Comment 13 for bug 1798475

@Slawek,
Thanks for the post.
http://logs.openstack.org/57/628057/2/gate/neutron-fullstack/fe4a1c0/logs/testr_results.html.gz
This one seems met a new issue I've not seen before.

In the journal.log, I noticed these logs:
http://logs.openstack.org/57/628057/2/gate/neutron-fullstack/fe4a1c0/logs/journal.log
Keepalived_vrrp 13885 is master at this time:
Jan 10 04:45:42 ubuntu-xenial-limestone-regionone-0001611222 Keepalived_vrrp[13885]: VRRP_Instance(VR_1) Transition to MASTER STATE
and Keepalived_vrrp 14083 is backup at this:
Jan 10 04:45:41 ubuntu-xenial-limestone-regionone-0001611222 Keepalived_vrrp[14083]: VRRP_Instance(VR_1) Entering BACKUP STATE

At this time, a state change happened.
Jan 10 04:46:01 ubuntu-xenial-limestone-regionone-0001611222 Keepalived_vrrp[14083]: VRRP_Instance(VR_1) Transition to MASTER STATE
Jan 10 04:46:01 ubuntu-xenial-limestone-regionone-0001611222 Keepalived_vrrp[13885]: VRRP_Instance(VR_1) Received advert with higher priority 50, ours 50
Jan 10 04:46:01 ubuntu-xenial-limestone-regionone-0001611222 Keepalived_vrrp[13885]: VRRP_Instance(VR_1) Entering BACKUP STATE

This LOG "Received advert with higher priority 50, ours 50" is really interesting, seems that 'non-preemptive' VRRP does not work as excepted.

The testing keepalived version is v1.3.9:
Jan 10 04:45:35 ubuntu-xenial-limestone-regionone-0001611222 Keepalived[13880]: Starting Keepalived v1.3.9 (10/21,2017)
Here are some github issues:
https://github.com/acassen/keepalived/issues/512
https://github.com/osixia/docker-keepalived/issues/20