Graceful failover or how to eliminate failover downtime in keepalived

Hi,

Whenever there is a failover between two machines with keepalived there is a small window of downtime (about 2 seconds), because the VIP is removed by keepalived first from node A, and added later by keepalived on node B. When a failover is started due to a priority election or scheduled maintenance it seems absurd to tolerate this two seconds of downtime. How do we properly prevent this downtime? 

I know this isn't a problem that is up to Keepalived to fix, but I am keen to learn how you deal with this problem. 

Examples:

1. Priority election (approx. two seconds downtime):

22:54:53 node1 Keepalived_vrrp[2477]: VRRP_Instance(VI_5) forcing a new MASTER election
22:54:53 node2 Keepalived_vrrp[2466]: VRRP_Instance(VI_5) Received higher prio advert
22:54:53 node2 Keepalived_vrrp[2466]: VRRP_Instance(VI_5) Entering BACKUP STATE
22:54:53 node2 Keepalived_vrrp[2466]: VRRP_Instance(VI_5) removing protocol VIPs.
22:54:53 node2 Keepalived_healthcheckers[2465]: Netlink reflector reports IP 10.0.0.100 removed
# Downtime started
22:54:54 node1 Keepalived_vrrp[2477]: VRRP_Instance(VI_5) Transition to MASTER STATE
22:54:55 node1 Keepalived_vrrp[2477]: VRRP_Instance(VI_5) Entering MASTER STATE
22:54:55 node1 Keepalived_vrrp[2477]: VRRP_Instance(VI_5) setting protocol VIPs.
# Still down!
22:54:55 node1 Keepalived_vrrp[2477]: VRRP_Instance(VI_5) Sending gratuitous ARPs on eth1 for 10.0.0.100
22:54:55 node1 Keepalived_healthcheckers[2476]: Netlink reflector reports IP 10.0.0.100 added
# Finally, we're up again!

2. Scheduled downtime (again, look at all the unnecessary downtime):

07:08:44 node1 Keepalived_vrrp[3045]: VRRP_Script(wsrepready) failed
07:08:45 node1 Keepalived_vrrp[3045]: VRRP_Instance(VI_5) Entering FAULT STATE
07:08:45 node1 Keepalived_vrrp[3045]: VRRP_Instance(VI_5) removing protocol VIPs.
07:08:45 node1 Keepalived_healthcheckers[3044]: Netlink reflector reports IP 10.0.0.100 removed
# Downtime started
07:08:45 node1 Keepalived_vrrp[3045]: VRRP_Instance(VI_5) Now in FAULT state
07:08:46 node2 Keepalived_vrrp[2466]: VRRP_Instance(VI_5) Transition to MASTER STATE
07:08:47 node2 Keepalived_vrrp[2466]: VRRP_Instance(VI_5) Entering MASTER STATE
07:08:47 node2 Keepalived_vrrp[2466]: VRRP_Instance(VI_5) setting protocol VIPs.
07:08:47 node2 Keepalived_vrrp[2466]: VRRP_Instance(VI_5) Sending gratuitous ARPs on eth1 for 10.0.0.100
07:08:47 node2 Keepalived_healthcheckers[2465]: Netlink reflector reports IP 10.0.0.100 added
# Up again! 


Best regards,

Pim



------------------------------------------------------------------------------
What NetFlow Analyzer can do for you? Monitors network bandwidth and traffic
patterns at an interface-level. Reveals which users, apps, and protocols are 
consuming the most bandwidth. Provides multi-vendor support for NetFlow, 
J-Flow, sFlow and other flows. Make informed decisions using capacity 
planning reports. https://ad.doubleclick.net/ddm/clk/305295220;132659582;e