16:04:21 #startmeeting Brahmaputra Release 16:04:21 Meeting started Tue Dec 15 16:04:21 2015 UTC. The chair is debrascott. Information about MeetBot at http://wiki.debian.org/MeetBot. 16:04:21 Useful Commands: #action #agreed #help #info #idea #link #topic. 16:04:21 The meeting name has been set to 'brahmaputra_release' 16:04:28 #topic roll call 16:04:35 #info to add to meeting minutes rpaik. The topic of the updte from the board meeting has been dleayed until the meeting on the 22nd. 16:04:50 oops sorry debra 16:05:08 https://global.gotomeeting.com/join/341956909 16:05:15 #info Chris Price (opnfvdocs) 16:05:18 #info Bryan Sullivan 16:05:21 #info n0ano (Don Dugger - Intel) 16:05:38 #info Mark Beierl (StorPerf) 16:05:48 #info Uli Kleber, Octopus 16:05:48 #info Ana Cunha (Yardstick) 16:05:53 #info Vikram Dham (QTIP) 16:06:00 #info Parviz Yegani 16:06:07 #info Howard (Parser,Multisite) 16:06:08 #info Jose Lausuch (Functest) 16:06:21 #info Fatih Degirmenci (Releng) 16:06:26 #info Simona Coppola (Intel) 16:06:31 #info Bin Hu (IPv6) 16:06:50 #info Billy O'Mahony 16:06:52 #info Gerald Kunzmann (DOCOMO) 16:06:57 #topic Agenda 16:07:04 #info Frank 16:07:09 #info Frank Brockners 16:07:09 #info Larry Lamers 16:07:24 #info Ashlee Young (onosfw) 16:07:29 International dial numbers for GTM don't seem to work 16:07:39 rpaik - could you check? 16:09:41 #info Stuart Mackie - OpenContrail 16:10:13 #Update CI/Config Files/Genesis 16:10:14 #link https://wiki.opnfv.org/genesis/configuration-files-discussion?&#common_config_files 16:11:26 #info config file proposals have been made, 3 have been agreed upon as starting point for each area. Now depends on people putting comments into Gerrit 16:13:32 #note that the files are not finalized, but just a starting point is agreed upon 16:14:17 #info Trevor Cooper (Pharos) 16:14:24 #topic installer plugin discussions 16:16:30 #info seems teams are now making progress on plugins 16:17:16 #topic any other blockers toward milestone E?? 16:18:15 #info Uli is concerned whether matrix will really allow us to release on Feb 2 16:18:50 if unit test don't exist 16:18:57 are they out? 16:19:10 or the other stuff 16:19:43 http://213.77.62.197/test_projects 16:20:23 #info some test cases are declared in database, but will get a very ugly jason file but progress of tests is not known 16:21:11 +++++++++1 16:21:17 what is our release criteria?? 16:21:19 #info What is the criteria for Milestone E; which projects need to provide unit test 16:21:30 fdegir: jump servers in PODs 5, 6, 8 will be available tomorrow 16:21:40 standalone projects can pass the milestone, etc 16:21:43 define "Unit Test" CI functional test or project code level self tests? 16:21:46 #info Yardstick will not be ready with all test cases by Milestone E 16:21:59 but the B-release itself 16:22:12 and not just features 16:22:14 http://213.77.62.197/test_projects//cases 16:22:15 example: http://213.77.62.197/test_projects/sfc/cases 16:22:18 we have bunch of other type of projects 16:23:05 Example from StorPerf: https://build.opnfv.org/ci/view/storperf/job/storperf-merge-master/ shows number of test cases and code coverage. Does this matter to Brahmaputra? 16:24:00 will the test case failures be taken into account when evaluating release-readiness? 16:24:05 which tests cases are allowed to fail? 16:24:21 Should be 0, IMHO 16:24:36 If the test cannot pass, then either fix the code or remove the test :) 16:24:55 mbeierl: this might end up not having any test cases 16:25:03 or having test cases based on what is tested 16:25:07 both are wrong 16:25:25 Again, define test case in this scenario. Code coverage or functionality coverage 16:26:24 fdegir: you are correct in that statement, and my statement is explicitly worded that way to make people think about why the test exists. 16:28:07 I think we must at least match what we released in Arno SR1 in order to release B 16:28:08 #info it is important for projects that are going to break features such as HA if you use them to document the limitations of the feature on the release. 16:28:53 #info in addition if a feature test case is not able to pass it should not be a documented feature of the release. 16:33:15 #link https://wiki.opnfv.org/releases/brahmaputra expectations on projects and capabilities for Brahmaputra 16:34:38 #link https://wiki.opnfv.org/releases/brahmaputra/release_plan 16:38:27 perfect question 16:39:35 yes ! 16:40:44 which combination we are going to do that exercise against? 16:41:04 for arno, we knew we have os + odl 16:41:08 what we have for b? 16:41:59 "if a feature test case is not able to pass it should not be a documented feature of the release." - what about if a feature doesn't have a test case it should not be a documented feature? :) 16:42:21 we need a basic combination 16:42:28 which we have been crying for weeks 16:43:24 Hay mbeierl, on the page https://wiki.opnfv.org/releases/brahmaputra it states that for a feature to be release in Brahmaputra it must be verified. That means a repeatable test case proving it. 16:44:57 awesome! 16:57:38 trevor_intel: thanks trevor_intel ! 16:57:57 looking forward to get my hands on them 17:00:28 Sorry, need to drop out now.... 17:01:16 #info continue test discussions this week in Pharos and Test meetings 17:01:22 #endmeeting