17:01:59 #startmeeting training-guides 17:02:00 Meeting started Mon Nov 24 17:01:59 2014 UTC and is due to finish in 60 minutes. The chair is sarob. Information about MeetBot at http://wiki.debian.org/MeetBot. 17:02:01 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 17:02:03 The meeting name has been set to 'training_guides' 17:02:31 good evening. 17:02:33 hi all 17:02:40 hello 17:02:45 Hello 17:02:49 Hello 17:02:56 Hello 17:04:09 I think we got the time this time 17:04:15 *right this time i mean 17:04:31 :) 17:04:34 #topic kilo plans 17:04:37 ShillaSaebi: I hear you - happend to me too few weeks ago 17:04:38 Right 17:05:16 there are few patches in the pipeline about enhancing existing docbook content 17:05:40 we should decide about that 17:05:46 I think its on the agenda too 17:06:05 But no bp or specs 17:06:23 sarob: ? 17:06:25 what sarob ? 17:07:08 I was hoping for a few bp and/or specs 17:07:26 To set the agenda for the kilo cycle 17:07:32 #link https://review.openstack.org/#/c/136633/ 17:08:04 #link https://blueprints.launchpad.net/openstack-training-guides/+spec/training-vm-in-public-cloud 17:08:08 do we use doc's spec repo or ours? 17:08:14 Ah didn't see that 17:08:35 We agreed with the docs team to use theirs 17:08:48 At least the last time we discussed it 17:09:27 sarob: I submitted the spec to our repo because that's where our other spec are. 17:09:33 sarob: are you going to move them? 17:10:03 Sure I can do that this morning 17:10:54 The ones I created are stale 17:11:37 sarob: so do we delete or move them? or move them to juno folder in spec for archive purposes? 17:11:58 On second look 17:12:31 Only shillasaebi spec is relevant still 17:12:54 If the team is okay with it 17:13:10 I'd like to just remove the whole specs dir 17:13:35 yeah I'm good with that 17:13:38 I'll double check with annegentle that we are still go to use docs-specs 17:13:49 Okay 17:14:12 Other thoughts? 17:14:57 sarob: I think the person writing up a spec should want to implement the feature (or at least believe it should be done). 17:15:03 hello 17:15:14 Agreed 17:15:32 rluethi: yes, but not exclusively 17:15:32 Evening dguitarbite 17:15:47 matjazp: what do you mean by exclusively? 17:15:58 others can join and help? 17:16:00 possibly 17:16:37 morning sarob 17:16:43 apologies for being late 17:16:47 all I'm saying is I wrote a spec with some reluctance because I don't think it should be done in Kilo. 17:16:54 I was a bit too deep into Unreal Engine 4 source code 17:17:36 rluethi: the b/p is registered in Kilo 17:17:45 does not mean that it should be implemented in Kilo 17:17:58 you could implement it or target the b/p for another release in future too 17:17:59 Right 17:18:05 or keep in on hold with appropriate reasons 17:18:16 We can document that stuff 17:18:23 dguitarbite: yeah, but why spend precious time speculating about random features that we might or might not do in the rather distant future? 17:18:36 Most important is this release work 17:18:46 Needs to be planned out 17:18:57 Through bp 17:18:59 rluethi: its not speculation 17:19:13 these are desired features etc. etc. and we prioritize them as per the time/feedback etc. 17:19:20 The docs team and TC 17:19:34 Need to know what we have planned 17:19:53 That's through the bp assigned to milestones 17:20:26 sounds good 17:20:32 sarob: from my side, I am almost ready with my patch 17:20:33 https://review.openstack.org/#/c/136774/ 17:20:37 this one is the dep. 17:20:57 for this patch https://review.openstack.org/#/c/134538/ for the basic install guides 17:21:01 after this I will push the spec 17:21:11 Okay 17:21:20 rluethi sent a mail to the ML, some of us commented about priorities.. do we have a consensus on that? 17:21:35 when did the mail go out 17:21:39 i don't remember seeing that one 17:22:01 11/21/2014 17:22:10 ShillaSaebi: http://lists.openstack.org/pipermail/openstack-docs/2014-November/005510.html 17:22:30 Ah right 17:22:38 matjazp: I am fine with the priorities as suggested by you. 17:22:42 btw: good work writing that down, roger :) 17:23:08 ok i did see this 17:23:16 yeah thanks for putting this together, I'm fine with these prios 17:23:46 I agree too. 17:24:33 rluethi: there you go - no BPs for you, as #1to #3 dont need them 17:25:00 the rest of them are my responsibilities which I will write up as b/ps :) 17:25:09 Weird I don't have that email 17:25:10 rluethi: you dont need to work that hard ;) 17:25:10 matjazp: perfect :) 17:25:27 I will write up #9 :) 17:25:30 rluethi: I knew you would be happy ;) 17:25:33 sarob: you replied to matjazp's reply. 17:25:57 sayali: do not forget #0! 17:26:56 Strange 17:27:11 I can't find it now 17:27:13 dguitarbite: :/ 17:27:32 Found it 17:27:41 * dguitarbite #9 does not exist, I figured you are joking 17:28:16 Need more coffee 17:28:16 dguitarbite: it does! I wrote back a mail adding #9 17:28:20 :P 17:28:31 ahh, ok :| 17:29:05 dguitarbite: ;D 17:29:05 sayali: does #9 need bp? 17:29:09 sayali: I am not sure if its a spec 17:29:12 matjazp: I agree 17:29:20 seems like a patch 17:29:20 may be its a subpoint to the already existing scripts b/p 17:29:25 may be for the host side scripts 17:29:52 rluethi: sayali do you guys agree to discuss about the specs in labs after the meeting on docs IRC channel? 17:30:03 well I thought it would be a bp if we want to have a variety of colors and variations for logsand stuff 17:30:04 we might need to figure out the number of specs that make sense 17:30:11 matjazp, dguitarbite: you'd be surprised how much is involved in making colors work well for the scripts. I looked into it before. 17:30:31 dguitarbite: sure, I've got an hour. 17:30:44 sayali: exactly, thats why I suggested to have a chat later on and finalize the b/p's required for labs section 17:30:58 * dguitarbite meeting need not be today but soonish 17:31:15 dguitarbite: I am not sure I can make it today after the meeting. 17:31:17 dguitarbite: It's not the number of specs, but the kind of features that need a spec. 17:31:42 I can't either :/ 17:31:57 dguitarbite: we could discuss it later this week if thats alright 17:32:04 rluethi: I meant to discuss these things, not to say that we just decide number of specs ... rather filter the features and classify them into required amount of specs 17:32:10 sayali: yes 17:32:20 dguitarbite: ah ok. 17:32:38 rluethi: it would be counterproductive otherwise 17:32:58 dguitarbite: so? that's not unheard of :). 17:33:04 ;) 17:34:08 lets move on 17:34:24 I would really like to talk about the point matjazp already brought up: 17:34:30 #link http://lists.openstack.org/pipermail/openstack-docs/2014-November/005518.html 17:34:45 we are getting lots of patches for docbook content. 17:34:47 why? 17:35:07 I thought that was going away for the most part. 17:35:13 may be because we are planning to remove them, people have finally started taking interest in that ? 17:35:32 is anyone trying to troll us? 17:36:17 rluethi: dguitarbite content will be copied from the docbook right? 17:36:19 we'll the move is not here, so we neet to figure out what is useful and what's not 17:36:21 dguitarbite: hardly. 17:36:27 well 17:36:37 How about I change the wiki to reflect our plans 17:36:47 The changes plans 17:36:49 Planned 17:36:50 ShillaSaebi? 17:36:52 some content update is still relevant 17:36:58 matjazp: around 1 or 2 pages are unique 17:37:04 which are virtualbox related 17:37:10 rest of them are repeating 17:37:14 so those should stay 17:37:15 the unique ones right 17:37:15 either copied from us or by us 17:37:25 ShillaSaebi: we could easliy push them into wikis 17:37:38 dguitarbite that sounds like a plan 17:37:48 the bigger question in this case is "Who will create/maintain the slides" 17:38:29 after my patch https://review.openstack.org/#/c/134538/ goes through 17:38:35 I still need a repo for quizzes, so I'll move all the relevant Q&A int RST format? 17:38:37 Well we need to get that patch agreed to first 17:38:41 the other question is: wasn't the pilot in CA supposed to start these days? what are they working with? sarob? 17:38:42 it will be 1-2 weeks before the docs folder is almost empty 17:39:03 Rluethi we started last thurs 17:39:13 sarob: so do they have slides? 17:39:24 sarob: this should work, annegentle is only worried about virtualbox content which is not included here 17:39:30 sarob: are you writing them as the course goes forward? 17:39:45 matjazp: about the quizzes, RST sounds good 17:40:01 I was hoping this was going to be a team effort 17:40:26 As the big patch gets approved 17:40:35 sarob: It won't be if you don't tell us :). 17:40:40 Then the slides for chap 3-4 are important 17:40:42 haha 17:40:44 sarob: I will be busy with creating exercises with osbash, openstack-koans and managing the infra for training guides 17:41:04 I'm motivated 17:41:06 if i can get someone to work together with me as part of a team effort, i will volunteer 17:41:06 I can review the slides :) 17:41:09 So I will start 17:41:15 just don't want to be the single point 17:41:19 ok 17:41:38 Nothing like making a fool of yourself 17:41:49 In front of 60 people 17:41:55 For motivation 17:41:59 haha 17:42:10 My big picture 17:42:23 Needs for running the training 17:42:28 Is simple 17:42:32 Slides 17:42:42 Test clusters 17:42:51 sarob: I can help out with the migration 17:42:52 Quizzes 17:42:58 track the content and syllabus 17:43:05 Cool 17:43:20 sarob: the training cluster in a public is not likely to happen this cycle (if ever). 17:43:32 The finer points of how we do these things are important 17:43:39 sarob: I sort of expect most trainees will turn up with Windows machines. 17:44:06 sarob: so based on your experience/feedback, we might have to elevate the priority for Windows batch scripts. 17:44:07 It's a mix of nothing, Linux, win, OS X 17:44:21 And tablets 17:44:24 Linux and OS X should be no problem. 17:44:42 The common thread is browser 17:44:44 sarob: tablets might be a challenge 17:44:47 tablet users can use them to read the install-guide :) 17:44:56 rluethi: hehe :D 17:45:17 If we can build clusters through scripts 17:45:26 And get copies of them 17:45:33 On public cloud 17:45:45 That's makes success 17:45:55 sarob: that was the spec roger has pushed as of now 17:45:58 For anyone that shows up 17:46:04 The main problem seems to be that we can't get all the OpenStack networking magic nested. 17:46:13 Yup 17:46:24 And I love him for it 17:46:53 That's pretty much a hard limit. Only way out is all-in-one. I tried to describe the issues in the spec. 17:47:09 I like what i read 17:47:35 If we spin all our goals back to enabling trainjng 17:47:37 sarob: rluethi lets ship it for the time being, depending on the cloud we could write the init scripts to run the osbash scripts and get a multi-node cluster running 17:47:51 s/cloud/public cloud/ 17:47:54 Our priorities are clearer 17:48:33 dguitarbite: let's talk about that later. 17:48:37 I will chk with annegentle on docs-specs use today 17:48:39 rluethi: yes 17:49:25 are labs working now as they should? 17:49:29 Then we can get a few more reviews on the patch and merge it to docs-specs kilo 17:49:58 matjazp: you tell me, you are testing them with your students. 17:50:15 rluethi: roger that 17:50:16 rluethi: as of now, we're still at devstack phase 17:50:40 matjazp: they are ready, there is a small issue and we have a work around for that 17:50:48 matjazp: to the best of my knowledge, the scripts are pretty solid now. 17:50:58 all-in-one 17:51:25 matjazp: ? 17:51:44 rluethi: devstack all in one 17:52:00 matjazp: I see. 17:52:21 dguitarbite: what issue? 17:52:48 matjazp: you just need to restart nova-compute service on compute node sometimes it is not running 17:53:00 rluethi: pushed a patch to fix that, hopefully it should work now 17:53:15 dguitarbite, matjazp: yeah, that should be fixed. 17:53:20 ok, that's not a biggie 17:53:42 nova-compute goes XXX for a couple of minutes after a controller reboot, but I think that's how OpenStack works. 17:53:44 matjazp: yes, sorry for that ... just a minor issue which is probably fixed 17:55:01 matjazp: the test script on gerrit gives you a good idea of some of the things that can go wrong and how to diagnose and fix them. 17:55:24 matjazp: awful code, though, sorry. 17:55:38 time check 5 mins remaining 17:55:41 it works really well 17:55:59 rluethi: https://review.openstack.org/#/c/131689/ ? 17:56:28 matjazp: that's the one. 17:56:41 rluethi: ok 17:57:38 matjazp: it works fine, the code is a bit ugly but it does the job well 17:58:23 I'm writing a ml on the priorities for associates training in sfbay user group 17:58:35 sarob: what is the start date? 17:58:43 Last thurs 17:58:48 ok 17:59:13 It should help with the team priorities 17:59:23 I'm asking reed to do the same 17:59:24 ok guys i have to sign off, have a hard stop at 1, have a good day 17:59:31 Me too 17:59:35 Cheers 17:59:39 see ya, times up 17:59:40 bye all 17:59:41 see ya 17:59:42 Have a good week! 17:59:42 bye 17:59:47 bye:) 17:59:47 sayali: roger lets meet on docs IRC channel 17:59:53 dguitarbite: k 17:59:58 to schedule the meeting 18:00:03 bye everyoen 18:00:07 #endmeeting