14:00:18 #startmeeting networking 14:00:19 Meeting started Tue Aug 29 14:00:18 2017 UTC and is due to finish in 60 minutes. The chair is jlibosva. Information about MeetBot at http://wiki.debian.org/MeetBot. 14:00:20 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 14:00:22 The meeting name has been set to 'networking' 14:00:24 hello o/ 14:00:44 hi 14:00:54 * jlibosva is glad he's not alone here 14:01:06 :) 14:01:08 #topic Announcements 14:01:27 jlibosva: well, my clock says :59 14:01:36 now it's :00 14:01:40 reminder: the ptg is in 2 weeks - https://etherpad.openstack.org/p/neutron-queens-ptg 14:01:56 I have already 02 14:01:59 haleyb: ^^ 14:02:19 * haleyb curses at ntp 14:02:49 you seem to have local ntp system not connected to the internet :) 14:02:49 next update: we've had a while now the stable/pike branch 14:03:31 so if you have any critical patches that need to go to Pike, you have 1 day left :) 14:03:37 The release is scheduled for tomorrow 14:03:50 any other announcements anybody? 14:05:10 let's move on then 14:05:12 #topic Bugs and gate failures 14:05:38 I think we haven't had a bug deputy for a while now 14:05:59 we will likely reconsider an approach of bug triaging for the next cycle 14:06:21 I think kevin also mentioned that in his PTL nomination 14:06:41 * mlavalle slides in 14:06:48 0/ 14:07:08 does anybody have any important bug that would be worth attention here? 14:07:14 * jlibosva looks at launchpad 14:07:56 https://bugs.launchpad.net/neutron/+bug/1618025 14:07:57 Launchpad bug 1618025 in neutron "MTU specified during neutron network create is not honored, always set to 1500" [Undecided,Invalid] - Assigned to Esha Seth (eshaseth) 14:08:05 * jlibosva is surprised no critical failures are happening with release date approaching 14:08:13 HoloIRCUser5: hi, thanks :) looking 14:08:37 Reedip here :) in the car so , Nick isn't set yet 14:08:40 Could anybody recheck https://review.openstack.org/#/c/497415/ if you see the gate failure. 14:09:15 HoloIRCUser5: I hope you're not driving :) this bug is quite old and was closed as invalid and there are no new actions there 14:09:28 hichihara: I'll keep an eye on it and recheck if needed 14:09:42 Default QoS is new feature for Pike but it's completely broken now. I think that we must fix. 14:09:49 mlavalle: Thanks! 14:09:55 No , at a halt :p 14:09:58 hichihara: OSC gate usually hit this issue. 14:11:01 it does not always happen from my experience 14:11:03 amotoki: I see. I didn't know why you push gate-failure tag into Launchpad. 14:11:32 hichihara: during exploring how to reproduce, you filed it :) 14:11:52 :) 14:11:55 jlibosva sorry this was the bug https://bugs.launchpad.net/neutron/+bug/1713499 14:11:56 Launchpad bug 1713499 in neutron "Cannot delete a neutron network, if the currently configured MTU is lower than the network's MTU" [Undecided,Incomplete] - Assigned to Reedip (reedip-banerjee) 14:11:57 I don't think it is critical for another RC. good backport candidate 14:12:24 amotoki: I think so 14:12:31 'it' is about QoS default policy 14:13:22 reedip: any update on bug 1713499? 14:13:23 bug 1713499 in neutron "Cannot delete a neutron network, if the currently configured MTU is lower than the network's MTU" [Undecided,Incomplete] https://launchpad.net/bugs/1713499 - Assigned to Reedip (reedip-banerjee) 14:15:06 amotoki I checked the behavior and found I cannot create the network with the specified MTU greater than MTU in the configuration file. But I would like the opinion of neutron cores on the steps performed by the bug logger in comment#5(I guess) 14:15:53 to me it sounds like a legitimate bug 14:16:03 IMHO decreasing the global MTU value is very rare operation process 14:16:43 it doesn't seem like a blocker since there is a workaround - don't change the value 14:16:44 But if someone does that, then should.the person just revert the global physnet MTU to delete any pre existing networks? 14:16:57 anyway we need to check scenarios this bug hits 14:17:12 amotoki: but the server should be protected either by failing/stating in the logs, that there are network with higher mtu than the one in config file, right? 14:17:28 jlibosva: sounds fair 14:17:55 jlibosva so you mean running a check at startup for network with MTU greater than current MTU specification? 14:18:31 jlibosva : and should it be restricted to any network (IMO it can be both self service abd provider ) 14:18:57 HoloIRCUser5: either that or not failing deleting a network because of that. I think the main point is that the network should be deletable if op changes the configuration 14:19:36 jlibosva deletion can be allowed but creation can remain blocked like before 14:19:58 Updates should also be blocked I think 14:20:01 at least, the issue reported in bug 1713499 looks a valid bug. there is no reason to preventing such network at least 14:20:02 bug 1713499 in neutron "Cannot delete a neutron network, if the currently configured MTU is lower than the network's MTU" [Undecided,Incomplete] https://launchpad.net/bugs/1713499 - Assigned to Reedip (reedip-banerjee) 14:20:58 I'll try to reproduce it on my setup, so we can flip the LP bug to confirmed 14:21:08 and we can discuss on the LP what is the expected behavior 14:21:17 +1 14:21:25 ++ 14:21:28 Sure jlibosva 14:21:31 +1 14:21:35 +1 14:21:38 HoloIRCUser5: thanks for bringing this up 14:21:54 :) 14:22:21 I think we can move on to the next topic 14:22:27 #topic OSC 14:22:38 amotoki: hi, wanna give any updates? 14:22:55 no updates on OSC side 14:23:09 we have several OSC plugin reviews like dynamic-rotuing 14:23:18 I think they are the next focus 14:23:44 good, thanks 14:23:53 #topic Neutron-lib 14:24:09 boden couldn't make it to the first half of the meeting but he dropped me an email 14:24:38 if folks have some spare cycles, there are some -lib patches but no critical ones 14:25:29 boden will send an email to ML once he revives all consumption patches for Neutron and other stadium projects 14:25:47 that's all he wanted to say about neutron-lib :) anybody has anything else? 14:27:04 sounds like we're good to move to the next topic 14:27:09 #topic docs 14:27:42 boden left me in email this LP bug https://bugs.launchpad.net/neutron/+bug/1684379 - it needs some attention 14:27:43 Launchpad bug 1684379 in neutron " Expose neutron api application as a wsgi script" [Medium,Triaged] 14:27:46 bobh: hi :) 14:27:51 bobh: sorry :) 14:27:53 boden: hi :) 14:28:02 hi 14:28:24 boden: I've covered lib and now we're in docs 14:28:59 jlibosva: ok I see you already mentioned the bug that could use some input 14:29:28 other than that, just FYI that I’ve been working through the doc defects in parallel with the lib work.. I hope to get the number down in the coming weeks 14:29:45 and that’s it :) 14:30:13 boden: thanks :) 14:30:24 anybody has anything related to docs? 14:30:41 Just to mention that in the PTG Monday and Tuesday 14:30:48 The docs team will get together 14:31:07 I plan to attend and invite folks to join in if you have interest in docs 14:31:27 maybe amotoki will join in :-) 14:31:43 will be there 14:31:49 Nice! 14:32:08 mlavalle: thanks for advertising :) 14:32:27 anything else? 14:32:37 not from me 14:32:43 #topic Open discussion 14:32:52 does anybody have a topic to discuss here? 14:32:58 Just a quick note, oslo is soon completing the removal of some deprecated options, as mentioned here: 14:33:01 http://lists.openstack.org/pipermail/openstack-dev/2017-August/121345.html 14:33:27 The work to be done is detailed in the attached etherpad: 14:33:28 https://etherpad.openstack.org/p/oslo-queens-tasks 14:33:46 If you know of anything that might be affected, now would be an excellent time to move to the newer options :) 14:35:27 on another topic, I was talking yesterday to some of my Huawei co-workers who won't be able to be at the PTG. Are we going to have some remote connection option for people not physically present? 14:35:51 janzian: thanks for the update :) 14:37:12 mlavalle: I guess that's up to us, if somebody is willing to sacrifice a laptop or tablet 14:37:32 on previous design sessions, I remember remote people participating via etherpad 14:37:37 jlibosva: ok, I'll talk to Kevin off-line 14:37:58 mlavalle: or maybe you can drop an email to foundation 14:38:11 That's a good idea. I'll follow up 14:38:21 any other topics to discuss? 14:39:21 ok, we can end earlier then 14:39:26 thanks everyone for showing up 14:39:27 #endmeeting