21:01:02 <kevinbenton> #startmeeting networking 21:01:03 <openstack> Meeting started Mon Jun 12 21:01:02 2017 UTC and is due to finish in 60 minutes. The chair is kevinbenton. Information about MeetBot at http://wiki.debian.org/MeetBot. 21:01:04 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 21:01:07 <openstack> The meeting name has been set to 'networking' 21:02:31 <kevinbenton> #topic Announcements 21:03:01 <Arkady_Kanevsky> hello PWG 21:03:15 <kevinbenton> ihrachys: what's the status of the gate ATM? 21:03:25 <ihrachys> hm. broken? 21:03:30 <ihrachys> mostly linuxbridge job failures 21:03:46 <ihrachys> which is at 40% failure rate 21:03:51 <ihrachys> this is bug 1696125 21:03:53 <openstack> bug 1696125 in OpenStack Compute (nova) "Detach interface failed - Unable to detach from guest transient domain (pike)" [High,Confirmed] https://launchpad.net/bugs/1696125 - Assigned to Matt Riedemann (mriedem) 21:04:07 <ihrachys> we have this revert that seem to help: https://review.openstack.org/#/c/472365/ but we don't know the culprit 21:04:14 <kevinbenton> ok. i'll check that out after the meeting and will merge that revert if I don't see anything obvious 21:04:16 <ihrachys> afaik nova collects some debug logs with additional patches atm 21:04:24 <haleyb> shows up as a failed to delete network/subnet 21:04:56 <kevinbenton> the issue is that it likely means there is a problem with the tap type on the nova side 21:05:25 <ihrachys> apart from this issue, I am not aware of anything else pressing, though ofc there are random breakages in functional tests and others. 21:05:33 <kevinbenton> which is unfortunate because at least one other plugin is using it (calico) 21:05:58 <kevinbenton> ok 21:06:20 <kevinbenton> does anyone have any announcements to make? 21:07:01 <kevinbenton> ok, moving on 21:07:03 <kevinbenton> #topic Blueprints 21:07:38 <kevinbenton> ihrachys: can you give a quick update of patches still needed for OVO ? 21:08:15 <ihrachys> the current queue can be seen at: https://review.openstack.org/#/q/status:open+project:openstack/neutron+branch:master+topic:bp/adopt-oslo-versioned-objects-for-db 21:09:02 <kevinbenton> ihrachys: is that enough to finish it though? 21:09:06 <ihrachys> of those interesting to point out are: secgroups: https://review.openstack.org/284738; port bindings: https://review.openstack.org/407868 and https://review.openstack.org/382037; router: https://review.openstack.org/307964 and https://review.openstack.org/466237 21:09:19 <ihrachys> oh you want the whole scope? 21:09:24 <ihrachys> no it's not enough 21:09:51 <ihrachys> the major bits left are network and subnet I believe 21:09:57 <kevinbenton> and port too? 21:09:59 <manjeets> ports 21:10:21 <ihrachys> ah right. but this one is blocked by bindings, so we are going up the stack. 21:10:30 <ihrachys> while networks and subnets are largely independent of l2 21:10:31 <kevinbenton> makes sense 21:10:50 <ihrachys> we have some WIPs but they are old 21:10:54 <ihrachys> f.e. subnet: https://review.openstack.org/321001 21:11:08 <ihrachys> (there was network but probably abandoned already) 21:11:16 <kevinbenton> at the moment we aren't actually dependent on this for live migrations, right? 21:11:33 <ihrachys> anyway, if someone feels like helping with remaining bits, you are more than welcome to pick one and drive it, we are ready to help 21:11:41 <kevinbenton> sorry 21:11:44 <kevinbenton> rolling upgrades* 21:12:05 <ihrachys> kevinbenton, not technically, but we need it for new migrations that may come up in the future. 21:12:47 <ihrachys> and speaking of all pieces, there is also a new neutron-db-manage command for data migration currently in the works but it's deeply WIP: https://review.openstack.org/432494 21:12:50 <kevinbenton> right, i just wanted to make sure we prioritize any objects that are needed if that was the case 21:13:27 <kevinbenton> ok 21:13:27 <ihrachys> port binding is I believe of more interest because of work jlibosva drives right now for multiple bindings 21:14:07 <kevinbenton> ihrachys: since he's offline, do you know where we are with fixing live migration for multiple binding types? 21:14:26 <kevinbenton> is someone working on the nova side for that? 21:14:45 <ihrachys> I saw him respinning the patches, but I am not sure how close we are. last time I checked, he had this and ovsfw bugs on his plate. 21:14:58 <ihrachys> afaik no one on nova side, but I can be wrong 21:15:04 <ihrachys> so it's a long shot 21:15:26 <kevinbenton> ok 21:15:34 <kevinbenton> maybe hirofumi can help with the ovsfw bugs 21:15:59 <ihrachys> I can pass your suggestion to him 21:16:18 <kevinbenton> hichihara: ^^ 21:16:24 <kevinbenton> ok 21:16:33 <kevinbenton> boden: are you around for a quick neutron-lib update? 21:16:45 <boden> hi yep 21:16:45 <kevinbenton> any other blueprints anyone wants to discuss? 21:16:50 <kevinbenton> #topic neutron-lib 21:17:03 <kevinbenton> boden: go ahead 21:17:34 <boden> most of the neutron-lib consumption patches have been held up with the gate issues; so not much to say there 21:18:12 <boden> as mentioned last week we are onto the “fun stuff” in regards to decoupling networking-ovn from neutron 21:18:24 <kevinbenton> boden: this is DB stuff? 21:18:26 <boden> in particular the db 21:18:27 <boden> ya 21:18:36 <kevinbenton> which db components are the big issues? 21:18:45 <boden> so here’s a spec to discuss how we can get this done 21:18:47 <boden> #link https://review.openstack.org/#/c/473531/ 21:19:04 <kevinbenton> ack 21:19:14 <boden> kevinbenton: I hope the spec has that info; if not please comment and I’ll update 21:19:26 <boden> that’s really all I have, unless other’s have something? 21:20:22 <kevinbenton> boden: thanks 21:20:33 <kevinbenton> #topic bugs 21:20:39 <kevinbenton> who was bug deputy last week? 21:20:44 <trevormc> me! 21:20:48 <mlavalle> trevormc was 21:20:50 <kevinbenton> trevormc: how did it go? :) 21:20:54 <trevormc> I got it ready for ya 21:20:55 <trevormc> 35 bugs total this last week, some were well known gate-failures and/or reported by cores who assigned the bug to themselves, I didn’t triage those. 21:21:05 <trevormc> From the one’s remaining, I marked 2 Invalid, 3 Confirmed, 1 Opinion, and 4 Incomplete 21:21:29 <trevormc> nothing to bring up in particular 21:21:33 <kevinbenton> cool. anything you need help triaging or anything high/critical we need to know about? 21:21:36 <kevinbenton> ack 21:22:07 <kevinbenton> do we have a volunteer for bug deputy this week? 21:22:44 <kevinbenton> armax: ^^ ? 21:22:45 <kevinbenton> :) 21:23:06 <kevinbenton> i can do it this week 21:23:34 <kevinbenton> #topic Docs 21:23:39 <mlavalle> hi 21:23:46 <kevinbenton> what's up doc? 21:23:49 <kevinbenton> :) 21:23:57 <mlavalle> so I have started attending the docs meeting with asettle 21:24:20 <mlavalle> here's the bugs open for Neutron 21:24:35 <mlavalle> #link https://bugs.launchpad.net/openstack-manuals/+bugs?field.tag=networking-guide 21:24:42 <mlavalle> this is the networking guide 21:24:51 <kevinbenton> these are DocImpact auto bugs, right? 21:24:58 <mlavalle> yeah 21:25:33 <mlavalle> amotoki_away has been working at the one at the top 21:25:45 <mlavalle> and he is pretty active in the docs meeting 21:25:58 <mlavalle> he is willing to help as much as he can 21:26:07 <mlavalle> on an on-going basis 21:26:27 <mlavalle> I encourage the team to look at bugs and help where possible 21:26:53 <kevinbenton> ++ 21:27:00 <mlavalle> also John was planning a restructuring of the networking guide 21:27:08 <mlavalle> #link https://etherpad.openstack.org/p/networking-guide-improvements 21:27:20 <kevinbenton> mlavalle: what's the status on the move into a neutron repo for docs? 21:27:21 <mlavalle> I plan to spend some time this week looking at this 21:27:47 <mlavalle> kevinbenton: last meeting not much was mentioned about that 21:28:05 <mlavalle> asettle was really interested on getting us working on the bugs 21:28:17 <kevinbenton> mlavalle: do you know if that is something they are targeting for pike? 21:28:20 <mlavalle> but I'll have a specific conversation with her this week 21:28:25 <kevinbenton> ok 21:28:31 <kevinbenton> just wanted to get an idea of timeline 21:28:50 <mlavalle> it's a good point. I'll have more next meeting on that 21:29:20 <kevinbenton> ok 21:29:40 <mlavalle> that's all I have this week 21:29:46 <hichihara> I noticed this mail http://lists.openstack.org/pipermail/openstack-dev/2017-June/118055.html 21:30:25 <hichihara> https://review.openstack.org/#/c/472275/ It looks like the spec proposes networking guide is put in neutron tree 21:30:51 <hichihara> I think that we need to watch the spec 21:31:21 <mlavalle> hichihara: thanks, will do :-) 21:31:33 <mlavalle> i'll add my comments to it 21:31:46 <kevinbenton> thanks 21:32:22 <kevinbenton> #topic OSC 21:32:30 <kevinbenton> anyone online that can give an update here? 21:33:18 <kevinbenton> ok 21:33:29 <kevinbenton> #topic Open Discussion 21:33:43 <kevinbenton> does anyone have anything else? 21:33:57 <yushiro> hi, We're trying to update logging API's patch. 21:34:23 <yushiro> I think we need to focus on reviewing these 2 patches. 21:34:23 <yushiro> B models and migration rules https://review.openstack.org/#/c/395483/ and Introduce logging api extension https://review.openstack.org/#/c/415817/ 21:34:41 <yushiro> Because, OSC plugin patch(CLI) is pending for them. Add Logging API commands for OSC plugin https://review.openstack.org/#/c/409819/ 21:34:57 <kevinbenton> ok. i'll take a look at those first two today 21:35:03 <yushiro> I hope to be reviewed these patches :) 21:35:06 <ihrachys> just updating that I will be off for next two weeks starting next Monday 21:35:08 <yushiro> kevinbenton, thanks a lot. 21:35:30 <mlavalle> ihrachys: have fun :-) 21:35:33 <kevinbenton> ihrachys: ouch. have fun 21:35:55 <ihrachys> will try my best 21:35:57 <yushiro> hichihara, thanks for your review! https://review.openstack.org/#/c/415817/ 21:36:18 <mlavalle> ihrachys: going anywhere fun? 21:36:23 <hichihara> You're welcome. I'll review again after meeting. 21:36:33 <ihrachys> mlavalle, one week yes, another is actually work, just not online 21:36:54 <yushiro> hichihara, yeah. Let's decide REST API format. 21:37:38 <kevinbenton> ok 21:37:52 <kevinbenton> if there isn't anything else let's end early? 21:37:58 <kevinbenton> not a question :) 21:38:04 <ihrachys> sure 21:38:11 <kevinbenton> ok, thanks everyone 21:38:15 <mlavalle> o/ 21:38:17 <kevinbenton> #endmeeting