The use case from our customers has been mostly for database (MySql) load balancing. If the master goes down then they want another master/slave on standby ready to receive traffic. In the simplest case, I think Neutron can achieve this with 2 pools with 1 node each. If pool #1 goes down then pool #2 becomes active. We currently solve this with the notion of primary and secondary nodes. If all primary nodes go down then secondary nodes become active.
Cheers, --Jorge From: Eugene Nikanorov <[email protected]<mailto:[email protected]>> Reply-To: "OpenStack Development Mailing List (not for usage questions)" <[email protected]<mailto:[email protected]>> Date: Thursday, March 20, 2014 11:35 AM To: "OpenStack Development Mailing List (not for usage questions)" <[email protected]<mailto:[email protected]>> Subject: Re: [openstack-dev] [Neutron][LBaaS] Requirements Wiki * Active/Passive Failover * I think this is solved with multiple pools. The multiple pools support that is coming with L7 rules is to support content-switching based on L7 HTTP information (URL, headers, etc.). There is no support today for an active vs. passive pool. I'm not sure that's the priority. It depends on if this is widely supported among vendors. A commercial load balancer that doesn't have high availability features? Is there really such a thing still being sold in 2014? ;) I might be missing something fundamental here, but we're talking about 'additional' HA at pool level? Why not just add nodes to the pool? Also, Jorge-- thanks for creating that page! I've made a few additions to it as well that I'd love to see prioritized. Stephen -- Stephen Balukoff Blue Box Group, LLC (800)613-4305 x807 _______________________________________________ OpenStack-dev mailing list [email protected]<mailto:[email protected]> http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev
_______________________________________________ OpenStack-dev mailing list [email protected] http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev
