12:58:46 #startmeeting OPNFV TSC Aug 29, 2017 12:58:46 Meeting started Tue Aug 29 12:58:46 2017 UTC. The chair is frankbrockners. Information about MeetBot at http://wiki.debian.org/MeetBot. 12:58:46 Useful Commands: #action #agreed #help #info #idea #link #topic. 12:58:46 The meeting name has been set to 'opnfv_tsc_aug_29__2017' 12:58:52 #rollcall 12:59:00 #info Frank Brockners 12:59:12 #chair rpaik 12:59:12 Current chairs: frankbrockners rpaik 12:59:20 #info Morgan Richomme 12:59:27 #info Juha Kosonen (proxy for Tapio Tallgren) 12:59:32 #info Bin Hu 12:59:44 #info Wenjing 12:59:48 #info hongbo 13:00:05 #info Fatih Degirmenci 13:00:21 #info Tim Irnich 13:00:34 #info Brian Skerry 13:00:38 #info Rossella Sblendido 13:00:47 #info Xavier Costa 13:01:04 we have a quorum 13:01:28 #topic approval of previous minutes 13:01:45 #info dneary 13:01:49 #info no comments/feedback on previous minutes, thus approved 13:01:57 #topic agenda bashing 13:02:15 #link https://wiki.opnfv.org/display/meetings/TSC#TSC-August29,2017 today's agenda 13:02:33 #info fuqiao 13:02:37 #info no other topics suggested 13:02:48 #topic Committers list for upcoming elections 13:03:06 #link https://wiki.opnfv.org/display/DEV/Eligible+Committers+for+Elections 13:04:09 #info corrections to this list sent to Ray by noon PT on Friday, Sept/1 13:04:32 #info nomination period (1 week) open from Sept/4th onwards 13:04:55 #info rpaik will follow up with an email on details 13:05:24 #info committer diversity (global theatres, gender, ..) would be greatly appreciated for the nominees 13:05:58 #topic Euphrates update 13:06:41 #info all projects reported for MS5 with exception of Parser project 13:07:25 #info Assumption is that Parser will not participate in Euphrates 13:09:14 #info dmcbride to contact Parser one more time 13:09:34 #info morgan_orange mentioned that Parser was in contact with functest this morning 13:09:56 #info for MS6 there are about 75% of projects reporting back 13:10:29 #info MS6 opens stable branch - MS7 (Sept/15) will close it 13:11:05 #info MS6 Exception requests need to be submitted by September 1st 13:11:58 #info Milestone exceptions for MS5 were all met and closed 13:12:16 #info MS5 exception requests from 3 projects (Armband, Calipso, FDS) have all been met 13:12:44 #topic LF IT/infra update 13:12:56 #info nothing to report 13:13:19 #info Jenkins system update still to be scheduled. No security updates scheduled yet. 13:13:32 #info once Centos 7.4 releases updates can be expected 13:13:46 #topic XCI update 13:14:16 #link https://wiki.opnfv.org/download/attachments/8687635/XCI_TSC_Report_August2017.pdf?api=v2 13:15:05 #info Bryan Sullivan 13:15:08 #info update from Fatih on current status of XCI (see slides) 13:16:41 #info fdegir notes that from his perspective, time devoted to testing has decreased over the past couple of OPNFV releases 13:17:25 fdegir, What is the Y axis on your graph? 13:18:01 #info What Fatih is describing is exactly what I raised concerns about in the tech-discuss email https://lists.opnfv.org/pipermail/opnfv-tech-discuss/2017-June/016821.html 13:19:30 #info I also provided some recommended actions to address those issues in that email 13:23:51 #info three OPNFV scenarios will use OpenStack and ODL master 13:27:34 #info I think X-CI will help us avoid problems that may arise later, but I'm not sure how it will help lower bugs arising from the packaging of scenario components into OPNFV releases, and how thoroughly we test that 13:29:46 #info I believe a lot of the reduction in quality (lowering scope of what is released, lateness of the release, etc) is related to that job of packaging and testing. I would like to see how X-CI will help that, more clearly. Or what otherwise OPNFV and the installer teams can do to address the points raised in the email ref'd above. 13:32:13 #info based on ODL and SFC scenarios, the XCI team will share their learnings (around Sep 15th/MS7) 13:33:38 #info currently 3 distros (Ubntu, CentOS, SUSE) are supported 13:35:46 #info The expansion of X-CI into more complex scenarios is a good step; has anyone stepped up to support an OCL scenario? 13:36:59 #info I will start looking into how to focus our scenario/functest contributions in the F release on the X-CI path; this may help address some of the issues we had in the past with the scalabiliy of supporting n distribution based scenarios. 13:39:33 #info fdegir notes that it's up to project teams to decide which upstream versions to use 13:40:13 #topic Creation review for Stor4NFV 13:40:18 #link https://wiki.opnfv.org/display/PROJ/Stor4NFV 13:40:27 frankbrockners, A pity - I had a question re XCI too 13:40:48 dneary - please take the question to the list 13:41:21 fdegir, You are supporting 3 scenarios now - are there plans to support others? And when you support more scenarios, are you confident that you will avoid the "big bang" effect as scenarios move to a new release? 13:41:56 dneary: Y axis is the quality 13:42:06 #info Shane_Wang notes that the goal of the project is investigate storage intensive workloads (e.g. vCDN) 13:42:36 bryan_att: openstack is deployed from master so we provide feedback on commit level 13:42:39 #info the project team will be working with OpenSDS & Ceph communities 13:42:51 bryan_att: but this doesn't prevent faults caused by packaging 13:43:06 bryan_att: sorry : openstack is deployed from source... 13:46:22 fdegir: how deep can X-CI get if we do a full deploy and n-scenario testing on every commit? Perhaps we should run the bulk of the scenarios once per day etc. 13:47:18 #info Heather asks whether Stor4NFV has any additional hardware requirements above what OPNFV has available 13:47:38 #info Shane_Wang notes that he does not anticipate any additional storage HW requirements 13:49:00 #info dneary asks what test methodology would be used - or whether storperf would be used off the shelf 13:49:19 bryan_att: the plan is to start with periodic runs for scenarios like how we are doing for upstream components 13:49:20 #info Shane_Wang answers that no special tests will be developed day1 13:49:47 bryan_att: but this increases the delta between last working version of the scenarios and currently tested version of it 13:50:00 bryan_att: making it harder to pinpoint the faulty commit 13:50:31 #info APEX and Compass are the initial installer targets 13:50:40 #info Shane_Wang mentions that initially integration with Apex and Compass is planned 13:50:43 bryan_att: once we get the basics up and running, we want to have more frequent runs and finally commit basis 13:51:35 #info dneary asks whether the project already identified ambassadors in the upstream projects (OpenSDS,...) 13:51:35 #info StorPerf does perform functional tests, other than by assuming it already works and then profiles the performance of the storage backend. Stor4NFV expands that scope from what I can see. 13:51:39 fdegir: OK, we'll start looking at getting it running locally so we can test scenarios and tests for possible inclusion. 13:51:58 #info sorry, StorPerf does NOT perform functional tests. 13:52:11 bryan_att: thanks! I'd start with virtual mini flavor to get a feeling with the sandbox 13:52:30 #info Shane_Wang mentions that several contributors/committers are also working in OpenSDS and/or Ceph 13:52:32 bryan_att: and we will let community know once we are done with pdf work and baremetal 13:52:39 bryan_att: so you can go baremetal 13:54:52 dneary: about your question with the big bang 13:54:54 #info following a question - folks suggested to avoid forking upstream repos but work upstream 13:55:28 #info recommendation is to the project team is maintain patches in OPNFV vs. forking upstream code 13:55:36 #startvote Does the TSC approve the creation of the Stor4NFV project? -1,0,+1 13:55:36 Begin voting on: Does the TSC approve the creation of the Stor4NFV project? Valid vote options are -1, 0, +1. 13:55:36 Vote using '#vote OPTION'. Only your last vote counts. 13:55:49 #vote +1 13:55:50 #vote +1 13:55:54 #vote +1 13:55:55 #vote +1 13:55:55 #vote +1 13:55:56 #vote +1 13:55:57 #vote +1 13:55:58 #vote +1 13:55:59 #vote +1 13:56:01 #vote +1 13:56:06 #vote +1 13:56:10 #vote +1 13:56:18 #vote +1 13:56:22 #vote + 13:56:22 hongbo75698745: + is not a valid option. Valid options are -1, 0, +1. 13:56:44 #vote +1 13:56:47 #endvote 13:56:47 Voted on "Does the TSC approve the creation of the Stor4NFV project?" Results are 13:56:47 +1 (14): hongbo75698745, frankbrockners, fuqiao, bh526r, dneary, timirnich, fdegir, morgan_orange, juhak_, bryan_att, rossella_s, XavierCosta, bjskerry, Wenjing 13:57:28 #endmeeting