18:00:29 #startmeeting tsc 18:00:29 Meeting started Thu Nov 5 18:00:29 2015 UTC. The chair is colindixon. Information about MeetBot at http://ci.openstack.org/meetbot.html. 18:00:29 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 18:00:29 The meeting name has been set to 'tsc' 18:00:35 #topic roll call and agenda bashing 18:00:36 #info colindixon 18:00:51 #link https://wiki.opendaylight.org/index.php?title=TSC:Main&oldid=38148#Agenda agenda in it's usaul place 18:01:37 #link https://meetings.opendaylight.org/opendaylight-meeting/2015/tsc/opendaylight-meeting-tsc.2015-10-22-17.00.html last week's minutes 18:02:01 #action colindixon to try to find somebody to help with documenting the general procedure for the platform upgrade from Helium to Lithium (for SR3) https://bugs.opendaylight.org/show_bug.cgi?id=3160 18:02:07 #info jmedved 18:02:28 #action ttcacik and zxiiro to work on getting feature-level code coverage numbers 18:03:05 #info dlenrow 18:03:16 #link https://bugs.opendaylight.org/show_bug.cgi?id=4519 the bug to get dsbenchmark out of coretutorials by offset 2 M4 18:03:19 #info LuisGomez 18:03:30 #action jmedved to talk to ttkacik about how to get the benchmarks into controller by offset 2 M4 (tracked by https://bugs.opendaylight.org/show_bug.cgi?id=4519) 18:04:00 #info mohnish anumala 18:04:38 #action rovarga, alagalah, mohnish, ttkacik, vishnoianil, ebrjohn, RPenno, abhijitkumbhare will try to draft a paragraph around the best practices including this discussion and feedback (to be gotten) from PTLS, they should also feel free to provide feedback around tools and technical solutions to help with upstream/downstream breakages 18:05:29 we have colindixon jmedved dlenrow LuisGomez mohnish_ edwarnicke which is quorum 18:05:42 #info edwarnicke 18:06:04 and the peasants rejoiced! 18:06:22 this overworked admin sure looks forward to an extra hour back ;) 18:06:26 #topic events 18:06:36 #chair phrobb_ anipbu phrobb dfarrell07 18:06:36 Current chairs: anipbu colindixon dfarrell07 phrobb phrobb_ 18:06:42 #info Daniel Farrell 18:06:57 #link https://www.opendaylight.org/global-events upcoming events 18:07:14 #info hackfest on Monday/Tuesday of next week, register if you haven't 18:07:28 #link https://wiki.opendaylight.org/view/Events/Beryllium_HackFest#Topics please add your attendance and/or topics here 18:07:59 #link https://www.regonline.com/Register/Checkin.aspx?EventID=1761135 register here if you haven't (it's free, but might run out of space) 18:08:29 #info phrobb says that there will be obviously some unconfernece topics too 18:08:46 #info there will be some remote participation, but phrobb suggests that we maybe use hangouts for smaller groups as it might work better 18:11:55 #info colindixon and phrobb give a summary of OpenStack tokyo, we had no talks, there was a huge attendance at the demo we gave, we should aim to have more talks in Austin 18:13:29 #topic TSC mailing list votes and discussions 18:13:45 #link https://lists.opendaylight.org/pipermail/tsc/2015-October/004026.html at-large election bylaw waiver discussion 18:14:13 #link https://lists.opendaylight.org/pipermail/tsc/2015-October/004021.html requirement to be in the release 18:14:49 #link https://lists.opendaylight.org/pipermail/tsc/2015-November/004098.html meeting next week despite the OPNFV summit 18:15:31 Get in a room @ venue for call? 18:16:00 #action phrobb to try to find a room for the TSC call under the assumption it's still happen 18:16:48 #topic TSC at-large elections 18:17:59 #info phrobb says that the discussion of how to avoid needing a permament exception to leave platinum designates in place has stalled a bit, and we still need to close on that 18:20:32 #info dfarrell07 says he wished that there were experts that could help us with governance and election design 18:20:57 #info colindixon and edwarnicke say that desiging good governance is hard and a contenttious/open problem 18:21:09 #topic beryllium 18:21:37 #info we're still wating on 5 projects to get M3 updates, they've been notified 18:21:57 #info we have several projects with action items that are being tracked 18:22:15 #info we have 8 projects that aren't in integration/distribution, they have also been notified 18:23:03 #info beryllium autorelease is currently blocking on NetIDE using an old version of config-parent, it will be adressed soon or they'll be dropped, also being blocked by dsbenchmark which is planned to be moved by offset 2 M4 18:23:33 https://docs.google.com/spreadsheets/d/1Kfp5HVQGydNegEjp6dD2V3XsUnqpice0bysWbdxABA4/edit#gid=1817245674 18:23:42 #info dfarrell07 asks if we can keep stats about missed deadlines for fugture needs 18:24:11 https://docs.google.com/spreadsheets/d/1KPpO9LH539Vlcoa4RvLa6PPCdLifi5JD-ihRhlybqeo/edit#gid=1241552017 18:25:30 #info colindixon and anipbu say we track how late milestone readouts are, but don't track deep on timeness of internal deliverables 18:25:36 Writing script to game the timeliness metric... 18:26:52 #info anipbu says he'll track when a project is 100% complete with a milestone starting with M4, to get better information 18:27:14 +1 to edwarnicke about anipbu 18:27:22 #info edwarnicke says this is incredibly helpful and he can't overstate how important what anipbu is doing and how it's doing 18:27:41 #info colindixon, abhijitkumbhare, dfarrell07 (and presumably many ohters) offer hearty +1s 18:28:06 #link https://wiki.opendaylight.org/view/Simultaneous_Release:Beryllium_Release_Plan#Project_Status the projecct tracking sheet is linked from teh main release plan here 18:28:28 #link https://docs.google.com/spreadsheets/d/1Kfp5HVQGydNegEjp6dD2V3XsUnqpice0bysWbdxABA4/edit#gid=1862710416 the actual sheet with the auto-computed milestone completely summary 18:28:54 https://docs.google.com/spreadsheets/d/1hci5TMUPyB6PX8Al-fwfVqvs5SQVa2wZLja_7rKWN6o/edit#gid=2030147923 18:29:27 #link https://wiki.opendaylight.org/view/Weather#Beryllium_Blocking_Bugs cross-project issues and blocking bugs are tracked in this sheet (linked from weather page) 18:29:54 #link https://wiki.opendaylight.org/view/Weather 18:30:23 #info we tried to merge the jersey/jaxrs patch last week, it turned to break a lot of people and we've reverted it 18:30:42 #info the structural container patch was merged, there was a controller bug, but it was fixed and remains in 18:38:23 #link https://docs.google.com/spreadsheets/d/1hci5TMUPyB6PX8Al-fwfVqvs5SQVa2wZLja_7rKWN6o/edit#gid=1368787047 <-- JDK8 tracking 18:39:02 #Info we're working on getting JDK8 to work at least for building and verify jobs, there are patches for all be 2-3 projects that get them to build with JDK8 onmaster, 2-3 of them seem to fail with non-JDK8 issues on master 18:39:11 #info anipbu has been tracking this and tracking it well 18:39:48 colindixon: quick question: this should work, right? log:set trace org.opendaylight.l2switch.* 18:39:49 #info it remains to be seen if we can get JDK8 and JDK7 autorelease to build for autorelease 18:41:28 #Info rovarga says his goal is to have Beryllium be compatible with both JDK7 adn JDK8 and then move to JDK8-only in Boron, which seems within grasp 18:41:50 #info rovarga says that's in part because YANG Tools would like to be able to use Jave 8 only features for YANG 1.1 java binding 18:42:07 #topic stable/lithium 18:42:18 #info the next SR is in two weeks with a cutoff of 11/15 for patches 18:42:36 #topic system integration and test 18:43:09 #info jamoluhrsen says that there are missing system test templates that they're still waiting on 18:43:34 #topic infrastructure 18:44:08 #info we had reboots of infra that happened last week due to rackspace host upgrated, caused angst on master jenkins, but it's fixed now 18:45:13 #info we now have more than 1000 jobs in a jenkins master (tykeal notes that OpenStack has more than 3000), restarts to jenkins now take a long time, just from starting jenkins to consider processing jobs takes 30 minutes, in total it takes 1-2 hours for jenkins restarts 18:45:59 #info some of these issues will go away as we move to nodepool and xuul 18:46:18 #link https://lists.opendaylight.org/pipermail/dev/2015-October/001325.html GPG keysigning info at the Beryllium Hackfest 18:47:11 #topic DIDM feature names 18:47:34 #link https://lists.opendaylight.org/pipermail/tsc/2015-October/003946.html mail thread 18:48:30 The good news is that HP no longer exists 18:48:41 dlenrow: :) 18:48:50 HPE :) 18:49:01 #link https://github.com/opendaylight/openflowplugin/search?utf8=%E2%9C%93&q=nicira it's worth noting that we have other features, the nicira extensions here 18:49:43 #info phrobb says that this isn't a legal question, because it will fall under fair use, but we have avoided having trademark names in project names from business reasons, thus would also logically apply here 18:49:47 This is a vendor specific device identifier/translator project no? 18:51:25 #info the consensus seems to be tending toward the naming is fine as long as it's really about a vendor product, not naming things that would would be confusing 18:54:38 #info cdub says that the general best practice is to use obvious name without obfuscation for vague business reasons, and it's different between being a low-level (driver-level) than a higher-level API or feature 18:54:47 +1 what cdub is saying 18:55:04 That matches my recollection of best practice in other multi-project foundations that allow vendor specific drivers 18:55:06 #info edwarnicke, colindixon, and dlenrow all say +1 18:55:23 +1 to cdub 18:56:48 (cdub and I are the same vote, for the sake of counting up support atm) 18:57:18 feature:install firewall-driver-teal-coloured-starts-with-A-three-letters-ASA 18:57:54 alagalah: Do not tempt my sense of the absurd ;) 18:58:40 #info it seems everyone agrees it would be different if the vendor-specific name was further up in the stack 18:59:48 #info mohnish_ says also, it would be really important for PTLs to separate vendor-specific code from vendor-agnositc code 19:00:34 Timebox? Huge agreement here! 19:01:17 #topic wiki revamp 19:02:45 #link https://wikistg.opendaylight.org/view/Main_Page odl-casey has been trying to do a bit of a visual and some content revamp of the wiki (username: daylight, password: ODLStageSiteAccess!) 19:02:56 #info it has a new theme, that's a bit cleaner 19:03:54 #Info the links of the main page have beeen grouped into "user", "developer" and "community" generally to make it easier for people to find what they need, epsecially for users (not ODL developers) who often end up and are confused by our wiki 19:04:15 #info odl-casey would love feedback, colindixon will start a thread 19:04:36 This is very nice 19:04:40 #action colindixon to start a thread on the wiki revamp 19:05:38 odl-casey: may want to include this user-facing-ish Deployment wiki in the "Installing, Running and Usage" section https://wiki.opendaylight.org/view/Deployment 19:08:01 +1 for time back 19:08:20 #topic cookies 19:08:30 dfarrell07, how current is that page? 19:08:38 #endmeeting