17:59:19 #startmeeting tsc 17:59:19 Meeting started Thu Nov 13 17:59:19 2014 UTC. The chair is colindixon. Information about MeetBot at http://ci.openstack.org/meetbot.html. 17:59:19 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 17:59:19 The meeting name has been set to 'tsc' 17:59:29 #topic agenda bashing and roll call 17:59:34 phrobb: yeah — but it’s not happening on all the webexes 17:59:41 #chair tbachman phrobb 17:59:41 Current chairs: colindixon phrobb tbachman 18:00:21 #info regXboi (irc so far) 18:00:40 #link https://wiki.opendaylight.org/view/TSC:Main#Agenda the agenda in the usual place 18:00:45 #chair regXboi 18:00:45 Current chairs: colindixon phrobb regXboi tbachman 18:00:58 #info regXboi (the audio half now) 18:01:06 phrobb: I saw the icon that you are talking but no audio however others are fine 18:01:23 #info kwatsen 18:02:09 * tbachman waits for rest of TSC members to info in 18:03:07 #info jmedved 18:03:17 #info colindixon 18:03:26 #info edwarnicke 18:03:31 #info LuisGomez 18:03:38 #action gzhao and edwarnicke to continue to work on getting Lithium auto-build to work 18:03:52 #action phrobb to start email thread on dates/times/activities for hackfest and integration-togetherness 18:03:56 WOW, meeting started on time... 18:03:58 * alagalah_ impressed 18:04:46 #info +1 to distributed hackfest 18:05:10 #undo 18:05:10 Removing item from minutes: 18:05:13 cdub: did you #info in? 18:05:26 #info opinion: regXboi +1s the concept of a distributed hackfest 18:05:35 Youcef: did you #info in? 18:05:36 #info the action from dlenrow, mohnish and mlemay to work on gathering and acrhiving use cases, I will assume that they’ll let us know when they want more time on the TSC 18:05:37 #info Youcef Laribi 18:05:40 #info Chris Wright 18:05:45 thx guys! 18:06:29 #action mlemay to Investigate usage of repotool manifest for inter-repo + tags management with zxiiro 18:06:38 LuisGomez: Keep me honest here mate 18:06:48 #info mohnish 18:06:49 #action colindixon to put together pieces for the Helium-SR1 release 18:07:05 #topic Canceling Upcoming TSC Meetings 18:07:30 #link https://wiki.opendaylight.org/view/OpenDaylight_Controller:Pulling,_Hacking,_and_Pushing_the_Code_from_the_CLI For alagalah action item 18:08:19 #info colindixon sent email to the list to cancel TSC meetings over the Thanksgiving, Christmas, and New Year breaks 18:08:46 #agreed TSC Meetings on Thanksgiving, Christmas, and New Years are all cancelled 18:08:48 #topic Updates 18:09:25 #info phrobb says that there’s nothing new on the board elections, but reminds committers to feel free to join the non-profit and vote 18:09:45 #info Board election closes November 18th 18:09:59 #info currently there are 30/142 committers that have signed up 18:10:32 #link https://wiki.opendaylight.org/view/Committer_Members_Board_Representation 18:10:36 #link https://wiki.opendaylight.org/view/Committer_Members_Board_Representation wiki detailing ODL committer board 18:10:45 cdub: just beat me to it ;) 18:10:55 #link https://linuxfoundation.echosign.com/public/hostedForm?formid=AI3XA256U2H44Y 18:10:59 #info colindixon asks TSC members to reach out to committers to encourage signing up 18:11:04 #info chris price (is late) 18:12:49 #info phrobb says that there are no large events in near future; looking at events for the new year 18:13:12 #info phrobb says ODL Summit Call For Papers is anticipated to open in January 18:13:18 edwarnicke: thx! :) 18:13:30 #info phrobb say ODL Summit Call for Papers likely to close in Feb 18:13:42 tbachman: Thank *you* for scribing so diligently :) 18:13:58 Does anyone else think that window is a little small ? 18:14:09 Or is that typical ? 18:14:13 #info jmedved asks if we have release notes for stable/helum 18:14:50 #info stable/helium isn’t currently building, due to outstanding patches 18:15:13 #link https://docs.google.com/spreadsheets/d/1PYxjiSYEks44uJByVO1P44rnI5xTJRulpKyrSsDQF9g/edit#gid=383947003 Spreadsheet for tracking stable/helium 18:15:28 #link https://docs.google.com/spreadsheets/d/1TRYposNDFPaKcySlvwkOXvfR6Anx2EFujlIjoTthhRY/edit#gid=829479065 18:15:53 #info dlenrow 18:15:55 gzhao: thx! 18:15:59 (was just looking for that ;) ) 18:16:14 tbachman: sure 18:16:17 edwarnicke: my pleasure (btw) :) 18:17:12 #info LuisGomez asks how we can improve our process/automation around release/version-bump because there is 2 weeks of disruption currently when doing this 18:17:18 #info tykeal says that the hope is that the autorelease job set will do the tagging, etc. automatically 18:17:21 phrobb: thx! 18:18:09 #info right now there’s a bit of “hand-crafting” to do version bumping due to project dependencies 18:18:46 #colindixon notes no single user can merge the tagging patches, and hand-crafting is needed to deal with the cyclic dependencies… scripting the hand-crafted work can be done but intricate and probably fragile… getting rid of cyclic depedencies would be "swell" 18:20:02 #topic Infrastructure 18:20:05 #info tykeal says that all silos are now in rackspace 18:20:16 #info congratulations to tykeal for getting all the silos into rackspace 18:20:28 yay! 18:20:33 #info gerrit didn’t get moved into rackspace last week, but that will be worked on tomorrow evening, with email sent to the list 18:21:33 #info tykeal says that all projects should be publishing to jenkins email list, rather than having this done per-project 18:22:21 #info tykeal says for projects that have dedicated jenkins email aliases, he will inject ~2 weeks a notification that the list will be decommissioned, and will remove the lists in January. 18:24:10 #actions tykeal will look through existing jenkins jobs to make sure they point to the right place 18:24:16 * ChrisPriceAB object to calling a vote 18:24:17 #agreed baring any unforseen reasons, the project specific bug/jenkins/gerrit mailing lists will be decommissioned in January 18:24:18 #topic Creation Reviews 18:24:26 #undo 18:24:26 Removing item from minutes: 18:24:33 #info Creation Review for SXP Project 18:24:37 dang 18:24:39 #undo 18:24:39 Removing item from minutes: 18:24:46 #topic Creation Review for SXP Project 18:25:06 #link https://wiki.opendaylight.org/view/Project_Proposals:SXP Project Proposal 18:25:48 #link https://wiki.opendaylight.org/images/5/5a/SXP_Specification_and_Architecture_v00.docx Specification and Architecture 18:26:17 #link https://wiki.opendaylight.org/images/d/d2/Sxp-odl-plugin.pdf Plugin proposal (pdf form) 18:26:52 #info the above link is the presentation that Miloslav is presenting 18:30:05 #info abhijitkumbhare asks if miroslav can describe the SXP protocol 18:30:19 #info SXP is a security protocol dedicated for delivery of security data 18:30:41 #info security data consists of security groups, with each group having multiple users registered 18:30:59 #info it delivers IP address bindings to security groups 18:31:18 #info The attached router/switch advertises these security bindings into the network 18:31:33 #link https://datatracker.ietf.org/doc/draft-smith-kandula-sxp/ 18:31:34 #info This will be something similar to BGP route reflection 18:31:44 edwarnicke: thx! 18:32:14 #info devices can then act on this security information 18:32:38 #info jmedved says this would be very useful as a renderer in groupbasedpolicy 18:32:44 #info colindixon asks how this would relate to opflex 18:33:00 #info jmedved says this doesn’t relate to opflex, but could be one of the renderers used in groupbasedpolicy 18:33:34 #info jmedved this just implements the protocol and the bindings database 18:33:56 an “offset 2” project :) 18:34:21 tbachman: can't be - it's not a leaf 18:34:58 #info LuisGomez asks if there will be REST APIs to connect this 18:35:14 #info colindixon notes that since there is an MD-SAL data store, the REST APIs will be present 18:35:38 regXboi: how is it not a leaf? 18:35:55 (i.e. what is it depending on) 18:36:40 they don’t have to pick what offset they’re at until they create a release plan for lithium 18:36:53 colindixon: fair point 18:36:54 or what depends on it tbachman regXboi 18:37:00 gbp 18:37:02 ah, good point 18:37:03 lol 18:37:07 they've said that 18:37:37 #info kwatsen asks if SXP has been implemented by any other vendors than Cisco 18:37:50 presumably the project will be dependent on yangtools, controller (for the MD-SAL), and odlparent 18:38:00 #info jmedved says he’s not sure, but there is interest from academia (e.g. Rice University) 18:38:45 #info colindixon says that it should be clarified that the scope is only things related to the SXP protocol 18:39:16 #info colindixon notes that there’s only a single committer on the intial committer list, and asks if that’s intentional? 18:39:52 #info miroslav says that there are 4 committers, but they are working out the details 18:40:46 #info regXboi and colindixon say it’s much easier to approve if there’s a full list of committers 18:41:21 #info edwarnicke says he’d like to see a link to the SXP IETF draft in the proposal 18:41:40 #info colindixon says there is a link to the IETF draft in the proposal 18:42:34 there is a link to the datatracker 18:42:56 Hang on 18:43:23 My basic point is, if we are going to request greater clarity (which is probably goodness) let us be as correct as possible so as to help the proposers to take the action :) 18:43:31 concrete, not correct 18:43:33 yes 18:43:40 Why do we care so much about the RFC ??? I thought the beauty, and much hooplah made in the press, about how OpenSource is the NEW way standards are made. 18:44:44 what was the point related to i-d? 18:44:49 #action miroslav to clarfiy the scope of the SXP proposal, add link to IETF doc-tracker, and provide full list of committers 18:48:13 #info link to datatracker is already on first line of proposal description https://wiki.opendaylight.org/view/Project_Proposals:SXP#Description 18:48:33 phrobb: thx! 18:49:21 #info mohnish asks how you would verify the functionality (i.e. against a device)? 18:49:26 I missed that question? 18:49:41 thanks for the notes tbachman, you rock 18:49:51 #info miroslav says this has only been tested against Cisco equipment 18:49:57 colindixon: my pleasure! :) 18:50:53 #info edwarnicke points out that with the openflowplugin, every time we test against other vendors, we run into interoperability/standardization issues 18:51:22 #info jmedved adds that there are actually committers, but they just weren’t listed in the proposal 18:53:42 #info in particular, the list of “resoruces committed” (4 people) should also be initial comitters 18:53:51 colindixon: thx! 18:54:09 * tbachman was leaving architectural space for phrobb’s startvote ;) 18:54:23 tbachman: you know 18:54:26 nature 18:54:29 Could we add an agenda item next week to talk about the 'user story' idea? I think if we adopt it we'd need to alter the project lifecycle to add it as a field 18:54:40 abhors 18:54:43 lol 18:54:43 +1 dlenrow's comment about it being something we should ask for, but not block on 18:54:45 a 18:54:47 vacuum 18:54:48 vacuum 18:54:50 damn 18:54:52 beat me 18:54:54 LOLZ 18:54:57 Lol 18:55:01 cdub: That would explain all the dirt on the carpeting outside ;) 18:55:05 #startvote Shall the TSC approve the SXP project to the Incubation state within OpenDaylight provisional to adding the full list of initial committers and the scope to be made clear as tied to the SXP protocol, and a trademark name search being satisfactorily completed? -1, 0, +1 18:55:05 Begin voting on: Shall the TSC approve the SXP project to the Incubation state within OpenDaylight provisional to adding the full list of initial committers and the scope to be made clear as tied to the SXP protocol, and a trademark name search being satisfactorily completed? Valid vote options are -1, 0, +1. 18:55:05 Vote using '#vote OPTION'. Only your last vote counts. 18:55:11 edwarnicke: heh 18:55:16 #vote +1 18:55:19 #vote +1 18:55:23 #vote +1 18:55:31 #vote +1 18:55:33 #vote +1 18:55:34 #vote +1 18:55:37 #vote +1 18:55:38 Could we please get the specifics of the initial committers in the vote? 18:56:01 #vote +1 18:56:25 #vote +1 18:56:33 #endvote 18:56:33 Voted on "Shall the TSC approve the SXP project to the Incubation state within OpenDaylight provisional to adding the full list of initial committers and the scope to be made clear as tied to the SXP protocol, and a trademark name search being satisfactorily completed?" Results are 18:56:33 +1 (9): dlenrow, regXboi, LuisGomez, ChrisPriceAB, edwarnicke, cdub, mohnish, colindixon, Youcef 18:57:38 #info kwatsen expresses concern about project proliferation for protocols that are only supported by a single vendor 18:57:42 #info note that the above vote was to merely move the 4 resources committed to also be initial committers, not a blank check for initial committers 18:58:19 #info cdub agrees, and says that odlforge could help here, but this all requires wider discussion 18:58:29 wahoo 18:58:33 colindixon: take it away 18:58:35 #topic Code Maturity and Lifecycle Issues 18:59:53 #topic Summer Intern Program 19:01:39 #info phrobb says that Summer Interns lock down their plans right around the end of the calendar year 19:02:39 #info ODL needs to start coming up with projects/topics, and identifying mentors for the program 19:02:54 #info The program would start May 19th, and end about the 3rd week of August 19:04:34 #info colindixon asks if there’s a wiki page for the program this year? 19:04:43 #info phrobb says he’ll be working on the wiki page this week 19:04:55 #action phrobb will create a wiki page by next week’s TSC meeting where we can put projects and mentors 19:06:56 #topic Certificate Authorities 19:07:51 #info jmedved says when you try doing builds from the MAC and pulling artifacts from nexus, it requires certificates 19:08:28 #info jmedved says that you need to manually install the certificates into your JDK/JVM, as the manuals on startcom are insufficient, and not as widespread as other certificate authorities 19:08:45 #info jmedved asks if we can use a more wider known certificate authority for our nexus? 19:09:16 #info jmedved says this is a MAC-specific problem (but hasn’t tried windows — works on Linux) 19:09:44 so startcom is pretty much a standard cert authority 19:10:11 #info tykeal says that he as an admin has no ability to change the certificate authority that the LF uses 19:10:11 is this a problem with the JRE distributed by mac? 19:10:15 *apple? 19:10:53 because my first guess is that apple is pruning the truststore ... 19:10:53 #action phrobb and tykeal to work with the LF staff to identify this as a problem that the LF has and start to investigate alternatives (i.e. raise this as a potential problem) 19:10:54 #action phrobb and tykeal to work with the LF staff to see if there is a reasonable solution here and to understand how easy/hard it would be to fix 19:10:59 colindixon: sorry 19:11:01 #undo 19:11:01 Removing item from minutes: 19:11:04 all good :-) 19:11:07 yours was better ;) 19:13:06 #info the issue is that downloading the archetype index, it points you to https 19:13:59 #info rovarga asks if this is an issue with the JVM distributed by apple 19:14:40 colindixon: you want to action that? 19:15:54 #action rovarga, jmedved, edwarnicke, tykeal, and phrobb will work on finding solution 19:16:24 colindixon: i believe you are right 19:16:50 #topic Progress on Technical Debt and pom files 19:17:22 #link https://wiki.opendaylight.org/view/CrossProject:HouseKeeping_Best_Practices_Group:Project_layout 19:17:57 #info rovarga says that the idea is to introduce a general best practices for both project layout and what stuff goes into which poim and why 19:18:13 poim? 19:18:20 #undo 19:18:20 Removing item from minutes: 19:18:27 #info rovarga says that the idea is to introduce a general best practices for both project layout and what stuff goes into which pom file and why 19:18:40 poim, it's a fancy way of saying poem ;) 19:18:44 tykeal: lol 19:19:20 #info The current wiki page provided in the link is WIP 19:20:38 #info colindixon says that the pom changes will facilitate version bumping between releases 19:21:48 rovarga: very nice! 19:22:01 #action rovarga to email alagalah to do a TWS on this topic 19:22:03 rovarga: thank you for taking stab at this 19:22:06 rovarga: hi there 19:22:11 thanks a ton! 19:22:27 rovarga: you da man 19:22:44 #topic Code Maturity and Lifecycle Issues 19:22:56 * tbachman wonders when we’ll move back to 1hr TSC ;) 19:23:25 #info colindixon asks if we’re willing to have principals on things such as how long we’ll support releases 19:23:37 #info colindixon would also like to discuss ODLForge 19:23:59 #info colindixon would also like to discuss the project promotion process 19:24:49 #topic ODLForge 19:25:04 #link https://pad.opendaylight.org/public/odlforge -- etherpad high-level ideas around ODLForge 19:25:25 #info cdub says it would be beneficial to have a space to do collaborative development to bring developers into the fold for the purpose of building up a project 19:25:46 #info cdub feels that many of the projects proposals are too empty, and that having code might help here 19:25:55 tykeal: thx! 19:26:00 :) 19:27:36 #info regXboi asks how are we talking about changing the project review process 19:27:37 odlforge link requires login. 19:27:50 mohnish: use your ODL credentials 19:27:55 #info cdub ODLForge and project reviews don’t have to be coupled 19:28:10 ok, thanks. 19:28:35 tbachman: can you #info regXboi attempts not to be human? ;) 19:28:44 #action regXboi to be human 19:28:44 lol 19:28:48 #undo 19:28:48 Removing item from minutes: 19:28:52 lol 19:28:58 #action regXboi to attempt to not be human 19:29:01 Lost audio. Cdub Forge vision will fix what's broken in project creation process 19:29:01 * tbachman forgot regXboi had been chaired ;) 19:29:27 #info regXboi doubts he'll be able to succeed in that action items 19:29:31 :) 19:30:18 #info tykeal says that the jenkins side would use Jenkins Job Builder as a separte silo 19:30:44 #info The projects that come in would get repositories under an ODL namespace, similar to how releng has theirs in a separate namespace 19:31:00 #info projects would be able to publish to their own place in the nexus repo 19:31:23 #info tykeal says building out that infra would make it quicker to bring new projects online 19:32:20 Will projects migrate easily from forge to approved repo? 19:32:22 #info tykeal says that setting up this infra isn’t that hard (could even do by next week’s TSC) 19:32:59 #info edwarnicke asks if it would be possible to set up something separately, in order to have a clear distinction 19:33:31 #info tykeal says they’ve refused import of history from outside repos, b/c they have no way of handling identity linkage from outside entities 19:33:55 namespace hierarchy 19:34:06 #info tykeal says that openstack also has their stackforge on their primary system 19:35:41 #info edwarnicke would like to have a clearly separated (by name — DNS) nexus server 19:36:19 #info edwarnicke would like to have the governance issues worked out, so that ODL projects don’t have dependencies on ODL Forge artifacts 19:37:08 not repo 19:37:11 namespace 19:42:08 * regXboi realizes that the microphone bleed through has given him a splitting headache :( 19:42:28 Is there an important difference that I'm missing between pulling a project from the Forge as a dep vs pulling ? 19:42:42 dfarrell07: Chris Wright is pointing to that 19:42:42 dfarrell07: there is no difference 19:42:58 dfarrell07: exception is...odlforge _feels_ more odl-ish 19:43:14 * dfarrell07 feels better, isn't as confused as he thought 19:44:00 Forge == GitHub 19:46:16 tykeal already wrote this down 19:47:10 heh, the chris's 19:47:11 cdub: Link? 19:47:19 Ah... I see it :) 19:47:32 edwarnicke: https://pad.opendaylight.org/public/odlforge 19:47:46 #link https://pad.opendaylight.org/public/odlforge <-- current proposal 19:47:56 cdub: We can discuss the time to have a call about it on the call about call avoidance ;) 19:48:05 #action edwarnicke cdub tykeal and others to come up with concrete plan for next week’s TSC on how to make ODL Forge amenable to all 19:48:15 suitable? 19:48:46 +1 19:48:53 tbachman: Looks right to me :) 19:49:02 :) 19:50:28 dlenrow: your gain is ... wow 19:50:38 19:50:50 dlenrow’s amp goes to 11 :) 19:50:55 ^^lol 19:52:28 Do we know when we’re going to move to 1hr meetings? 19:53:22 hahah 19:53:29 #topic singularity clause 19:53:36 #link http://www.opendaylight.org/project/tsc/tsc-policy 19:53:38 http://en.wikipedia.org/wiki/Technological_singularity 19:54:06 alagalah: Skynet is out of scope ;) 19:54:18 it's networking isn't? 19:54:41 tykeal: Firecontrol systems are out of scope ;) 19:54:48 heh 19:55:04 * tbachman just wants sharks with laser beams on their heads 19:55:22 colindixon: I think that is an impossibility 19:55:29 tbachman: I think there may be an aquatic exception ;) 19:55:36 lol 19:55:40 http://blogs.westword.com/showandtell/05shark.jpg 19:55:49 nice 19:55:56 * edwarnicke wonders if perhaps we should start a tradition of April 1 petitions to the board 19:56:10 * regXboi has one 19:56:19 * edwarnicke has many :) 19:56:32 * regXboi still wants the "add #sand command to IRC petitions" 19:56:36 Including edits to the scope to allow for control systems for lasers beams on the heads of sharks :) 19:56:41 controller for tcp over carrier pigeon? 19:57:37 tykeal: Rest interfaces for feeding the carrier pidgeons 19:59:17 regXboi: you sure it's a chicken and not a pigeon? 20:00:46 #info cdub, colindixon and edwarnicke talk about how we can encourage (and later require) projects to move though the lifecycle process 20:01:10 colindixon: thx 20:01:48 #endmeeting