15:00:30 #startmeeting neutron_l3 15:00:31 Meeting started Thu Jul 13 15:00:30 2017 UTC and is due to finish in 60 minutes. The chair is mlavalle. Information about MeetBot at http://wiki.debian.org/MeetBot. 15:00:32 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 15:00:34 The meeting name has been set to 'neutron_l3' 15:00:50 #chair haleyb 15:00:50 Current chairs: haleyb mlavalle 15:01:07 hi 15:01:09 #chair Swami 15:01:10 Current chairs: Swami haleyb mlavalle 15:01:13 hi 15:01:30 #topic Announcements 15:01:53 Pike-3 is less than two weeks away.... 15:02:14 #link https://releases.openstack.org/pike/schedule.html 15:02:25 July 24th - 28th 15:02:34 another cycle practically gone 15:03:07 mlavalle: time runs fast 15:03:22 And the PTG is fast approaching. September 11th - 15th 15:03:33 i though time was relative? if we speed up it will slow down :) 15:03:43 #link https://www.openstack.org/ptg/ 15:03:46 haleyb: agreed 15:03:57 ahhh, haleyb is a special relativity fan 15:04:09 * haleyb has been watching Einstein documentary recently 15:04:26 * mlavalle and his wife have been doing the same 15:05:09 i have put in a request for the PTG, should be approved 15:05:36 Finally, tomorrow, July 14th is the deadline for submissions for the Sidney Summit talks 15:05:57 I may or may not attend PTG, depends upon our approval. 15:06:41 Any other announcements? 15:06:45 mlavalle: did you submit anything for Sydney? I didn't think i'd be attending so spent time doing reviews 15:07:24 I am going to make a submission with Swami. an update of routed networks progress, including floating ips 15:07:42 +1 15:07:57 haleyb: we were planning to include the DVR fast path support for routed-networks 15:08:05 yeap 15:08:29 Swami: nice 15:08:34 any other announcements? 15:09:06 ok, let's move on 15:09:11 #topic Bugs 15:09:24 Swami: please fire away 15:09:29 mlavalle: thanks 15:09:42 We have two new bugs for DVR that was filed last week. 15:10:04 #link https://bugs.launchpad.net/neutron/+bug/1702790 15:10:06 Launchpad bug 1702790 in neutron "DVR Router update task fails when agent restarts" [Undecided,In progress] - Assigned to Swaminathan Vasudevan (swaminathan-vasudevan) 15:10:15 I have a patch up for review for this bug. 15:10:19 Let me post the link 15:10:46 #link https://review.openstack.org/#/c/481321/ 15:11:01 haleyb: mlavalle: needs review 15:11:36 ack 15:11:58 haleyb: thanks 15:12:07 Swami: is that linuxbridge test failure related to the patchset? 15:12:23 mlavalle: don't think so. 15:12:37 mlavalle: I have seen that failure in other patches as well. 15:12:41 Swami: yeah, I just looked at them 15:13:09 I have been seeing those failures since yesterday in several patches 15:13:09 The next bug is 15:13:12 #link https://bugs.launchpad.net/neutron/+bug/1702769 15:13:13 Launchpad bug 1702769 in neutron "Binding info for DVR port not found error seen when notify_l2pop_port_wiring is called with DVR routers" [Undecided,In progress] - Assigned to Swaminathan Vasudevan (swaminathan-vasudevan) 15:13:37 I also have a patch to address this bug. 15:14:00 #link https://review.openstack.org/#/c/481311/ 15:14:08 This needs review as well 15:15:16 Those are the only two bugs that we have for this week. 15:15:37 Since pike 3 is approaching, we need to review and push the allowed_address_pair patch. 15:16:01 #link https://bugs.launchpad.net/neutron/+bug/1583694 15:16:02 Launchpad bug 1583694 in neutron "[RFE] DVR support for Allowed_address_pair port that are bound to multiple ACTIVE VM ports" [Wishlist,In progress] - Assigned to Swaminathan Vasudevan (swaminathan-vasudevan) 15:16:26 haleyb: mlavalle: could anyone of you tag this bug for the Pike3, right now it does not have any milestone attached. 15:16:49 #link https://review.openstack.org/#/c/466434/ 15:17:01 done 15:17:08 #link https://review.openstack.org/#/c/437986/ 15:17:34 Both these patches needs review. The server side is almost done and I need another +2 on it. 15:17:40 Swami: i think i just had one nit on the first review 15:18:01 haleyb: let me know if I need to fix it and I can do a churn 15:18:37 Swami: what about the other question? was just this morning 15:18:49 from david shaughnessy 15:18:51 from davidsh 15:19:03 haleyb: I have not taken a look at it. 15:20:13 i'll take a look at the agent patch again today 15:20:24 haleyb: mlavalle: the function naming was changed based on the previous comments from kevin. 15:20:37 haleyb: but if it is misleading I can change it again. 15:21:00 Swami: Ack, I'll read the conversations on it. it was just a small nit. 15:21:09 Swami: let me look, but kevin did give the +2 :) 15:22:07 davidsha: thanks for your comments. I will go through it once again to see what is best and will update the patch. 15:22:19 mlavalle: back to you. 15:22:28 Swami: thanks for the update 15:23:05 First one I have today is https://bugs.launchpad.net/neutron/+bug/1683227 15:23:07 Launchpad bug 1683227 in neutron "test_legacy_router test cases sometimes fail with: Cannot find device "qg-*" when moving it into a namespace" [High,Confirmed] - Assigned to Miguel Lavalle (minsel) 15:23:35 I came up with a good Kibana query for this one 15:23:50 it is happening at a rate of about once a week 15:24:04 in the functional tests 15:24:48 we add a router interface (it may be qr-* or qg-*) to br-int 15:25:23 ovs logs the addition of the interface 15:25:41 the interface is handled by the test 15:26:02 and the when trying to move the interface to the namespace, we don't find it 15:26:32 after the failure, we can see ovs logging the addition of the inerface to the bridge a second time 15:27:34 here's a combination/summary of the ovs and functional test logs: 15:27:40 #link http://paste.openstack.org/show/615193/ 15:28:16 I checked with otherwiseguy to see if he had any suggestions. he didn't have any 15:28:17 sounds odd 15:29:22 might be time for an email to ovs-dev ML? 15:29:22 we have seen it before. in fact, the method failing has a temporrary fix by kevinbenton while we get to the bottom of the failure 15:29:46 that was a few months ago 15:30:16 so yeah, yesterday I filed a bug with bugs@openvswitch.org 15:30:39 hidden port syndrome or something like that 15:30:55 and this morning I confirmed in the ovs irc channel that is the correct way to report bugs 15:31:05 haleyb: that's exactly it 15:31:56 I could move the addition of the interface to the namespace inside kevinbenton's temporary fix 15:32:12 but I think it is best to try to get to the bottom of it 15:32:48 especially since it is not happening that often, so it is not pressing 15:33:34 short-term moving the neutron code might be good idea since getting an OVS in and through distros could take time 15:33:48 but if it's not often we can wait 15:34:15 hopefully that made sense, bad spelling 15:34:45 yeah, unless it becomes more pressing, I'd like to wait 15:35:56 if there are no more comments, let's move on 15:36:06 Next one is https://bugs.launchpad.net/neutron/+bug/1694764 15:36:08 Launchpad bug 1694764 in neutron "test_metadata_proxy_respawned failed to spawn metadata proxy" [High,Confirmed] - Assigned to Miguel Lavalle (minsel) 15:36:30 I tried to find ocurrences of this one in Kibana 15:36:38 I couldn't find any 15:36:54 I updated the bug with the query I am using 15:37:14 and I will monitor Kibana for a few days 15:38:23 any other bugs we should discuss today? 15:39:28 ok, moving on 15:39:33 #topic DNS 15:40:11 we made progress with the series of 3 patchsets 15:40:30 https://review.openstack.org/#/c/457035/ was approved. Thanks haleyb 15:41:04 this one depeneds on https://review.openstack.org/#/c/457101/ that needs one more +2 15:41:30 and the last one awaiting reviews is https://review.openstack.org/#/c/468697/ 15:41:48 I addressed kevinbenton comments on this last one 15:41:57 that's it 15:42:06 #topic Open Agenda 15:42:09 mlavalle: there were comments on that last one from hirofumi 15:42:44 err, https://review.openstack.org/#/c/457101/ 15:43:31 ok, I will respond to those 15:43:41 any other topics? 15:44:10 mlavalle: I don't have any other topic to discuss 15:44:17 i am out next week if you're looking for me, that's all for me 15:44:29 going to the beach? 15:44:35 haleyb: golf 15:44:54 both, am i that predictable? :) 15:45:03 have fun :-) 15:45:08 haleyb: based on your past history 15:45:13 haleyb: enjoy 15:45:27 it doesn't matter as long as you are having fun, haleyb :-) 15:45:37 I'm going to be drafting an RFE for the openflow PoC next week, just for a heads up. 15:45:38 thanks, i just follow the cooler around 15:45:53 davidsha: that's cool 15:45:56 thanks! 15:46:00 davidsha: thanks 15:46:36 ok guys, talk to you next week 15:46:40 #endmeeting