        }
    }

    real_server 10.1.88.79 23{
        weight 1
        TCP_CHECK {
            connect_timeout 3
            connect_port 23
        }
    }

    real_server 10.1.88.80 23 {
        weight 1
        TCP_CHECK {
            connect_timeout 3
            connect_port 23
        }
    }
}

The configuration in 77 server

! Configuration File for keepalived

global_defs {
   notification_email {
     [email protected]
     [email protected]
     [email protected]
   }
   notification_email_from [email protected]
   smtp_server 192.168.200.1
   smtp_connect_timeout 30
   router_id LVS_DEVEL
}

vrrp_instance VI_1 {
    state BACKUP
    interface eth0
    virtual_router_id 101 
    priority 90
    advert_int 1
    authentication {
        auth_type PASS
        auth_pass 2013
    }
    virtual_ipaddress {
        10.1.88.81
    }
}

virtual_server 10.1.88.81 23 {
    delay_loop 6
    lb_algo rr
    lb_kind DR
    nat_mask 255.255.255.0
    persistence_timeout 50
    protocol TCP

    real_server 10.1.88.78 23 {
        weight 1
        TCP_CHECK {
            connect_timeout 3
            connect_port 23
        }
    }

    real_server 10.1.88.79 23{
        weight 1
        TCP_CHECK {
            connect_timeout 3
            connect_port 23
        }
    }

    real_server 10.1.88.80 23 {
        weight 1
        TCP_CHECK {
            connect_timeout 3
            connect_port 23
        }
    }
}

The message from 76 server

Jun 26 11:06:41 node0 Keepalived_vrrp[1264]: VRRP_Instance(VI_1) sending 0 priority
Jun 26 11:06:41 node0 Keepalived[1262]: Stopping Keepalived v1.2.7 (02/21,2013)
Jun 26 11:07:44 node0 Keepalived[1261]: Starting Keepalived v1.2.7 (02/21,2013)
Jun 26 11:07:44 node0 Keepalived[1262]: Starting Healthcheck child process, pid=1263
Jun 26 11:07:44 node0 Keepalived[1262]: Starting VRRP child process, pid=1264
Jun 26 11:07:44 node0 Keepalived_vrrp[1264]: Interface queue is empty
Jun 26 11:07:44 node0 Keepalived_vrrp[1264]: Netlink reflector reports IP 10.1.88.76 added
Jun 26 11:07:44 node0 Keepalived_vrrp[1264]: Netlink reflector reports IP fe80::5054:ff:fe8f:97f6 added
Jun 26 11:07:44 node0 Keepalived_vrrp[1264]: Registering Kernel netlink reflector
Jun 26 11:07:44 node0 Keepalived_vrrp[1264]: Registering Kernel netlink command channel
Jun 26 11:07:44 node0 Keepalived_vrrp[1264]: Registering gratuitous ARP shared channel
Jun 26 11:07:44 node0 Keepalived_healthcheckers[1263]: Interface queue is empty
Jun 26 11:07:44 node0 Keepalived_healthcheckers[1263]: Netlink reflector reports IP 10.1.88.76 added
Jun 26 11:07:44 node0 Keepalived_healthcheckers[1263]: Netlink reflector reports IP fe80::5054:ff:fe8f:97f6 added
Jun 26 11:07:44 node0 Keepalived_healthcheckers[1263]: Registering Kernel netlink reflector
Jun 26 11:07:44 node0 Keepalived_healthcheckers[1263]: Registering Kernel netlink command channel
Jun 26 11:07:44 node0 Keepalived_healthcheckers[1263]: Opening file '/etc/keepalived/keepalived.conf'.
Jun 26 11:07:44 node0 Keepalived_vrrp[1264]: Opening file '/etc/keepalived/keepalived.conf'.
Jun 26 11:07:44 node0 Keepalived_vrrp[1264]: Configuration is using : 63254 Bytes
Jun 26 11:07:44 node0 Keepalived_vrrp[1264]: Using LinkWatch kernel netlink reflector...
Jun 26 11:07:44 node0 Keepalived_healthcheckers[1263]: Configuration is using : 17141 Bytes
Jun 26 serial8250: too much work for irq4
11:07:44 node0 Keepalived_vrrp[1264]: VRRP sockpool: [ifindex(2), proto(112), fd(11,12)]
Jun 26 11:07:44 node0 Keepalived_healthcheckers[1263]: Using LinkWatch kernel netlink reflector...
Jun 26 11:07:44 node0 Keepalived_healthcheckers[1263]: Activating healthchecker for service [10.1.88.78]:23
Jun 26 11:07:44 node0 Keepalived_healthcheckers[1263]: Activating healthchecker for service [10.1.88.79]:23
Jun 26 11:07:44 node0 Keepalived_healthcheckers[1263]: Activating healthchecker for service [10.1.88.80]:23
Jun 26 11:07:45 node0 Keepalived_vrrp[1264]: VRRP_Instance(VI_1) Transition to MASTER STATE
Jun 26 11:07:46 node0 Keepalived_vrrp[1264]: VRRP_Instance(VI_1) Entering MASTER STATE
Jun 26 11:07:46 node0 Keepalived_vrrp[1264]: VRRP_Instance(VI_1) setting protocol VIPs.
Jun 26 11:07:46 node0 Keepalived_vrrp[1264]: VRRP_Instance(VI_1) Sending gratuitous ARPs on eth0 for 10.1.88.81
Jun 26 11:07:46 node0 Keepalived_healthcheckers[1263]: Netlink reflector reports IP 10.1.88.81 added
Jun 26 11:07:51 node0 Keepalived_vrrp[1264]: VRRP_Instance(VI_1) Sending gratuitous ARPs on eth0 for 10.1.88.81
[root@node0 ~]# date
Wed Jun 26 11:08:08 CST 2013

the message from 77 server

Jun 26 11:07:50 node1 Keepalived[1284]: Starting Keepalived v1.2.7 (02/21,2013)
Jun 26 11:07:50 node1 Keepalived[1285]: Starting Healthcheck child process, pid=1286
Jun 26 11:07:50 node1 Keepalived[1285]: Starting VRRP child process, pid=1287
Jun 26 11:07:50 node1 Keepalived_vrrp[1287]: Interface queue is empty
Jun 26 11:07:50 node1 Keepalived_vrrp[1287]: Netlink reflector reports IP 10.1.88.77 added
Jun 26 11:07:50 node1 Keepalived_vrrp[1287]: Netlink reflector reports IP fe80::5054:ff:fee6:54b7 added
Jun 26 11:07:50 node1 Keepalived_vrrp[1287]: Registering Kernel netlink reflector
Jun 26 11:07:50 node1 Keepalived_vrrp[1287]: Registering Kernel netlink command channel
Jun 26 11:07:50 node1 Keepalived_vrrp[1287]: Registering gratuitous ARP shared channel
Jun 26 11:07:50 node1 Keepalived_healthcheckers[1286]: Interface queue is empty
Jun 26 11:07:50 node1 Keepalived_vrrp[1287]: Opening file '/etc/keepalived/keepalived.conf'.
Jun 26 11:07:50 node1 Keepalived_healthcheckers[1286]: Netlink reflector reports IP 10.1.88.77 added
Jun 26 11:07:50 node1 Keepalived_healthcheckers[1286]: Netlink reflector reports IP fe80::5054:ff:fee6:54b7 added
Jun 26 11:07:50 node1 Keepalived_healthcheckers[1286]: Registering Kernel netlink reflector
Jun 26 11:07:50 node1 Keepalived_healthcheckers[1286]: Registering Kernel netlink command channel
Jun 26 11:07:50 node1 Keepalived_healthcheckers[1286]: Opening file '/etc/keepalived/keepalived.conf'.
Jun 26 11:07:50 node1 Keepalived_vrrp[1287]: Configuration is using : 63252 Bytes
Jun 26 11:07:50 node1 Keepalived_vrrp[1287]: Using LinkWatch kernel netlink reflector...
Jun 26 11:07:50 node1 Keepalived_healthcheckers[1286]: Configuration is using : 17139 Bytes
Jun 26 11:07:50 node1 Keepalived_vrrp[1287]: VRRP_Instance(VI_1) Entering BACKUP STATE
Jun 26 11:07:50 node1 Keepalived_vrrp[1287]: VRRP sockpool: [ifindex(2), proto(112), fd(11,12)]
Jun 26 11:07:50 node1 Keepalived_healthcheckers[1286]: Using LinkWatch kernel netlink reflector...
Jun 26 11:07:50 node1 Keepalived_healthcheckers[1286]: Activating healthchecker for service [10.1.88.78]:23
Jun 26 11:07:50 node1 Keepalived_healthcheckers[1286]: Activating healthchecker for service [10.1.88.79]:23
Jun 26 11:07:50 node1 Keepalived_healthcheckers[1286]: Activating healthchecker for service [10.1.88.80]:23
Jun 26 11:07:53 node1 Keepalived_vrrp[1287]: VRRP_Instance(VI_1) Transition to MASTER STATE
Jun 26 11:07:54 node1 Keepalived_vrrp[1287]: VRRP_Instance(VI_1) Entering MASTER STATE
Jun 26 11:07:54 node1 Keepalived_vrrp[1287]: VRRP_Instance(VI_1) setting protocol VIPs.
Jun 26 11:07:54 node1 Keepalived_vrrp[1287]: VRRP_Instance(VI_1) Sending gratuitous ARPs on eth0 for 10.1.88.81
Jun 26 11:07:54 node1 Keepalived_healthcheckers[1286]: Netlink reflector reports IP 10.1.88.81 added
Jun 26 11:07:59 node1 Keepalived_vrrp[1287]: VRRP_Instance(VI_1) Sending gratuitous ARPs on eth0 for 10.1.88.81
1
  • I found the solution of this problem. The master virtual_router_id should equal to backup's. Jun 26, 2013 at 6:39

1 Answer 1

0

You need to allow vrrp traffic between the two hosts so they can run the heartbeat.

3

Not the answer you're looking for? Browse other questions tagged .