18:00:18 #startmeeting tsc 18:00:18 Meeting started Thu Feb 5 18:00:18 2015 UTC. The chair is colindixon. Information about MeetBot at http://ci.openstack.org/meetbot.html. 18:00:18 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 18:00:18 The meeting name has been set to 'tsc' 18:00:22 #topic roll call and agenda 18:00:38 #chair alagalah regXboi edwarnicke tbachman alagalah dfarrell07 18:00:38 Current chairs: alagalah colindixon dfarrell07 edwarnicke regXboi tbachman 18:00:39 #info alagalah for jmedved 18:00:43 #info colindixon 18:00:47 #info edwarnicke for fun 18:00:47 #info regXboi <- only IRC for now 18:00:54 #link https://wiki.opendaylight.org/view/TSC:Main#Agenda Today’s TSC agenda 18:00:54 #unchair regXboi 18:00:54 Current chairs: alagalah colindixon dfarrell07 edwarnicke tbachman 18:01:07 regXboi: is the anti-chair 18:01:08 #link https://wiki.opendaylight.org/view/TSC:Main#Agenda the agenda in it’s usual place 18:01:13 #info kwatsen 18:01:18 colindixon: I'm going to try and multiplex, but it means I'll be IRC only - making me chair makes no sense 18:01:19 #https://meetings.opendaylight.org/opendaylight-meeting/2015/tsc/opendaylight-meeting-tsc.2015-01-29-18.00.html Last week’s meetin gminutes 18:01:25 #info LuisGomez 18:01:26 #link https://meetings.opendaylight.org/opendaylight-meeting/2015/tsc/opendaylight-meeting-tsc.2015-01-29-18.00.html last week’s minutes for action items 18:01:39 #chair phrobb 18:01:39 Current chairs: alagalah colindixon dfarrell07 edwarnicke phrobb tbachman 18:01:44 * tbachman notes that regXboi either has a lot of meetings, or a few meetings that happen to be at the same time ;) 18:01:44 lol 18:01:49 #info Youcef Laribi 18:01:58 tbachman: this is an emergency things 18:02:01 #info dlenrow 18:02:01 ah 18:02:03 sorry to hear 18:02:08 and I do mean an things 18:02:08 #action colindixon to continue to follow AD-SAL deprecation between VTN, OVSDB, and controller (are ther others?) 18:02:16 #actoin colindixon and TSC to elaborate on proper procedure for removal of committers from a project 18:02:23 oops 18:02:27 #action gzhao colindixon to work out what the next Helium SR and the target dates 18:02:35 #undo 18:02:35 Removing item from minutes: 18:02:37 #undo 18:02:37 Removing item from minutes: 18:02:39 oops 18:02:52 #action colindixon to continue to follow AD-SAL deprecation between VTN, OVSDB, and controller (are ther others?) 18:03:01 #action colindixon and TSC to elaborate on proper procedure for removal of committers from a project 18:03:08 #action gzhao colindixon to work out what the next Helium SR and the target dates 18:03:16 meetbot no-likey actoin 18:04:12 colindixon: you have 7 18:04:32 #topic updates 18:04:55 #info phrobb says deadline for CFP is tomorrow, 2/6/2015 18:05:04 #info there are 20 submissions so far 18:05:18 * dfarrell07 will work on submission today, last second as always 18:05:25 dfarrell07: me too 18:05:31 tbachman: is rocking the minutes :) 18:05:32 #info phrobb says we’re locked down for the Marriott Santa Clara for April 15/16 for Lithium Milestone sync up 18:05:40 tbachman is epic 18:05:41 colindixon: dfarrell07: me three 18:05:45 colindixon: lol 18:05:46 #info abhijitkumbhare for Chris Price 18:05:46 lol 18:05:56 #info mohnish anumala 18:06:13 #info CFP is for the main conference tutorials 18:06:17 oops 18:06:19 #undo 18:06:19 Removing item from minutes: 18:06:28 #info edwarnicke asks how we’re doing on the tutorials 18:06:32 #info phrobb wasn’t sure about those yet 18:06:50 continue without me if you can function 18:07:21 sound of elevator music playing 18:07:43 and now back to our regularly scheduled programming 18:07:44 we’re at 8 18:08:07 ah, 10 18:08:08 lol 18:08:09 Yeah, I'm super snobby about not doing tourist stuff like bus tour (I'm not japanese). Understand why you would with a kid though... 18:08:17 tbachman: off by 2 error 18:08:20 lol 18:08:24 better than my usual ;) 18:08:50 I think that was just echo colindixon 18:09:00 #info chris wright 18:09:25 #topic Lithium and Stable Helium updates 18:09:34 #info zxiiro says we’re still in the process of tagging SR2 18:09:36 The TSC goes all the way to 11 18:09:47 now with cdub we’re at 11 out of 13 18:10:03 #info we’re missing 2 projects that just need to tag and 5 other projects that need to merge SR2 stats 18:10:06 #info gzhao says that we have 43 projects in Lithium, 3 offset 0, 16 in 1, the rest in offset 2 18:10:29 #info gzhao says 4 projects are in yellow status — two are missing questions SDNi and OpenContrail 18:10:52 #info SXP hasn’t set up JJB and hasn’t finalized release plan, and SNBI hasn’t finalized their release plan 18:10:54 Flakey WiFi so more like 10.5 18:11:14 #info gzhao says SNBI will finalize release plan on February 13th 18:11:18 #info colindixon asks if they are an offset 1 or 2 18:11:26 #info gzhao says SNBI is an offset 1 project 18:12:07 #action gzhao to look at dependency spreadsheet to see if SNBI has any dependencies, and communicate with SNBI about this 18:12:51 #link https://wiki.opendaylight.org/view/Simultaneous_Release:Lithium_Release_Plan#Offset_1_Projects <-- offset 1 status 18:13:41 #info gzhao says we have only one project — SXP — doesn’t support clustering 18:13:48 if anyone else on the TSC wants to comment on this, would it would be welcome 18:13:49 There are many things I would like to express, but most are highly inappropriate :) 18:14:09 #info gzhao says 4 projects has special requirements for CI, and 3 projects won’t use LF CI for testing 18:14:24 #info gzhao is working on the dependency diagram, as per action item from prior TSC 18:14:26 #action gzhao will follow up with the SNBI project to have them convey to the release staff and TSC any impact if any their delay will have on other projects and a plan to to get them back on track 18:14:31 #link https://wiki.opendaylight.org/view/Simultaneous_Release:Lithium_Release_Plan#Project_Dependency_Diagram < -- WIP dependency diagram 18:14:39 Thank you for all that great work, gzhao :) 18:14:48 dfarrell07: +1k 18:14:49 dfarrell07: thanks 18:15:18 gzhao: same 18:15:25 #info phrobb asks what does the TSC want to do with IRC meetings in the future — do we want to do them the week before the milestone? 18:16:57 #info colindixon says it’s useful to get folks together on IRC, asks if every other week, every week is best? 18:17:20 #info abhijitkumbhare asks for every week 18:17:51 #info LuisGomez likes the sync up 18:18:11 #info edwarnicke likes to have these kept on the google calendar, so whatever is decided, asks to have it updated there 18:18:24 #info colindixon recommends 1/2-hour meeting every week; if it seems excessive, we’ll modify 18:18:41 * ebrjohn laughing at imagining Ed "Wandering about..." 18:18:43 #info mohnish asks if we can slide it a bit later 18:19:09 #info colindixon asks if there are other opinions on the time change 18:19:44 ebrjohn: You imply that I am ever doing anything more defined than Wandering :) 18:19:49 * tykeal suddenly has the soundtrack from Les Mis running through his head "rise up!" 18:20:00 #info rovarga says they’d probably be okay with 1/2 slide (7:30 am PST) 18:20:01 * edwarnicke notes that as rovarga does not sleep, he may not be the best representative of his geography ;) 18:20:16 edwarnicke: I’ve always suspected that of rovarga 18:20:20 and you ;) 18:20:22 * tykeal figures that edwarnicke knows this since he doesn't sleep either 18:20:23 * rovarga notes edwarnicke has a twisted perception of reality :) 18:20:28 tykeal: is there anyway to get #barracades for meetbot and #storm ? 18:20:35 lol 18:20:49 #agreed The weekly IRC meeting for Lithium sync up will be 7:30 am PST 18:21:24 #topic System Integration and Test 18:21:42 #info LuisGomez reminds projects to move to SR2 tagging and version bumping, as this affects integration testing 18:21:58 register richardi 18:22:19 *cough* dlenrow change your password and do that in private ;) 18:22:20 * tbachman wonders if dlenrow is programming in mnemonic assmembly 18:22:45 I thought this was asm for dummies 18:22:51 #info colindixon asks if we can set this up as an opt-in or opt-out fashion 18:22:53 woops 18:23:05 dlenrow: lol 18:23:24 I am running SR2 for all of my work, any reason I shouldn't? 18:23:46 #info edwarnicke says the history is that people tend to opt-in 18:24:22 * edwarnicke thinks deep thoughts about libertarian paternalism 18:24:44 * tbachman is lost on all this history 18:24:59 #link http://en.wikipedia.org/wiki/Soft_paternalism 18:25:06 snoble: Always use the latest release, imho :) 18:25:29 #action colindixon to send email to allow projects to opt-out rather than opt-in for version bumping process 18:27:06 #undo 18:27:06 Removing item from minutes: 18:27:09 lol 18:27:21 got colindixon’s attention ;) 18:27:30 #action colindixon to coordinate figuring out how to set up opt-in/opt-out automatic version bump patches and tags 18:27:33 #undo 18:27:33 Removing item from minutes: 18:27:45 #action colindixon, zxiiro and others to coordinate figuring out how to set up opt-in/opt-out automatic version bump patches and tags 18:27:57 colindixon: thx! 18:28:31 #topic Infrastructure 18:28:48 #info zxiiro says that there are about 32 projects in JJB 18:28:58 #info zxiiro is working with hideyuki to get VTN at the moment 18:29:10 #info zxiiro is waiting to hear from the remaining projects 18:29:11 thanks zxiiro, this is hugely helpful 18:29:37 zxiiro helps me a lots around JJB! 18:30:05 #info tykeal says that odlforge is being affected by the addition of projects 18:30:07 #info VPN service project has resources, yay! We’re still waiting on MAPLE for IPR reasons 18:30:09 hideyuki: zxiiro helps everybody out around a lot of things :) 18:30:18 edwarnicke: +1k to that as well 18:30:27 edwarnicke: :) 18:30:46 #topic Creation Reviews 18:31:09 #link https://lists.opendaylight.org/pipermail/tsc/2015-February/002499.html PCMM would like permission to join the Lithium release late 18:31:47 #info colindixon prefers not to vote on this without any attendance of contributors or committers from the PCMM project 18:31:51 #info edwarnicke agrees 18:32:35 #info odl-patches.sh script now available in scripts/odl-patches in the releng/autorelease repo 18:32:43 #agreed Action item to review addition of PCMM to Lithium release is pushed out a week, due to no one from PCMM in attendance 18:32:54 #topic Helium Lifetime Poll results 18:33:22 zxiiro: colindixon who is actioned to send out the email to see if project leads have opinions about the patching of release stuff? 18:33:29 #info phrobb says that there was confusion in the poll about what the “current” release; in his mind, that would be what has been released, meaning “Helium" 18:33:32 edwarnicke: that’s me 18:33:43 edwarnicke: unless you want to do it :-) 18:33:56 colindixon: Cool :) You felt appropriate to me for that, but I'm starting to feel guilty about your AI count ;) 18:34:04 colindixon: Not that guilty :) 18:34:14 * colindixon is the Coordinator in Chief 18:34:32 * edwarnicke suppresses a large number of feline related comments 18:34:44 #info phrobb says there were a large quantity fo projects that didn’t respond, and sent a clarification of what the current release is to projects this a.m., and asks what they think they can support 18:35:02 To be clear my patch currently creates all the commits in all the repos giving a directory that contains the patches produced but autorelease-worker job. I still have to go in and "git push origin HEAD:refs/for/master" all the patches to Gerrit but that's an in progress automation step 18:36:03 audio went dark again for me… anyone else? 18:36:04 #info colindixon wants to schedule more stability releases for Helium until we get past the Lithium release, and has worked with phrobb to come up with dates that wouldn’t stall Lithium 18:36:10 phrobb: okay for me 18:36:23 phrobb: colorado snow again? ;) 18:36:31 zxiiro: Would you pushing in that manner cause the firing of the appropriate jobs in Jenkins for verify and merge? 18:36:32 #action phrobb will send out sample dates for later Helium SRs that try to avoid critical dates in Lithium 18:37:02 * tbachman worries when he can’t hear phrobb laugh 18:37:06 nice getting an action item while I couldn't defend myself :-D 18:37:08 edwarnicke: i'm going in through the standard interface that people submit patches today. A committer still needs to click teh "review" and "submit" button, and yes verify and merge still needs to be run 18:37:10 lol 18:37:29 zxiiro: I mean if projects were to opt into allowing releng to push the version bump stuff 18:37:30 edwarnicke: we're just asking for permission to allow us to be allowed to click the "review" and "submit" button on behalf of projects who are slow on doing it :) 18:37:42 zxiiro: Cool, that answers my question :) 18:37:47 #info rovarga says the decision to schedule the next SRs should have come before SR2 went out; when SR2 was frozen, they thought that was it for the stability branch, and now they have to look at whether they can backport patches from master 18:37:59 #info colindixon says that was part of the intent of the poll 18:38:16 #info colindixon asks if rovarga feels this is the wrong thing to do 18:38:24 #info rovarga advises to be more explicit next time around 18:38:40 edwarnicke: yep, so everything will work the same as it does today, i'm not bypassing any steps. We just want to help speed up the part that currently requires the project committers to click the buttons in Gerrit. 18:38:59 zxiiro: Cool, just wanted to get it clear in my head :) I appreciate your clarification :) 18:39:14 was that cdub? 18:39:19 tbachman: yes 18:39:23 #info cdub agrees with colindixon 18:39:25 lol 18:40:08 #info colindixon encourages PL’s to express opinions on the continued stable releases for Helium 18:41:37 #info rovarga prefers that the release after Lithium is out to become the official code freeze, and not to expect anything more from that service release 18:41:47 * regXboi thanks tbachman for the useful minutes as he is still IRC only 18:41:57 regXboi: my pleasure! 18:42:41 tbachman: gives gold star service http://www.katiyana.com/Logos/GoldStarService.jpg ;) 18:42:51 #info colindixon says that we’re not sure whether we’re in agreement on what the code freeze should be, based on the poll (and confusion around “Current”) 18:43:22 #topic Use Cases 18:43:57 #info colindixon says that there’s been a lot of talk in the Board meetings for ODL to meet Use Cases, rather than be just a collection of code 18:44:35 #info colindixon says mohnish, mlemay, mcohn, dmm, dlenrow, et. al. have been discussing this so that ODL can do something out of the box to meet a use case 18:45:05 #info colindixon says that even when there’s agreement on things that matter, such as neutron, in the past it’s been difficult to get resources to commit in order to move these things forward 18:45:20 colindixon: is there agreement on neutron? 18:45:27 #info The board is trying to find a way to encourage such participation, possibly through use cases 18:45:33 * regXboi doesn't think he understands what a "neutron" use case means 18:45:40 regXboi: I think this is more general 18:45:41 regXboi: +1 18:46:02 neutron is just a favorite ;) 18:46:02 I understand what service chaining/nfv *might* mean 18:46:10 neutron-on-top-of-what? :) 18:46:10 but neutron is too broad for a single use case 18:46:21 rovarga: +1 !!!! exactly !!!! 18:46:39 #info edwarnicke says using the distribution question to address the use case question is going about it wrong 18:46:57 honestly, what we need is cookbooks 18:47:00 #info colindixon says maybe we should declare that 1 or 2 use cases per release are things that we’d like to target 18:47:06 rovarga: Joy of Cooking? 18:47:19 * tbachman looks up recipe for “cookies" 18:47:20 tbachman: roasting.. various protocols :-D 18:47:27 tykeal: that one was for you 18:47:30 Julie Childs take on how to use ODL? 18:47:31 rovarga: lol 18:47:31 * edwarnicke has an excellent recipe for chocolate 18:47:46 edwarnicke: does it involve smuggled butter? 18:47:52 :-) 18:47:53 s/Julie/Julia/ 18:48:01 Java has julia childs "clean as you go" model 18:48:10 heh 18:48:16 #info edwarnicke says he feels we’re better off just having a single feature tied to a single use case, rather than a distribution question 18:48:33 #info colindixon says the question as to whether it’s a feature or a distribution is fine 18:48:53 well... the problem is that usually use cases need to be customized... which requires more of a hands-on tutorial of customization 18:48:55 #info colindixon asks if we can agree on some common goals for Lithium 18:49:20 #info edwarnicke says there are goals that we’ve targeted before, but haven’t been properly realized 18:49:41 #info edwarnicke actually said that for most common goals, we *have* pulled together in the past 18:49:47 #info but perhaps not at other times 18:49:51 #info RajeevK likes the idea of common goals, where ODL stands behind a few use cases; asks how we can get to those use cases 18:49:57 edwarnicke: thanks, and sorry for the mis-statement 18:50:41 #info edwarnicke says this goes to the question that people ask is how the TSC directs resources; notes the TSC can’t direct resources 18:51:08 #info colindixon says would creation of use cases encourage them to participate to make that happen 18:51:17 * regXboi wonders if the board understands that fact 18:51:32 regXboi: I routinely remind them of that 18:51:44 colindixon: that doesn't mean they *understand* that fact 18:51:57 :-( 18:52:09 #infor RajeevK asks what uses cases we would focus on, how many, and how would we track progress 18:52:16 colindixon: thx! 18:52:22 Speaking as the Integration committer who's driving (as of very recently) Neutron integration stuff, we 18:52:22 #info RajeevK asks what uses cases we would focus on, how many, and how would we track progress 18:52:31 echo? 18:52:32 we're already down for committing resources 18:52:44 ah 18:52:46 infor 18:52:59 + 18:53:25 Freedom != anarchy;. Collective good != fascism 18:54:34 the use cases are nice, but unless we get deployment scenarios ("How do I"), I don't think they are that useful 18:54:35 dlenrow: Fish != Grateful Dead 18:54:44 Jerry!!!!!! 18:54:48 :) 18:54:49 #info colindixon summarizes someone’s statement, says we should also track what we are doing, so that we drive advice and insight as to what ODL is good at beyond what we call attention to 18:54:51 * regXboi thinks he is missing discussion 18:54:58 Old and In the Way :) 18:55:08 what did you call me? 18:55:10 Would it make sense to in Beryllium start asking projects the use cases they are trying to solve and their deployment snares? 18:55:13 senarios 18:55:14 dlenrow: lolz 18:55:14 dlenrow: Way 18:55:15 :) 18:55:21 tbachman: someone=mohnish. :-) 18:55:25 mohnish: thx! 18:55:28 That preserves bottom up, but provides the visibility you are suggesting 18:55:30 colindixon: unfortunately it is quite hard to document something which has not been coded up 18:55:32 I couldn’t tell — you must be caller 9? 18:55:34 edwarnicke: The problem is that would be per-project, not controller wide (multiproject) 18:55:45 #undo 18:55:45 Removing item from minutes: 18:55:48 alagalah: I see no reason that projects couldn't collaborate on use cases 18:55:53 alagalah: In fact I think that's ideal :) 18:55:54 #info mohnish says we should also track what we are doing, so that we drive advice and insight as to what ODL is good at beyond what we call attention to 18:55:59 edwarnicke: Neither do I but that isn't what you said mate 18:56:02 Wouldn't use case automation and documentation be along the lines of a deliverable for the integration project? 18:56:12 * edwarnicke is not always good with his words :) 18:56:26 * alagalah I'm in a bit of a glass-house for wordification 18:56:35 #info colindixon suggests the first use case would be neutron with network virtualization, as there are already tests and CI infrastructure from openstack that would allow us to get that up and running quickly 18:56:56 so I have very strong opinions about that 18:57:01 and they aren't positive ones 18:57:04 colindixon: Is OPNFV at a state we can integrate with ? 18:57:05 #info edwarnicke asks if we can throw this out to the project leads 18:57:19 alagalah: what does that mean? 18:57:28 OPNFV - may be not yet 18:57:30 ShaunWackerly: To some extent, yes, I think so. I also think that's a nuanced question and would need a discussion to really answer it 18:57:43 alagalah: OPNFV isn't a thing, it's integration, and odl is part of the opnfv integration w/ openstack 18:57:46 * tykeal is pretty certain the OPNFV is not at a state for integration. They don't have build infra online yet even as they've been wrangling that 18:57:47 cdub_: Is the OPNFV project at say API freeze for arguements sake so it maybe a use case... maybe Be 18:57:55 #info colindixon agrees with edwarnicke on contacting the project leads 18:57:56 cdub_: I see, thanks 18:57:58 OPNFV will suck down LIthium and try to make it do something good 18:57:59 alagalah: We're working on that *right now* 18:58:02 alagalah: there's no API freeze 18:58:12 Thanks all :) 18:58:16 R1 pulls LIthium and integraTES 18:58:22 #info regXboi expresses doubts about the suggested first use case 18:58:40 we have a chicken and egg w/ users 18:58:50 cdub: are we back to cooking? 18:58:51 cdub_: Are you proposing bacon? 18:59:00 * edwarnicke mmm... delicious bacon 18:59:06 we can't clearly tell our users what we do, so it's hard to get user input 18:59:11 I wouldn't put OPNFV in the same bucket as Neutron and SFC ;) 18:59:17 tbachman: unintentionally! 18:59:26 lol 18:59:29 cdub: we do SDN (sarcasm) 18:59:37 #info I think that NFV is a critical use case for us in Lithium :) 18:59:39 NetVirt, SFC are good ones IMO 18:59:41 dlenrow: cool, i have a slide on that 18:59:49 * phrobb notes that cookbooks as gifts at the summit is now a contender… every recipe with bacon…. 18:59:57 phrobb: and cookies! 18:59:57 lol 19:00:00 * regXboi goes ugh 19:00:03 (SDN means anything to anybody) 19:00:06 #info bacon cookies, in fact 19:00:08 bacon infused cookies 19:00:09 cdub_: +1 19:00:10 cdub_: I suggested to the ironic team that the createed a project called sarcastic 19:00:16 phrobb: that was why the ugh 19:00:17 * tbachman had to info that last one 19:00:18 * edwarnicke *loves* the ODL cookbook swag idea 19:00:18 +1 to RajeevK (NetVirt & SFC) 19:00:21 dlenrow: haha 19:00:22 Bacon cookies are over-rated :( 19:00:26 * edwarnicke notes we need to also work in unicorns and rainbows 19:00:31 i worked on a project called sardonic 19:00:42 cdub_: how was the morale? 19:00:42 perfect 19:00:46 are we done? - i.e. can I run to another meeting? 19:00:56 regXboi: judging by this IRC, maybe 19:01:08 bacon cookies? http://www.joeydevilla.com/wordpress/wp-content/uploads/2007/12/bacon_chocolate_chip_cookies_with_maple_cinnamon_glaze.png 19:01:16 tbachman: I got my info in on the whole use case proposal 19:01:22 so it's in the minutes 19:01:34 colindixon: that was a while ago? 19:01:34 lol 19:01:36 Oh 19:01:38 sardonix actually 19:01:53 tbachman: it was bitter (project failed ;) 19:01:53 #action colindixon to start a discussion around possible use cases as focus areas for lithium on the mailing list 19:02:02 evolve the toaster to an oven and theen ... :) 19:02:09 colindixon: thx 19:02:13 beat me to it 19:02:20 rovarga: I bet we could make cookies in a toaster ;) 19:02:21 cdub_: lol 19:02:38 * rovarga is willing to review patches towards that :) 19:02:43 * tbachman wonders how these minutes will appear to the outside 19:02:55 SHOULD in RFC verbiage 19:03:02 tbachman: the minutes probably fine, the full logs though... 19:03:04 tbachman: hopefully delicuous :) 19:03:23 * tykeal notes he didn't stick any of the picture links in as #info lines 19:03:30 or #link lines for that matter 19:04:18 edwarnicke: that higher level structure is called system architecture and we need it badly 19:04:40 #endmeeting