Re: Keepalived-devel Digest, Vol 96, Issue 3

Hi everybody,

I'm new to keepalived and I've been searching a lot to try to fix
keepalived using physical dev's mac address instead of the vmac, which
cause that after a change in master-slave state, all clients are
disconected untill arp cache is cleaned.
I've seen many posts about sysctl but none of them solved this for me:

net.ipv4.conf.all.arp_ignore=1
net.ipv4.conf.all.arp_announce=2


sysctl -w net.ipv4.conf.vrrp/50.rp_filter=0
sysctl -w net.ipv4.conf.vrrp/51.rp_filter=0

2: ens160: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast
state UP group default qlen 1000
    link/ether 00:50:56:01:04:16 brd ff:ff:ff:ff:ff:ff
    inet 192.168.110.110/24 brd 192.168.110.255 scope global ens160
       valid_lft forever preferred_lft forever
    inet6 fe80::250:56ff:fe01:416/64 scope link
       valid_lft forever preferred_lft forever
13: vrrp.50@ens160: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc
noqueue state UNKNOWN group default
    link/ether 00:00:5e:00:01:32 brd ff:ff:ff:ff:ff:ff
    inet6 fe80::200:5eff:fe00:132/64 scope link
       valid_lft forever preferred_lft forever
14: vrrp.51@ens160: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc
noqueue state UNKNOWN group default
    link/ether 00:00:5e:00:01:33 brd ff:ff:ff:ff:ff:ff
    inet 192.168.110.106/32 scope global vrrp.51
       valid_lft forever preferred_lft forever
    inet6 fe80::200:5eff:fe00:133/64 scope link
       valid_lft forever preferred_lft forever


arping -I eth1 192.168.110.106
ARPING 192.168.110.106 from 192.168.110.254 eth1
Unicast reply from 192.168.110.106 [00:50:56:01:04:16]  5.359ms

Any hints?
Regards,


El 23/06/14 08:53, [email protected] escribió:
> Send Keepalived-devel mailing list submissions to
> 	[email protected]
>
> To subscribe or unsubscribe via the World Wide Web, visit
> 	https://lists.sourceforge.net/lists/listinfo/keepalived-devel
> or, via email, send a message with subject or body 'help' to
> 	[email protected]
>
> You can reach the person managing the list at
> 	[email protected]
>
> When replying, please edit your Subject line so it is more specific
> than "Re: Contents of Keepalived-devel digest..."
>
>
> Today's Topics:
>
>    1. Failover to backup router when VS lost quorum
>       (Pim Rupert (Lemonbit))
>    2. Re: Failover to backup router when VS lost quorum
>       (Alexey Andrianov)
>    3. [acassen/keepalived] 0a83ac: vrrp: Fix a memory leak while
>       dropping incoming IP... (Alexandre Cassen)
>    4. [acassen/keepalived] c1b686: vrrp: Extend IPSEC-AH auth to
>       support unicast (Alexandre Cassen)
>    5. Re: vmac mode send out arp request with MAC of	physical
>       interface (Alexandre Cassen)
>    6. Re: VMAC, vmac_xmit_base, and IGMP (Alexandre Cassen)
>
>
> ----------------------------------------------------------------------
>
> Message: 1
> Date: Mon, 16 Jun 2014 14:33:54 +0200
> From: "Pim Rupert (Lemonbit)" <[email protected]>
> Subject: [Keepalived-devel] Failover to backup router when VS lost
> 	quorum
> To: [email protected]
> Message-ID: <[email protected]>
> Content-Type: text/plain; charset=us-ascii
>
> Hi,
>
> We have a setup of two keepalived routers (one master, one backup) which load balance traffic via LVS. Sometimes we experience failures from the Keepalived_healthcheckers to all of the real servers on *one router only*. During such a failure we would like to failover to the other keepalived router. 
>
> How do we achieve a router failover when quorum is lost on an LVS virtual server? 
>
> Can we use the 'quorum_down' directive for this? What script should we call to force a failover to the other keepalived router? 
>
> Kind regards,
>
> Pim Rupert
>
>
> ------------------------------
>
> Message: 2
> Date: Mon, 16 Jun 2014 18:50:09 +0400
> From: Alexey Andrianov <[email protected]>
> Subject: Re: [Keepalived-devel] Failover to backup router when VS lost
> 	quorum
> To: "Pim Rupert (Lemonbit)" <[email protected]>,
> 	[email protected]
> Message-ID: <[email protected]>
> Content-Type: text/plain; charset=ISO-8859-1; format=flowed
>
> Hi.
>
> You could use some IGP routing protocol to announce your VIP from both 
> routers, and setup quorum_up / quorum_down handlers to bring the 
> interface up and down to trigger the announce.
>
> 16.06.2014 16:33, Pim Rupert (Lemonbit) wrote:
>> Hi,
>>
>> We have a setup of two keepalived routers (one master, one backup) which load balance traffic via LVS. Sometimes we experience failures from the Keepalived_healthcheckers to all of the real servers on *one router only*. During such a failure we would like to failover to the other keepalived router.
>>
>> How do we achieve a router failover when quorum is lost on an LVS virtual server?
>>
>> Can we use the 'quorum_down' directive for this? What script should we call to force a failover to the other keepalived router?


------------------------------------------------------------------------------
HPCC Systems Open Source Big Data Platform from LexisNexis Risk Solutions
Find What Matters Most in Your Big Data with HPCC Systems
Open Source. Fast. Scalable. Simple. Ideal for Dirty Data.
Leverages Graph Analysis for Fast Processing & Easy Data Exploration
http://p.sf.net/sfu/hpccsystems