01:01:02 #startmeeting tricircle 01:01:03 Meeting started Wed Dec 13 01:01:02 2017 UTC and is due to finish in 60 minutes. The chair is zhiyuan. Information about MeetBot at http://wiki.debian.org/MeetBot. 01:01:04 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 01:01:06 The meeting name has been set to 'tricircle' 01:01:52 i see our our unit test fails, there may be some changes in neutron code that cause the problem 01:03:03 do you have time to fix it? 01:03:21 #link https://review.openstack.org/#/c/396351/ 01:03:43 i already checked and tried to fix it 01:04:46 thanks Yipei! 01:05:56 adding [Urgent] tag in the commit message title so we will merge it soon once the unit tests pass 01:06:13 ok, got it 01:07:25 next let's update the development progress 01:08:04 I've just +1 on the distinguish source of request patch, sorry for the late comment 01:08:32 I wait this pr to be merged. 01:09:14 i think joe will merge it once he sees the update 01:09:37 but we also need to fix the current unit test failure first 01:10:02 i am not hurry.^.^ 01:10:40 what about the spec for resource deletion reliability? it seems that it hasn't been submitted yet 01:11:13 oh, I know it have finished. I will ask CRH. 01:11:32 look like he does not in here now. 01:12:24 if he does not have time . I pull it a little later. 01:13:20 fine, have you discussed with zhuangzhuang offline? 01:13:33 oh, ronghui is online now 01:13:42 hello 01:13:54 we have discuss the spec 01:14:12 they are waiting for see it. 01:14:23 yeah 01:14:23 i think you have finished already. 01:15:15 ok i will finish the specc this week 01:15:15 because my code already finished.^.^ 01:16:13 hopefully the spec won't change too much during the review 01:16:37 never mind. 01:17:02 1) i am working on the spec for l3 networking, 2) for lbaas, plan to join weekly meeting of octavia to discuss installation automation. 01:17:46 what's the time of the weekly meeting? 01:18:05 20:00 UTC on Wednesdays 01:18:50 oh, it's 4 am beijing time 01:19:57 excuese me,hello everyone. zxh and me is fixing a bug about port-delete. 01:20:06 i was busy with some personal business last week, so i cannot always be online to discuss with the octavia team 01:20:07 but i have some confusions that what the meaning of shadow network and shadow port. sorry to disturb all. 01:21:33 hi zhu 01:21:42 zhu is a student new to here. 01:21:46 welcome 01:22:04 Yipei, that's fine, take your personal business first. I think the automation is not that hurry 01:22:05 Are there some documents to introduce these conceptions? 01:22:36 hi zhu, welcome 01:22:53 thank you , zhiyuan. 01:22:57 hi 01:23:19 for the shadow network and port mechanism, we have a spec discussing about it 01:23:25 https://github.com/openstack/tricircle/blob/master/specs/pike/cross-neutron-vxlan-networking.rst 01:23:29 ok, thanks, zhiyuan 01:24:00 to zhiyuan, i met a problem, i create vm in region1 and delete port when creating vm in region2, then delete vm in region2, when in region1 again to create the vm, there will be a virtualInterfaceCreateException error. if set vif_plugging_is_fatal=false in nova-cpu.conf, create vm success but ping failed. 01:25:50 ok , thank you, zhiyuan. 01:25:53 what about the vm creation in region2, success or not? 01:26:14 success 01:26:35 the port is also deleted in region1, is that right? 01:28:01 use delete_resources function in central to delete port in region1 01:29:15 delete_resources funtion in which module? 01:30:13 tricircle.common.client 01:31:22 _get_client(region_name).delete_resource(constants.RT_PORT, t_ctx, port['id']) 01:32:18 port will be normal delete in region1 01:32:55 oh, i think the resource mapping also need to be deleted 01:33:36 also deleted use e 01:33:54 use core.delete_resources 01:37:03 hmm, so the procedure looks fine 01:38:23 what is the time between you create the vm in region2 and delete vm, recreate in region1? 01:40:56 delete vm in active status, list server is null, recreate vm in region1. 01:43:11 so this doesn't seem a concurrent problem 01:43:52 could you provide me more context about the error via mail? and also your fixsing code 01:44:01 s/fixsing/fixing 01:44:59 good, thanks 01:48:52 is ronghui still online? 01:49:33 yes 01:50:01 is your spec ready to be submitted? 01:50:35 no is there anything need to add ? 01:51:55 I think we can pull it first. let all people to review.maybe it is not very perfect. 01:52:11 +1, song 01:52:18 ok 01:52:36 thanks to RongHUI. 01:53:12 is it fine for you to submit the patch by the weekend? 01:53:21 yes 01:54:45 good, thanks ronghui 01:55:41 so far for the progress update, any other topics? 01:56:07 no for me 01:56:13 no for me 01:56:52 no 01:58:00 ok, thanks for attending the meeting 01:58:05 #endmeeting