17:00:11 #startmeeting training-guides 17:00:13 Meeting started Mon Aug 11 17:00:11 2014 UTC and is due to finish in 60 minutes. The chair is rluethi. Information about MeetBot at http://wiki.debian.org/MeetBot. 17:00:14 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 17:00:17 The meeting name has been set to 'training_guides' 17:00:20 hello 17:00:22 roll call 17:00:28 hi 17:00:33 Hello 17:00:34 sarob cannot be with us today. 17:00:35 Hi 17:00:45 hi everyone. 17:01:00 hello guys 17:01:15 dbite: update on docs? 17:01:21 yes 17:01:28 set the topic? 17:01:34 *can you set the topic 17:01:38 #topic docs 17:02:06 #info docs team was previously called stable team 17:02:33 basic install guide needs to be updated for Icehouse 17:03:03 some of the sections like ML2 plugin etc. are missing and older OVS plugin for Neutron is still in use 17:03:12 sarob_: hello 17:03:27 Hey. I'm on a short break 17:03:33 yes. a comparison with the openstack-manuals install-guide for icehouse will show the necessary changes. 17:03:50 is the Comcast team there with us? 17:04:08 I'm here 17:04:10 dbite: for basic install guide, just report relevant bugs... so we can all chip in 17:04:47 I agree with rluethi, for getting better idea we can compare with install guides under openstack-manuals project 17:05:01 matjazp: Yes, I will start logging bugs 17:05:29 MeganR: I you guys want to help out, you don't need lots of prior knowledge for this task. 17:05:34 MeganR: have you guys been working on the install guides part? Last time we discussed we were planning on migrating to something very close to install guides 17:06:09 rluethi: we would be happy to help out - especially if not a lot of previous knowledge is needed 17:06:15 sayali_: also can you get in touch with MeganR and her team to get better idea for working on install-guides 17:06:31 dbite, alright 17:06:59 MeganR: it has a learning curve, but with the current labs section its easier to test out the install guides 17:07:01 MeganR: there shouldn't be. Everything in training-guides that deviates from the openstack-manuals install-guide is suspicious. 17:07:30 to be more precise basic-install-guides under training-guides 17:07:39 ok - we can discuss this in more detail offline, I will look at scheduling some time on Thursday, when Shilla is back 17:07:48 what dbite said :). 17:07:51 sure thing 17:08:02 rluethi: you just freaked me out ;) 17:08:33 any questions for the docs sub-team? 17:09:46 rluethi: lets move on 17:10:05 #topic upstream 17:10:18 I got a sec 17:10:32 sarob: cool. your stage. 17:10:45 I can't get epicslide to build on osx 17:10:56 Only ubuntu 17:11:05 sarob_: what about landslide? 17:11:11 So I PDF support will be delayed a bit 17:11:18 Landslide is it 17:11:19 matjazp: the problem with landslide is PDF support 17:11:37 So HTML slides only for right now 17:12:02 Id like to get some feedback on the dir structure 17:12:11 sarob: any idea how hard it would be to fix epicslide? 17:12:14 For the current upstream patch 17:12:37 dbite: maybe we can ask Foundation for help with PrinceXML for PDF geenration? I know its not opensource, but in the mean time... it gets things moving 17:12:39 Rluethi: not sure. Something to work on longer term 17:13:00 Short term we only need HTML I believe 17:13:04 matjazp: we need to push opensource as much as possible 17:13:18 Dbite: agreed 17:13:32 #info epicslide doesn't build on OS X. We need to stay with landslide for the time being (which means no PDF generation). 17:13:38 sarob_: I think we should proceed with landslide and try to get the PDF support in eventually 17:13:55 sarob_: dbite: sure, but I see it as a tool--- much like Oxygen for XML editing... 17:14:02 Gotta go 17:14:08 Agreed 17:14:13 matjazp: Oxygen is another problem, not an example to follow :). 17:14:16 Cheers 17:14:27 matjazp: Oxygen is not exactly a good example in this case 17:14:32 sarob_:bye 17:14:40 I use VIM to write XML code 17:14:45 yes, I know.. but it gets the job done 17:14:50 sarob: we can't give such tools to volunteers and students, that's bad. 17:14:53 with LandSlide we officially lock down 17:15:04 the RST is there to be used by people 17:15:16 so someone can volunteer in and use PrinceXML 17:15:23 but we should try to keep it out of upstream code 17:15:36 does it make sense? 17:16:09 #action sarob will try to get upstream material merged soonish. 17:16:28 dbite: no, I mean that PrinceXML can be used only at Foundations's servers, for ppl to load PDFs... not as a tool that everyone needs to have on theirs coms 17:16:44 ok, makes sense 17:16:59 we just need to deliver PDFs to DL 17:17:02 I would suggest asking this to Sarob during the next meeting 17:17:06 OK 17:17:07 sorry, I got everyone off track. Any comments on the directory structure? 17:17:22 directory structure for upstream materials? 17:17:27 I think what reed has proposed is a good direction to start with 17:17:34 for the presentation 17:18:10 rluethi: just start the ball rolling, we can change it later ;) 17:18:21 rluethi, afaik landslide generates pdf 17:18:24 if anyone else is interested, have a look here: 17:18:29 #link https://review.openstack.org/#/c/100999/ 17:18:48 reed: I thought it only does so with PrinceXML. 17:19:14 reed: did you try to do it without PrinceXML? 17:19:16 rluethi, you may be right, I never tried pdf 17:19:27 reed: there's your problem then :). 17:19:36 i don't see the point of pdf anyway :) 17:19:53 reed: nice for handouts, I guess. 17:19:54 reed: students like to DL pdfs... 17:20:02 rluethi: exactly 17:20:04 can we not use another technology/project for generating PDF's ? 17:20:13 students are wrong :) 17:20:18 rluethi: they are bugging me if I don't give them materials for DL 17:20:21 reed: they are the future ;) 17:20:35 god save us from that future :) 17:20:42 hehe 17:20:48 * rluethi sighs 17:20:53 moving on? 17:21:08 #topic testing 17:21:14 matjazp: ? 17:21:27 I started with associate guide 17:21:34 missing quizzes 17:21:47 I have bugs reported 17:22:11 matjazp: you reported bugs or you just have them? 17:22:50 rluethi: I reported bugs, so ppl looking for something to contribute can chip in 17:23:05 matjazp: are you getting help? 17:23:10 We could use bartaz.github.io/impress.js/#/bored for the presentation 17:23:22 rluethi: no, not at the moment 17:24:01 MeganR: that may be another thing your folks might be able to help out, learning stuff in the process. 17:24:18 ok, we will add it to the list! 17:24:48 matjazp: could you give them some help getting started on tests? 17:24:50 sayali_: can you take a lead on this? 17:24:53 maybe we can all report as many bugs as we can, so ppl can see where they can contribute 17:25:05 rluethi: yes, sure 17:25:39 dbite, sure, we can generate pdfs too using that 17:25:40 great. all: please remember to see the "low hanging fruit" tag for bugs that are easy. 17:25:52 what's the proper tag again? 17:27:27 low-hanging-fruit 17:27:42 is it? cool, thanks! 17:27:56 anything else for testing? 17:28:00 nope 17:28:19 #topic infrastructure 17:28:38 dbite? 17:28:41 #info training-guides team mark low-hanging-fruits for easy bugs 17:28:58 we have good progress for the labs 17:29:05 controller node is almost done 17:29:21 I would suggest running the osbash.sh script for generating multi-node environment 17:29:26 but its not ready yet 17:30:03 dbite: but if I remember correctly, base disks are finished, right? 17:30:20 matjazp: never finished, but working. 17:30:27 what do you mean by base disks to be precise? 17:30:32 its not finished 17:30:58 rluethi: yes... as long its functional, its ok ;) 17:31:16 matjazp: the openstack deployed on it is not yet functional 17:31:19 the only change in basedisks in the past weeks was moving to Ubuntu 14.04 17:32:26 however, we are at a point where several services should be up and running after the scripts have run. 17:32:37 to be precise, we could claim that base disk is ready 17:32:43 you can test the services and see if they work. 17:33:13 openstack deployed on it is only completed for Keystone, Glance for Controller Node 17:33:27 Neutron and Nova for controller node are WIP 17:33:35 rluethi: dbite: yes, I'm looking at your code and scripts from time to time.. nice progress 17:34:04 I try to keep Vagrant working, but I'm not testing very often. 17:34:12 It will lack some features. 17:34:19 matjazp: thanks, but it needs lot of work 17:34:30 I don't know yet how to work automated snapshots into vagrant. 17:35:08 dbite: anything else on labs/infrastructure? 17:35:50 rluethi: I know I read somewhere about snapshost in wagrant.. If I find it, will forward 17:36:11 nope 17:36:32 matjazp: there was a plugin, I tested it. but we don't have host-side scripting on Vagrant. 17:36:56 matjazp: we have to finish the scripts first, anyway. 17:37:06 #topic misc 17:37:19 anybody confused, shocked, depressed? 17:37:36 any questions/issues you would like to have discussed? 17:37:42 or puzzled or just landed here by mistake? 17:37:56 dbite: then I don't care :). 17:38:11 haha :) 17:38:12 can you think about impress.js for the lanslide issue? 17:38:45 sayali_: have you tried both? how do they compare? 17:38:46 do you mean landslide? 17:38:55 yes dbite 17:39:03 sayali_: how does it generate PDFs? 17:39:07 does it support PDF features? 17:39:11 bartaz.github.io/impress.js/#/bored 17:39:35 rluethi, I haven't tried landslide, but impress is pretty good. It supports pdf too 17:40:02 sayali_: I know impress (and similar tools) look cool, but easy of use for slide authors is very important. 17:40:41 if it can be created using MD or RST its should be tested out 17:40:41 sayali_: impress might be too complex. if you have experience that shows otherwise, at least I would be interested. 17:41:18 rluethi, you can generate it using MD which is pretty easy 17:41:48 I could look into it further and check the complexity 17:42:24 sayali_: upstream materials should land in the repo soon, maybe you could translate it (or some of it) to impress. 17:42:37 sayali_: that would give us something to compare. 17:43:03 sayali_: working PDF generation would definitely be a plus, since that's our current problem with landslide. 17:43:34 rluethi, ok I will put up an initial patch to show how it works 17:44:28 #action sayali will create a patch to help us evaluate impress.js as a possible alternative to landslide. 17:44:40 anything else? 17:45:14 thanks everyone for being here! 17:45:19 #endmeeting