14:00:19 <jlibosva> #startmeeting networking
14:00:23 <openstack> Meeting started Tue Jul 18 14:00:19 2017 UTC and is due to finish in 60 minutes.  The chair is jlibosva. Information about MeetBot at http://wiki.debian.org/MeetBot.
14:00:25 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
14:00:27 <openstack> The meeting name has been set to 'networking'
14:00:28 <jlibosva> hi everyone!
14:00:41 <boden> howdy
14:00:45 <jlibosva> we're gonna do neutron-lib first today :)
14:00:49 <jlibosva> #topic neutron-lib
14:00:54 <jlibosva> boden: hello :)
14:00:59 <emagana> hello all networking folks!
14:01:04 <jlibosva> emagana: welcome!
14:01:07 <boden> hi, and thanks for beng flexible on the time slot for lib
14:01:16 <boden> only 1 thing I want to mention quick
14:01:27 <trevormc> o/
14:01:31 <hichihara> hi
14:01:35 <hoangcx_> o/
14:01:43 <boden> last week we discussed a new release for neutron-lib… 1.8.0 got released over the weekend perhaps a little too quickly
14:02:00 <boden> so we are proposing a 1.9.0 to pick up some of the latest changes
14:02:00 <emagana> jlibosva: Just stopping by to read what's going on. It has been a while  :-)
14:02:07 <boden> #link https://review.openstack.org/#/c/484475/
14:02:27 <boden> feel free to weigh in on that patch if you have comments
14:02:30 <emagana> maybe I could start fixing some bugs.. anything about typos.. that's my style!
14:02:36 <jlibosva> boden: is there gonna be yet another release before pike m3 cut?
14:02:55 <boden> I think this week is our deadline, so I’m hoping not
14:03:11 <jlibosva> ok :-/
14:03:43 <boden> that’s really all I have.. otehr than thanks to all who have been helping with lib :)
14:03:53 <jlibosva> boden: thanks for updates
14:03:54 <amotoki> for clarification, this week is a last moment for release for non-client libs and neutron-lib is one of them.
14:04:09 <amotoki> (as boden and jlibosva already said so)
14:04:32 <jlibosva> let's move to announcements then
14:04:38 <jlibosva> #topic announcements
14:05:25 <jlibosva> so what amotoki said above :) this week is a deadline for releases of non-client libs
14:05:35 <jlibosva> I don't have any announcements
14:05:42 <jlibosva> does anybody have any?
14:06:19 <jlibosva> let's move on then
14:06:21 <jlibosva> #topic blueprints
14:06:37 <jlibosva> does anybody have a blueprint that requires attention?
14:06:50 <jlibosva> if not, I have one :)
14:07:14 <jlibosva> I'm working on https://bugs.launchpad.net/neutron/+bug/1580880
14:07:16 <openstack> Launchpad bug 1580880 in neutron "[RFE] Distributed Portbinding for all port types" [Wishlist,In progress] - Assigned to Jakub Libosvar (libosvar)
14:07:30 <jlibosva> which requires an api definition to neutron-lib https://review.openstack.org/#/c/484389/3
14:07:43 <jlibosva> thanks boden and hichihara for being very quick at reviewing
14:07:59 <jlibosva> boden: hichihara do you think it has a chance to get into 1.9.0 proposed release?
14:08:21 <hichihara> jlibosva: Of course
14:08:31 <jlibosva> I'll try to address comments between this mtg and neutron CI
14:08:43 <jlibosva> hichihara: ok, thanks. Stay tuned ;)
14:08:51 <hichihara> jlibosva: I'll review quickly if you push new version
14:09:02 <jlibosva> hichihara: great, thank you very much
14:09:36 <jlibosva> does anybody have any other bp or wants to raise a flag ?
14:09:56 <amotoki> even if we cannot make it in neutron-lib, I think we can ship some feature in neutron (by having the API definition in neutron repo)
14:10:06 <jlibosva> alright
14:10:13 <hichihara> I think so
14:10:23 <jlibosva> I hope for the best and will try to make it to lib
14:10:39 <amotoki> totally agree
14:10:46 <hoangcx_> jlibosva: Just FYI, All necessary patches for logging are ready for review: https://review.openstack.org/#/q/topic:bug/1468366+AND+status:open
14:10:57 <jlibosva> unless nobody has anything regarding blueprints we can jump to next topic
14:11:09 <jlibosva> hoangcx_: thanks, does it get proper attention?
14:11:21 <hoangcx_> jlibosva: Yeah, I think so :-)
14:11:21 <jlibosva> hoangcx_: not just from me (I know it doesn't)
14:11:50 <jlibosva> hoangcx_: thanks for bringing this up
14:11:54 <annp> hoangcx_: thank a lot for bring it.
14:12:39 <jlibosva> ok, let's move on
14:12:41 <jlibosva> #topic Bugs and gate failures
14:13:05 <jlibosva> looking at logs from last team meeting I think kevinbenton was a bug deputy?
14:13:18 <hichihara> Right
14:14:04 <jlibosva> there were only a few bugs reported
14:14:11 <jlibosva> I think this is worth to mention here: https://bugs.launchpad.net/neutron/+bug/1704000
14:14:12 <openstack> Launchpad bug 1704000 in neutron "Sometimes OVO unit tests clash on non-unique attributes" [High,Confirmed]
14:14:21 <jlibosva> it's a gate-failure in unittests without any owner
14:14:51 <jlibosva> Ihar already provided some hint on the fix
14:14:56 <jlibosva> does anybody want to take care of this one?
14:15:56 <jlibosva> nobody? :)
14:16:20 <jlibosva> ok, if I find some cycles, I'll try to craft some patch
14:17:14 <jlibosva> I noticed last week gate wasn't really stable, mostly failing on volume operations (which is not really Neutron). Does anybody know any details about such failures
14:17:35 <jlibosva> also graphana doesn't look nice, we'll probably talk about that later today at Neutron CI meeting in ~2 hours
14:18:21 * jlibosva feels alone here
14:19:04 <hichihara> I saw similar failure too
14:19:20 <hichihara> I reported one bug https://bugs.launchpad.net/cinder/+bug/1703512 I noticed
14:19:21 <openstack> Launchpad bug 1703512 in Cinder "tempest.api.volume.test_volumes_snapshots.VolumesSnapshotTestJSON failed" [Undecided,New]
14:19:41 <jlibosva> hichihara: cool! thanks :)
14:20:00 <trevormc> yeah I also saw the same failures.
14:20:22 <hichihara> I believe cinder guy will catch it...
14:20:33 <jlibosva> if that reproduces, we can send a patch to elastic recheck to get some stats about hits
14:21:05 <jlibosva> we don't have a deputy for this week
14:21:12 <jlibosva> do we have any volunteers in the room?
14:21:13 <amotoki> I also see a frequent failure of gate-neutron-dsvm-functional-ubuntu-xenial in https://review.openstack.org/#/c/468262/
14:21:39 <amotoki> is it related to a bug we discuss here?
14:22:14 <jlibosva> amotoki: you mean in CI results on that patch?
14:22:24 <amotoki> jlibosva: yes
14:22:46 <amotoki> it is just tempest decorator change, but it requires four rechecks before the merge
14:23:07 <trevormc> I'm seeing  tempest.api.compute.servers.test_device_tagging.DeviceTaggingTestV2_42.test_device_tagging failing on ps4
14:23:10 <amotoki> 'four' looks wrong. it required more
14:23:20 <jlibosva> amotoki: do you have an example?
14:23:28 <amotoki> https://review.openstack.org/#/c/468262/
14:23:55 <amotoki> I haven't investigated which test(s) failed
14:24:10 <jlibosva> I'm trying to find some failed functional job
14:25:07 <amotoki> it might be a good example
14:25:08 <hichihara> A failure looks like test_firewall on the patch http://logs.openstack.org/62/468262/4/gate/gate-neutron-dsvm-functional-ubuntu-xenial/d289223/testr_results.html.gz
14:25:45 <hichihara> I think that we need bug report
14:26:41 <amotoki> which of our three would like to file a report?
14:26:58 <jlibosva> oh, I think I saw some failures where killing nc wasn't successful or something
14:27:03 <jlibosva> cause process was probably gone
14:27:16 <ihrachys> I triaged all late failures in func tests: https://etherpad.openstack.org/p/neutron-functional-gate-failures-july
14:27:26 <ihrachys> it's basically job timeouts, and firewall test
14:27:29 <ihrachys> could be same thing
14:27:33 <jlibosva> ihrachys: cool!
14:27:39 <jlibosva> the amount of failures, not cool :-/
14:27:47 <ihrachys> yeah, it's just gate failures
14:27:50 <ihrachys> no check queue
14:27:50 <amotoki> perhaps ihrachys knows all
14:27:54 <ihrachys> since a week
14:27:57 <jlibosva> weird, firewall tests should be perfect :-P
14:28:08 <ihrachys> jlibosva, weird right???
14:28:15 <ihrachys> :)
14:28:25 <hichihara> I see
14:28:51 <ihrachys> something brings workers down
14:28:51 <jlibosva> was there any update of ubuntu perhaps?
14:29:03 <ihrachys> I see 'Killed' in console
14:29:23 <ihrachys> wondering if it may be smth like a memory exhaustion
14:29:35 <ihrachys> I am not aware of OS update
14:30:47 <jlibosva> ihrachys: are there any LP bugs related?
14:30:51 <jlibosva> for both issues
14:31:14 <ihrachys> no I was going to report them now
14:31:18 <jlibosva> ok, thanks
14:31:23 <ihrachys> but then learned there is a meeting :-x
14:32:21 <jlibosva> sorry for the news :)
14:32:30 <jlibosva> I'll look at the firewall failures
14:32:56 <jlibosva> I experienced something similar during development of those where closing connection on client ended nc listening process on the other side
14:33:03 <jlibosva> but that was because of missing parameter
14:33:15 <jlibosva> maybe there was a change in nc, anyway, I'll try to look at it
14:33:22 <jlibosva> any other failures?
14:33:35 <ihrachys> I would think that whatever nc issue is about, the larger issue is test runner dying ungracefully
14:34:05 <ihrachys> if you mean gate, I am not aware of anything else pressing. there was that cinder streak of breakages but that should be tackled now.
14:34:21 <jlibosva> ihrachys: do you have any lp handy?
14:34:28 <jlibosva> ihrachys: hichihara opened a bug against it
14:34:28 <ihrachys> for cinder?
14:34:33 <jlibosva> maybe it's a dup
14:34:35 <jlibosva> ihrachys: yep
14:34:57 <jlibosva> ihrachys: https://bugs.launchpad.net/cinder/+bug/1703512
14:34:58 <openstack> Launchpad bug 1703512 in Cinder "tempest.api.volume.test_volumes_snapshots.VolumesSnapshotTestJSON failed" [Undecided,New]
14:35:52 <ihrachys> sec
14:36:08 <ihrachys> I think it was 'fixed' with https://review.openstack.org/#/c/484019/
14:36:18 <ihrachys> v
14:36:19 <ihrachys> https://bugs.launchpad.net/cinder/+bug/1694509
14:36:21 <openstack> Launchpad bug 1694509 in tooz "eventlet errors and cannot switch to a different thread" [High,Confirmed]
14:36:50 <ihrachys> they are still looking for a proper fix
14:37:00 <ihrachys> but for now, we just blocked the release of tooz
14:37:32 <jlibosva> ok, thanks. I left a comment the 1703512 to not forget
14:37:37 <jlibosva> Let's move on
14:37:43 <jlibosva> I don't see mlavalle here
14:38:03 <jlibosva> so maybe we skip docs topic unless somebody has anything to say
14:38:08 <amotoki> jlibosva: do you want to cover the doc stuff?
14:38:17 <jlibosva> amotoki: yep, do you have anything?
14:38:44 <amotoki> thanks all for moving doc-migration stuffs forward fast.
14:38:50 <jlibosva> #topic docs
14:39:27 <amotoki> in the neutron repo, we have very good progress in doc-migration effort.
14:39:33 <amotoki> https://review.openstack.org/#/q/status:open+project:openstack/neutron+branch:master+topic:doc-migration
14:39:48 <amotoki> these three are remaining stuffs.
14:40:14 <amotoki> as neutron stadium, we have several things remaining.
14:40:33 <amotoki> stadium projects needs to do similar things.
14:41:09 <amotoki> I am working closely with the docs team and will send the status report and some reminder tomorrow.
14:41:20 <amotoki> generally we have good progresses :)
14:41:30 <amotoki> that's from me
14:41:54 <jlibosva> amotoki: good, thanks for updates
14:42:23 <jlibosva> #topic Transition to OSC
14:42:25 <jlibosva> amotoki: hi :)
14:43:08 <amotoki> regarding OSC, we don't have progress from several weeks ago but we have only two things remaining
14:43:15 <amotoki> tag support and L3 agent support
14:43:38 <amotoki> both got +2 from one OSC core and need a few comment to be addressed.
14:43:57 <amotoki> so I think we can have both in pike release. I will take care of them
14:44:05 <amotoki> it is in the OSC side.
14:44:24 <jlibosva> awesome :)
14:44:32 <amotoki> in the neutronclient repo side, we have several OSC support reviews from the stadium projects.
14:45:03 <amotoki> I think SFC support is close to merge.
14:45:39 <amotoki> dynamic-routing stuff is another candidate but needs some more reviews.
14:45:49 <amotoki> I will take care of it this week.
14:46:19 <amotoki> another announcement is neutronclient has the same feature freeze next week.
14:46:20 <hoangcx_> amotoki: logging is also ready for your revisit :-)
14:46:44 <amotoki> hoangcx_: it needs to be along with the main stuff in the neutron repo.
14:47:01 <jlibosva> amotoki: also I have a client patch for multiple port bindings rfe :)
14:47:04 <jlibosva> just reminding myself
14:47:29 <hoangcx_> amotoki: Understood
14:47:56 <amotoki> I think we consider neutronclient patches if we think neutron patches are close to be merged.
14:48:45 <jlibosva> makes sense
14:48:56 <amotoki> we can quickly merge them once corresponding neutron patches get in shape.
14:49:03 <amotoki> feel free to ping me then
14:49:11 <jlibosva> alright, thanks amotoki
14:49:14 <jlibosva> and thanks for updates
14:49:15 <hoangcx_> amotoki: Thanks
14:49:20 <jlibosva> #topic open discussion
14:49:29 <jlibosva> I don't see any specific topic in the wiki
14:49:38 <jlibosva> does anybody have a topic to discuss here, we have 10 minutes
14:50:47 <jlibosva> if not, we can end earlier and do some quality coding/reviewing ;)
14:50:55 <amotoki> do we start to gather PTG topics? I haven't checked the recent neutron meetings.
14:51:20 <jlibosva> oh, I haven't seen any email from kevin
14:51:24 <jlibosva> ihrachys: ^^ do you know?
14:52:23 <jlibosva> and nothing in previous two meetings
14:52:53 <amotoki> we have two months to PTG and enough time. it is just a question.
14:53:20 <jlibosva> so I think we havent' started yet
14:53:31 <jlibosva> I noticed we have reserved the XL room there, so that's nice :)
14:54:26 <jlibosva> ok, if there is nothing else, then we can end this meeting
14:54:32 <jlibosva> thanks all for coming! :)
14:54:34 <jlibosva> #endmeeting