============================================ #opendaylight-clustering: clustering_hackers ============================================ Meeting started by moizer at 15:58:28 UTC. The full logs are available at http://meetings.opendaylight.org/opendaylight-clustering/2015/clustering_hackers/opendaylight-clustering-clustering_hackers.2015-02-24-15.58.log.html . Meeting summary --------------- * Make CDS the default datastore (moizer, 16:46:43) * Make CDS the default datastore (moizer, 16:47:02) * Need to get an answer on the following (moizer, 16:47:34) * 1. How can we make a single node secure (bug 2736) (moizer, 16:48:00) * 2. Why are some csit tests faiing (bug 2737) (moizer, 16:48:51) * 3. Performance has to be on-par with IMDS. Jan and Mark to test performance again. (moizer, 16:49:18) * Some other concerns (moizer, 16:49:56) * rovarga - are we ready for the huge influx of issues when we made CDS default as happened with IMDS in Hydrogen/Helium? (moizer, 16:50:39) * moizer - can IMDS be used in production? (moizer, 16:51:04) * rovarga - IMDS should be used for operational, CDS for config. This kind of hybrid deployment will be possible with the new DOM API. (moizer, 16:51:58) * jmedved - since the new DOM API will be ready for use only post Lithium our only choice for now is CDS. (moizer, 16:52:51) * tpantelis - once CDS is enabled in production we need to have a way to backup the data. (moizer, 16:53:45) * - We need to think through the design - created bug 2752 for tracking (moizer, 16:54:46) * ttkacik says we are at feature freeze so this cannot go into Lithium (moizer, 16:55:45) * bug review (moizer, 16:56:11) * most major items taken care of - the most important bugs now for Lithium are now the ones gating enabling CDS for single node deployment by default (moizer, 16:57:28) * moved one bug about network segmentation to Beryllium (moizer, 16:58:09) Meeting ended at 16:58:40 UTC. People present (lines said) --------------------------- * moizer (20) * odl_meetbot (3) Generated by `MeetBot`_ 0.1.4