01:01:31 #startmeeting tricircle 01:01:32 Meeting started Wed Jun 14 01:01:31 2017 UTC and is due to finish in 60 minutes. The chair is zhiyuan. Information about MeetBot at http://wiki.debian.org/MeetBot. 01:01:33 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 01:01:35 The meeting name has been set to 'tricircle' 01:01:47 #topic rollcall 01:01:54 #info zhiyuan 01:02:05 #info xiulin 01:02:21 #info Yipei 01:03:00 #topic progress update 01:03:13 so let's update our delelopment progress 01:03:46 for me, most of my submitted bug fixes have been merged 01:04:02 for lbaas, i have some new trouble 01:04:36 I have also submitted a patch to improve our smoke test 01:04:44 to Yipei, what's the problem? 01:04:56 i'am modify sfc's create,delete patch, and implement sfc's update patch 01:05:12 after creating a lb, the amphora can be startup successfully 01:05:28 but its IP in management network is not ping-able 01:05:37 to xiulin, Joe told me last night that our sfc deletion had some problems. I will talk to you offline. 01:06:12 making the vip cannot be plugged via AmphoraAPI 01:06:23 can it retrieve the IP via DHCP? 01:06:27 yes 01:07:10 i can see it through neutron port list in local and central neutron 01:07:11 so the VM can ping the DHCP port, but not the interfaces in management network, is that right? 01:08:54 i think the vm can ping the dhcp port, but its interface in management network is not ping-able in default network namespace 01:09:29 is it because of the security group? 01:09:32 in original octavia, its interface in management network is ping-able from default network namespace 01:09:51 seem like it's one way ping-able 01:10:59 VM can ping default network namespace, default network namespace cannot ping VM 01:11:04 hi, ronghui 01:11:58 to zhiyuan, the vm can ping the dhcp port in management network namespace 01:12:04 hello 01:12:14 to ronghui, we are updating the development progress 01:12:23 but default network namespace cannot ping vm 01:12:44 to yipei, yes, so I say "one way ping-able" 01:13:17 ok 01:13:25 to zhiyuan, i see 01:13:43 it's possible if there are some rules rejecting the income packets to VM 01:14:23 so you mean the security group of the interface? 01:15:06 to yipei, yes, or say the security group of the port 01:15:36 i see 01:15:46 i will check it 01:17:08 ok 01:18:32 to ronghui, I see xiongqiu has updated the patch for QoS, but the py35 the fails. 01:19:25 yes i will tell him to update the patch and fix this problem 01:19:44 thanks :) 01:20:20 ok, let's move to the next topic 01:20:34 #topic storyboard 01:21:02 in the last meeting, Joe mentioned storyboard. have you tried it out? 01:26:08 Storyboard introduces story and task. We can map story to bug/blueprint in launchpad, so task can help us trace the work items for solving a bug or implementing a blueprint 01:27:11 and the comment of storyboard supports markdown 01:28:57 it is sounds good 01:29:49 similar to bug/blueprint in launchpad, we can also link our patch to the story page, here is how to do that 01:30:11 #link https://docs.openstack.org/infra/manual/developers.html#development-workflow 01:31:39 to xiulin, yes, but it seems that we cannot add attachement to the comments in storyboard, in launchpad we can do that.` 01:32:21 What a pity 01:35:14 yesterday I tried to create a story in the story sandbox, we can compare the difference between these two tools: 01:35:21 https://bugs.launchpad.net/tricircle/+bug/1691918 01:35:22 Launchpad bug 1691918 in Tricircle "Local FIP not created after central FIP creation" [Undecided,Fix released] - Assigned to Zhiyuan Cai (luckyvega-g) 01:35:25 https://storyboard-dev.openstack.org/#!/story/121 01:36:17 tricircle is not registered in the storyboard, so the story is under openstack-dev/sandbox project 01:37:37 the story's UI look good 01:38:12 agree :) 01:39:24 maybe after Joe is back we can start a vote whether to migrate to storyboard 01:39:59 #topic open discussion 01:40:08 other topic? 01:40:32 no for me 01:40:35 no for me 01:42:02 for sfc's deleteion, what about judge other port chain is use the ppg and fc ,if use ,not delete 01:43:30 not that easy, there may be two port chains using the same ppg being deleted 01:44:00 they both see the ppg is used by other pc, so not delete the ppg 01:45:31 actually we don't need to check, we will get a exception if the ppg is in use 01:46:03 so one method is to retry deletion 01:46:29 delete, if get exception, delete again after some sleep 01:48:05 if get a exception, mean other port chain in use, why need retry,? 01:48:54 similar reason, if two port chains are being deleted, both of them may get the exception 01:49:24 ok, understand 01:49:39 that's why we need some sleep before the retry 01:49:40 i will try it 01:49:56 fine :) 01:50:28 ok, let's end the meeting 01:50:33 #endmeeting