15:01:12 #startmeeting neutron_upgrades 15:01:12 Meeting started Mon Jun 6 15:01:12 2016 UTC and is due to finish in 60 minutes. The chair is ihrachys. Information about MeetBot at http://wiki.debian.org/MeetBot. 15:01:13 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 15:01:16 The meeting name has been set to 'neutron_upgrades' 15:01:48 bauzas, dansmith: honestly, this is what I feel like doing: let me rebase cdent's patches for the object work and put them in a state I'm happy with. Let cdent be able to focus on the REST API changes that will be needed after the amendment to the generic-resource-pools spec, and let us not try to boil the ocean too much this week and just try to get agreement on the definition of the 15:01:54 InventoryList and AllocationList objects. 15:02:01 hi upgrades folks! 15:02:37 o/ 15:03:16 upgrade-o-nauts 15:03:26 or upgrade-o-nuts 15:03:33 meh :) 15:03:36 john-davidge_: howdy! 15:03:42 korzen: hello 15:04:02 ihrachys, hello 15:04:08 I've just joined 15:04:19 * ihrachys waves at rossella_s 15:04:25 hello sorry for being late 15:04:53 np, we are starting late :) 15:05:02 ok, let's get it rolling 15:05:05 I am a lucky girl 15:05:06 #topic Actions from the last meeting 15:05:16 "korzen to send a breaking patch to validate multinode grenade job catches rolling breakages" 15:05:21 korzen: did it happen? 15:05:48 ihrachys, sadly not yet but I have a patch ready for push 15:06:01 I will push it and write the report on ML 15:06:20 nice 15:06:27 "korzen to prepare first bi-weekly status update for ML" 15:06:30 that actually happened! 15:06:45 i can confirm. good job korzen! :) 15:06:45 general idea is to modify the RPC method interface for neutron server and agent communication in incompatible way... 15:06:48 #link http://lists.openstack.org/pipermail/openstack-dev/2016-June/096512.html 15:07:06 korzen: yeah, like drop an attribute from notification payload 15:07:17 "ihrachys to update neutron meeting open agenda with ovo status update" 15:07:26 that did *not* happen, but I have time till tomorrow ;) 15:07:41 I will refer to korzen's report for the most part 15:07:48 "everyone to switch from 'ovo' topic to bug/XXXX or bp/XXXX where XXXX is targeted for Newton" 15:07:59 * ihrachys looks at gerrit 15:08:01 #link https://review.openstack.org/#/q/topic:ovo+status:open 15:08:19 I see some stuff that is relevant to the effort but has wrong topics. 15:08:37 I guess I need to walk through the list and make sure topics are correct 15:08:47 ihrachys, I should update my patches...but not sure if it's worth reviewing them, they are still kind of WIP 15:08:59 rossella_s: if it's WIP, it's ok to keep it as-is for now 15:09:10 the main point of changing is getting attention from reviewers 15:09:19 ihrachys, that's what I thought 15:09:21 I am about to split my Subnet OVO eval patch and submit it with proper bp 15:09:23 if we don't want that for a patch, it may be better to keep it under radar 15:09:58 #action ihrachys to walk thru topic:ovo and change topics where applicable 15:10:18 #topic Partial Multinode Grenade 15:10:51 I haven't heard any progress on that one. sc68cal, did you have a change to look at expanding the testing to DVR and other stuff from that etherpad? 15:10:57 *chance 15:11:18 sorry, haven't 15:11:36 np, let's just move on 15:11:42 #topic Object implementation 15:12:09 ok, I was looking at the blocker from API layer where I thought page_reverse is broken 15:12:19 it could not be less trues 15:12:19 *true 15:12:49 turned out it's my brain having a bug, and I just incorrectly interpreted a goal of page_reverse 15:13:00 so long story short, no bug and we just need some api test coverage 15:13:05 #link https://review.openstack.org/318270 some tests 15:13:24 and with that, I assume api test coverage is enough for us to start switching core resources to objects 15:13:48 we just need to cover them with existing tests, which is a matter of inheritance. 15:14:21 and btw, we landed sorting/pagination support in objects! 15:14:22 #link https://review.openstack.org/300055 15:14:29 ihrachys, cool 15:14:30 ! 15:14:32 so from that perspective, we should be good to proceed 15:14:56 now the focus probably moves to patches that adopt existing resources in db code 15:15:04 ihrachys, I will catch up with the reviews 15:15:04 * ihrachys will get back to those patches this week. 15:15:17 rossella_s: thanks, I would really love to get those api tests in 15:15:35 ihrachys, will do later tonight 15:16:00 any interesting patches to discuss in this regard? do we have mergeable pieces in review that adopt objects? 15:17:01 ihrachys, no news on my side unfortunately...we should get the sec group patch in 15:17:30 rossella_s: does it have db side? or just object definitions/ 15:17:32 I will synch with sayali and see if she can work on it or if we should find volunteers 15:17:54 rossella_s: thanks for handling that! we should make sure no piece is stuck to deliver as planned. 15:18:06 ihrachys, it's just the ovo introduction 15:18:51 rossella_s: it's my thinking that we should focus on object adoption, would be great to see objects used in code and not just hanging out there. 15:19:12 ihrachys, I totally agree, I commented that on some ovo patches 15:19:15 without adoption, we risk to chase new features. 15:19:21 agreed 15:19:22 ok, good to be on same side 15:19:34 I will publish some Subnet usage this week 15:19:43 korzen, great 15:20:19 I know that jlibosva was looking at refactoring resource hooks our of common_db_mixin but I don't see anything in review queue so far 15:20:49 I heard that he may need to sync with korzen on that one. 15:20:54 ihrachys, I've seen that it was added to subnetpool 15:21:20 oh right. 15:21:22 #link https://review.openstack.org/#/c/300056/4/neutron/objects/base.py 15:21:49 something we may consider moving out of this patch since it seems like a separate piece of code 15:21:49 yes 15:22:00 for people who want to help I'd like to ask for help for the port extensions and the port object...instead of adding new ovo I think it's more productive to finish what we have 15:22:26 rossella_s: absolutely. what's the base link to port object? 15:22:46 #link https://review.openstack.org/253641 15:23:08 ok, do we have a list of todos in this regard? 15:23:13 we still need some work for the binding object... 15:23:19 ihrachys, good idea I can update the team page 15:23:27 maybe even putting it in commit message could help to understand the scope 15:24:01 ihrachys, another good point, will do 15:24:30 rossella_s: or this way. not sure about others, but I don't see the whole picture for the object as you probably do, so documenting your vision would be the 1st thing to do 15:24:54 #action rossella_s to come up with specific list of TODOs for port object 15:25:02 ihrachys, right, I will do that because I realized that I want to work on that but I keep postponing 15:25:12 ihrachys, thanks for the TODO 15:25:48 any more specific stuff for objects that would require team attention? 15:26:27 ok, let's move on 15:26:32 #topic Open Discussion 15:26:48 one thing to note, seems like mid-cycle date is moved 15:26:55 rossella_s: does it cover your concern with the original date? 15:27:22 ihrachys, yes the email was clear 15:27:34 ihrachys, still don't know if I will be able to attend, you people? 15:27:58 rossella_s: nope, I still did not get ack from my management, and I would need a visa for that. we'll see. 15:28:24 i didn't ask for permission either 15:28:27 I guess we may have some focused discussion there on objects if we'll have some and if people attend; otherwise, no big dea. 15:28:29 *deal 15:28:56 indeed, it would be useful to know who's attending to understand if we can have that discussion 15:29:07 I assume HenryG will ;) 15:29:26 ok folks, anything more to cover? 15:29:34 or we cut it short :) 15:30:06 the latter ;) 15:30:11 thanks folks! 15:30:11 #endmeeting