16:00:24 #startmeeting INFRA Working group weekly meeting 16:00:24 Meeting started Wed Nov 9 16:00:24 2016 UTC. The chair is uli-k. Information about MeetBot at http://wiki.debian.org/MeetBot. 16:00:24 Useful Commands: #action #agreed #help #info #idea #link #topic. 16:00:24 The meeting name has been set to 'infra_working_group_weekly_meeting' 16:00:28 #info Aric Gardner 16:01:05 Hi everybody, I can't start the GTM, some proxy issues..... 16:01:18 info wu tianwei 16:02:11 info Jingbo Hao 16:02:34 aricg or jmorgan1, can you open the GTM? 16:02:55 #info wu tianwei 16:03:21 #info Jack Morgan 16:03:48 #info Jose Lausuch 16:04:16 jmorgan1, can you open the GTM? I am blocked by proxy to login 16:04:21 #info Trevor Bramwell 16:04:30 uli-k: just a sec 16:04:59 #info Aimee Ukasick 16:05:10 #info David McBride 16:08:10 hw_wutianwei, are you on gotomeeting? 16:08:19 I can not open the GTM either. 16:08:35 You should be able to connect now. 16:09:19 #link https://global.gotomeeting.com/join/773318405 16:09:56 may beMy network is not good, so I can not join the gotomeeting. 16:10:24 OK. Jack explained shortly. I can share some information with you tomorrow. 16:10:28 I will discuss in the channel. is this ok? 16:10:35 sure 16:11:56 OK let's come back to the topic later (or we have a discussion tomorrow) 16:12:08 #topic Scenario based POD allocation 16:12:37 let's come back to the topic later 16:18:39 Need to step away for a few minutes 16:18:44 dmcbride: no problem 16:19:01 DanSmithEricsson: You audio is extremely faint. :( 16:19:17 #info ChrisPB supports the idea 16:19:32 ok.. i think we need some alignment here 16:19:40 cause what you guys are talking about does not come into play in the approach we are talking 16:19:45 we are not dealing with anything at the OS level 16:19:54 #info Jonas reminds us that this was tried in Arno and the issue before was clean up scripts were needed and sometimes they failed 16:19:58 the approhac for Dynamic POD means that 16:20:04 this discussion in now moot 16:21:16 sure.. we have a slide 16:21:39 we have slides that we showed a while back and Fatih has expanded on them 16:21:43 #topic Scenarios and common config files 16:21:50 #info Jack suggests we schedule F2F time at Plugfest/hackfest 16:23:28 #info Uli share some slides to discuss the problem and reason for proposal 16:27:39 #info Jonas explains that installers already have such a file but those are not common 16:28:33 #info Jonas also explains that for a new feature the installer specific file needs to be updated as well 16:30:20 it was a bit unclear that last statement.. Jack - did you say a "File with all Installer parameters listed across the board" and then we slice it away for each installer 16:30:30 or only for "Feature parameters" that are introduced? 16:30:31 #info Uli mentions that the scenario config file is still a WIP and needs to have missing options added in 16:31:19 DanSmithEricsson: i just meant that we create a template file which would have all possible supported feature for opnfv 16:31:26 cooo 16:31:34 cool*.. 16:31:52 DanSmithEricsson: then installers might take this file and adapt it to their installer - including just those features it supports 16:33:02 ok.. so you want a Umbrella that lists "all available features in OPNFV" (good - i like that for a bunch of reasons) - then from there, an "installer" can utline a subset of that list (what they support).. from there, that supportted list then outlines what "scenario combinations" (which are comprised of configs and features" can be "run" on a given setup? 16:33:11 (or something like that? ) - i like it :) 16:40:28 but this Piece is going to be read from a "ORDER" in the Bookig tool 16:40:37 and then you will have an inventory across your Distributed Lab setup 16:40:56 so when you "request a scneario" for test and the scneario is of type "ARM" then it will only ordre that to a POD that is able to support hat 16:40:56 that 16:42:06 I'm back 16:44:46 Sorry - the voice is really bad.. 16:44:59 but my question is that by making a SCENARIO description file 16:45:06 are we saying that this assumes CI/CD runs 16:45:18 and thus you "must use CI/CD with a scenario" 16:45:19 ? 16:45:55 ok.. so that is fine 16:46:02 we are saying that scenario implicity means a CI/CD run 16:53:26 #info lively discussion followed covering several details on implementaions 16:53:59 #action Jack to collect feedback on the patch and include them 16:55:28 #info several people mentioned their concern over the file name 16:55:57 #info need to find some combination of features to provide some hint to the scenario contents 16:57:44 #topic Use artifacts to update WIKI of Pharos Labs 16:59:16 I got disconnected 17:00:34 #info aricg proposes to just use direct links instead since there are many problems with the plugin 17:01:03 agree to discuss that on the hackfest 17:01:12 * jmorgan1 leaves the building 17:02:11 #endmeeting