16:01:31 #startmeeting integration 16:01:31 Meeting started Thu May 19 16:01:31 2016 UTC. The chair is jamoluhrsen. Information about MeetBot at http://ci.openstack.org/meetbot.html. 16:01:31 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 16:01:31 The meeting name has been set to 'integration' 16:01:39 running to fill my coffee... 16:02:29 running late. be there in about 5 minutes 16:03:58 LuisGomez, dfarrell07: my webex is not starting.... 16:04:22 jamoluhrsen: ack 16:04:56 #topic packaging 16:05:25 packaging intern started now and going well. working on .deb packaging project 16:06:03 #info packaging intern started now and going well. working on .deb packaging project 16:06:44 #info BeSR2 release artifacts to be complete this week. 16:07:03 #info good progress with different installers in OPNFV. up to now it's mostly only been APEX 16:08:03 #action dfarrell07 to update int/pack's trello 16:09:09 #info cluster test and perf call still at 9am on Fridays. 16:09:26 Does Bluejeans work on Linux? 16:09:26 #action jamoluhrsen to setup a bluejeans meeting for 8:30am going forward starting in two weeks 16:10:12 #topic distribution 16:10:36 #info karaf upgrade complete and we ended up with some broken distribution related jobs 16:11:04 #info one problem is that sshpass stopped working which affects bin/client 16:11:51 #info sshpass problem not happening in every environment, but does happen in releng 16:12:19 #info there are people working on this, and probably some workaround with some config or removal of a file 16:13:29 #info other issue was that after karaf upgrade, jetty started failing to start. iotdm needed to be removed from -all- features to fix this 16:15:37 #info autorelease was failing because sfc was still depending on ovsdb version of netvirt, so needed to move to netvirt features as ovsdb and netvirt split 16:16:25 #info autorelease currently failing because vxlan adapter issue. faas project. vratko sent email about this one. 16:17:06 #info autorelease didm project still not part of autorelease after being removed several weeks ago. 16:17:59 #info autorelease successful build for all projects is part of M4 requirement, but ideally projects should try to always be there and building successfully. 16:19:36 #topic builder 16:20:01 #info private cloud is under test, and going well. 16:20:20 #info private cloud sandbox will be migrated to soon (maybe this weekend) 16:20:42 #info everyone using sandbox will need to pull in some patches in order to use new sandbox 16:20:51 #link https://git.opendaylight.org/gerrit/#/q/status:open+topic:rpc-migration rpc-migration patches 16:23:22 #info we need to reduce the resources each VM uses in the jobs we are using. zxiiro plans to investigate this. 16:24:06 #info biggest impact is if we can reduce system requirements on builder vms, mininet, etc. 16:24:59 #info we could reduce mininet vms right now, but we'll still need larger mininet systems for things like scale tests 16:26:59 #info do *NOT* abort jobs more than once. just click the little red x one time. more than once will leave VMs hung and wasting resources. 16:27:40 #info tykeal thinks the sandbox migration could start happening friday night (5/20) 16:29:00 #info if anyone notices anything in new sandbox that is not the same as old sandbox, make it known. 16:29:36 #info releng private cloud will migrate a few weeks after sandbox migration assuming all is well in new sandbox 16:31:16 #info new private cloud will not let us view robot logs in browser remotely. we have to save the logs locally and open locally. 16:31:35 #info might have workaround for viewing logs, or they might get posted to nexus eventually. 16:31:52 #topic release 16:33:09 #info M3 offset 2 coming up. seems like the toughest milestone. maybe we can do a TWS working group to answer questions 16:33:49 #topic misc 16:34:16 #info what do we want to do with the old Integration repo? 16:35:14 #info archived projects have locked repos, meaning read-only. we can unlock to do one off changes. we can move it, but non-trivial to do. 16:35:50 #info the problem is that people can clone and use this old archived repo, and can cause confusion, especially to new people. 16:36:19 #info LuisGomez likes idea to remove all repo files and add one readme that points users to the new repos. 16:36:54 #info tykeal asks that a committer do this. 16:37:39 #info anipbu agrees with this idea, but we should do this with all archived projects. 16:38:34 #info Integration is a special case, because the project is archived but there is still the same work ongoing in a new repo 16:39:48 #info vratko points out that we need to do this to all branches. 16:40:00 #action jamoluhrsen to double check with TSC that this is ok. 16:42:12 https://wiki.opendaylight.org/view/Archive_Proposals/Integration#Archive_Location 16:42:58 #link https://wiki.opendaylight.org/view/Archive_Proposals/Integration#Archive_Location needs to be updated if we do this repo cleaning 16:44:58 #info jamoluhrsen asks how can we start getting our wiki pages indexed so they can be publically searched. 16:45:11 #info zxiiro notes that phrobb had a good reason to not have this. 16:45:29 #action jamoluhrsen to email phrobb and CaseyODL to get this enabled if possible 16:46:48 #action anipbu to survey all projects to see if they want to be indexed and present that info to LF folks 16:47:37 We purposely blocked searching of the wiki because people were ending up on content from helium and hydrogen 16:47:57 #topic system test waivers 16:47:59 Some pages should be indexed. 16:48:08 But only top-level pages 16:48:35 Project pages should be indexed. But I'm not sure that it's been added yet 16:48:40 I will take that as an action 16:49:10 anipbu: ^ 16:49:31 #info DLUX has the basics of what they need to do system tests. jamoluhrsen recommends we do not give a waiver 16:52:44 #info ODLPARENT granted system test waiver 16:53:34 #info MDSAL granted system test waiver 16:54:37 #link https://lists.opendaylight.org/pipermail/integration-dev/2016-May/006768.html reasons why we may want to rethink what is valid for system tests. maybe something yang could be. 16:54:46 #info YANG granted system test waiver 16:55:14 #info OPENFLOWJAVA granted system test waiver 16:55:56 #link https://wiki.opendaylight.org/view/Simultaneous_Release/Boron/Waiver/System_Test <--- List of current system test waivers 16:56:18 #info YANGIDE granted system waiver because there is no infra to run their system tests 16:58:15 #info anipbu would like a way that projects report external system test results. 16:58:26 https://wiki.opendaylight.org/view/CrossProject:Integration_Group:Feature_Integration_System_Test_Template 17:00:25 #info Be projects still not requesting waiver, but maybe wont need one (OpFlex, SNMP4SDN, TTP, TCPMD5) 17:00:46 #endmeeting