[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]


Hi all,


I’m wondering if I could find some VRRP expert who could help me.. ^^


I’m trying to set UP an High Availability Service for a MTA.

I’ve installed two box with debian lenny and a MTA and i‘m trying to set keepalived and those server.

I want to set both server as backup, so the first UP will be elected MASTER, and then this avoid double switch when a faulting node become available again.


Everything seems working fine except that I could not disable premption.



Package version

ii  ipvsadm                           1:1.24-2.1               Linux Virtual Server support programs

ii  keepalived                        1.1.15-1                 Failover and monitoring daemon for LVS clusters


Here is my conf:


global_defs {

        router_id LVS_HA_2



vrrp_instance VI_1 {

        state BACKUP

        interface eth0

        virtual_router_id 13

        priority 25

        authentication {

                auth_type PASS

                auth_pass mailinblack


        virtual_ipaddress {

       brd dev eth0 label eth0:1


        Nopreempt               # try to avoid preemption with BACKUP state

        notify_master "/root/notify.sh MASTER"

        notify_backup "/root/notify.sh BACKUP"

        notify_fault "/root/notify.sh FAULT"

        #notify /root/notify.sh




virtual_server 25 {

        delay_loop 10

        lb_algo wrr

        lb_kind DR

        persistence_timeout 10

        protocol TCP

        ha_suspend              # I disable heal checker is current node is not available

        real_server 25 {

                weigth 25

                #inhibit_on_failure                       # I try both whith and whithout this and that makes no change

                notify_up "/root/notify_rs.sh UP MTA"

                notify_down "/root/notify_rs.sh DOWN MTA"

                SMTP_CHECK {

                        # Postfix IN

                        host {


                                connect_port 25



                        connect_timeout 10

                        retry 3

                        delay_before_retry 20

                        helo_name "High Availability Checker"







My script allow me to switch from backup and slave when healtchecker detects a failure (de activate and activate network link)


If someone could help me or point me in the right direction..   ;)


Tanks in advance



Reply to: