16:59:26 #startmeeting tsc 16:59:26 Meeting started Thu Oct 30 16:59:26 2014 UTC. The chair is colindixon. Information about MeetBot at http://ci.openstack.org/meetbot.html. 16:59:26 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 16:59:26 The meeting name has been set to 'tsc' 16:59:34 #topic agenda bashing and roll call 16:59:36 #info colindixon 16:59:53 #chair regXboi 16:59:53 Current chairs: colindixon regXboi 17:00:03 #info Kent Watsen 17:00:38 #info regXboi 17:00:49 colindixon: did you #link the agenda? 17:01:02 #info Chris Price 17:01:09 #link https://wiki.opendaylight.org/view/TSC:Main#Agenda 17:01:12 * ChrisPriceAB can scribe a bit too 17:01:18 #chair ChrisPriceAB 17:01:18 Current chairs: ChrisPriceAB colindixon regXboi 17:01:30 wow a volunteer! sign him up! sign him up! 17:01:46 * ChrisPriceAB haha 17:02:07 * ChrisPriceAB prepare for some good old fashioned "Australian English" 17:02:36 cdub: are you around? 17:02:42 #info Luis Gomez 17:03:13 having some trouble getting onto webex 17:03:19 #info mohnish 17:04:37 colindixon: yes, but can't join audio 17:04:46 #info edwarnicke 17:06:03 #info jmedved 17:06:31 #info it appears that action items (with the exception of master being stuck) have been addressed 17:06:42 #action gzhao and edwarnicke to continue to work on getting master (Lithium) auto-release back unstuck 17:07:00 #link https://meetings.opendaylight.org/opendaylight-meeting/2014/tsc/opendaylight-meeting-tsc.2014-10-23-16.59.html old action items 17:07:23 audio is old school :) 17:07:24 ChrisPriceAB, cdub: do you guys have audio yet 17:07:38 no, having some trouble with our webex still. soon to be there I hope 17:08:43 #topic new TSC members 17:09:01 colindixon: i'm going to pass to flaviof 17:09:03 welcome Luis! 17:09:08 Welcome LuisGomez ! :) 17:09:08 welcome luis 17:09:14 welcome luis 17:09:15 #info regXboi welcomes luis 17:09:17 #info welcome LuisGomez to the TSC 17:09:26 thanks 17:09:30 welcome luis 17:10:02 #info LuisGomez is replacing dmm as brocade's platinum member designate 17:10:57 #topic Updates 17:11:05 #info no updates on events 17:11:34 #info updates for system integration/testing and stable/helium both from Luis 17:12:29 #info plan to have proposal for integration/testing requirements my next TSC call 17:12:50 #info primary pain point continues to be infrastructure 17:13:34 #info all integration pieces for stable/helium are now complete - continuous testing of stable/helium is now ongoing 17:13:37 * regXboi cheers 17:14:20 #info there is a RC for stable/helium .... status hopefully forthcoming in a linked email? 17:14:21 #link https://jenkins.opendaylight.org/odlautorelease/job/autorelease-helium-worker/ the job building helium stable pre-release 17:14:32 colindixon: thinks 17:14:36 er thanks :) 17:14:40 * regXboi doesn't think :) 17:16:29 #action gzhao/luis to send mail to discuss that the last day to have a patch merged to stable/helium is sunday 11/2. 17:16:36 #link https://docs.google.com/spreadsheets/d/1PYxjiSYEks44uJByVO1P44rnI5xTJRulpKyrSsDQF9g/edit#gid=1307630876 17:16:45 #info project status 17:16:55 #link https://lists.opendaylight.org/pipermail/release/2014-October/000777.html the mail saying when the final artifacts will be cut 17:16:56 #undo 17:16:56 Removing item from minutes: 17:17:00 ouch 17:17:05 #link https://lists.opendaylight.org/pipermail/release/2014-October/000777.html the mail saying when the final artifacts will be cut 17:18:32 #action gzhao to send email to release/discuss that stable/helium RC will be cut on monday and projects should test it and report as TSC will vote on that RC on next thursday's meeting 17:19:20 https://wiki.opendaylight.org/view/Simultaneous_Release:Cutting_Stability_Branches 17:19:32 #info question on what the completed through tab of project status means 17:19:42 lost audio 17:19:50 #info answer this how far through the cutting stabilty branches process 17:20:10 #link https://wiki.opendaylight.org/view/Simultaneous_Release:Cutting_Stability_Branches cutting stability branches process 17:20:23 yo: keith! 17:20:39 #chair alagalah 17:20:39 Current chairs: ChrisPriceAB alagalah colindixon regXboi 17:20:55 regXboi: yeah... thanks 17:21:00 #link https://wiki.opendaylight.org/view/Simultaneous_Release:Cutting_Stability_Branches 17:21:11 #undo 17:21:11 Removing item from minutes: 17:21:15 #undo 17:21:42 #action colindixon to talk to LF marketing around co-ordinating any PR around stable release (in a week) 17:22:04 #info no Creation Reviews and no Committer Promotions 17:22:34 #topic ovsdb splitting in two 17:23:01 do we need a #link? 17:23:06 of the email? 17:23:25 #info Srini describes that the OVSDB project contains two components, north and southbound today. 17:24:14 #info There is an ongoing discussion about splitting the project into two parts, he proposes a project called OVSDB plugin and openstack 17:24:20 would this require a scope (re-) definition? 17:24:22 * ChrisPriceAB I think 17:24:52 #info it is not clear that this discussion has been public ... i.e. we're still looking for public artifacts 17:25:06 #info Colindixon states that any project can split any time prior to M3 of a release plan. Pending creation reviews of the proprosed projects 17:25:21 ChrisPriceAB: I assume that scope definition would be part of that 17:25:37 * ChrisPriceAB yep, as part of the proposals to be approved 17:25:56 * ChrisPriceAB missed ed's statements. 17:26:26 ChrisPriceAB, alagalah: the current draft Lithium release plan requires that the split off projects to pass creation review normally, and thus would have scope discussion 17:26:55 Thread: https://lists.opendaylight.org/pipermail/ovsdb-dev/2014-October/000750.html 17:27:12 #info there is a patch planned for the openflowplugin project to include the "Nicira extentions" currently in OVSDB. 17:27:30 #link https://lists.opendaylight.org/pipermail/ovsdb-dev/2014-October/000750.html ovsdb split proposal thread 17:27:48 * ChrisPriceAB thanks regxboi 17:28:41 colindixon: Might it make sense to have a very *short* standing 'FYI' item on the agenda for folks to bring up things like this that aren't quite TSC business (yet) but folks think the TSC should know about 17:28:58 edwarnicke: yes, it does 17:29:56 #action colindixon to make the possibitliy of adding new topics for agenda bashing more explicit on the TSC agenda 17:30:03 #topic OPNFV 17:30:12 #link https://www.opnfv.org/ 17:30:53 #link https://wiki.opnfv.org/ the wiki (analogous to our OpenDayilght wiki) 17:33:14 #info OPNFV is about 1.5 months old and plans to work with “upstream” projects, e.g., in OpenStack and OpenDaylight to define an Open Platform for NFV 17:33:31 #info the intent is that most (maybe all) code will go to upstream projects 17:33:49 #info similar to ODL, there are open TSC meetings and feel free to join in 17:34:30 this is by far the quietest TSC meeting with edwarnicke on the call that I can remember 17:34:51 wow 17:34:56 did you really just type that? 17:35:03 ChrisPriceAB: Any tie in between opnfv.org and any ONF WGs ? Do you guys talk ?? Espec. re standards etc ? 17:36:00 ChrisPriceAB: ha ha 17:36:03 * alagalah Aussie joke 17:36:55 ChrisPriceAB: Thanks mate... 17:38:52 Thanks colindixon, appreciate the opportunity to spread the word. ;) 17:40:08 Some might beg to differ edwarkicke! 17:41:09 #topic infrastructure update 17:41:19 #info there are templates in the releng/builder repo 17:41:40 #info projects that want want to try it out should contact Thanh 17:42:09 #action colindixon to add infrastucture to the list of deafualt update 17:42:18 #info the directories contain a readme that will allow early adopters to spawn a jenkins docker image and get started with jenkins builder 17:42:31 What is Thanh's IRC nick ? 17:42:50 alagalah: zxiiro 17:42:53 alagalah: zxiiro 17:43:47 ty 17:43:47 zxiiro: We need docker support in Jenkins etc in GBP so count us in as guinea pigs 17:44:31 zxiiro: 17:45:01 zxiiro: TTP is also a decent project, very simple 17:45:30 colindixon: asks if projects can be both in their jenkins silo and the new unified silo at the same time 17:45:50 #info colindixon asks if projects can be both in their jenkins silo and the new unified silo at the same time 17:46:55 #info the answer is that should be possible with a few possible quirks, for instance two jobs pushing to nexus from two places and two jobs voting in gerrit 17:47:45 #info colindixon volunteers TTP and alagalah volunteers GBP to participate 17:48:00 #info a question asked: "is there a target date for when projects that want to move should have moved?" 17:48:12 #info answer: no target date set as of now 17:49:07 #info skipping progress on technical debt/integration to ... 17:49:14 #topic lithium release plan 17:49:30 alagalah: to be clear the docker image i mentioned is to start a local jenkins instance so that you can test Job builder against it. Not a jenkins instance that runs docker 17:49:43 zxiiro: Aaaahhhhh 17:49:50 zxiiro: Thanks for the clarification, mate 17:49:52 #link https://wiki.opendaylight.org/view/Simultaneous_Release:DRAFT_Lithium_Release_Plan_ckd 17:50:09 changes: https://wiki.opendaylight.org/index.php?title=Simultaneous_Release%3ADRAFT_Lithium_Release_Plan_ckd&diff=22026&oldid=21676 17:50:14 #info previous link is draft lithium release plan 17:50:33 #link https://wiki.opendaylight.org/index.php?title=Simultaneous_Release%3ADRAFT_Lithium_Release_Plan_ckd&diff=22026&oldid=21676 changes to draft plan 17:50:35 Does anyone mind if I change the main wiki page to link in https://wiki.opendaylight.org/view/Simultaneous_Release:DRAFT_Lithium_Release_Plan_ckd instead of the older one that is there currently? 17:50:54 edwarnicke: I'm ok with that 17:52:58 #info colindixon noticed that the last date for project joining is a friday, which doesn't quite line up with the thursday TSC schedule 17:54:07 * regXboi wonders if the TSC has hit the halloween candy early and is now on a sugar crash :) 17:59:29 yeah 18:00:29 #info discussion of API declaration at M1 between colindixon and edwarnicke 18:00:29 #undo 18:00:29 Removing item from minutes: 18:00:29 #info discussion of API declaration at M2 between colindixon and edwarnicke 18:02:39 #action colindixon and edwarnicke to craft language around API declaration at M2 to achieve something useful but not overbearing 18:04:26 #link https://wiki.opendaylight.org/view/Simultaneous_Release:DRAFT_Release_Plan_2014_Template the per-project release plan template 18:04:33 #info question of what the default state should be. colindixon feels that if all APIs are enunciated, default status is less important 18:07:25 #info point that the release plan template now includes a section to be filled out by other teams that see a dependency 18:10:51 #info edwarnicke asks if we should make it possible for external projects, e.g., OPNFV to make requests of projects as part of negotiation 18:10:52 I think this is wrong 18:11:10 alagalah: please speak up? 18:11:12 External folks should be doing this via the Project Lead, and the team updates the Release plan 18:11:16 I can't, no mike 18:11:55 alagalah: interesting thought, so is there a reason not to allow that to be documented in the same way as internal project requests? 18:11:59 I’m honestly not sure 18:12:11 Unless I am misunderstanding, a project's release plan should only be udpated by members of the project 18:12:23 What we are talking about is the flow of information. 18:12:26 dlenrow: asks if there should be an opnfv project to help with this 18:12:27 ah 18:12:29 I see 18:12:52 my thought was to only have one section that others opened to start and log a negotiatoin 18:12:54 Otherwise someone update's it, you as the project lead don't realise... and a 3rd party reads it and takes it as gospel 18:12:59 similar to the API execptoin project 18:13:04 process 18:13:05 Ok.... I was perhaps misunderstanding 18:15:02 #info dlenrow says his intent is to try to get OPNFV to create a project in ODL to manage this 18:15:04 huzzah @ upstream first 18:15:20 #info uri and edwarnicke point out that they really want to push the code into ODL directly 18:15:58 alagalah: did you follow the API exception process in Helium? 18:17:05 I missed the Request part... 18:17:17 Right, just creating a wiki-thing === danger 18:17:30 BTW I am mike-less ... 18:18:27 #action colindixon to add a note that you need to send a mail along with putting the request on the providing project’s release plan for API negotiation 18:18:45 #info abhijitkumbhare asks if we should use bugzilla to track these things 18:19:14 #info edwarnicke and colindixon say there should probabably be a way to link to a bug representing the request 18:21:55 #info next subtopic: should there be requirements sonar code coverage/quality beyond reporting 18:23:46 #info Jan represents his and robert's view that we should be having a requirement to ensure overall code coverage/quality 18:24:06 * colindixon picks up the pen 18:24:56 #info regXboi asks (1) what about non-java code? (2) it seems to make senes that we report it in Lithium with the express intent of requiring coverage and lack of findbugs Beryllium 18:25:20 #info colindixon agrees with everything Jan said (as does regXboi) 18:27:05 #info colindixon also notes that blind reaching for code coverage can do more harm than good 18:27:41 #info edwarnicke notes that he likes the principle of reporting first before requiring 18:29:09 #info edwarnicke also notes that code coverage is a double edge sword, all it does is tell you that some measured behavior of the code hasn’t changed, it can actually bake in bugs rather than find them 18:31:48 #info regXboi proposes the idea of encouraging the key infrastructure projects to voluntairly try to hit certain code coverage and findbugs in their projects 18:32:08 #Info jmedved, colindixon, and edwarnicke all say this sounds like the right approach 18:33:12 #info jmedved, colindixon, and edwarnicke all say this sounds like the right approach (apologies for double if that happens) 18:33:50 #info regXboi also adds that we should note that doing this will be viewed positively when it comes to moving along the lifecycle phases 18:34:40 #info flaviof asks if there’s a way to codify good logging processes as part of this and make that something we can encourage 18:35:26 #link https://wiki.opendaylight.org/view/Logging_Best_Practices 18:36:10 * regXboi raises hand 18:36:20 regXboi: you’re next 18:37:02 #info colindixon proposes maybe opening logging bugs, edwarnicke extends this to having a logging component in all projects so that we can track how bad the logging problem is as the bugs come in 18:37:23 edwarnicke; +1 18:37:28 #info regXboi asks if we start to tighten up the logging practices release, by release 18:41:09 #action colindixon to work with gzhao to try to get project to create logging components in each project 18:48:44 #info a debate about whether all the the features need to be KDC (known, doable, confirmable) before the release starts 18:48:48 Apologies, gotta drop 18:48:58 #info the key point seems to be around documentation requirements 18:52:29 #info edwarnicke is of the opinion that if it’s not KDC at the start of the release, it shouldn’t be required as part of this release plan 18:52:48 #info colindixon says that can be carried to an extreme where the release plan doesn’t ever start 18:53:41 #action colindixon to reach out to integration and documentation to write requirements ASAP 18:57:26 #info edwarnicke also wants to see a full 4 weeks between when we approve the release plan and M1 19:00:32 #info edwarnicke wants to see all the requirements explained clearly *before* we approve the release plan, and points out that the TSC has been bad at meeting deadlines to provide thing in the future 19:00:52 #action regXboi to create a draft version of documentation requirements for Lithium 19:01:53 ty all. evening from europe 19:03:22 #action colindixon to make sure we make it clear when the developer pre-milestone meeting 19:03:26 #endmeeting