13:01:18 #startmeeting tricircle 13:01:19 Meeting started Wed May 18 13:01:18 2016 UTC and is due to finish in 60 minutes. The chair is joehuang. Information about MeetBot at http://wiki.debian.org/MeetBot. 13:01:20 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 13:01:23 The meeting name has been set to 'tricircle' 13:01:51 hello, Zhiyuan is attending the OSCON in Austin, USA 13:02:03 so he will not attend the meeting 13:03:01 and he will have a joint session with google engineer to talk about tricircle and ubernets(Kubernets cluster federation) in OSCON 13:03:14 #topic rollcall 13:03:20 #info joehuang 13:03:28 #info Ronghui 13:03:52 #info Yipei 13:04:44 #topic cross pod L2 networking and dynamic pod binding 13:05:13 we have several round review of these specifications 13:05:47 do you have any questions in implementation and comments 13:07:38 i submit a patch to add resource_affinity_tag 13:07:47 but encounter following errors 13:07:53 It is likely that your change has a merge conflict. You may resolve it in the working tree now as described above and then run 'git review' again, or if you do not want to resolve it yet (note that the change can not merge until the conflict is resolved) you may run 'git rebase --abort' then 'git review -R' to upload the change without rebasing. 13:08:17 wait for a moment qiongqiu's network is not stable 13:08:30 because your base has been changed 13:08:41 you have to do the code rebase 13:09:07 and you'd better to have your own branch, but not directly modified in the master branch 13:09:25 yes, i create a new branch 13:11:11 you can try 'git rebase --abort' 13:11:13 before you run this command 13:11:29 please backup your modification 13:11:44 then git clone the latest code 13:12:03 and create your own branch and switch to this branch 13:12:11 ok, got it 13:12:17 then edit your code, and test it 13:12:42 after everything is ok, including pep8, py27 test... 13:13:03 ok, got it 13:13:08 then git add/commit/review 13:13:41 i execute git commit -a -amend 13:13:48 then git review 13:14:30 no, for the first time, don't use -amend 13:15:37 ok 13:16:55 hi. Sorry for late 13:17:02 you are welcome 13:17:18 i think i still need to do some research on measurement 13:20:46 measurement is not the key 13:20:57 the key is the framework 13:22:36 is dynamic pod bingding visible to end user? 13:22:54 no 13:23:07 it's not visible to end user 13:23:24 only admin can query the binding information 13:23:29 you mean it is just suppled to admin? 13:23:47 ok, i understand 13:24:20 any doubts on L2 networking? 13:24:52 L2 networking is so complex ~ 13:24:59 that is our feeling 13:25:45 yes, it's a little complex 13:26:16 what's the concrete issues for you 13:27:43 first should we have to change the network model of pod for text different use case 13:29:08 it's configuration to use vlan or vxlan, of course if vlan is used, you have to make the provider network support this 13:30:26 there are lots of bugs reported recently 13:30:29 https://bugs.launchpad.net/tricircle 13:31:28 would you like to fix some of them 13:32:37 through the bug fix, and try to run tricircle, you will become family with the code 13:32:44 ok,i will try it 13:32:52 sorry, familar 13:33:14 ok 13:34:32 good to see the new patches from you 13:36:03 so any other topics? 13:36:43 i have a question 13:36:53 please 13:36:56 about the dynamic pod bingding 13:37:32 and Yipei, please register a blueprint in https://blueprints.launchpad.net/tricircle for dynamic pod binding 13:37:51 when the dynamic pod bingding happend , whether it is effect on VMs? 13:38:03 and link the bp in the spec commit message, just like that in cross pod L2 networking 13:38:27 just for later new provisioned VMs/Volumes 13:38:48 #action register a blueprint in https://blueprints.launchpad.net/tricircle for dynamic pod binding 13:38:51 ok 13:39:09 #action link the bp in the spec commit message and spec, just like that in cross pod L2 networking 13:39:36 more questions, ronghui? 13:39:43 A situation occurred 13:39:53 that at the same PM 13:40:18 the VM will in different pod ? 13:40:51 what's PM, you mean baremetal machine? or physical server? 13:41:07 physical server 13:41:27 no physical server only belongs to one bottom pod 13:41:46 no, physical server only belongs to one bottom pod 13:41:47 before dynamic pod bingding the PM will in pod1 bingding on AZ1 13:42:08 but bottom pod not only belongs to one AZ 13:42:12 right? 13:42:24 @joe, there already exists a blueprint of dynamic-pod-binding 13:42:26 no, one pod only inside one AZ 13:42:56 i mean that bottom pod not just belongs to one AZ 13:43:28 ok, please add link in the spec 13:43:37 ok 13:43:37 and commit message 13:43:47 it just belongs to one AZ but the bingding will change the belong to 13:44:07 one bottom pod just belongs to one AZ 13:44:38 no, the binding is about the relationship of tenant and pod 13:44:44 i am not allowed to edit the spec url 13:44:45 when dynamic pod bingding happened it also not change? 13:45:22 to Yipei, you need to change the asignee 13:45:32 yes 13:45:38 to Ronghui, yes 13:46:04 AZ>>>>>pod>>>>tenent the change happened in pod>>>>>tenent? 13:47:02 a tenant can be bound to different pod in different AZ 13:47:25 you can read the google doc blueprint for the relationship of these concepts 13:47:42 ok i know 13:48:00 i confuse the migration and bound.. 13:48:12 ok, let's end the meeting 13:48:18 ok 13:48:36 how to change the assignee 13:49:25 have you log on to the site, if yes, you should be able to add yourself as the asignee 13:49:46 yes i have logged on the site 13:50:35 I have asigned the bp to you 13:51:06 ok, see you 13:51:10 i works 13:51:11 #endmeeting