13:00:45 <joehuang> #startmeeting tricircle 13:00:47 <openstack> Meeting started Wed May 11 13:00:45 2016 UTC and is due to finish in 60 minutes. The chair is joehuang. Information about MeetBot at http://wiki.debian.org/MeetBot. 13:00:48 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 13:00:50 <openstack> The meeting name has been set to 'tricircle' 13:01:04 <joehuang> hello 13:01:12 <joehuang> #topic rollcall 13:01:23 <joehuang> #info joehuang 13:01:27 <Ronghui> #info Ronghui 13:01:35 <Yipei> #info Yipei 13:01:49 <zhiyuan__> #info zhiyuan 13:02:25 <joehuang> #topic L2 networking 13:03:13 <Ronghui> please wait for Xiongqiu 13:03:17 <joehuang> we have several round review and spec update 13:03:19 <joehuang> ok 13:04:06 <Ronghui> he is comming 13:04:13 <lxq> hi 13:04:17 <lxq> sorry 13:04:23 <lxq> i forgot it 13:04:36 <joehuang> welcome 13:05:06 <joehuang> just found that the spec is updated 13:05:08 <Ronghui> he is trying to solove the problem of patch 13:05:17 <joehuang> ok 13:06:01 <joehuang> would like to know whether there is somethine still not clear 13:06:13 <joehuang> in cross pod L2 networking 13:07:33 <joehuang> #link https://review.openstack.org/#/c/304540/9/specs/cross-pod-l2-networking.rst 13:08:34 <Ronghui> we are trying to write some simple test and it will help us to understanding quickly 13:08:34 <lxq> i will try to write shared VLAN and local network 13:09:24 <joehuang> great, good to start from test cases 13:09:48 <lxq> :), i will. 13:10:11 <joehuang> shared VLAN and local network are easy to get onboard 13:10:37 <joehuang> #info start from test cases, and shared VLAN, local network 13:11:18 <joehuang> the commit message seems will lead to some comments 13:11:49 <joehuang> if you remember that fengpan has already comment on this description 13:12:35 <Ronghui> i forget the format of commit 13:12:56 <Ronghui> so update the spec for 4 times. sorry 13:13:43 <joehuang> it's not unusaual :) 13:13:52 <joehuang> https://review.openstack.org/#/c/304540/5/specs/cross-pod-l2-networking.rst 13:15:06 <joehuang> I suggest you update the commit message to stress on no cross OpenStack L2 networking automation yet 13:15:33 <Ronghui> ok 13:15:48 <zhiyuan__> I found some trailing space in the spec 13:16:22 <Ronghui> i will update as soon as possible 13:17:05 <joehuang> I have one idea on the development of cross pod L2 networking feature 13:17:48 <joehuang> how about to have a work group, just like DVR feature developed in Neutron 13:18:18 <lxq> a work group? what does it mean? 13:19:18 <joehuang> several engineers working on the same big feature, just like DVR and cross pod L2 networking 13:19:20 <Ronghui> u mean create a network group just that service for work group 13:20:07 <joehuang> it's a virtual team 13:20:21 <joehuang> virtual sub-team 13:20:30 <joehuang> more exactly 13:20:49 <lxq> how it works? 13:22:21 <joehuang> more frequent discussion inside the sub-team, and one sub-team leader to lead the development 13:23:07 <lxq_> my net is unstable 13:23:23 <joehuang> understand 13:24:12 <zhiyuan__> members in work group still discuss in #openstack-tricircle? 13:24:18 <joehuang> yes 13:25:06 <joehuang> and daily regular meeting inside the sub-team 13:25:20 <joehuang> to discuss what's the challenge, how to solve it 13:25:45 <zhiyuan__> i see. do we need to publish the discussion in #openstack-tricircle? 13:26:04 <lxq_> it sounds good 13:26:11 <joehuang> it's encouraged to do so, we are open community 13:27:09 <zhiyuan__> not sure if we can config #openstack-tricircle to use meeting-robot 13:27:10 <joehuang> yes, weekly meeting is too long to wait for feature development 13:27:32 <joehuang> understanding your concerns 13:27:49 <zhiyuan__> the idea of running work group is greate 13:28:00 <zhiyuan__> s/greate/great 13:28:04 <joehuang> or use etherpad for persistence 13:28:38 <zhiyuan__> i prefer etherpad :) 13:28:45 <joehuang> etherpad also support online chat 13:29:03 <lxq_> we can discuss the network featrue step by step in work group 13:29:40 <Ronghui> prefer etherpad 13:29:44 <joehuang> of course, if the team discuss in daily basis 13:29:45 <lxq_> agree with zhiyuan 13:30:18 <joehuang> #info to setup a working group( virtual sub-team ) for cross pod L2 networking 13:30:23 <lxq_> etherpad is good for me,xchat is unstable 13:30:29 <Yipei> etherpad+1 13:31:00 <Ronghui> same situation for me 13:31:01 <joehuang> #info use etherpad for daily meeting, time is suggested not to longer than one hour 13:31:56 <Yipei> how to get informed? through mail or channel #openstack-tricircle? 13:32:02 <zhiyuan__> send a mail to schedule a time? 13:32:08 <joehuang> I would recommend Zhiyuan as the virtual sub-team leader for the cross pod L2 networking 13:32:20 <lxq_> ok 13:32:22 <joehuang> yes, publish it in the maillist 13:32:22 <Ronghui> great 13:32:23 <Yipei> +1 13:32:23 <zhiyuan__> not all members are in the meeting currently 13:32:37 <joehuang> and the sub-team is also open 13:32:51 <joehuang> if you want to submit patch for cross pod L2 networking 13:33:18 <joehuang> than just join the sub-team daily meeting, and work on patch, and discuss 13:33:36 <zhiyuan__> i am going to attend OSCON next week so it may impact the daily meeting 13:33:59 <joehuang> we don't close the door for others to contribute 13:34:18 <joehuang> it's ok, you are free to set the schedule 13:35:21 <joehuang> if nothing to discuss, you can close the meeting quickly, may be one second 13:36:22 <joehuang> #info Zhiyuan as the virtual sub-team leader of cross pod L2 networking 13:36:47 <Yipei> i do not understand why we need to store historical pod binding records 13:37:11 <Yipei> i think store current active ones is enough 13:37:15 <joehuang> Zhiyuan, if you feel you have no time for the meeting, you can cancel one 13:37:44 <joehuang> it's for histroial retrospect 13:38:09 <joehuang> it can be added later 13:38:34 <zhiyuan__> ok, since i need some time to prepare the talk in OSCON 13:38:36 <joehuang> you know that in internet, everything is recorded 13:38:56 <Yipei> got it, i think maybe we can store then in another table 13:39:10 <Yipei> instead of storing them with active ones 13:39:11 <joehuang> to zhiyuan, no problem. You don't have to do it immediatly, it's just one idea to run this feature 13:39:22 <lxq_> when is the record deleted? 13:39:48 <joehuang> for deleted record, only tag is marked, the record is still there 13:40:23 <zhiyuan__> i see 13:40:46 <joehuang> to Yipei, it could be like that 13:41:10 <joehuang> it would be better to record start time and end time 13:41:22 <Yipei> ok, got it. i will take it into consideration in the future 13:42:13 <joehuang> ok, so for dynamic pod binding, Yipei has prepared a lot 13:44:02 <Yipei> i think i still need to understand more details, i will send mail to you guys for help 13:44:32 <joehuang> #info store historical pod binding information for retrospect, and including start, end time 13:45:14 <joehuang> you are active in m-l, very good 13:45:45 <joehuang> to zhiyuan, how many features lack to pass some basic tempest test 13:45:57 <joehuang> still lack 13:46:30 <lxq_> ... what does m-l mean? :) 13:47:10 <Yipei> mailing-list 13:47:30 <lxq_> thanks 13:47:31 <zhiyuan__> to pass server-create test, we need to improve server controller to support more parameters like admin-pass, key-pair 13:47:56 <zhiyuan__> and server-group which i am working on 13:48:27 <joehuang> seems a lot of work 13:48:35 <zhiyuan__> we also lack many server action operation 13:48:58 <joehuang> I can try to fix some server actions 13:49:07 <joehuang> and volume actions 13:50:06 <joehuang> ok any other topics? 13:50:35 <joehuang> so let's end the meeting 13:50:40 <Yipei> no for me 13:50:55 <joehuang> #endmeeting