17:01:14 #startmeeting Designate 17:01:15 Meeting started Wed Apr 16 17:01:14 2014 UTC and is due to finish in 60 minutes. The chair is kiall. Information about MeetBot at http://wiki.debian.org/MeetBot. 17:01:16 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 17:01:18 The meeting name has been set to 'designate' 17:01:25 Heya - who's about? :) 17:01:27 o/ 17:01:29 here 17:01:38 howdy 17:01:43 o/ 17:01:52 here 17:02:22 #topic Review action items from last week 17:02:29 here 17:02:51 2 items - 1 I was told was one after I added it, and other was to organize the hangout. So.. Both done. 17:02:54 o/ 17:03:03 #topic DNSpy update 17:03:12 I missed who added this item? :) 17:03:32 Joe added this 17:03:32 I added this 17:03:48 Unfortunately emmanuel is not here at the moment, as he has the best detail. 17:04:17 But from what he mentioned, notify works well with DNS Pythin 17:04:18 I know he is working through it and having some success… hasn't hit any major issues. 17:04:22 jmcbride: ah, will he be here before the end of the meet? If so, Let's come back to it/ 17:04:32 No not today 17:04:48 no, but my thought was mostly around what we want to do with dnspy and the diy 17:04:55 AXFR works as a client he needs to checkout as a server 17:05:00 vinod: Okay.. 17:05:20 I think he will be done soon and I thought I'd bring it up so we have what we need on diy to make a decision 17:05:26 kiall: How’s the DIY coming? 17:06:18 betsy: exactly as it was a few months ago - I'm confident it's relatively easy to implement, I think it's a *slightly* better option, but am open to both DIY and DNSpy 17:06:53 I would imagine the two could probably be swapped out without to much impact. 17:07:00 You might have noticed the # of reviews from me has jumped over the last week, finally managing to organize time etc :) 17:07:20 I know. I’m impressed 17:07:48 especially with the late nighters like yesterday :-) 17:08:00 jmcbride: yea, swapping one for the other should actually be easy enough 17:08:40 OK, I think we can probably move on. The main takeaway I think is the DNSpy might be an option to go with initially for mDNS then swap out later in the event we are not ready to address DIY yet. 17:08:47 Anyway - I'm personally open to both routes, I do think DIY is going to give us more control - but that's very much "It'll be useful at some point in the future" thing 17:09:20 jmcbride: yea, essentially the same as my thinking :) 17:09:26 sweet 17:10:08 Once the next BP (also the next topic) is implemented, we should be pretty close to being able to actually implement these choices 17:10:27 So - Anything else on this topic before we move on? 17:10:41 No 17:10:50 #topic Clarify details mdns-structured-rdata 17:10:55 #link https://wiki.openstack.org/wiki/Designate/Blueprints/MiniDNS#Update_the_Record_object_to_make_available_the_structured.2C_as_well_as_textual_representations_of_the_records.27s_RData 17:11:19 I wanted to clarify my understanding before I get started on this. 17:11:22 Currently designate/objects has an object for record. This object (designate.objects.Record) would provide from_text, to_text methods. designate.objects.base would provide a from_dict, to_dict methods. The various types of records supported ("A", "AAAA", "CNAME", "MX", "SRV", "TXT", "SPF", "NS", "PTR", "SSHFP") would each have a separate type that inherit from designate.objects.Record. Storage's get_record, find_record and find_records would be ab 17:13:56 vinod: Sorry, misundersood the first time a wrote a totally wrong comment.. Typing again ;) 17:15:15 So - Yes, for e.g. "MXRecord" would inherit from Record, it would provide (to|from)_text methods (for the API), and in the next BP, (to|from)_dnpy methods 17:16:00 Storage's methods would then be updated to accept/return the appropriate object, rather than the generic "Record" class 17:16:27 ok. the next part of my question 17:16:28 For storage's create_record, update_record to take in one of these record types (rather than a dict that as it is currently) and be able to create/update it, we would need to modify either the central and/or api layer to supply one of the record types rather than a dict. Should this be done as another blueprint? 17:17:32 Eventually, we should end up in a place where we can go recordset = storage.get_recordset(12345) ; recordset.to_dnspy(), which that would then be pushed out over the wire in response to queries etc 17:17:59 vinod: I actually consider that a bug in the o.m switchover, previous to the o.m migration, our own objects 100% passed over the wire to the API layer 17:18:44 So .. I do think that bug needs to be fixed so we can avoid unnecessary dict<-->object conversions where there not needed 17:19:03 I was actually looking at it about an hour ago, but haven't found the cause yet 17:19:16 kiall: o.m switchover? 17:19:27 oslo.messageing 17:19:48 OK. I thought it was oslo something, but couldn’t think what the m was for 17:19:51 The switch from the depreciated oslo-incubator "rpc" stuff to the replacement oslo.messaging lib 17:20:06 Sorry - should have explained :) 17:20:19 np 17:20:37 ok kiall i will let you fix that bug - since you have more knowledge there 17:21:47 Sure :) 17:22:36 Okay .. So, vinod - I assume this means you're going to take that BP? 17:22:42 yes 17:23:09 Cool - I'll get that bug sorted as soon as I can - Hopefully today. 17:23:14 also wanted to check - is Graham doing the api validation for the various record types? 17:24:01 vinod: not at the moment - just the current objects - but when that BP merges, I can update 17:24:01 asking since kiall mentioned something to this effect yesterday 17:24:05 mugsie was adding validation pieces to your objects stuff, which would let us centralize the validations into something that can be used by both the API and Central and mDNS etc 17:24:47 i.e. anything based on those objects could then be validated, including the RRType specific objects 17:25:09 yeah, it would just be updating schemas 17:25:57 Okay - Anything else on this topic before we move onto Open Discussion? (Short agenda today) 17:26:26 Everything clarified 17:27:02 #topic Open Discussion 17:27:41 Any other topics today? :) 17:27:46 Design time at the summit 17:28:00 Ah yes - I saw that got scheduled, but I cant see when it was scheduled for! 17:28:11 mugsie: did you get an email or anything? 17:28:16 Mugsie started on this topic and he and I have been looking into finding more time to review blueprints, etc. 17:28:38 kiall: nope, nothing 17:28:55 I assume I will get an email from Beth at some point 17:29:08 Mugsie: what was the name of the program/event you submitted it under? 17:29:20 Design Summit 17:29:28 OK. Another option for time is https://www.openstack.org/blog/2014/03/call-for-proposals-open-source-openstack-summit/ 17:29:33 yeah 17:29:44 I would say we go for it 17:29:54 "it’s an opportunity for these projects to hold small design summits with their community members." 17:30:12 The more time available, the better. I'll go ahead and submit. 17:30:17 It sounds like that's the same thing you did mugsie? 17:30:27 nope, its a new side track 17:30:32 not part of the design summit 17:30:39 we got offical design summit time 17:30:56 and this would be extra time -its a full 1/2 day sesion 17:31:10 How much time did you secure? 17:31:20 I can work wit jmcbride to put in an application for the new side session 17:31:32 I don't know - I have never been to a design summit 17:31:43 kiall: any idea how long thre design sessions go on for> 17:31:44 ?* 17:31:45 mugsie: there all different lengths from memory 17:32:07 Isay we go for it - the more exposure the better 17:32:10 I say* 17:32:15 summit.openstack.org/cfp/details/196 says "Scheduled"... I'm assuming that means there's a schedule *somewhere* 17:32:36 yeah - but I don't think it is public yet 17:32:43 they said they would let us know by friday 17:32:47 OK, lets go for it. Can we assume the usual suspects from HP, RedHat, eBay and Rackspace will make it? 17:32:57 me anyway 17:33:02 yes 17:33:56 jmcbride: yep, definitely 17:34:05 jmcbride: you free after this? we can put it in this evening if possible 17:34:46 #action Joe Work with mugsie to get an application in for https://www.openstack.org/blog/2014/03/call-for-proposals-open-source-openstack-summit/ 17:34:51 cool 17:34:58 yes, I'm free 17:36:16 anythign else? 17:36:31 I’m good 17:36:59 Not from me, I'm already semi-distracted right now (dragged into another conversation as I'm apparently the only person who can still login to the server.. heh) 17:37:14 :) 17:37:47 nothing from me 17:38:15 Cool, Okay.. vinod I'll ping you once I've figured out the bug :) 17:38:32 Thanks all :) See you in #openstack-dns .... 17:38:41 #endmeeting