15:03:42 #startmeeting ceilometer 15:03:43 Meeting started Thu May 7 15:03:42 2015 UTC and is due to finish in 60 minutes. The chair is gordc. Information about MeetBot at http://wiki.debian.org/MeetBot. 15:03:44 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 15:03:46 The meeting name has been set to 'ceilometer' 15:03:52 o/ 15:03:52 o/ 15:03:53 sorry folks. 15:03:55 o/ 15:03:56 o/ 15:03:59 o/ 15:04:00 was doing some house cleaning 15:04:11 o/ 15:04:25 gordc has hoffa's body 15:04:38 o/ 15:04:45 cdent: i'll google that reference later.lol 15:04:52 #topic summit session proposals - VOTING DONE 15:05:10 so i've shut down the voting for design session proposals 15:05:18 http://libertydesignsummit.sched.org/type/design+summit/Ceilometer#.VUttSdvsl4s 15:05:21 o/ 15:05:25 o/ 15:05:37 above is the proposed schedule 15:06:08 o/ 15:06:11 gordc: why many with the same title? 15:06:15 the 'fishbowl' large rooms will cover componentisation and event alarm topics as they had the most interest 15:06:36 i can't edit the 'work session' titles 15:06:49 ttx: ^ i assume that is normal 15:07:11 for the work session slots you'll need to dig in 15:07:21 https://docs.google.com/spreadsheets/d/1vlEe75uZc5faytpvIMPrQdViOOsRLnfeThk_Rp_lIvI/edit#gid=0 15:07:44 you can also find the plan in the spreadsheet where it says 'GORD TYPING' 15:07:52 could you include the etherpad links in the description as well for work sessions 15:08:11 prad: yep. will build etherpads today 15:08:39 if anyone notices conflicts please shout ASAP. 15:08:51 the schedule should be finalised by next week. 15:09:11 gordc: for the fishbowl, do you think it would be better to have slides due to the large audience? 15:09:39 we don't want a presentation, we want a conversation 15:09:41 fabiog: you should probably minimise the slides... the layout as i understand it is a circle 15:09:55 the screen will most be used for capturing notes on etherpad 15:10:19 gordc: I think the idea of a generic title for working sessions was to gently discourage "lurking" 15:10:26 I think some figures on whiteboard or in a presentation can be helpful, but otherwise I agree with gordc and cdent 15:10:45 ... at least that was a notion expressed at a prior cross-project meeting 15:10:53 fabiog: i guess you could consider visuals if they are too complex to draw. 15:11:08 eglynn: ah cool cool 15:11:27 eglynn: yeap, I think this was an outcome from Paris, that we should minimize the number of people who are just lurking in the room 15:11:30 gordc: yes. I think at least the current state design and the desired one should be in a driagram 15:11:57 fabiog: +1, it sounds good as a starting point 15:13:00 fabiog: ok. if you believe it's something beyond the realm of a drawing then i guess a slide might work... but i don't think we should encourage it. 15:13:14 fabiog: I think the point is to figure out the desired one not present it. 15:13:16 tbh i hope the plan is simple enough that it doesn't need a powerpoint slide to explain 15:13:28 I think we should go into that discussion without a preconceived design 15:13:36 cdent: + 15:14:18 i should also point out the time... it'll be early... so i guess hold back on the drinks tuesday night. 15:14:31 cdent: I think having a starting point is beneficial, we onlu have 50 min 15:15:09 fabiog: i should also point out we have a open slot wed at 3:30pm 15:15:13 as long as it is presented as a straw man, okay, but really I think we need to talk about something more fundamental than "we want ceilo to look like this" 15:15:34 we will use that as a carry over time for small room discussions 15:15:50 gordc: yes, good idea 15:16:07 cdent: I think we should start with the more fundamental things before any ppt or whiteboard diagram 15:16:42 ildikov: like what? 15:17:00 for those sharing time, ie. cdent/fabio. r-mibu/llu i'll let you guys discuss how you want to organise it. 15:17:10 feel free to bring me in if you want input. 15:18:07 but for the fishbowl items we should probably discuss the broader items... and we can drill down further in work sessions if needed. 15:18:36 <_nadya_> should we visualize meters/ events and should we do data analysis. These questions should be raised again, imho 15:19:30 * idegtiarov agree with _nadya_ 15:19:31 _nadya_: is that a separate topic you want to propose? 15:20:19 _nadya_: based on the one line description above, it's probably something that should exist separately (like gnocchi) for now. 15:20:52 <_nadya_> gordc: no, I think it may be discussed on cdent's session, about the future 15:20:55 _nadya_: but i'ts something we can discuss in one of open discussoin slots or the final meetup slot 15:20:57 fabiog: I meant that what we would like to do with Ceilo in general, what is the mission of this module, etc. 15:21:08 _nadya_: ah i see. kk 15:22:17 ildikov: yes, but we need to propose a guideline, otherwise we won't achieve anything in 50 min. 15:22:53 fabiog: i think you and cdent can come up with some discussion points before hand to guide discussion 15:23:01 ildikov: it could be a list of questions too. to facilitate the discussion 15:23:20 fabiog: gordc: well, if we know already that it will take longer, then we could mark the free sessions as it is already marked as we can continue the dicussion there 15:23:26 gordc: cdent I will schedule a meeting to talk about it 15:23:35 fabiog: kk 15:23:55 cdent: you are in Ireland, right? 15:24:13 ildikov: i think i already did (unofficially)... it's in the brackets 15:24:23 _nadya_: that topic brings us far IMHO, but I agree that it belongs to cdent/fabiog's session to discuss the idea 15:24:35 UK, send me some email and we'll work out what to do 15:24:39 gordc: yeap, that is what I tried to refer to 15:25:13 ildikov: sure i'll change that slot to officially say "deal with cdent's madness" :) 15:25:40 NO COOKIES FOR GORDC 15:25:57 cdent: fabiog: an etherpad with a rough guideline prior to the session would be useful 15:26:08 yes, that makes sense 15:26:08 gordc: LOL :) 15:26:11 cdent: i'll just go to store.. /me hopes for sale. 15:26:52 fabiog: cdent: please use this: https://etherpad.openstack.org/p/liberty-ceilometer-identity 15:26:58 i assume there isn't one created already 15:27:50 fabiog: cdent: i forgot but do you think your topic links to any other teams? 15:27:53 ops? 15:28:02 there are the notes that I made when putting the thing on the spreadsheet, but those were supposed to be background to figuring out a guide, so I'll just link to that from this and do more outliney stuff on the new one 15:28:12 its partially in response to the ops stuff the prior day 15:28:32 as in "what strategies can we use to make the changes they require of us" 15:28:51 gordc: I think if we would bring this up on an ops session it would be an endless discussion 15:28:52 cdent: kk. i'm just asking because i can also make the session appear in other projects schedules... so there's more eyes 15:28:54 (with the general notion that breaking-things-up is a way to go for a variety of reasons) 15:29:22 ildikov: yeah... i took off original ops link because i was worried of that. we can play it by ear. 15:29:23 gordc the idea is for it to involve input from other people, so yeah, I guess so 15:29:46 It needs to be more more people than just us who are talking right now or why bother? 15:29:51 cdent: ok. i'll add tag. 15:30:28 for the other topics. please let me know if you'd like a link to another project. 15:30:44 gordc: also is relevant for billing people ... 15:30:59 cdent: in case of topics like this it usually not helps that much to have a huge amount of people involved from the beginning 15:31:27 fabiog: doesn't seem to be a 'billing' stream. so i think it might be covered by 'ops' 15:31:33 ildikov: I know that is usually the case but the entire point of doing this is to make it clear that people have a vision of how things are and should be that is not aligned 15:31:39 either with each other or with the user base 15:31:49 * cdent just gave away the secrete 15:31:51 whoops 15:31:54 ildikov: i think it might be ok. a 9am start tends to have a smaller audience anyways 15:32:35 cdent: everyone will have an opinion that where and what we should do, you will not find a matching one, so it's better to involve more people, when at least this team tends to agree 15:32:40 gordc: yeah, depends on the party success the night before .. 15:32:43 gordc: good point :) 15:32:53 this team _doesn't_ agree! 15:33:14 and we need to 15:33:15 cdent: right not it does not, but we should get there once, IMHO 15:33:16 thus the session 15:34:14 * gordc isn't overly concerned with complete agreement... just a common general direction. 15:34:54 gordc: this was the tends to agree part of my sentence :) 15:35:06 ildikov: :) 15:35:52 we all good? i will add the etherpads to the schedule within next 30mins... you can all post there afterwards 15:35:58 yup 15:36:17 gordc: I think yeap 15:36:32 cool cool 15:36:41 cdent i just noticed your etherpad. i'll use yours 15:36:53 i'll give it a minute but we'll move on 15:37:55 #topic recurring: Gnocchi status 15:38:05 jd: sileht: cdent: anything? 15:38:15 1.0.0 released 15:38:26 cool stuf 15:38:26 gordc: yes it is normal. See my email introducing Cheddar 15:38:27 gordc, I have rework the influxdb driver to work with the latest release binding binding 15:38:30 some nice noise on the os-dev list from tim bell and crew wanting to test it 15:38:38 ttx: cool cool. thanks. 15:38:56 cdent: no pressure. 15:39:07 indeed 15:40:39 #topic open discussion 15:41:55 so... we all good? :) i'll give it a min 15:42:43 see you later folks! 15:42:46 #endmeeting