17:01:02 #startmeeting Designate 17:01:03 Meeting started Wed Oct 5 17:01:02 2016 UTC and is due to finish in 60 minutes. The chair is mugsie. Information about MeetBot at http://wiki.debian.org/MeetBot. 17:01:04 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 17:01:06 The meeting name has been set to 'designate' 17:01:09 #topic Roll Call 17:01:11 o/ 17:01:12 o/ 17:01:14 O/ 17:01:16 o/ 17:01:34 Kiall: federico3 pglass ping 17:01:57 #topic Announcemnets 17:02:42 so, RC2 was tagged, and will be the release arifact 17:03:06 #topic Action items from last week 17:03:12 Kiall timsim - review https://review.openstack.org/#/c/374308/ 17:03:19 done 17:03:32 mugsie bp ae6e84810033a02a8c4e6d729958333c682e8b4a Resolve description parameter in v2:set floating ip 17:03:35 #action mugsie bp ae6e84810033a02a8c4e6d729958333c682e8b4a Resolve description parameter in v2:set floating ip 17:03:46 #topic Bug Triagge 17:03:59 o/ 17:04:03 timsim: :) 17:04:05 https://bugs.launchpad.net/designate/+bug/1627941 I think is waiting on mugsie 17:04:07 Launchpad bug 1627941 in Designate "Zone transfer accept is forbidden for the non admin users " [Undecided,New] - Assigned to Graham Hayes (grahamhayes) 17:04:22 Now for actual new ones: 17:04:23 https://bugs.launchpad.net/designate/+bug/1630666 17:04:24 Launchpad bug 1630666 in Designate "monasca-statsd is a non-optional import even when disabled" [Undecided,New] 17:04:47 yeah 17:04:55 I saw that just before the meeting 17:05:12 I kinda think not installing our requirements.txt is bad 17:05:37 Yeah isn't it better to instal something you don't need then to not have a thing you might need and have to go hunt it down? 17:05:55 I would have thought so 17:05:57 Yea, but it probably should be optional.. kinda like pymysql is 17:06:02 I've had to do that for some memcached thing before. 17:06:06 and that 17:06:09 and it suckss 17:06:20 Because they you have to explicitly spell it out and pin a version in automation 17:06:29 Yeah 17:06:49 dmsimard: is the reporter 17:06:55 Yea ... We could use setup.cfg optional requirement groups.. but openstack infra isn't set up to handle them 17:07:03 e.g. no auto updates as requirements versions change 17:07:26 hey, just joined 17:07:30 I don't have much time, have to step away 17:07:38 yeah - I perfer to have this in requirements, as it is under development, and not as an option 17:07:38 I'm just saying this sounds like it should be an optional backend.. just like when I install Glance, it doesn't require me to install Ceph or Swift because my choice is the file backend. 17:07:47 as we dont have anything else that can slot in there right now 17:07:48 Pretend designate implements metrics collections with, I don't know, gnocchi/ceilometer and such. Will you require that as well ? 17:07:56 It's not really requiring that you install that hough, just a client? 17:08:24 dmsimard: Yea, makes sense... We do the same with e.g. mysql. But.. Trying to find a balance between just works, works with -infra tools, and least footprint is .. yea ;) 17:08:59 Same with keystoneauth1/keystonemiddleware, python-memcached 17:09:07 One danger with the extra req is the extra deps it might bring with it 17:09:08 I'll leave that up to you, just user/developer/deployer feedback :) 17:09:30 I have to step out for now but I'll be available to discuss later if you have any questions 17:09:36 For me, I'd love to see the requirements process handle optional requirement groups - that solves this problem "the right way" 17:09:39 OK, thanks dmsimard 17:09:41 But generally, adding clients doesn't bother me, especially if it keeps deps out config management 17:10:13 Plus, it's currently required in the code. So it's not really optional. That should probably change, but still. 17:10:13 yeah - I can see both sides 17:10:46 Why don't we tag this low, no release, and see if requirements folks have plans around optional groups 17:10:48 what do we think? move the bug to "this should be optional" ? 17:10:55 Kiall: +1 17:11:01 Yeah +1 Kiall 17:11:24 +1 17:11:29 Who wants to do that? 17:11:34 OK, - do what Kiall suggested 17:11:37 Give them an action 17:11:41 I can talk to -requrements 17:11:55 Alright 17:11:56 #action mugsie talk to requirements team about optional requireemtns 17:12:03 I closed this out before the meeting https://bugs.launchpad.net/designate/+bug/1628800 17:12:05 Launchpad bug 1628800 in Designate "Designate Blacklist :Blacklisted zones patterns are not checked while create zone using admin." [Undecided,Invalid] 17:12:20 +1 17:12:22 https://bugs.launchpad.net/python-designateclient/+bug/1630186 17:12:23 Launchpad bug 1630186 in python-designateclient "'openstack zone list --all-projects' failing with unexpected exception." [Undecided,New] 17:12:28 Seems legit 17:12:28 policy allows it, you can customize if needs be 17:12:29 -_- 17:12:54 tahts new 17:13:05 and legit, as timsim says. 17:13:06 high, triaged 17:13:19 Though - I'm nearly sure we've seen that one before and fixed it. 17:13:23 cool. That's it 17:13:25 did we ? 17:14:15 I think so - but I don't remember what we were talking about 5mins ago, so..... 17:14:20 #topic Backports - Kiall 17:14:26 #link http://paste.openstack.org/show/584510/ 17:14:32 Does ^ look complete? 17:14:40 yeah 17:14:42 its been slow 17:14:43 As usual, take a few, nominate backports 17:14:52 277dcae Merge "Infoblox: Reverse lookup zone creation fails" 17:14:55 (already done) 17:15:00 I think that ios back 17:15:07 Yeah, and that's probably it, yeah? 17:15:08 thats it really 17:15:23 Yep - I think that's it, nothing else warrants a backport in that list./ 17:15:36 #topic Design Summit 17:15:42 #link https://docs.google.com/spreadsheets/d/1TQ-RSlbiBBEclkonIbfUP7R1ExZSJylF1uiEKV2G_Cw/pubhtml?gid=1107826458&single=true 17:15:53 this is the allocation of rooms 17:15:55 we have 6 17:16:01 one is a big room 17:16:13 so the usual "meet DNS team" 17:16:20 and then we have 5 40 mins slots 17:16:32 #link https://etherpad.openstack.org/p/designate-ocata-summit-planning 17:16:39 ^ is pretty light 17:17:17 We can nominate them as "Working Sessions" - I saw a few teams do that last time 17:17:27 but I would like a few named sessions 17:17:39 there is on for the NS propegation 17:17:41 one* 17:17:45 One thing that might be cool is queuing work outside of retries 17:18:21 That was kind of in the worker spec, being able to queue different types of things independently and have logic to work on them 17:18:28 But there wasn't a simple solution 17:18:43 yeah. it is also a bigger change than I would like to see in Ocata 17:18:49 Yeah, worker model enables it. The job spec further enables it as well 17:19:10 Just throwing it out there! 17:19:25 :) 17:19:46 any other suggestions? 17:20:20 Beer! 17:20:27 OK, if people think of them, ping me with them 17:20:29 soon 17:20:30 :) 17:20:38 What's the deadline mugsie ? 17:20:42 I will be setting the schedule real soon 17:20:42 * elarson isn't going to be there so don't listen 17:20:56 friday AFAIK 17:21:00 cool 17:21:04 I'll try and think of more things 17:21:07 ++ 17:21:46 #topic Pike PTG (Project Teams Gathering) 17:22:04 We have been asked if we will be attending the first PTG 17:22:10 it will be in Austin 17:22:12 nope 17:22:17 Atlanta 17:22:28 Feb 20-24 17:22:44 it is envisaged that this will replace Mid Cycles 17:23:06 Hm. I'm betting you guys would have a hard time getting there? 17:23:20 If I go, I will be paying my own way 17:23:52 go, talk to people internally, and find out if you can go 17:24:05 I need to give the foundation an answer by the 16th 17:24:21 so I will be pinging you about it 17:24:43 Alright, we'll give it a shot. 17:25:01 cool 17:25:06 #topic Open Discussion 17:25:14 any off agenda items? 17:25:28 going once 17:25:33 twice 17:25:41 thrice 17:25:45 gone! 17:25:48 #endmeeting