Hi Eugene, Having an aggregate call to get all of the stats and statuses is good, but we should also keep the ability to retrieve statistics or the status of individual resources IMHO.
Thanks Youcef From: Eugene Nikanorov [mailto:[email protected]] Sent: Wednesday, March 05, 2014 12:42 PM To: OpenStack Development Mailing List Subject: [openstack-dev] [Neutron][LBaaS] Health monitoring and statistics for complex LB configurations. Hi community, Another interesting questions were raised during object model discussion about how pool statistics and health monitoring should be used in case of multiple vips sharing one pool. Right now we can query statistics for the pool, and some data like in/out bytes and request count will be returned. If we had several vips sharing the pool, what kind of statistics would make sense for the user? The options are: 1) aggregated statistics for the pool, e.g. statistics of all requests that has hit the pool through any VIP 2) per-vip statistics for the pool. Depending on the answer, the statistics workflow will be different. The good option of getting the statistics and health status could be to query it through the vip and get it for the whole logical instance, e.g. a call like: lb-vip-statistics-get --vip-id vip_id the would result in json that returns statistics for every pool associated with the vip, plus operational status of all members for the pools associated with that VIP. Looking forward to your feedback. Thanks, Eugene.
_______________________________________________ OpenStack-dev mailing list [email protected] http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev
