16:00:35 #startmeeting fuel 16:00:35 #chair xarses 16:00:35 Meeting started Thu Oct 1 16:00:35 2015 UTC and is due to finish in 60 minutes. The chair is xarses. Information about MeetBot at http://wiki.debian.org/MeetBot. 16:00:35 Todays Agenda: 16:00:35 #link https://etherpad.openstack.org/p/fuel-weekly-meeting-agenda 16:00:35 Who's here? 16:00:36 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 16:00:39 The meeting name has been set to 'fuel' 16:00:40 Current chairs: xarses 16:00:41 hi 16:00:48 o/ 16:00:50 hi 16:00:52 hi 16:01:13 hi 16:02:18 #topic action items from last week 16:02:24 mihgen to find bug about missing neutron options found by scale 16:02:32 sorry folks, I spent some time and was unable to find that bug.. Let's just pay attention when we sync something with upstream, that we don't lose our settings which we made based on testing 16:03:13 that's it on this topic.. 16:03:19 thanks 16:03:29 #topic MAINTAINERS: let's go ahead and extend the list (mihgen) 16:04:08 it's pity that we don't have many ppl today .. most are busy with another meeting ( anyway.. 16:04:13 First patch is merged to fuel-web: 16:04:18 #link https://review.openstack.org/#/c/225457/ 16:04:25 I call everyone to create this file in every repo, and fill it in by next IRC meeting. We need to move on with our code review improvements, and this is one of the steps. 16:04:37 Then, maintainers will need to review code first before asking cores to review. 16:04:38 I'll use http://lists.openstack.org/pipermail/openstack-dev/2015-September/075008.html as a source on who is core in which repo, and assign an action item for the first core from each repo to create MAINTAINERS file for that repo. 16:04:51 If you folks see that someone is on vacation, etc. - please pick it up as being another core / maintainer of the repo. Let's get it all done by next IRC meeting, so we can go repo by repo and ensure that files are there. 16:05:22 angdraug: do you agree here.. .? 16:05:29 mihgen: do we have some template for that? 16:05:42 let's use patch which I've shared as template 16:05:45 mihgen: sorry, i see. 16:06:17 mihgen sorry I didn't catch. How we can add maintainers now? 16:06:36 just go ahead and propose patch to that file in fuel-web 16:06:42 or just add in other repos 16:06:47 ok 16:06:54 or do you mean how you identify particular people? 16:07:12 I think bulat can be a maintainer of fuel-createmirror 16:07:29 ashtokolov, that's not on a stackforge/fuel-* repo 16:07:50 mattymo: fuel-createmirror is there 16:07:54 mihgen, who is maintainer of nailgun? Who is maintainer of specs? I know your goal is to mark a maintainer for everything... but there's only a few areas where it's black and white 16:07:56 well then you propose him 16:08:17 mattymo_: there can be few maintainers 16:08:29 let's bring gray areas to discussion and decide 16:08:43 mihgen, https://github.com/stackforge/fuel-createmirror ? no... it's https://review.fuel-infra.org/#/admin/projects/packages/centos6/fuel-createmirror 16:08:46 it's actually ideal situation when there a few maintainers 16:08:54 mattymo_ I think it should be on a stackforge/fuel-* repo anyway 16:08:55 https://github.com/stackforge/fuel-mirror 16:09:07 mihgen: +1 on MAINTAINERS 16:09:24 it has already been moved to fuel-mirror 16:09:27 mihgen, we should have an etherpad, all top level dirs for all our main repos, and we should fill it out 16:09:36 rather than inviting random 1-20 people to create reviews 16:09:54 mattymo_: I don't want to invite random 16:09:56 cores! 16:10:10 they must know whom to propose to maintainers 16:10:21 they merge other people's code 16:10:23 mwhahaha, you're core. you should know all :) 16:10:57 mattymo_: -1 16:11:19 mattymo_: so I don't know why we'd have etherpad 16:11:30 if core of particular repo can just go ahead and propose patch 16:11:32 all cores combined will definitely know all maintainers 16:11:35 then get people review it 16:11:42 including future maintainers 16:11:52 doesn't mean a single core would know everything 16:11:53 so it should go with approve 16:12:41 mattymo_: do you have any particular thing it mind why it may not work.. ? 16:13:07 mihgen, because your patch sat for a week without much involvement from cores. I'm seeing a bystander effect 16:13:23 "EVERYONE do this" results in nobody does anything 16:13:33 anyway, move on 16:13:49 mattymo_: I'll assign first core of each repo, as I said.. did you read my messages above?))) 16:13:55 one guy per repo 16:14:01 lets give out some actions 16:14:01 who will be responsible to make it done 16:14:48 xarses: you can give an action item for me to assign first core to create/fill in MAINTAINERS file for the repo core reviewer is responsible for 16:15:02 " I call everyone to create this file in every repo" != first core of each 16:15:30 assign the first core? 16:15:56 mattymo_: sorry. I fixed myself in next sentence: 16:15:58 "I'll use http://lists.openstack.org/pipermail/openstack-dev/2015-September/075008.html as a source on who is core in which repo, and assign an action item for the first core from each repo to create MAINTAINERS file for that repo." 16:16:42 #action mihgen will task a core from each repo to start the first version of the MAINTAINERS file 16:16:48 moving on then? 16:17:05 yes please 16:17:08 #topic team structure policy: can we get more +2 and merge it.. ? (mihgen) 16:17:19 #link https://review.openstack.org/#/c/225376/ 16:17:37 angdraug, you are as an acting PTL, in my opinion should make a decision if we need to wait some more time / ask particular cores to +1, and finally merge it. This doc is ~2 weeks around, so I'd assume that we can move on with mergin... 16:18:33 +1 16:19:25 angdraug: ^^ ? 16:19:26 the deadline to merge it is 8th 16:19:38 so we still have a few days 16:20:00 I don't want to have some last-minute -1's so lets agree that the last day to provide feedback is 5th 16:20:03 did we mention it in openstack-dev ML .. ? 16:20:09 oh yes 16:20:14 oh sorry then 16:20:18 I didn't notice 16:20:36 #link http://lists.openstack.org/pipermail/openstack-dev/2015-September/074984.html 16:20:39 ok, so I hope everyone had a chance to read already 16:20:59 then let's move on.. 16:21:25 yes 16:21:26 #topic clean up in core reviewers group: http://lists.openstack.org/pipermail/openstack-dev/2015-September/075008.html (mihgen) 16:22:07 so folks it's been 10 days since I proposed it 16:22:15 I think we can go ahead with it 16:22:24 any objections? 16:22:33 mihgen: what do you think about keeping me on fuel-docs-core? 16:22:46 I'd task fuel-infra team to make it happen 16:23:03 :) 16:23:14 angdraug: you are #19 there, so sorry.. 16:23:31 if you go back to review stats, other cores will re-propose you?) 16:23:52 #link http://stackalytics.com/report/contribution/fuel-docs/180 16:24:01 action on angdraug then? 16:24:07 no thanks 16:24:23 xarses: yes, on angdraug to execute as he is leading fuel-infra 16:24:37 no thanks on going back up in review stats, that is 16:24:46 yes on action for fuel-infra 16:24:57 #action angdraug will task fuel-infra team to update core groups 16:25:15 fuel-docs team does come to me for help occasionally, but I'm not likely to be able to do enough reviews there to keep up with the top 10 16:26:11 moving on? 16:26:21 #topic announcement! 7.0! 16:26:38 what's there to say? 7.0 is out! 16:26:47 I'd like to take a moment to thank every one for their work on 7.0, it went out the door 16:26:52 links shared? 16:26:56 due to every one's hard work 16:26:58 #link http://lists.openstack.org/pipermail/openstack-dev/2015-October/075932.html 16:26:58 https://www.fuel-infra.org/ 16:27:09 #link https://www.fuel-infra.org/ 16:27:19 yes. This should be the best release we had ever in terms of quality 16:27:46 hopefully it works when you install it this time (j/k) 16:27:50 =) 16:28:01 not just quality 16:28:27 we managed to 1) improve quality; 2) reduce time between releases; 3) integrate new OpenStack release; 4) add a whole bunch of cool features! 16:28:41 8) 16:28:51 5) and do it in time 16:29:19 bookwar: that's (2) :p 16:29:34 but yes, great job everyone! 16:29:47 thanks folks. Let's keep it up! 16:29:50 #topic open discuss 16:30:03 thats it for the scheduled topics, if any one has anything to raise 16:30:10 otherwise we can close early 16:30:12 IBerezovskiy: around? 16:30:17 xarses have we decided about Q^A questions for PTL? 16:30:19 angdraug, yep 16:30:21 Q&A* 16:30:28 I was wondering if we should talk again about librarian 16:30:49 mattymo_: you can propose some =) 16:31:08 also when should we schedule that, next meeting? 16:31:47 angdraug, mwhahaha have sent email yesterday about our meeting results and provided detailed description of new workflow for puppet modules 16:32:54 it was a private mail though, so worth reiterating here 16:33:07 mwhahaha: you were going to update the wiki page on librarian, right? 16:33:10 yes 16:33:22 it is on my list of things to do to provide more details on the wiki 16:33:32 #link https://wiki.openstack.org/wiki/Fuel/Library_and_Upstream_Modules 16:33:46 the email was just adding more details on things already provided in that page 16:34:05 additionally there were concerns around the puppetfile and people merging items that bypass fuel-library core review 16:34:15 one thing in particular that still worries me is setting the expectation on what is and is not ok to merge into fuel-infra clones of upstream modules 16:34:25 I've created a script to validate Puppetfiles based on our policy around upstream modules. The goal is to turn this into a jenkins job so we do not accidentally merge changes that do not follow our guidelines around upstream modules. 16:34:25 #link https://review.openstack.org/#/c/229605/ 16:34:46 We'll need to figure out a list of things that are OK to merge on fuel-infra and what is not 16:34:52 and that should be put in the wiki as well 16:34:57 yup 16:35:12 I'll try to come up with a starting point when i update the wiki and send an email 16:36:12 thanks 16:36:45 anything else? 16:36:49 I have something 16:37:00 rmoe: go for it :) 16:37:05 I'd like to give a short update on some work I've been doing with network templates 16:37:17 One of the things I've been working on recently has been using network 16:37:17 templates to deploy environments with only two networks, one for admin traffic and one 16:37:17 for all OpenStack traffic. I've also created a template to use a single interface for 16:37:17 everything, admin + OpenStack. 16:37:17 There are preliminary instructions for configuring these templates here: https://review.openstack.org/#/c/229948 16:37:18 It requires applying a few patches to 7.0 environments but all but one of the required changes 16:37:20 have been merged so 8.0 ISOs require less effort. 16:37:22 HA works fine and all OpenStack components appear to work correctly. The only thing I haven't tested 16:37:25 yet is Ceph. Hopefully this serves as a good example of how flexible network templates 16:37:27 are. 16:37:41 this came from frequent requests from customers (and potential customers) for this kind of network configuration 16:38:30 funny that originally the idea of network templates came from the need to _add_ more networks :) 16:38:55 rmoe: cool 16:38:58 Hi 16:38:59 it's pretty cool that you've made it to scale in both directions 16:39:04 cool 16:39:11 let's complete it and merge template someone 16:39:17 so people can easily reuse it! 16:39:45 rmoe, can we make it so these templates are part of our CI coverage? 16:39:45 but there is even more demand on one interface for compute node 16:39:55 if it needs code from 8.0 does it mean we can't merge this documentation to fuel-docs stable/7.0? 16:40:06 it works in 7.0 but it's not easy 16:40:18 not as easy as 8.0 will be 16:40:34 we may want to document it in stable/7.0, and provide instructions what patches to apply if needed 16:40:38 there were some important changes that didn't make 7.0 (like the ability to update network groups with the cli) 16:40:46 or just better - document it in development guide 16:41:14 rmoe: that particular one should be targeted to 7.0-updates in my opinion 16:41:20 I agree 16:41:39 can you please ensure it target that milestone and it's clearly explained why it's important... ? 16:41:52 yes, I'll do that 16:42:32 thanks 16:42:55 anything else? 16:43:56 ok, thanks every one 16:44:04 #endmeeting