14:00:51 #startmeeting neutron_upgrades 14:00:52 Meeting started Thu Oct 12 14:00:51 2017 UTC and is due to finish in 60 minutes. The chair is ihrachys. Information about MeetBot at http://wiki.debian.org/MeetBot. 14:00:53 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 14:00:56 The meeting name has been set to 'neutron_upgrades' 14:01:15 o/ 14:01:19 hi 14:01:43 o/ 14:01:49 #topic OVO patches 14:01:49 https://review.openstack.org/#/q/status:open+project:openstack/neutron+branch:master+topic:bp/adopt-oslo-versioned-objects-for-db 14:02:41 https://review.openstack.org/#/c/321001/ "objects: get, update and delete converted to Subnet OVO usage" 14:02:57 slaweq is still working on it. there are multiple comments to tackle. 14:03:14 next is https://review.openstack.org/#/c/495810/ "Use Agent OVO in agents_db and test_agents_db" 14:03:54 again, multiple comments there to address 14:04:06 ihrachys, sorry, this week i don't have time to work on it. Next week i'll back to it. 14:04:25 that's fine, thanks for the update 14:04:35 next is https://review.openstack.org/#/c/382037/ "Integration of Port Binding Level OVO" 14:05:04 again, slaweq is working on it, no one reviewed so if you have cycles, it may make sense to give him initial feedback 14:05:27 next is https://review.openstack.org/#/c/466237/ "Change join relationship between routerport and router" 14:05:48 I already +2d it, now waiting for someone to chime in with the second one (I don't think it will take a lot of time) 14:06:05 next is https://review.openstack.org/#/c/396351/ "Integration of Floating IP OVO" 14:06:06 is Kevin around recently? 14:06:11 lujinluo, I don't think so 14:06:14 if so, i can try to ping him 14:06:16 seems like he just left 14:06:16 i see 14:06:17 :) 14:06:41 though he was at indoctrination session at his new company so maybe was consumed 14:07:32 that's true.. if Kevin is not available, who else can we try to ping? garyk? 14:07:59 i mean, we still need at lease one more +2.. 14:08:03 garyk is usually quick to jump at those. we can also try haleyb or mlavalle since it's somewhat l3 related 14:08:23 ack. will contact them tmr or so 14:08:38 several people periodically look through the list of patches with one +2 so it shouldn't be a huge problem 14:08:53 good to know that! 14:09:40 so back to the floating ip patch, I see you respinned it lately. I will have a look during the week 14:09:49 thanks 14:09:59 anything pending there? 14:10:03 or it's all good? 14:10:12 i think zuul is broken? 14:10:22 cause jenkins gave +1 but zuul gave -1 14:11:03 other than that, i think it is all fine but it would be appreciated that you can take a closer look 14:11:31 I believe zuulv3 had some issue with devstack-gate yesterday 14:11:36 that they postponed migration to it 14:12:31 yeah, seems we need to give it a bit more time 14:12:59 lujinluo, I think it shouldn't even affect potential merging 14:13:07 it just votes -1 but doesn't gate 14:13:26 (unless they switched to it already, they change plans every day) 14:13:33 they == infra 14:13:39 lol, i see.. 14:13:44 to your question in https://review.openstack.org/#/c/396351/34/neutron/db/l3_dvr_db.py 14:13:57 about how it's possible that floating IP is removed in between lines 14:14:14 sorry not that one file... 14:14:28 https://review.openstack.org/#/c/396351/34/neutron/db/l3_db.py@1317 14:14:54 this is because line 1304 is executed under a different transaction than line 1317 14:15:18 see where the scope of the first transaction opened in line 1277 ends 14:17:23 hmm, i see. before ovo transition the same db object is carried to line 1319 14:17:56 right, but it's assumed to be already pre-cached 14:18:03 in the session 14:18:08 all its relationships 14:18:30 and 'apply_funcs' is not supposed to modify the model, just read attributes 14:18:48 in this case, if we cannot fetch any ovo object in line 1317, do we need to throw exceptions? 14:19:30 if floatingip is no longer there, there is no point to proceed to any attributes 14:21:26 in line 1306, we already update the port. I wonder why we just don't return the payload from the update. 14:22:10 maybe the dns integration code after update_port call changes something that requires triggering extensions again, but I doubt that since it's postcommit 14:23:05 i need to double check but i somehow remember that some methods in resource_extend.apply_funcs() do modify floatingip_dict 14:23:33 dict? yeah. but we should already have them triggered inside update_port 14:23:56 the result of update_port (that we currently just drop) should have everything already 14:24:04 (I assume) 14:24:15 ok, i will double check tmr.. 14:25:52 ok, I will too 14:26:00 next is https://review.openstack.org/#/c/501190/ "Integration of L3HARouterAgentPortBinding in ml2/drivers/l2pop/db.py" 14:26:23 I see Vu posted a new patchset as I asked 14:26:40 I actually started playing with it once but then something slipped and I forgot :) 14:27:23 LOL 14:27:40 looking at CI results, it seems to me there is no serious issue there, just unit tests need some refinement 14:28:10 I will have another look 14:28:26 next is https://review.openstack.org/#/c/501925/ "Use Agent OVO in l3_agentschedulers_db" 14:29:14 I left some comments there that Vu will need to follow up on 14:29:43 other patches in the queue seem to be not ready for reviews 14:30:34 #topic Other 14:31:07 I don't have anything to discuss beyond that, I think we make good progress, and I will try to review asap whenever patches are respinned 14:31:45 i don't have anything to discuss either 14:31:52 ihrachys, I have a small question 14:31:56 annp, shoot 14:32:17 Can we get list objects by filter a list of object id? 14:32:34 same as https://review.openstack.org/#/c/468265/39/neutron/services/logapi/common/db_api.py@119 14:33:12 I'm quite confused about that. :) 14:34:04 annp, it should work that way, yes 14:34:15 not everyone is aware we can do that 14:34:29 both in sqlalchemy code and, by transition, for OVO objects 14:35:01 ihrachys, yeah. Thanks. That's all from me 14:35:08 I commented there 14:35:15 ok great, let's call it a day then 14:35:19 #endmeeting