17:01:45 #startmeeting Integration 17:01:45 Meeting started Thu Dec 17 17:01:45 2015 UTC. The chair is dfarrell07. Information about MeetBot at http://ci.openstack.org/meetbot.html. 17:01:45 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 17:01:45 The meeting name has been set to 'integration' 17:02:11 #info Jamo is not here this week, dfarrell07 and Luis to run meeting 17:03:01 #topic News 17:04:05 #info Boron Dev Design Forum will be 2/29-3/1 in San Fran 17:04:41 #info Vratko sent a large update to the Integration/Test/Test Code Guidelines wiki. Others should also make contribs, give feedback. 17:04:54 #link https://wiki.opendaylight.org/view/Integration/Test/Test_Code_Guidelines Int/Test (Int/*?) code guidelines being refactoring 17:08:32 #topic Int/Dist Updates 17:08:38 #info zxiiro and dfrarell07 briefly talked about maybe trying to include ODL's systemd .service file in the distribution 17:08:44 #link https://github.com/opendaylight/integration-packaging/tree/master/rpm/unitfiles Current location of ODL's systemd .service files 17:13:11 #info distro was broken for a while, errors going to karaf console vs log issue was fixed by a patch from jamo, going to be painful to debug at this point (remove some projects, try to build, try with different set of projects, repeat) 17:13:40 #info Discussion about Jenkins delay, how to speed it up or reduce load 17:15:34 #info tykeal points out that good smoke tests to run often, then full stuff less frequently, would help a lot 17:17:25 #undo 17:17:25 Removing item from minutes: 17:17:27 #undo 17:17:27 Removing item from minutes: 17:17:35 #topic RelEng Updates 17:17:40 #info Discussion about Jenkins delay, how to speed it up or reduce load 17:17:43 #info tykeal points out that good smoke tests to run often, then full stuff less frequently, would help a lot 17:17:52 #info Since there's a lot of overlap between RelEng and Integration/*, both ppl and work, giving RelEng a portion of the Int/* meeting 17:17:59 #info Thanh has autorelease running as of Monday 17:18:06 #link https://nexus.opendaylight.org/content/repositories/staging/org/opendaylight/integration/distribution-karaf/0.4.0-Test-beryllium-v201512150335/ First Be autorelease artifacts (tmp repo) 17:18:15 #info dfarrell07's large JJB docs refactoring patch has been merged, more edits welcome in follow-up patches 17:18:19 #link https://nexus.opendaylight.org/content/sites/site/org.opendaylight.releng.builder/master/jenkins.html Current JJB docs 17:19:20 Lori Jakab proposed a change to integration/test: LISP: Separate Lithium and Beryllium tests https://git.opendaylight.org/gerrit/31304 17:29:16 #info Vratko points out that some people push many, many patch-sets for new changes. When they don't need tests, they should mark that frequently-updating change as a draft to avoid kicking off verify jobs. 17:30:20 #action Vratko to get that feedback/tip about reducing verify jobs using drafts into a wiki 17:31:00 #topic Int/Pack Updates 17:31:11 #info dfarrell07 created "Experimental Release Preview" (ERP) ODL RPM based on the first successful Be autorelease build artifacts 17:33:47 #info ERP RPM is designed to allow OPNFV testing projects to start consuming ODL Be well before OPNFV Brahmaputra, which gives them confidence to include a flag to install OPNFV with ODL Be vs Li 17:33:54 #info sudo yum install -y http://cbs.centos.org/repos/nfv7-opendaylight-4-testing/x86_64/os/Packages/opendaylight-4.0.0-1.20151215033.el7.noarch.rpm 17:34:40 #link https://github.com/dfarrell07/vagrant-opendaylight/blob/1c69f5839f2f15965557c32521c4250fc805939a/Vagrantfile#L74-L92 CentOS 7 Vagrant box that installs ERP via repo config 17:34:47 #link https://git.opendaylight.org/gerrit/#/c/31376/1/rpm/example_repo_configs/opendaylight-4-testing.repo To-be-merged example repo config for installing Be ERPs 17:34:54 #info dfarrell07 would really appreciate it if someone would create a .deb package for ODL 17:35:08 #topic General updates, questions, etc 17:35:36 #info Vratko asks if there's a good way to auto-remove a feature from autorelease if it's removed from integration 17:38:12 #info Discussion about having autorelease working from the very start of Boron 17:38:47 #info zxiiro points out that master currently works, so we should have a working autorelease once we branch 17:40:19 #info tykeal points out that autorelease doesn't exist until mid-Li, hence the world-always-broken state before now 17:40:29 #undo 17:40:29 Removing item from minutes: 17:40:38 #info tykeal points out that autorelease didn't exist until mid-Li, hence the world-always-broken state before now 17:41:10 #info Discussion about new autorelease verify jobs that should help break autorelease breakages 17:41:56 #info LuisGomez points out that we need to make it clear that breaking autorelease == bad and we'll kick you out until you stop 17:42:21 beatings will continue until... ;) 17:42:25 ;) 17:43:17 #info LuisGomez points out that we need to make it clear when new projects need to be in autorelease. Maybe m2, since need to be in distribution by m3? 17:45:14 #info Discussion about opflex isn't really a part of the release "train" as zxiiro puts it, shouldn't really be a part of the SR 17:46:12 #info LuisGomez, dfarrell07, zxiiro and others want to make it clear to new projects that being part of the SR isn't a bad thing if you don't need it, no need to take part if bad fit 17:46:35 #info Int/Dist hasn't sent out M4, anipbu points out 17:47:10 #action edwarnicke (or some other Int/Dist person?) to send Int/Dist M4 17:47:31 #info Discussion about how autorelease likely shouldn't be part of the SR, since every field is NA 17:48:28 #info Vratko points out that Int/Dist needs to have a primary contact other than PTL 17:50:13 anipbu: maybe we should build a tool that people can run and answer the milestone questions and it automatically sends teh email to the mailing list for them 17:50:28 #agreed Next two Int/* meetings (12/24 and 12/31) are canceled due to Christmas and New Years 17:50:28 anipbu: one issue i always have with the template is I forget where to get it 17:51:02 zxiiro: i would like to put such an app in spectrometer do that. 17:51:30 zxiiro: that was one of the reasons I joined spectrometer =) 17:51:47 #info dfarrell07 asks if we can start using JIRA for issue tracking, not sure if it's still a pilot or ready-for-use 17:51:55 #info dfarrell07 is interested in getting Int/*'s many Meetbot minutes dirs `cp/mv`'d into a single one 17:52:05 #info dfarrell07 says it makes it hard to find minutes for a date, have to know which dir to check 17:52:14 #link https://meetings.opendaylight.org/opendaylight-integration/2015/ Meetings minutes for this IRC chan, split incorrectly into many different dirs b/c of inconsistent meeting name 17:52:40 #link https://trello.com/b/ACYMpTVD/opendaylight-integration Trello for Int/* 17:52:43 dfarrell07: JIRA is still in pilot there's open questions that must be answered still 17:52:59 tykeal: yeah, I thought there was something like that 17:53:15 tykeal: not a prio I guess, but I'd be great to start using it 17:53:21 it'd* 17:53:35 dfarrell07: what's the thing about meetbot minutes? 17:54:00 tykeal: they are spread out in many dirs with various names based on how we started the meeting 17:54:07 tykeal: so it's hard to find anything 17:54:20 yeah, I can't make it happen for a subset of meetings 17:55:17 tykeal: I'd rather incorrectly drop things into the same dir than have them spread out like they are, imho 17:55:57 dfarrell07: open a ticket ;) 17:56:14 tykeal: need to get agreement from this group first :) 17:56:19 tykeal: but yeah, +1, will do 17:56:20 lol 17:57:49 https://meetings.opendaylight.org/opendaylight-integration/2015/ 17:59:20 * tykeal notes the TSC call starts in less then 1 minute 17:59:32 #action dfarrell07 to raise helpdesk ticket to refactor meeting minutes dir structure as described above 17:59:36 LuisGomez: TSC 18:01:33 #endmeeting