15:00:38 #startmeeting neutron_l3 15:00:38 Meeting started Thu Jun 22 15:00:38 2017 UTC and is due to finish in 60 minutes. The chair is mlavalle. Information about MeetBot at http://wiki.debian.org/MeetBot. 15:00:39 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 15:00:42 The meeting name has been set to 'neutron_l3' 15:00:46 hi 15:00:54 Hi 15:01:02 hi 15:01:05 #chair haleyb Swami 15:01:05 Current chairs: Swami haleyb mlavalle 15:01:20 Hi everyone. Thanks for attending 15:01:37 #topic Announcements 15:02:18 Reminder that the next milestone is Pike-3, July 24 - 28 15:02:40 #link https://releases.openstack.org/pike/schedule.html 15:03:58 Also call for papers for Sidney Summit is open until July 14th 15:04:22 mlavalle: thanks for the information. 15:04:57 In case you are interested 15:05:03 Any other annoucements? 15:05:24 nothing from me. 15:05:48 I have nothing. 15:05:59 cool, let's move on 15:06:07 #topic Bugs 15:06:18 Swami: the stage is yours 15:06:39 mlavalle: thanks 15:06:59 New bugs this week for DVR. 15:07:18 #link https://bugs.launchpad.net/neutron/+bug/1698517 15:07:19 Launchpad bug 1698517 in neutron "DVR FIP namespace missing default route after upgrade" [Critical,In progress] - Assigned to Kevin Benton (kevinbenton) 15:07:49 This bug is seen only in Ocata and Newton on upgrades from mitaka to newton or ocata. 15:08:04 This bug was triaged and we did not see any issue with the master branch. 15:08:20 A patch is there for review both in ocata and newton branch. 15:08:38 #link https://review.openstack.org/#/c/475108/ 15:08:54 please review it if not reviewed yet. 15:09:22 it needs one more +2 15:09:32 Yes, mlavalle - please 15:09:49 yes, i was waiting for Swami's triage 15:10:34 Then I'll let haleyb review it 15:10:49 haleyb: yes I have triaged it, and calling the create_rtr_2_fip_link on every iteration will not be a bad idea, since it always checks for the existence before it proceeds. 15:11:00 Swami: the only thing I'm hesitant about is submitting just to stable branches, i'll double-check the steps there 15:11:42 haleyb: The alternate option is to called _add_gateway on every iteration, but both takes the same hit. Ok, kevin was ok with that I hope. 15:12:00 haleyb: thanks 15:12:18 The next in the list is 15:12:20 Swami: i just want to make sure i follow the process 15:12:21 #link https://bugs.launchpad.net/neutron/+bug/1695140 15:12:22 Launchpad bug 1695140 in neutron "SRC without FIP,DEST with FIP case is broken for dvr-multinode" [Undecided,New] - Assigned to Brian Haley (brian-haley) 15:12:51 haleyb: after the regression fix patch merged are we seeing this problem still in mutlinode. 15:13:26 #link https://review.openstack.org/#/c/470063/ 15:13:28 Swami: i haven't checked, will look today 15:14:02 haleyb: ok if you can verify it and update the bug, that would be great. If there is a bug, then I will try to fix it. 15:14:33 The next one is 15:14:36 #link https://bugs.launchpad.net/neutron/+bug/1682345 15:14:37 Launchpad bug 1682345 in neutron "neutron metering agent does not work in DVR mode" [Undecided,In progress] - Assigned to Swaminathan Vasudevan (swaminathan-vasudevan) 15:15:00 I have posted a patch. Still wip, I need to add a test case to validate the notification. 15:15:22 #link https://review.openstack.org/#/c/475077/ 15:15:56 The next in the list is 15:16:23 #link https://bugs.launchpad.net/neutron/+bug/1682228 15:16:24 Launchpad bug 1682228 in neutron "can't cross address scopes with DVR " [Undecided,New] 15:17:09 haleyb: This bug should be closed, since the fast path exit patches have merged. Can you check with kevin on this and close this or are we missing any other use case here. 15:17:54 Swami: sure, i hadn't noticed it merged, :) 15:18:10 haleyb: thanks 15:18:36 There are two patches that needs review for the Floatingip support for allowed_address_pairs, can you guys review it. 15:18:39 #link https://review.openstack.org/#/c/466434/ 15:18:56 #link https://review.openstack.org/#/c/437986/ 15:19:38 These two patches are required for the RFE and two other related bugs. 15:19:53 We should target this for Pike3. 15:20:02 right, and the follow-on to do centralized FIP, right? 15:20:20 haleyb: yes, add the config option to that code, should be straight forward. 15:20:43 that's all I had for now. 15:21:19 Swami: the centralized fip, meaning that moving it to the network node if the compute doesn't have connectivity? 15:21:47 haleyb: mlavalle: We are seeing a scale issue with the DVR routers and neutron. At large scale around 8000 vms, the rpc loop to update the ports are getting slow. 15:21:52 mlavalle: yes 15:22:08 haleyb: mlavalle: have anyone else seen the same issue. 15:22:32 Swami: check with kevinbenton about that rfe 15:22:36 The rpc loop to update_device_list is taking longer time based on the number of devices. 15:22:55 mlavalle: ok, will check. I saw some update to it, but did not pay attention. 15:23:43 this one, right? https://bugs.launchpad.net/neutron/+bug/1667877 15:23:44 Launchpad bug 1667877 in neutron "[RFE] Allow DVR for E/W while leaving N/S centralized" [Wishlist,Triaged] 15:23:47 Swami: i haven't seen that issue. I'd open a bug, could be the way it's reading devices, might be time to use pyroute2 in that code path 15:24:45 mlavalle: ye 15:25:01 better check wit kevinbenton about the next step 15:25:08 mlavalle: ok 15:25:32 haleyb: yes it takes a lot more time to update the device state. 15:26:07 Ok, that's all I had for now. 15:26:12 mlavalle: back to you 15:26:21 Thanks Swami for the update 15:26:33 next bug is https://bugs.launchpad.net/neutron/+bug/1667877 15:26:34 Launchpad bug 1667877 in neutron "[RFE] Allow DVR for E/W while leaving N/S centralized" [Wishlist,Triaged] 15:26:52 for this one, we merged a fix yesterday 15:27:13 bug # might be wrong ^^ 15:27:36 sorry, here it is https://bugs.launchpad.net/neutron/+bug/1655567 15:27:38 Launchpad bug 1655567 in neutron "DHCPv6 failures with "address already allocated in subnet" error" [High,Fix released] - Assigned to Miguel Lavalle (minsel) 15:27:51 the fix was https://bugs.launchpad.net/neutron/+bug/1655567 15:28:05 the fix was https://review.openstack.org/469669 15:28:44 I spent sometime checking logstash for more ocurrences of IpAddressAlreadyAllocated 15:29:32 So far I have only seen them is connection with a test that expects the exception: http://git.openstack.org/cgit/openstack/tempest/tree/tempest/api/network/test_dhcp_ipv6.py#n324 15:30:04 so I will keep an eye on it over the next few days before declaring final victory 15:30:47 Next one is https://bugs.launchpad.net/neutron/+bug/1683227 15:30:49 Launchpad bug 1683227 in neutron "test_connection_from_same_address_scope failed with: Cannot find device "qg-74372988-7f"" [High,Confirmed] - Assigned to Miguel Lavalle (minsel) 15:30:59 I've started working on this one 15:31:10 no much progress yet to report 15:31:47 And the last one I have today is https://bugs.launchpad.net/neutron/+bug/1694764 15:31:47 Launchpad bug 1694764 in neutron "test_metadata_proxy_respawned failed to spawn metadata proxy" [High,Confirmed] 15:31:53 No owner yet 15:32:18 I'll take it over over the next few days 15:32:30 Any other bugs from the team 15:32:36 ? 15:32:52 no 15:33:04 no 15:33:11 #topic Open Agenda 15:33:24 any other topics we should cover today? 15:33:32 maybe davidsha ? 15:33:57 mlavalle: I fine thanks, just interested in seeing the work you guys are doing :) 15:34:08 Thanks 15:34:46 if there are no other topics, let's return 25 minutes to everybody's agenda.... 15:34:53 Thanks for attending 15:34:57 #endmeeting