Are you using ucarp or keepalived to manage the VIP address? Basically, are you rebooting the load balancer, which everything is configured to use?
John On Tuesday, January 13, 2015 at 5:04 AM, Pedro Sousa wrote: > Hi all, > > I have 3 nodes that are loadbalacing some API Openstack Based Services. When > I reboot one of my servers to test HA, and when that server comes online I > start getting 504 errors, specially with nova-api, glance and keystone APIS. > This only happens If I reboot the server. > > Then to recover from that errors I have to restart all my services in all the > servers, without rebooting, running openstack-service restart. > > I'm providing my haproxy conf in attach case someone can assist me. > > Thank you. > > > > > _______________________________________________ > OpenStack-operators mailing list > OpenStack-operators@lists.openstack.org > (mailto:OpenStack-operators@lists.openstack.org) > http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-operators > > > > > Attachments: > - haproxy.cfg >
_______________________________________________ OpenStack-operators mailing list OpenStack-operators@lists.openstack.org http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-operators
