Re: wrong real server ip in connection

"Vladimir S. Nikeshin" <[email protected]>
Newsgroups gmane.linux.keepalived.devel
Organization ponch.ru
Message-ID <[email protected]>
The problem was in -j REDIRECT iptables rules on nodes (to prevent
loop).

В Sat, 17 Oct 2015 23:02:05 +0300

> Hello!
> 
> First of all - Sorry, if wrong maillist.
> I come up with strange situation and i dont really know, if its by
> design or something is wrong in my configuration. so pls help)
> 
> i have three node self-balancing keepalived cluster for smtp.
> 
> server1 195.111.111.164 195.111.111.168
> server2 195.111.111.165 195.111.111.169
> server3 195.111.111.166 195.111.111.170
> 
> two vip address 195.111.111.177 and 195.111.111.176
> 
> problem:
> 
> when i do "telnet 195.111.111.176 25" i got balanced to address
> from "virtual_server 195.111.111.176" group .164 .165 .166(i can see
> in mail server log, which ip i am connected to). It ok, correct.
> 
> but with "telnet 195.111.111.177 25", i am expecting to see in
> mail log, that connection goes to ip .168 or .169. or 170. But they,
> somehow, again goes to .164 .165 .166. Is it correct behavior? I wish
> it is not. Any ideas how to fix?
> 
> ======
> two vrrp instaces:
> 
> vrrp_instance VI_domain1 {
>     state MASTER
>     interface eth3
>     lvs_sync_daemon_inteface eth3
>     virtual_router_id 101
>     priority 200
>     advert_int 2
>     virtual_ipaddress {
>         195.111.111.176/27 dev bond0
>     }
> }
> 
> vrrp_instance VI_domain2 {
>     state MASTER
>     interface eth3
>     lvs_sync_daemon_inteface eth3
>     virtual_router_id 102
>     priority 200
>     advert_int 2
>     virtual_ipaddress {
>         195.111.1111.177/27 dev bond0
>     }
> }
> 
> =====
> virtual_server 195.111.111.176 25 {
>     delay_loop 30
>     lb_algo wrr
>     lb_kind DR
>     protocol TCP
>     persistence_timeout 50
> 
>     real_server 195.111.111.164 25 {
>         weight 100
>         SMTP_CHECK {
>                 connect_timeout 3
>                 retry 1
>                 delay_before_retry 2
>                 helo_name keepalived
>         }
>     }
> 
>     real_server 195.111.111.165 25 {
>         weight 100
>         SMTP_CHECK {
>                 connect_timeout 3
>                 retry 1
>                 delay_before_retry 2
>                 helo_name keepalived
>         }
>     }
> 
>     real_server 195.111.111.166 25 {
>         weight 100
>         SMTP_CHECK {
>                 connect_timeout 3
>                 retry 1
>                 delay_before_retry 2
>                 helo_name keepalived
>         }
>     }
> }
> 
> virtual_server 195.111.111.177 25 {
>     delay_loop 30
>     lb_algo wrr
>     lb_kind DR
>     protocol TCP
>     persistence_timeout 50
> 
>     real_server 195.111.111.168 25 {
>         weight 100
>         SMTP_CHECK {
>                 connect_timeout 3
>                 retry 1
>                 delay_before_retry 2
>                 helo_name keepalived
>         }
>     }
> 
>     real_server 195.111.111.169 25 {
>         weight 100
>         SMTP_CHECK {
>                 connect_timeout 3
>                 retry 1
>                 delay_before_retry 2
>                 helo_name keepalived
>         }
>     }
> 
>     real_server 195.111.111.170 25 {
>         weight 100
>         SMTP_CHECK {
>                 connect_timeout 3
>                 retry 1
>                 delay_before_retry 2
>                 helo_name keepalived
>         }
>     }
> }
> 
> ------------------------------------------------------------------------------
> _______________________________________________
> Keepalived-devel mailing list
> [email protected]
> https://lists.sourceforge.net/lists/listinfo/keepalived-devel

------------------------------------------------------------------------------
_______________________________________________
Keepalived-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/keepalived-devel
lmpx.com only provides a reader for public news (NNTP) servers. It is not affiliated with the servers or forums shown here and is not responsible for the content of articles, which is written by their respective authors.