14:00:24 #startmeeting networking 14:00:29 hi 14:00:30 hi 14:00:30 Meeting started Tue Aug 1 14:00:24 2017 UTC and is due to finish in 60 minutes. The chair is jlibosva. Information about MeetBot at http://wiki.debian.org/MeetBot. 14:00:31 hi 14:00:32 hello everybody 14:00:32 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 14:00:34 The meeting name has been set to 'networking' 14:00:41 #topic Announcements 14:01:12 The pike 3 milestone happened last week 14:01:59 If you're working on a feature that didn't make it to p3 milestone and you think it should go to Pike, you can ask for a feature freeze exception 14:02:25 next one 14:02:35 dev mail on FFE http://lists.openstack.org/pipermail/openstack-dev/2017-July/120340.html 14:02:58 amotoki: thanks :) 14:03:14 PTG is happening next month in Denver, CO 14:03:33 please see email from Kevin: http://lists.openstack.org/pipermail/openstack-dev/2017-July/120270.html 14:04:04 there is a link to etherpad where, if you're gonna attend the event, you can write down your name. It also serves as a place to collect topics to discuss 14:04:32 The deadline to get travel support is this week 14:05:23 I also saw an email that User Committee nomination was started: http://lists.openstack.org/pipermail/openstack-dev/2017-August/120393.html 14:05:38 so if you feel like, you can self-nominate yourself :) 14:05:48 that's all from me, anybody has anything to announce? 14:06:15 in addition, PTL self-nomination week 14:07:02 http://lists.openstack.org/pipermail/openstack-dev/2017-August/120383.html 14:07:41 if nobody has other announcements to make, we can move on 14:07:57 #topic Blueprints 14:08:18 I'm not sure there is anything to discuss here as we're in FF :) 14:08:43 so unless anybody has anything to say about blueprints, we will skip this topic 14:09:13 ok 14:09:15 #topic Bugs and gate failures 14:09:25 amotoki: you were the bug deputy? 14:09:31 i was bug deputy 14:09:33 aah 14:09:38 haleyb: hi :) 14:09:53 * jlibosva mixed Jun and Jul 14:09:55 i think the wiki page is not updated recently :( 14:10:07 yeah, I'll update it after meeting 14:10:42 i think most bugs have been triaged, i'm going through the backlog now 14:11:34 the one that's still an issue is sometimes there are multi-node job failures, mostly non-voting jobs 14:11:46 https://bugs.launchpad.net/neutron/+bug/1707003 14:11:46 Launchpad bug 1707003 in neutron "gate-tempest-dsvm-neutron-dvr-ha-multinode-full-ubuntu-xenial-nv job has a very high failure rate" [High,Confirmed] - Assigned to Brian Haley (brian-haley) 14:11:56 i own that and continue to work on it 14:12:39 haleyb: are there any particular test that are failing? 14:13:02 jlibosva: it's sub-node setup failure, we talked about it in last week's CI meeting 14:13:39 ah, so more related to infra building stuff than neutron? 14:14:25 ok, if that has an assignee, then it's good :) 14:14:31 yes, seems so, there was a patch a while ago, but it needs updating. hopefully there is nothing in neutron 14:14:53 kevin has also been working on log file cleanup, https://bugs.launchpad.net/neutron/+bug/1707307 14:14:53 Launchpad bug 1707307 in neutron "Neutron log obesity epidepmic" [High,In progress] - Assigned to Kevin Benton (kevinbenton) 14:15:28 has a few patches out, but if anyone notices other non-essential log messages feel free to remove them 14:15:55 looks like ovsdbapp is chatty :) 14:16:08 slightly related to non-voting jobs instability, I sent an email last week calling for a help with one job: http://lists.openstack.org/pipermail/openstack-dev/2017-July/120294.html 14:16:30 did you get any help? 14:16:38 it reveled some real issues in trunk/ovs fw, I'm looking at the bug 14:16:46 https://bugs.launchpad.net/neutron/+bug/1707339 14:16:46 Launchpad bug 1707339 in neutron "test_trunk_subport_lifecycle fails on subport down timeout" [High,In progress] - Assigned to Armando Migliaccio (armando-migliaccio) 14:17:09 haleyb: yep, slawek started looking at qos and armax looked at trunk 14:17:54 but it would be good to get moar help, so if anybody has spare cycles, there are some failures that need love ;) 14:18:11 haleyb: thanks for updates about bugs 14:18:17 i did just notice a spike in the functional job, about 30% now 14:18:27 i hadn't refreshed grafana this morning 14:18:42 http://grafana.openstack.org/dashboard/db/neutron-failure-rate?panelId=8&fullscreen 14:18:52 func tests are timeouts I believe; I think jlibosva was going to recheck a testing patch to trigger a worker killed 14:19:23 yep, I got a hit on my patch but that didn't revealed the killer 14:19:32 it looks like the kill was not called via CLI 14:19:48 jlibosva: is there a bug yet? 14:19:51 patch here: https://review.openstack.org/#/c/487065/ 14:20:10 haleyb: probably not, we should open one 14:20:29 there is this etherpad ihrachys created for all func failures: https://etherpad.openstack.org/p/neutron-functional-gate-failures-july 14:21:13 any ideas how to nail down what is killing process? 14:21:35 the weird thing is that it always happens during firewall tests 14:21:36 you say smth not calling .execute() kills testers? 14:21:57 something not calling "/usr/bin/kill" 14:22:08 which execute() is a subset of 14:23:20 ok maybe it's killall :) 14:23:26 do we log all commands? 14:23:35 in tests we do 14:24:32 ok, so it's not just kill that you capture 14:24:39 anyway, I don't think we will make progress right now 14:24:45 let's report the bug as first step 14:24:49 I will take it 14:24:51 yeah, sounds good 14:24:53 ihrachys: thanks 14:25:02 I was just wondering if anybody has a good trick for it :) 14:25:29 ok, let's move on from functional failures, we can talk about that at Neutron CI meeting in ~1.5hrs 14:25:52 anybody wants to be a bug deputy for this week? 14:27:02 it looks like it's not the most favourite role :) 14:27:31 i will put names on dart board and throw... 14:27:50 haleyb: tell me who won 14:28:12 or we can open an rfe to openstack bot, to pick names randomly :-p 14:28:18 if nobody wants it, then I'll take it 14:28:34 #info jlibosva is a bug deputy for the week of Jul-31 14:28:35 maybe worth talking to PTL if there is a problem of getting candidates 14:28:58 good idea 14:29:01 let's move on then 14:29:15 #topic Docs 14:29:35 amotoki: do you have any updates? I don't see mlavalle here 14:29:55 we have good progress on doc-migration :) 14:30:20 neutron itself have completed the effort. 14:30:38 we have some projects still having patches on doc-migration 14:30:39 See 14:30:48 "Missing URLs" at https://doughellmann.com/doc-migration/ 14:30:52 nice :) 14:31:04 we only have four projects remaining 14:31:19 I believe time will solve it 14:31:49 I already post patches on these remaining things. 14:32:02 thats' the update on doc-migration 14:32:08 sounds good, thanks 14:32:33 amotoki: do you think you should reach out to people from those four projects or are you good doing it yourself? 14:32:34 at some time, we need to discuss on the future of LBaaS document with octavia team 14:32:58 jlibosva: i already sent the required patches to the four projects 14:33:16 ok :) 14:33:30 Anybody has any other updates regarding docs? 14:34:00 if not, we can move to osc 14:34:02 #topic Transition to OSC 14:34:11 amotoki: hi :) how are the tags and routers doing? 14:34:25 both landed :) yay 14:34:31 wow 14:34:44 so does it mean all resources now have support in osc? 14:35:01 at now, we've migrated all regarding neutron stuff 14:35:18 the remaining part is about advanced services (as OSC plugin) 14:35:46 I think we can officially discuss and declare when we drop neutron CLI 14:36:23 maybe could be a topic at etherpad for the ptg? :) 14:36:31 will add it 14:36:36 amotoki: thanks 14:36:48 amotoki: and thanks for updates and good news :) 14:36:59 I think we can move to the lib 14:36:59 thanks everyone 14:37:10 #topic neutron-lib 14:37:17 boden: hello :) 14:37:21 hi 14:37:33 just a few minor FYIs 14:38:11 now that neutron-lib 1.9.1 is out, you should see some “consumption” patches coming down the pipeline 14:38:22 there are a number of them out already; and more in the works 14:38:43 rather than sending a note to the ML for each, I was just going to get them all out there and send a single note 14:38:58 so, any help reviewing is appreciated :) 14:39:06 since we are in freeze mode, we probably don't rush with them 14:39:23 ihrachys: ah, so they won’t be able to get in for pike? 14:39:43 isn't 1.9.1 the one that's gonna be released with pike? 14:39:54 boden, it's probably a question for PTL, but my understanding is that now we allow features with exceptions approved by drivers team + bug fixes 14:40:13 jlibosva: yes that’s my understanding, but here I was talking aobut neutron and other project patches (not lib) 14:40:19 ihrachys: ack thanks 14:40:19 that being said, I saw patches landing that are of other type 14:40:27 boden: aah, sorry, I misunderstood 14:40:28 I assume those are mistakes 14:40:42 ihrachys: I’ll follow up with kevin later 14:40:53 refactoring is really a gray zone 14:41:10 if it help us backports to Pike a lot, it might be acceptable. 14:41:32 anyway it loosk like a question to kevin 14:41:38 that’s really all I have that’s worth taking everyones time 14:41:47 I'll have a chat with him tomorrow EMEA morning, I'll deliver the message 14:41:55 boden: thanks for updates 14:42:05 anybody has anything to neutron-lib? 14:42:40 #topic open discussion 14:43:03 any topics to discuss here or we can end the meeting earlier :) 14:43:30 go for it 14:44:27 thanks everyone for coming :) 14:44:31 #endmeeting