17:00:05 <phrobb> #startmeeting OpenDaylight TSC Meeting 2014-05-08 17:00:05 <odl_meetbot> Meeting started Thu May 8 17:00:05 2014 UTC. The chair is phrobb. Information about MeetBot at http://ci.openstack.org/meetbot.html. 17:00:05 <odl_meetbot> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 17:00:05 <odl_meetbot> The meeting name has been set to 'opendaylight_tsc_meeting_2014_05_08' 17:00:08 <regXboi> running late 17:00:24 <regXboi> dmm: I do not (as far as I know) have delegation today 17:01:45 <phrobb> #topic TSC members please #info in 17:02:13 * regXboi waits for webex 17:02:30 <RobDolin> #info Rob Dolin proxy for Rajeev Nagar (Microsoft) 17:02:56 <cdub> #info Chris Wright 17:03:04 <ChrisOShea> #Info Chris O'Shea proxy for Chris Price (Ericsson) 17:04:04 <dmm> @RobDolin: welcome 17:04:16 <Kent> #info Kent Watsen 17:04:37 <dmm> #info dmm 17:05:05 * regXboi now watches the spinning ball :( 17:05:14 <alagalah> regXboi: Its candy hitting a fan 17:05:30 <edwarnicke> I'm having some webex issues... still trying 17:07:17 <phrobb> #chair regXboi edwarnicke RobDolin 17:07:17 <odl_meetbot> Current chairs: RobDolin edwarnicke phrobb regXboi 17:07:43 <phrobb> #topic Agenda Bashing 17:07:47 <dmm> #link https://wiki.opendaylight.org/view/TSC:Main#Agenda_Summary (agenda) 17:08:06 <edwarnicke> #info Ed Warnicke 17:08:06 <regXboi> #info regXboi for IBM ( I *do* hold the proxy today) 17:08:19 * edwarnicke feels sorry for our poor beleaguered agenda 17:08:26 <edwarnicke> always being bashed ;) 17:08:35 <regXboi> edwarnicke: really? 17:08:51 <phrobb> #topic Event Updates 17:09:17 <phrobb> #link http://www.opendaylight.org/events/2014-05-12/openstack-summit-atlanta 17:09:37 <cdub> could be ksh'ing, zsh'ing, dashing 17:09:46 <regXboi> #info dmm mentions that TSC members will be at OpenStack next week and hopes for running the call next week from the summit 17:09:47 <kwatsen> #info Kent Watsen 17:10:03 <alagalah> cdub: Oh man. 17:10:05 <phrobb> #topic At-Large TSC member election update 17:10:17 <regXboi> phrobb: you get the topics, I'll take the info 17:10:18 * cdub hands dmm an airhorn 17:10:53 * edwarnicke breaks out his protective ear gear 17:11:23 <regXboi> #info board meeting was held last night and the "option 1" proposal was discussed (removing the "unless otherwise represented" language for platinum members). 17:11:38 <regXboi> #info board approved with the stipulation that this clause only applies to the forthcoming election 17:11:59 <regXboi> #info so by-laws return to the original form for the next election 17:13:17 <phrobb> #topic Committer Promotions 17:13:39 <dmm> #link https://wiki.opendaylight.org/view/Group_Policy:Main (Group Policy) 17:13:50 <Madhu> phrobb: is it committer promotions or replacement ? 17:13:54 <regXboi> #info proposal: keith burns and rob adams to committer for group based policy 17:14:05 <Madhu> the email sounded like replacing committers due to unavailability of existing commiters 17:14:10 <regXboi> madhu: that is a good question 17:14:21 <regXboi> er Madhu: that is a good question 17:14:22 <phrobb> not sure.. just copied the agenda item 17:16:49 <readams> I don't see why we need to remove anyone 17:17:36 <Madhu> it is not about removing anyone actuallyy 17:17:40 <cdub> #link https://lists.opendaylight.org/pipermail/tsc/2014-May/001172.html (relevant list thread) 17:17:47 <regXboi> cdub: thanks 17:17:57 <Madhu> the question was due to previous decisions to add committers 17:18:17 <Madhu> and questions raised earlier about contributions etc. 17:18:26 <Madhu> for existing projects 17:22:11 <cdub> #info luis asks, can a committer voluntarily drop committer responsbility 17:22:29 * cdub was going to suggest that we add that to the charter 17:22:35 <cdub> but it's offtopic for this 17:22:57 <phrobb> #startvote "Shall we promote Kieth Burns and Rob Adams to Committer of Group Based Policy Project? 1, 0, -1 17:22:57 <odl_meetbot> Begin voting on: "Shall we promote Kieth Burns and Rob Adams to Committer of Group Based Policy Project? Valid vote options are 1, 0, -1. 17:22:57 <odl_meetbot> Vote using '#vote OPTION'. Only your last vote counts. 17:23:07 <cdub> #info we don't have charter mechanism for that, while could be useful, it's offtopic 17:23:07 <RobDolin> #vote +1 17:23:07 <odl_meetbot> RobDolin: +1 is not a valid option. Valid options are 1, 0, -1. 17:23:11 <dmm> #vote +1 17:23:11 <odl_meetbot> dmm: +1 is not a valid option. Valid options are 1, 0, -1. 17:23:12 <cdub> #vote +1 17:23:12 <odl_meetbot> cdub: +1 is not a valid option. Valid options are 1, 0, -1. 17:23:13 <edwarnicke> #vote +1 17:23:13 <odl_meetbot> edwarnicke: +1 is not a valid option. Valid options are 1, 0, -1. 17:23:16 <ChrisOShea> #vote 1 17:23:17 <regXboi> #vote 0 17:23:20 <RobDolin> @vote 1 17:23:20 <cdub> #vote 1 17:23:25 <kwatsen> #vote 0 17:23:26 <dmm> #vote 1 17:23:33 <RobDolin> Sorry, I voted "+1" instead of "1" 17:23:35 <cdub> turns out i can't read 17:23:35 <edwarnicke> #vote 1 17:23:49 <phrobb> #endvote 17:23:49 <odl_meetbot> Voted on ""Shall we promote Kieth Burns and Rob Adams to Committer of Group Based Policy Project?" Results are 17:23:49 <odl_meetbot> 1 (4): cdub, ChrisOShea, dmm, edwarnicke 17:23:49 <odl_meetbot> 0 (2): kwatsen, regXboi 17:23:54 <edwarnicke> cdub: evidently my literacy is also questionable ;) 17:24:09 <alagalah> Thanks everyone! 17:24:10 <RobDolin> It's 4-0-2 17:24:11 <cdub> edwarnicke: maybe it's a prereq for this vote? 17:24:29 <readams> Thanks folks 17:24:40 <RobDolin> Congratulations Keith Burns and Rob Adams 17:24:50 <phrobb> #agreed Keith Burns and Rob Adams are now Committers on the Group Based Policy Project 17:24:56 <RobDolin> #info Ryan Moats (IBM) suggests this should be a special case and not a precedent 17:24:57 <alagalah> RobDolin: Thanks sir 17:25:01 <Madhu> regXboi: this is exactly what i was worried about :) 17:25:06 * edwarnicke suggestions we add #congratulations to meetbot :) 17:25:18 * regXboi hugs meetbot 17:25:23 <phrobb> #topic Creation Reviews 17:25:23 <tbachman> edwarnicke: or #suggestions ;) 17:25:38 <edwarnicke> tbachman: #happydance :) 17:25:41 <tbachman> lol 17:26:09 <cdub> i'm sure meetbot has a #patchesaccepted policy 17:26:22 <phrobb> #topic Creation-Review: ODL-SDNi App 17:26:45 <edwarnicke> cdub: I'm sure it does... but the results may prove too... entertaining ;) 17:26:53 <dkehnx> which code scan product? 17:27:10 <edwarnicke> dkehnx: Check in with Phil... he handles it 17:27:12 <cdub> presentation that does not have a link does not exist 17:27:38 <dkehnx> phrobb, which code scan product was being referred to? 17:27:45 <RobDolin> #link https://wiki.opendaylight.org/view/Project_Proposals:ODL-SDNi_App (SDNi Project Proposal) 17:27:51 <cdub> #link https://wiki.opendaylight.org/view/File:SDNiAppProposal1.ppt <-- slides 17:27:55 <phrobb> @dkehnx Combination of FOSSology and Sonatype Component LifeCycle Manager 17:28:17 <dkehnx> phrobb, thx 17:30:12 <regXboi> #info question about whether this is hardcoded or user defined 17:32:49 <regXboi> phrobb: please scribe - I have some questions :) 17:33:00 <phrobb> regXboi: you got it 17:33:33 <phrobb> Q: regXboi asks "since this isn't clustering, are you assuming you are federating controllers in some way?" 17:33:58 <phrobb> #info Q: regXboi asks "since this isn't clustering, are you assuming you are federating controllers in some way?" 17:34:49 <phrobb> #Info Q in particular what does this project add above existing clustering capabilities? 17:35:25 <phrobb> #info A We are trying to establish a federation of controllers - different controllers from different domains 17:38:49 <alagalah> I'm curious how you get "the best possible route" when network topology is in the "Extendable to:" section... 17:43:11 <phrobb> #info Q: Concern this is federation based on BGP - issues with securing the data flowing between the domains 17:43:33 <phrobb> #info Q what is the plan to secure the trust-path? 17:45:10 <phrobb> #info A: This is a known issue with BGP, with solutions being worked on 17:46:18 <regXboi> phrobb: thanks for scribing - as I said I had questions :) 17:46:41 <rovarga> #info I think sacurity of this is being handled in the sidr IETF working group 17:46:47 <phrobb> #info there are dependencies on existing ODL-BGPCEP project and those are being taken into account in the release plan 17:47:13 <edwarnicke> #link https://wiki.opendaylight.org/view/Simultaneous_Release:Helium_Release_Plan#Participating_Projects - place to list yourself and your release plan (by Monday 5/12) if you want to join Helium 17:48:10 <phrobb> #info Q: Is this the first proposal that has described cross-domain federation? 17:48:17 <colindixon> I’ll jump in after dmm 17:48:36 <regXboi> #info A: yes and for opendaylight, things are bottom up 17:49:00 <phrobb> #info A: This is the first federation type of proposal, everything before has been more clustering focused… 17:50:37 <phrobb> #info With a clustering datastore working, similar functionality is exists but having a more federation-view is also a valid methodology. 17:51:08 <phrobb> #info Is there any provision for different controllers having similar/compatible functionality? 17:53:21 <phrobb> #info Current plan is to share the data provided in this presentation. More information can be shared in the future 17:54:00 <phrobb> #startvote Shall SDNi be moved to Incubation? +1, 0 -1 17:54:00 <odl_meetbot> Begin voting on: Shall SDNi be moved to Incubation? Valid vote options are +1, 0, -1. 17:54:00 <odl_meetbot> Vote using '#vote OPTION'. Only your last vote counts. 17:54:08 <regXboi> #vote 1 17:54:08 <odl_meetbot> regXboi: 1 is not a valid option. Valid options are +1, 0, -1. 17:54:09 <edwarnicke> #vote 1 17:54:09 <odl_meetbot> edwarnicke: 1 is not a valid option. Valid options are +1, 0, -1. 17:54:13 <regXboi> #vote +1 17:54:14 <dmm> #vote 1 17:54:14 <odl_meetbot> dmm: 1 is not a valid option. Valid options are +1, 0, -1. 17:54:17 <edwarnicke> #vote +1 17:54:29 <dmm> #vote +1 17:54:31 <ChrisOShea> #vote +1 17:54:38 <kwatsen> #vote +1 17:54:43 <cdub> #vote +1 17:54:59 <phrobb> #endvote 17:54:59 <odl_meetbot> Voted on "Shall SDNi be moved to Incubation?" Results are 17:54:59 <odl_meetbot> +1 (6): ChrisOShea, regXboi, dmm, edwarnicke, cdub, kwatsen 17:55:18 <edwarnicke> Who is ChrisOShea? 17:55:21 <regXboi> congratulations! 17:55:28 <ChrisOShea> Filling in for Chris Prices 17:55:29 <cdub> edwarnicke: ChrisPrice's proxy 17:55:29 <edwarnicke> Oh :) 17:55:36 <edwarnicke> Apologies :) Missed that 17:55:38 <phrobb> #agreed ODL-SDNi is moved to Incubation 17:55:42 <regXboi> nice! 17:56:09 <phrobb> #topic Creation Review - South Bound Plugin To The OpenContrail Platform 17:57:22 <cdub> we need a link for the slides 17:57:46 <regXboi> dmm: can we get a #link for slides? 17:58:42 <regXboi> dmm: thanks 17:58:50 <kwatsen> #info I have to upload the slides to the wiki first 17:59:15 <RobDolin> Here's the project proposal: 17:59:17 <RobDolin> #link https://wiki.opendaylight.org/view/Project_Proposals:Southbound_plugin_to_the_OpenContrail_platform (Southbound plugin to the OpenContrail platform Proposal) 18:02:28 <Madhu> is it targeted for Virt edition only ? 18:03:04 <cdub> for proposlas, there are no editions 18:03:20 <cdub> heh, a new kind of sla, a proposla 18:04:12 <colindixon> #info LuisGomez notes that the proposal seems to propose a completely vertically integrated approach with calls going from the Neutron plugin to their SB plugin and then to OpenContrail, he asks if there are any plans for more integration in the future. 18:04:26 <Madhu> cdub: thanks 18:04:44 <colindixon> #info answer is that they may integrate more in the future, but for now the veritcal integration is the plan for the first release efforts 18:05:22 <phrobb> thanks colindixon 18:05:30 <colindixon> phrobb: no worries 18:10:23 <phrobb> #info Question raised on added value expected from including OpenContrail plugin within OpenDaylight? 18:10:28 <colindixon> thanks 18:11:39 <phrobb> #info A: there have been customers wanting to manage OpenContrail-manged networks along with other types of networks all from OpenDaylight 18:12:39 <Madhu> am confused on this policy 18:13:01 <Madhu> doesnt openDOVE or VTN fall under the same question ? 18:13:02 <networkstatic> southbound REST_API sounds weird 18:13:16 <phrobb> #info Q: Given that Board has a Vendor-Neutral policy, can the TSC decide on this name? 18:14:13 <phrobb> #info A: dmm states that we will be only voting on the project. The name of the project is still TBD until the Board decides on OpenDaylight's "Vendor Neutral" policy 18:17:41 <networkstatic> would like to know what else they are interested in contributing too. Would be sweet to see more then just the sbound driver. 18:18:06 <phrobb> #startvote "Shall the TSC give conditional approval the Southbound Plugin to the OpenContrail platform" to Incubation.. two conditions 1) the name to be used for the project not decided, and 2) a successful code scan? +1, 0, -1 18:18:06 <odl_meetbot> Begin voting on: "Shall the TSC give conditional approval the Southbound Plugin to the OpenContrail platform" to Incubation.. two conditions 1) the name to be used for the project not decided, and 2) a successful code scan? Valid vote options are +1, 0, -1. 18:18:06 <odl_meetbot> Vote using '#vote OPTION'. Only your last vote counts. 18:18:27 <kwatsen> #vote 1 18:18:27 <odl_meetbot> kwatsen: 1 is not a valid option. Valid options are +1, 0, -1. 18:18:28 <regXboi> #vote +1 18:18:28 <edwarnicke> #vote +1 18:18:33 <RobDolin> #vote +1 18:18:33 <kwatsen> #vote +1 18:18:34 <ChrisOShea> #vote +1 18:18:39 <dmm> #vote +1 18:19:11 <cdub> #vote 0 18:19:16 <RobDolin> #endvote 18:19:16 <odl_meetbot> Voted on ""Shall the TSC give conditional approval the Southbound Plugin to the OpenContrail platform" to Incubation.. two conditions 1) the name to be used for the project not decided, and 2) a successful code scan?" Results are 18:19:16 <odl_meetbot> 0 (1): cdub 18:19:16 <odl_meetbot> +1 (6): ChrisOShea, regXboi, edwarnicke, dmm, kwatsen, RobDolin 18:20:25 <RobDolin> #agreed Conditional approval of project proposal 18:20:32 <phrobb> #topic Hydrogen Stable Release 18:21:14 <phrobb> #link https://wiki.opendaylight.org/view/CrossProject:Stable_Release 18:21:38 <cdub> #link https://lists.opendaylight.org/pipermail/tsc/2014-May/001156.html <-- mail thread 18:24:22 <phrobb> #info discussion ongoing on list. Proposal made to not Cherrypick all changes to master. Instead pick a point in time and copy master to stable-branch then pull non-bug-fix patches from the stable copy. Discussion is ongoing 18:27:59 <phrobb> #info cdub suggests maybe getting the specific projects that are having the problem of cherrypickig bug-fixes from master to do the mass-copy. Other projects that have not had that much delta or cross-project-dependencies can just cherrypick the bug-fixes. 18:28:27 <rovarga> so I think the ask is to do this for yangtools and controller 18:30:22 <cdub> rovarga: it's wider than that 18:30:49 <cdub> rovarga: at least Abhijit replied effective +1 as well 18:32:28 <rovarga> cdub: yes, I know. yangtools and controller are the critical ones, as they are the ones blocking all others 18:32:36 <cdub> Madhu: it looks like this: hydrogen->patch->pactch->patch->patch->patch[now] 18:33:10 <cdub> now name them A and B (A==stable B==not), hydrogen->A->B->A->B-> 18:33:24 <cdub> you can eityher start w/ dropping B's or applying A's 18:33:32 <cdub> personally, makes no sense to drop B's 18:33:49 <cdub> but the projects are saying this works 18:35:30 <edwarnicke> dmm: I'm good 18:36:21 <phrobb> #topic Developer Best Practices 18:36:32 <regXboi> #link https://wiki.opendaylight.org/view/Developer_Best_Practices 18:37:51 <Madhu> cdub: if that works for projects... thats fine. but end of the day it has to be stable. 18:38:01 <cdub> Madhu: totally agree 18:38:04 <cdub> Madhu: time will tell 18:38:09 <Madhu> but it is weird on the approach we are taking. 18:38:16 <cdub> Madhu: shows the the modulraity is...not 18:38:17 <Madhu> maybe this learning is good for helium post 18:38:42 <Madhu> to start this stable/helium right away and never have to rewind or having this discussion half way 18:38:51 <Madhu> and many wasted efforts by many 18:39:08 <Madhu> cdub: i don't want to blame modularity :) 18:39:12 <cdub> yes, it's in the release planning for helium 18:39:18 <cdub> Madhu: why not? it's true 18:39:51 <raghu67_> Agree. If we create the stable/helium concurrent with the helium release and require double commits for fixes, this should not be an issue 18:40:27 <Madhu> cdub: i was not clear. modularity will help if we follow that in practice. 18:41:24 <cdub> Madhu: ah, yup. agree. and it's _always_ tough to do cross-service (interdependent) changes...bugs should capture that inter-dependency 18:41:46 <phrobb> #info request is for TSC to advocate/strongly-encourage all projects in OpenDaylight to comply with 2 recommendations: 1) Build Time Testing, and 2) Comments documenting the code 18:42:56 <cdub> turns out it's not apple pie...it's a apple brown betty 18:43:41 <phrobb> #info comments made that this is "Motherhood and Apple pie", *but* we should probably look at a stylecheck rule that documentation is done with to succeed a build cycle 18:45:17 <phrobb> #info On commenting - not having comments is the exception not the norm… ie it must be explained why the comment is not there. 18:46:29 <colindixon> “Always code as if the person who ends up maintaining your code is a violent psychopath who knows where you live.” http://c2.com/cgi/wiki?CodeForTheMaintainer 18:46:42 <cdub> heh 18:46:54 <ChrsPriceAB> -> Austin ! 18:48:25 <colindixon> I don’t think the TSC *can* mandate these things, but this is the venue where we can raise the issue and put pressure on people in the ways that we can 18:50:13 <cdub> ok...i mandate that we stop (ab)using "motherhood and apple pie" 18:50:20 <rovarga> colindixon: yes. or, as I tell my devs: "imagine you will come back to this pile in 5 years and try to make sense of it" 18:50:22 <Madhu> +1000 18:50:32 <Madhu> have never heard of so many times someone saying that :) 18:51:10 <colindixon> I’m 99% sure that devinavery wasn’t trying to say the TSC should mandate this, but should raise attention to the fact that, at least for some people, joining from the outside, we haven’t been doing as good a job as we should to make code accessible 18:51:25 <colindixon> and here are some concete suggestions 18:52:47 <rovarga> I think we need to create a 'cookbook' of recommendations on which committers agree 18:52:53 <cdub> #link https://pad.opendaylight.org/p/DevelopmentBestPractices <-- here's what i wrote down as things to consider 18:52:57 <phrobb> #info devinavery to continue this discussion on the list 18:53:04 <rovarga> cookbook as in -- how do I do this easily 18:53:09 <edwarnicke> devinavery: How can I help you on getting some checkstyle stuff going? 18:53:15 <phrobb> action devinavery to continue this discussion on the list 18:53:15 <rovarga> and let committers enforce it how they feel it's appropriate 18:53:29 <edwarnicke> I have a checkstyle bundle that can be be included as a maven dependency 18:53:36 <colindixon> #action devinavery will send a mail out to start this discussion about code best practices 18:53:38 <edwarnicke> We could start from that 18:53:41 <phrobb> #topic ovsdb.neutron in SP edition 18:53:53 <cdub> edwarnicke: checkstyle is busted for me, can you help me debug it? 18:54:39 <phrobb> #info madhu asks TSC " given the interest from the users, should we add ovsdb integration project to the SP edition?" 18:55:36 <devinavery> edwarnicke: I wasn't actually suggesting stylechecking. I believe the ball has to lie with the committers to encourage comments and unit tests. I believe at this point we would spend more time mandating and building controls instead of just commenting and testing. :) Will add more in e-mailto the lists 18:56:25 <ChrsPriceAB> I would ask that the TSC review the "release vehicles" and their intentions for the release vehicles before Helium. 18:56:44 <phrobb> #info dmm notes that this request is also to add ovsdb to Hydrogen SP release 18:56:45 <dmm> @ChrisPriceAB -- that's in the release plan 18:57:10 <ChrsPriceAB> thanks, will check it. only able to follow on IRC today. :) 19:00:18 <phrobb> #info dmm suggests that we take this discussion to the List to further explore ovsdb inclusion into stable hydrogen release of the SP edition 19:00:33 <phrobb> #endmeeting