17:00:29 <colindixon> #startmeeting tsc 17:00:29 <odl_meetbot> Meeting started Thu May 12 17:00:29 2016 UTC. The chair is colindixon. Information about MeetBot at http://ci.openstack.org/meetbot.html. 17:00:29 <odl_meetbot> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 17:00:29 <odl_meetbot> The meeting name has been set to 'tsc' 17:00:39 <colindixon> #topic agenda bashing and roll call 17:00:40 <colindixon> TSC members please #info in 17:00:44 <colindixon> #info colindixon 17:00:49 <colindixon> #link https://wiki.opendaylight.org/index.php?title=TSC:Main&oldid=46153#Agenda today's agenda 17:00:58 <ChrisPriceAB> #info Chris Price 17:01:00 <colindixon> #link https://meetings.opendaylight.org/opendaylight-meeting/2016/tsc/opendaylight-meeting-tsc.2016-05-05-17.00.html last week's meeting minutes 17:01:14 <gkaempfer> #info gkaempfer 17:02:05 <vishnoianil> #info vishnoianil 17:02:47 <edwarnicke> #info edwarnicke 17:03:02 <dfarrell07> #info Daniel Farrell 17:03:19 <abhijitkumbhare> #info abhijitkumbhare 17:03:29 <mohnish_> #info mohnish anumala 17:03:42 <colindixon> #action colindixon to test signing gerrit commits per tykeal's instructions 17:03:46 <colindixon> #action colindixon to try to find somebody to help with documenting the general procedure for the platform upgrade to Boron? 17:03:53 <colindixon> #action colindixon to drive the MPLS-TP creation reviews to conclusion 17:04:17 <colindixon> #link https://lists.opendaylight.org/pipermail/release/2016-May/006524.html colindixon asked if offset 0 projects could have stable features 17:04:47 <ttkacik_mobile> #info tony tkacik 17:04:50 <colindixon> #action dfarrell07 to drive forward tooling and planning for the next TSC election 17:05:10 <dfarrell07> ^^ack, working on it 17:05:42 <tykeal> no 17:06:00 <colindixon> #topic TSC mailing list discussions and votes 17:06:15 <colindixon> #link https://lists.opendaylight.org/pipermail/tsc/2016-March/004956.html MPLS-TP Service creation review 17:06:34 <zxiiro> colindixon: I'd like to talk about autorelease workflow changes but lets do it after the beryllium updates 17:06:42 <colindixon> #link https://lists.opendaylight.org/pipermail/tsc/2016-April/005033.html MPLS-TP Solution creation review 17:08:05 <tykeal> do not want to type that 17:08:15 <colindixon> #Info colindixon and vishnoianil say this might point toward eventually having nonsense words as project names 17:08:32 <zxiiro> your project will be assigned a number in which you've joined 17:08:38 <dfarrell07> zxiiro: lolol 17:08:41 <tykeal> hehe 17:09:39 <colindixon> #link https://lists.opendaylight.org/pipermail/tsc/2016-May/005149.html split-out projects and lifecycle states 17:09:44 <vishnoianil> generate uuid :P 17:09:57 <tykeal> we shall now start naming projects after different species of nudibranch ;) https://en.wikipedia.org/wiki/Nudibranch 17:10:21 <tykeal> hehe 17:10:49 <abhijitkumbhare> In my prev company we used to have internal product names from characters in Sesame Street :) 17:12:10 <colindixon> #info the consensus there seemed to be that split-out projects should go through a graduation review if they want to be mature, that should be copy and pasting if it's really just a mature component of the prior one 17:12:56 <colindixon> #info abhijitkumbhare points out that he'd like to see the rought age of the part being spinning out, that would help 17:13:50 <colindixon> #info ChrisPriceAB asks how committers on the new project should work 17:15:05 <colindixon> #info edwarnicke says his take is that people on the new project should be the people with a history of contributions on that part, and thus would be most comfortable maintaining it, that means they might be new committers, and might only contain a subset of the old ones 17:16:31 <colindixon> #info colindixon says in the past, he's just asked people to provide the principle and reasons behind thier committer list and then evaluate it from there 17:19:14 <colindixon> #action colindixon to try to collate the notes and expectations about split-out projects and record them somewhere if we have consensus 17:20:07 <colindixon> #Info vishnoianil asks if we could have a template for split-outs to make this easy to figure out if we need discussion or not 17:22:18 <colindixon> #info people note that depending what's going to change, they might have different feelings on if a split-out should be mature. colindixon notes if they weren't splitting out, this question likely wouldn't come up. 17:22:32 <abhijitkumbhare> +1 edwarnicke 17:23:04 * edwarnicke is good up to 10 dimensions ;) 17:23:06 <colindixon> #info ChrisPriceAB notes that the mature state is at least as much about the team and people as well as the code. colindixon agrees and notes these two axes are the source of a lot of the problems we have. 17:23:32 <colindixon> #info mohnish_ asks if it would thus make sense to wait a release, or if it can really be done immediately after the split-out 17:24:04 <colindixon> #Info abhijitkumbhare notes that his take is any project can ask for any review any time, the TSC can decline to approve it or not without needing hard rules about it 17:27:04 <colindixon> #info ChrisPriceAB notes that he things just saying that "if you are a mature project and you split-out a project, it may or may not be able to become a mature project based on TSC input" 17:27:22 <colindixon> #info edwarnicke adds he's happy with that as long as it ends with "...but it can have a combined creation review and graduation review" 17:27:24 <colindixon> #topic events 17:27:37 <colindixon> #link https://www.opendaylight.org/global-events 17:27:45 * edwarnicke favors using the tools we have were applicable ;) 17:28:20 <colindixon> #link https://lists.opendaylight.org/pipermail/tsc/2016-May/005201.html the hackfest will be 6/29-30 17:28:38 <colindixon> #link https://www.opendaylight.org/events/2016-09-27-000000-2016-09-29-000000/opendaylight-summit-2016 CFP for the ODL summit closes TODAY 17:28:53 <dfarrell07> #info dfarrell07 is at the OPNFV plugfest representing ODL, going well 17:29:10 <colindixon> #link https://www.opendaylight.org/events/2016-06-21-000000/opendaylight-european-mini-summit ODL mini-summit with the OPNFV summit 6/21 17:29:21 <colindixon> #topic boron 17:29:36 <anipbu> #info ODLPARENT team is planning to merge the patch for upgrading Karaf to 3.0.6 this week or soon. 17:29:43 <anipbu> #link https://git.opendaylight.org/gerrit/#/c/26323/ <---- Patch to be merged for Karaf 3.0.6 17:29:54 <anipbu> #info The Boron Autorelease job failed to build aaa-authz-service. We are currently working with their team to identify the root cause. 17:29:58 <anipbu> #link https://lists.opendaylight.org/pipermail/aaa-dev/2016-May/000919.html <--- Email Thread discussing AAA Boron Autorelease Breaking 17:30:02 <anipbu> #link https://jenkins.opendaylight.org/releng/view/autorelease/job/autorelease-release-boron/119/org.opendaylight.aaa$aaa-authz-service/console <--- AAA Boron Autorelease Breaking 17:30:41 <anipbu> #info If want to have stable features and stable distribution, we will need offset 0 projects to graduate to mature and satisfy other stable feature requirements. 17:30:45 <colindixon> #info anipbu asks when we need a drop-dead date for when we need to have a stable distribution 17:30:46 <anipbu> #link https://lists.opendaylight.org/pipermail/release/2016-May/006524.html <---- Colin's Email regarding Stables Feature in Offset 0 17:30:51 <anipbu> #link https://wiki.opendaylight.org/view/Simultaneous_Release:Boron_Release_Plan#Stable_and_Extended_Features <---- Boron Stable Feature Requirements 17:33:19 <colindixon> #info colindixon asks what a reasonable line in the sand would be for when we would need stable features to meaningfully have a stable distribution 17:33:39 <colindixon> #info vratko says there are lots of tests: longevity, cluster, scale, etc. 17:34:58 <colindixon> #Info colindixon points out that he things the openflow plugin testing covers a lot of the offset 0 projects, e.g., RESTONF, MD-SAL, YANG tools, etc. 17:35:08 <colindixon> #info vratko points out the config subsytem probably isn't there 17:35:26 <colindixon> #info LuisGomez says the integration team should look at this and prepare a set of gaps 17:35:59 <colindixon> #info rovarga says his main concern is that the APIs have enough leeway to change behavior if need be, so that they can maintain it 17:38:16 <colindixon> #action rovarga to follow up with an e-mail about his concerns about the stable feature requirements 17:38:23 <LuisGomez> we have kernel projects representation in the integration group as well as understanding on the tests performed by all projects so this is the perfect place to start looking at stable feature gaps 17:38:32 <colindixon> #info anipbu asks if RC0 woudl be too late, colindixon says he things M5 is likely as late as we could go 17:39:02 <colindixon> #info dfarrell07 asks if vratko can give his opinion, vratko says M3 offset 2 is the curent line in the sand, but if we want to push it out, we could 17:39:35 <colindixon> #info vratko says M3 would make him happier, but M5 would be doable 17:40:24 <colindixon> #action LuisGomez and Vratko to check with the integration team about finding the gaps we have in our current testing 17:41:35 <colindixon> #topic boron documentation review subgroup 17:41:53 <colindixon> #info colindixon would love help reading boron docs for sanity, asks for volunteers 17:42:02 <colindixon> #info abhijitkumbhare and ChrisPriceAB speak up 17:42:16 <colindixon> #action colindixon to send a mail about a documentation review subgroup 17:42:21 <colindixon> #Info dfarrell07 also speaks up 17:42:46 <colindixon> #Info it should be noted that dfarrell07 and ChrisPriceAB say that they might not have many cycles 17:44:00 <abhijitkumbhare> I will not have much time in June-July (will be in traveling then) 17:44:36 <ChrisPriceAB> If you want help with RST and tooling we have a full toolchain in OPNFV that works pretty well, can help out. 17:44:54 <ChrisPriceAB> &me knows an RST toolchain ninja or two 17:46:36 <ChrisPriceAB> it's also a common method across open source projects and easy for onboarding... 17:46:56 <colindixon> #action colindixon to send e-mail about possibly docuemntation moving readthedocs, sphinx, and ReStructuredText, as opposed to using asciidoctor, and asciidoc which have caused us some pain 17:52:23 <colindixon> #action colindixon to schedule a TWS on documentation structure, toolchains and needs 17:53:03 <jamoluhrsen> colindixon: sys/int/test update is that we are removing all CSIT jobs from distribution-test (thats the job that gates our release) unless the csit jobs meet our system test expectations. this will allow for consumers to know (more easily) if the distribution is up to snuff or not. risk is we lose test coverage, but it's too unwieldy to continue as is. LuisGomez can comment further if needed. 17:53:09 <jamoluhrsen> colindixon: logging off. 17:53:13 <colindixon> thanks 17:54:00 <colindixon> #info rovaraga notes that his worry is less conversion than training developers and maintaining docs 17:55:05 <colindixon> #Info colindixon notes the there is a PoC (from zxiiro) which has the getting started guide into the new format, and ane examle that pulls in the releng docs as well 17:55:10 <ChrisPriceAB> colindixon; need to drop off. 17:55:16 <colindixon> ChrisPriceAB: noted thanks 17:55:19 <colindixon> jamoluhrsen: thanks 17:55:52 <colindixon> #info ttkacik_mobile notes that we need to integration with javadocs no matter what we do 17:56:57 <colindixon> #topic stable/beryllium 17:57:00 <anipbu> #info Beryllium SR2 has been approved to release: https://lists.opendaylight.org/pipermail/tsc/2016-May/005225.html 17:57:01 <anipbu> #link https://wiki.opendaylight.org/view/Simultaneous_Release:Beryllium_Release_Plan#Beryllium_SR2_Download <--- Download Beryllium SR2 at this Link 17:57:05 <anipbu> #link https://wiki.opendaylight.org/view/Simultaneous_Release/Beryllium/SR2/Status <--- Beryllium SR2 Status 17:57:08 <zxiiro> #link https://lists.opendaylight.org/pipermail/release/2016-April/006299.html Autorelease Workflow Idea 17:58:46 <colindixon> #topic system integration and test 17:58:56 <colindixon> #info sys/int/test update is that we are removing all CSIT jobs from distribution-test (thats the job that gates our release) unless the csit jobs meet our system test expectations. this will allow for consumers to know (more easily) if the distribution is up to snuff or not. risk is we lose test coverage, but it's too unwieldy to continue as is. LuisGomez can comment further if needed. 18:00:08 <colindixon> #info this is basically to have a set of tests that passes with high probability without random failures, so that we can not have to have projects sign off on random test failrues every time we run a test 18:00:29 <colindixon> #action jamoluhrsen to send mail about this moving of non-compliant tests out of distribution-test 18:00:59 <colindixon> #topic infrastructure 18:01:34 <colindixon> #info trying to move sandbox to the private cloud, waiting on rackspace 18:01:39 <colindixon> #topic TSC chair elections 18:01:45 <colindixon> #link https://lists.opendaylight.org/pipermail/tsc/2016-May/005242.html call for nominations here 18:02:14 <colindixon> #topic cookies 18:02:17 <colindixon> #endmeeting