14:01:00 #startmeeting neutron_upgrades 14:01:01 Meeting started Thu Oct 26 14:01:00 2017 UTC and is due to finish in 60 minutes. The chair is ihrachys. Information about MeetBot at http://wiki.debian.org/MeetBot. 14:01:03 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 14:01:06 The meeting name has been set to 'neutron_upgrades' 14:01:14 o/ 14:01:16 lujinluo, TuanVu, annp o/ 14:01:23 hi 14:01:28 Hi everybody :) 14:01:38 ihrachys: are you getting better now? 14:01:52 yes, same question 14:01:58 yeah, I am mostly back on track health wise. not work backlog wise but I am getting there. 14:02:04 thanks for asking. 14:02:15 happy to know that ;) 14:02:17 great 14:02:40 i'm happy to hear that ihrachys 14:02:44 :) 14:02:52 thanks :) 14:02:54 #topic OVO patches 14:02:58 https://review.openstack.org/#/q/status:open+project:openstack/neutron+branch:master+topic:bp/adopt-oslo-versioned-objects-for-db 14:03:05 walking top to bottom... 14:03:15 https://review.openstack.org/#/c/501925/ "Use Agent OVO in l3_agentschedulers_db" 14:03:51 TuanVu, you respinned lately. but it's ready for reviews now? 14:04:01 yes, it is 14:04:27 ihrachys, this patch is waiting you :) 14:04:27 thanks for your review, I've already updated the patch as your suggestion 14:05:01 it has been able to pass all the tests and if you don't mind, please help to take a look 14:05:25 thank you, An-san for your great help on this patch, too 14:05:55 ok, thanks a lot, as usual, I start to build a list for OVO reviews for today 14:06:09 next is https://review.openstack.org/#/c/396351/ "Integration of Floating IP OVO" 14:06:38 it's also respinned lately, should be ready I guess 14:06:48 we will need https://review.openstack.org/#/c/512908/ before looking at floatingip. it fixes your concern of passing None to get_objects() 14:07:10 please check that first. 14:07:19 nice. so we had a bug in base OVO classes? 14:07:35 or is it a problem with all queries? 14:07:37 tuanvu, you're welcome. :) 14:07:44 more specific in db/_model_query.py 14:07:56 neutron/db/_model_query.py is common to a lot of code, not just OVO 14:08:18 so the bug applies to all queries.. 14:08:31 cool, I will have a look asap 14:08:48 thanks ;) 14:09:04 next is https://review.openstack.org/#/c/407868/ "Integration of (Distributed) Port Binding OVO" 14:09:31 it failed some dvr tempest tests 14:09:38 not ready yet 14:10:05 but could you check https://review.openstack.org/#/c/512944/ ? 14:10:13 the LIKE statement bug 14:10:30 I need to use LIKE get_objects in the port binding patch 14:12:15 ok. I removed an unneeded __init__ there and +W'd it just now 14:13:04 cool. thanks 14:13:18 then we can revisit port binding later 14:13:18 as for the port patch itself, you say it's not ready 14:13:27 correct. 14:13:33 next is https://review.openstack.org/#/c/501190/ "Integration of L3HARouterAgentPortBinding in ml2/drivers/l2pop/db.py" 14:13:50 An-san has helped to solve concern from Miguel 14:14:04 it's ready for review now 14:14:28 I was slow to reply, and seems that Miguel was quicker :) 14:14:40 I am a slug lol 14:14:52 LOL 14:15:19 no way 14:15:31 except the issue already fixed, are there any more issues there? 14:15:43 we really appreciate your in-depth review 14:16:03 no problem at this moment 14:16:14 ok, will look again 14:16:30 thanks :) 14:16:38 next is https://review.openstack.org/#/c/495810/ "Use Agent OVO in agents_db and test_agents_db" 14:17:43 all the concerns from Slawek and Miguel have been solved 14:17:52 it's ready for review now 14:17:55 seems like it's ready to have another look. I see you listed tests that fail without the json workaround 14:18:04 I've also added list of tests that fail 14:18:10 yes 14:19:42 I will have a look, I would like to play with a failing test case locally to understand how we get there 14:19:53 next is https://review.openstack.org/#/c/321001/ "objects: get, update and delete converted to Subnet OVO usage" 14:20:14 slaweq respinned it again lately, I expect it to be ready for merging since the last time it was small things 14:21:11 other patches are in conflict/wip 14:21:23 do we need to discuss any of those? any blockers there? 14:22:04 it would be appreciated if you can check the port binding patch 14:22:22 i had the same dvr tempest failures about 7 or 8 months ago 14:22:30 which one is it? 14:22:40 but if i run tempest locally, nothing breaks.. 14:22:53 https://review.openstack.org/#/c/407868/ 14:23:04 the failed tempest tests are all related to dvr 14:23:43 not sure if i am running tempest in the correct way 14:24:40 do you configure for dvr though? 14:24:46 I mean, neutron itself 14:25:06 ouch, i think no.. 14:25:26 will update the configuration file and try again 14:25:44 :( thanks for the heads up 14:26:19 yeah, the error is specific to dvr port binding model: http://logs.openstack.org/68/407868/63/check/legacy-tempest-dsvm-neutron-dvr/4b005ac/logs/screen-q-svc.txt.gz?level=TRACE#_Oct_25_07_47_50_385819 14:26:35 so you need to instruct neutron to use dvr for routers to hit it 14:26:56 ack. Will update my devstack tmr 14:27:34 ok. is there anything else in the list that needs specific attention? 14:27:59 no more from me 14:28:52 TuanVu: annp you two? 14:29:01 hmm 14:29:18 regarding to https://review.openstack.org/#/c/507772/ 14:29:31 Use Network OVO in db_base_plugin 14:29:53 there’s problem with missing RBAC attribute for Network 14:30:00 honestly, I don’t have much experience with RBAC, so really appreciate any help 14:31:21 AttributeError: 'Network' object has no attribute 'network_id' 14:31:31 that's in one of unit test failures 14:32:00 yes 14:32:55 well, shouldn't you then use 'id' instead? 14:33:05 in https://review.openstack.org/#/c/507772/4/neutron/objects/network.py@263 14:33:37 though I thought that shared attribute is driven by a metaclass 14:33:38 +1 14:34:12 thanks, I'll update it 14:34:30 tomorrow I'll have a closer look 14:34:33 we have https://github.com/openstack/neutron/blob/master/neutron/objects/rbac_db.py#L37 that should implement 'shared' attribute 14:34:51 I think it may even just work for you, not sure why _load_shared is needed in the patch 14:36:15 but the code itself is incorrect. I would suggest to start with pep8, then py27 - this error, maybe completely remove _load_shared and see how it behaves. the assumption was that network object already supports 'shared' attribute. 14:36:29 if not, we may have a bug in existing rbac metaclass 14:37:08 I mean this metaclass: https://github.com/openstack/neutron/blob/master/neutron/objects/rbac_db.py#L268 14:37:09 thank you, Ihar. I'll follow your suggestion 14:37:12 or mixin above it 14:37:15 ok 14:37:54 anything else in the list where you are puzzled? 14:38:22 nothing else from me 14:38:32 me too 14:38:44 ok we are good then it seems 14:38:52 thanks folks for joining that late. 14:39:08 btw, the meeting time, would -1h work better for you? 14:39:15 thanks a ton for your great help, Ihar 14:39:21 yeah, that would be better 14:39:24 I think it would, for you are in Asia 14:39:28 ihrachys, thanks for your guidance and revew. 14:39:29 but won't that be too early for you? 14:39:36 yes, I agree 14:39:48 ok, I have a slot freed before that one, I will have a look if there is a meeting channel free at that time 14:39:55 no it's fine for me, I start at 6am :) 14:40:00 oh one more thing besides the meeting time 14:40:07 wow, that is really early! 14:40:10 no guarantees that we will move, but I will look into it and update you 14:40:21 lujinluo, what's the other thing? 14:40:21 sure, thanks for the efforts ;) 14:40:34 my boss assigned another colleague to help with OVO, hungpv 14:40:37 thank you, Ihar 14:40:42 He will pick up router I guess 14:41:17 with router being picked up, i guess we will have owners for all ovo objects, right? 14:41:24 oh that's very cool. looking forward to see even more patches. 14:41:44 all major I believe. I will have to review the list again to see if we missed anything. 14:41:48 Hi, yes I'll take care of router patch 14:42:07 hungpv, heya, it's a pleasure to have you onboard 14:42:12 And hope that can upload for review next week 14:42:22 cool! 14:42:26 Thank you all for your kind 14:42:48 that's all from me! 14:43:01 ok. I guess that's it. we'll meet in gerrit and next week same time, unless I am able to move the slot earlier. 14:43:05 cheers! 14:43:07 #endmeeting