18:04:48 #startmeeting tsc 18:04:48 Meeting started Thu Jan 5 18:04:48 2017 UTC. The chair is colindixon. Information about MeetBot at http://ci.openstack.org/meetbot.html. 18:04:48 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 18:04:48 The meeting name has been set to 'tsc' 18:04:49 #topic agenda bashing and roll call 18:04:52 #info edwarnicke 18:04:53 #link https://wiki.opendaylight.org/index.php?title=TSC:Main&oldid=52461#Agenda 18:04:54 #info Vratko Polak proxying for Luis Gomez 18:04:54 #info skitt 18:04:54 #info Thanh 18:04:55 #Info colindixon 18:04:56 #info jamoluhrsen 18:04:57 #info anipbu 18:04:57 #info Hideyuki 18:04:57 #info abhijitkumbhare 18:04:58 #info shague 18:05:16 #link https://meetings.opendaylight.org/opendaylight-meeting/2016/tsc/opendaylight-meeting-tsc.2016-12-23-04.30.html last week's meeting minutes 18:05:55 #action phrobb to follow up with LF folks about a TSC mixer at the Leadership Summit 18:06:03 rovarga: hmm... I was using the toll-free number. that's not working so the toll number is fine. 18:06:08 #action colindixon, zxiiro and phrobb to come up with a proposal for tracking project activity in a positive way 18:06:08 #action colindixon and zxiiro to work on trying to build tools to help with what's failing on failures in our jenkins jobs 18:06:10 #action colindixon to try to drive some/all of the ongoing TSC discussions to completion 18:06:11 #action phrobb and tykeal to look into an ODL infra micro-datacenter in a box to make things work better 18:06:11 #action colindixon, vishnoianil, gzhao, anipbu and others to work to advertise APAC-friendly TSC calls to the India and China communities, as well as projects/PTLs in Asia 18:06:43 #topic TSC mailing list votes and discssions 18:06:45 #info rovarga 18:06:54 #link https://lists.opendaylight.org/pipermail/tsc/2016-November/006368.html projects with few/no active committers 18:07:11 #link https://lists.opendaylight.org/pipermail/tsc/2016-November/006298.html Delivering to consumer project off-cadence (ODP & OPNFV collaboration) 18:07:42 #link https://lists.opendaylight.org/pipermail/release/2017-January/009034.html flagship features in Carbon 18:09:13 #action colindixon to try to close the loop with OPNFV on what they need for their next release to figure out what they actually need from us for that 18:11:39 #info dfarrell07 notes that there was a poll about what version of OpenStack OPNFV was going to use and so that might be useful input for how ODL/OPNFV interact 18:13:05 OPNFV: Tapio Tallgren, TSC Chair 18:13:27 from https://www.opnfv.org/about/governance/technical-steering-committee 18:13:34 #link https://lists.opendaylight.org/pipermail/tsc/2016-December/006485.html Carbon hackfest date/location 18:14:09 #info unless something else happens, the plan is to have no big, formal mid-cycle hackfest, there is a resonable chance that a member organziation will host one 18:14:28 #topic events 18:14:32 #link https://wiki.opendaylight.org/view/Events:Main <-- link to updated events wiki page where info will be posted ongoing 18:14:37 #link https://www.opendaylight.org/global-events 18:16:35 #info the wiki page above is the more TSC and ODL-internal useful calendar, with lots of events listed and where it's easy to edit 18:17:06 #info open source leadership summit at lake tahoe Feb 14-16 18:17:15 #info possible ODL carbon hackfest being planned 18:17:44 #info ONS 2017 CF closes 1/21, please submit, lots of work being done to make this more truly about open networking in general and not all ON.lab all the time 18:18:37 #info NFV world congress, OpenStack summit, OPNFV summit, Containercon/Linuxcon/cloudopen, all listed there with more details 18:19:21 #topic boron 18:19:26 #info nothing this week 18:19:29 #topic carbon 18:19:37 #chair phrobb anipbu 18:19:37 Current chairs: anipbu colindixon phrobb 18:20:02 #link https://lists.opendaylight.org/pipermail/release/2016-December/009011.html <--- Autorelease issues with VTN has beenn resolved. 18:20:08 #link https://lists.opendaylight.org/pipermail/release/2017-January/009023.html <--- Autorelease issues with BIER has beenn resolved. 18:20:14 #link https://lists.opendaylight.org/pipermail/release/2017-January/009032.html <--- Autorelease Build Job currently failing on OF-CONFIG. Currently waiting on folks from ZTE to identify a fix. 18:20:31 #link https://lists.opendaylight.org/pipermail/release/2016-December/008994.html <--- Still waiting on M3 Status from 4 Projects: aaa, controller, mdsal, netconf 18:20:50 #link https://lists.opendaylight.org/pipermail/release/2017-January/009022.html <--- Periodic Jobs Removed 18:20:55 #link https://lists.opendaylight.org/pipermail/tsc/2017-January/006498.html <--- Long Job Queues 18:21:09 #info Mostly impact netvirt. Waiting to see if other projects have been affected and if any impact on release. 18:21:21 #link https://lists.opendaylight.org/pipermail/release/2017-January/009024.html <--- Karaf 4 Migration Effort. 18:22:00 #info Karaf 4 migration effort will be a critical endeaver in Carbon. Need to identify if Karaf 4 is a REQUIREMENT to remain in the Carbon Simultaneous Release Plan. 18:22:21 #info If Karaf 4 is a requirement, we need to identify the timeline (M4/M5/RC0?), support documentation (step-by-step wiki page), and TWS recording 18:22:32 #info Two Migration: migration to singlefeaturetest, migration to karaf 4 18:22:39 #info We need to poll projects on available resources. If no resources available, what are the options? 2 distributions, drop from release, communnity generates patches? 18:24:50 #info colindixon says that we really have to get away from Karaf 3 and onto Karaf 4 for, at the very lease, security reasons 18:25:24 #undo 18:25:24 Removing item from minutes: 18:25:29 #info colindixon says that we really have to get away from Karaf 3 and onto Karaf 4 for, at the very least, security reasons 18:26:42 #info this would also get us on the current Karaf timeline, which means we get features faster from them 18:28:02 #startvote shall the TSC move that we will produce _only_ a Karaf 4 distribution for Carbon and deal with how to get us across that line? -1,0,+1 18:28:02 Begin voting on: shall the TSC move that we will produce _only_ a Karaf 4 distribution for Carbon and deal with how to get us across that line? Valid vote options are -1, 0, +1. 18:28:02 Vote using '#vote OPTION'. Only your last vote counts. 18:28:17 #vote +1 18:28:19 #vote +1 18:28:20 #vote +1 18:28:26 #vote +1 18:28:34 #vote +1 18:29:01 #vote +1 18:29:05 #vote +1 18:29:06 #vote +1 18:29:29 #vote +1 18:30:51 anyone else want to chime in? 18:31:00 that's 9/13, so it carries 18:31:15 vishnoianil ? 18:31:32 #vote +1 18:32:34 going once 18:32:42 .... 18:32:45 going twice 18:32:46 .... 18:32:50 #endvote 18:32:50 Voted on "shall the TSC move that we will produce _only_ a Karaf 4 distribution for Carbon and deal with how to get us across that line?" Results are 18:32:50 +1 (10): rovarga, skitt, anipbu, shague, hideyuki, colindixon, abhijitkumbhare, zxiiro, vrpolak, jamoluhrsen 18:33:18 #agreed the Carbon release of OpenDaylight will ship with Karaf 4 only, mostly for security reasons, the TSC takes ownership for getting us across that line 18:34:42 #action skitt and others to create a step-by-step guide and minimum required steps on th wiki 18:35:11 #link https://git.opendaylight.org/gerrit/49550 <-- yangtools patch 18:36:39 #info vishnoianil asks if it has to be done all at once 18:37:50 #info as skitt explained in the e-mail, the only requirment is that we move in dependency order and nobody should have to be broken ever, we have a wave of karaf3+4 patches cascade through and then remove karaf3 as a wave going backward if we want 18:39:49 #info skitt notes that Karaf 4.0.9, which is the starting point is supposed to be released this weekend 18:40:40 #info skitt notes that if 4.0.9 doesn't ship this weekend, we'll start working against our own pre-release version of Karaf 4.0.9 18:40:52 it may help to get a primary contact in each project that will work on karaf 4 (most likely PTL) 18:41:57 anipbu: yes, we need to wrangle the projects 18:42:13 anipbu: at least to know if we are people out in the woods 18:42:14 #topic Nitrogen DDF timing 18:43:27 #info our current releas has 5/11 as a target, and there are other events coming around that time with OpenStack and memorial day 18:43:30 Karaf 4: There are consequences for release plan. It contains conditions such as "features in integration" and "csit tests running", but currently we have no karaf4 integration and no karaf4 csit. 18:43:30 #undo 18:43:30 Removing item from minutes: 18:43:36 #info our current release plan has 5/11 as a target, and there are other events coming around that time with OpenStack and memorial day 18:44:53 #info phrobb earlier in the week of the 22nd (which is tight against our current release epsecially given that we current release during the OpenStack summit) and around May 31st/June 1st would be OK as well, but then it comes up against the OPNFV summit pretty quickly 18:45:22 #action phrobb to send mail about Nitrogen DDF timing 18:45:23 really, really painful 18:45:47 #topic jenkins queue issues 18:46:06 Prem_: are you here? how critcial is it to do this today? 18:46:39 #link https://lists.opendaylight.org/pipermail/tsc/2017-January/006498.html shague bring up current issues 18:47:03 #info zxiiro is looking at using OpenStack Heat instead of JClouds, which might help 18:47:52 #info tykeal notes that it's really just CSIT jobs that are the problem, so projects that gate on CSIT jobs are being hit hard 18:49:41 It would be good if we can complete it today :) 18:49:45 tykeal: I have also seen an uptake in SSH failures ... is that related to the queues ? 18:49:59 rovarga not certain 18:50:48 Prem_: I'll try 18:51:33 colindixon: Thanks! I will probably take 5 mins to present 18:54:42 #Info a lot of discussion about CSIT jobs and how we might help to fix this 18:55:36 Proposal page - https://wiki.opendaylight.org/view/Project_Proposals:COE_integration 18:55:40 #topic Container Orchestration Proposal 18:55:52 #link https://wiki.opendaylight.org/view/Project_Proposals:COE_integration 18:56:01 #link https://lists.opendaylight.org/pipermail/project-proposals/2016-December/000563.html proposed on 12/15 18:56:09 and we are just moving the discussion about gating with CSIT to another time and we still have a giant jenkins queue right now and has been there all week and no immediate solution. 18:56:39 phrobb, tykeal: can we set up time to talk about this tomorrow? 18:56:56 as long as it's afternoon PST I can do it 18:56:57 can we prioritize CSIT jobs triggered by some projects, e.g., those gating on them, for now? 18:57:10 tykeal: after noon? or afternoon? 18:57:22 much after noon starts to be hard for east coast folks, which shague and others are 18:57:28 noon+ 18:57:37 I'm in meetings till just a bit before noon 18:57:52 1pm PST? 18:57:56 I can do that 18:58:01 shague, jamoluhrsen, tykeal, zxiiro, could we start at 1p PST? 18:58:07 +1 18:58:26 somebody send an invite? 18:58:55 colindixon, I'll do it. bluejeans. 18:58:58 colindixon: works for me 18:59:04 how does that work - from 12am jobs are at normal priority, then at 1pm they bump up? What happens the next day? 18:59:26 oh sorry, buys I was mixing colin's question with the call times 18:59:42 #info Prem_ presents the project for providing interfaces for OpenDaylight to be able to provide networking features for container orchestration systems, e.g., Kubernetes 19:01:49 #info the approach is to integrate with Kubernetes CNI directly and also via Kuryr 19:02:20 thats' a long, long list of inital committers… 19:02:57 #info developers from ericsson, redhat, intel, cienna, HPE and inocybe committed 19:03:20 colindixon: exactly my concern 19:05:35 #info Prem_ notes that the intent is to map everythning onto existing networking virtualization components, not create new ones, it will use existing Neutron implementations as the base 19:05:58 #info colindixon asks if the project really needs 15 committers? Prem_ says he'll update that with something more reasable 19:06:53 #info vishnoianil asks if it it will be in Carbon? Prem_ says they would like to be, but that's up to the TSC 19:07:41 #info Prem_ adds that the simple integration with Kuryr is probably pretty easy to get done in Carbon 19:10:18 #info colindixon and edwarnicke ask Prem_ if this is the right committer list? 19:12:02 #info Prem_ notes that the large committer list is mostly because this is moving a lot of different parts and integrating them 19:12:19 #startvote shall the TSC approve the COE project as an incubation project? -1,0,+1 19:12:19 Begin voting on: shall the TSC approve the COE project as an incubation project? Valid vote options are -1, 0, +1. 19:12:19 Vote using '#vote OPTION'. Only your last vote counts. 19:12:43 are there committers listed that are only for the kubernetes side? that doesn't matter here right? 19:12:51 #vote +1 19:12:52 #vote +1 19:12:55 #vote +1 19:12:55 #vote +1 19:12:55 #vote +1 19:12:56 #vote +1 19:12:58 #vote +1 19:13:03 #vote +1 19:13:06 #vote 0 19:13:07 #vote 0 19:13:14 #vote +1 19:13:14 #vote +1 19:13:19 #endvote 19:13:19 Voted on "shall the TSC approve the COE project as an incubation project?" Results are 19:13:19 0 (2): vrpolak, rovarga 19:13:19 +1 (10): skitt, zxiiro, edwarnicke, shague, hideyuki, colindixon, jamoluhrsen, anipbu, vishnoianil, abhijitkumbhare 19:13:30 #agreed CoE is an incubtion project 19:13:33 #topic cookies 19:13:35 #endmeeting