13:01:52 <joehuang> #startmeeting tricircle
13:01:53 <openstack> Meeting started Wed Apr 13 13:01:52 2016 UTC and is due to finish in 60 minutes.  The chair is joehuang. Information about MeetBot at http://wiki.debian.org/MeetBot.
13:01:54 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
13:01:57 <openstack> The meeting name has been set to 'tricircle'
13:02:13 <joehuang> #topic rollcall
13:02:19 <joehuang> #info joehuang
13:02:34 <zhiyuan_> #info zhiyuan
13:02:40 <RHcao> #info RHcao
13:02:41 <Yipei> #info Yipei
13:02:48 <_lxq_> #info _lxq_
13:03:07 <joehuang> #topic tempest
13:03:22 <ztang> #info ztang
13:03:44 <joehuang> hello,
13:04:22 <joehuang> it's a good opportunity to integrate tricircle into OPNFV
13:04:39 <ztang> good
13:05:28 <joehuang> and IPv6 project in OPNFV wants to leverage the cross OpenStack L2/L3 networking capability
13:05:58 <joehuang> for multisite IPv6 vRouters deployed in different OpenStack instances
13:06:18 <joehuang> https://wiki.opnfv.org/
13:06:40 <RHcao> great
13:06:40 <joehuang> in Release D
13:07:09 <joehuang> Currently the Release C is planned to release in August
13:07:27 <zhiyuan_> when will release D come out?
13:08:00 <joehuang> We want tricircle could be integrated into release C, reduce the integration with IPv6 in release D
13:08:40 <joehuang> Release D maybe started after August, but I guess, will be based on OpenStack Newton
13:08:54 <joehuang> and release in next Feb or Mar
13:10:05 <joehuang> Now a lots of basic functionalities like VM/Volume operation, for example reboot, shutdown, ... not implemented
13:10:57 <joehuang> and more important, the system test using tempest also not implemented
13:11:07 <zhiyuan_> yes, at least we need to implement DELETE operation to run tempest
13:11:36 <joehuang> #link https://etherpad.openstack.org/p/TricircleToDo
13:12:02 <joehuang> the priorities of basic VM/Volume operation is promted
13:12:08 <joehuang> and tempest too
13:14:33 <joehuang> to Zhiyuan, agree, even basic CRUD operation on VM/Volume
13:15:20 <joehuang> #info tempest and VM/Volume CRUD, operation priorities prompted
13:16:09 <zhiyuan_> Are any members interested in implementing these features?
13:16:37 <joehuang> yes, please
13:18:47 <RHcao> we can try it but the process will be slowly
13:19:07 <RHcao> and hope members help us
13:19:21 <joehuang> which one
13:19:38 <joehuang> I think we can use co-author mechanism
13:19:54 <RHcao> CRUD of V M/volume
13:20:21 <RHcao> co-author will good for our team
13:21:07 <ztang> CRUD is for relational DB?
13:21:21 <joehuang> Volume has implemented CRD, but no update
13:21:39 <joehuang> no, forward the command to bottom pod
13:22:03 <joehuang> not in DB
13:22:28 <zhiyuan_> how to run "co-author"? someone submits a patch then others help to modify it?
13:23:01 <ztang> OK. I see
13:23:33 <joehuang> One small function as an example,
13:23:57 <joehuang> and the newbie follow this pattern to write other similar functions
13:25:14 <zhiyuan_> oh i see
13:25:36 <joehuang> For example, write one "reboot", and then the follower need to write similar functions for other purpose
13:26:42 <ztang> this way is good for our beginners
13:26:47 <joehuang> I am not sure this will work or not
13:27:00 <joehuang> but it'll be a good try
13:27:10 <RHcao> should we talk the case of CRUD by VM/volume
13:27:25 <joehuang> ok
13:27:53 <joehuang> Zhiyuan, for VM, support C, but no U and D
13:28:23 <joehuang> and with get_all but no get_one?
13:28:34 <RHcao> it is diff from traditional openstack
13:28:44 <zhiyuan_> yes, currently only boot is implemented
13:29:04 <zhiyuan_> i think show is implemented
13:29:29 <joehuang> ok
13:29:56 <RHcao> zhiyuan  is currently just support on pod (boot VM)?
13:30:04 <RHcao> one pod?
13:30:16 <joehuang> no, multiple pods
13:30:27 <zhiyuan_> boot VM with AZ parameter
13:30:27 <joehuang> otherwise L3 networking is not feasible
13:30:51 <RHcao> at present the node2 just failed
13:31:05 <RHcao> i try many times
13:31:10 <joehuang> to Yipei, when will your spec be ready for review.
13:31:21 <Yipei> before this weekend
13:31:31 <joehuang> Please continue discussion, I'll leave for a while
13:32:06 <RHcao> @zhiyuan now the enviroment of  mutiple pods  is ok?
13:32:45 <Yipei> @RHcao, what is the error?
13:33:24 <ztang> do you clear the reason for the failure of the node 2, RHcao?
13:34:04 <RHcao> i check the log and the reason come from keystone
13:35:10 <RHcao> the bug that zhiyuan submit (1540802) also happened in my environment
13:35:19 <zhiyuan_> @RHcao, can you successfully boot a VM in Pod2 directly?
13:35:27 <RHcao> no
13:35:37 <zhiyuan_> nova boot --os-region-name Pod2 ...
13:35:49 <zhiyuan_> to tell nova client to send to request to bottom OpenStack
13:36:21 <zhiyuan_> @RHcao, that is a bug of DevStack
13:36:27 <RHcao> i fix the stack.sh  but failed to run the devstack
13:37:00 <zhiyuan_> ? so devstack has not started in Pod2?
13:37:07 <RHcao> yes
13:37:41 <_lxq_> maybe we can try it in vm
13:37:55 <zhiyuan_> what's the error stopping devstack in Pod2 to start?
13:38:02 <zhiyuan_> have you checked the log?
13:38:11 <ztang> i agree with lxq
13:38:59 <zhiyuan_> yes, you can use virtualbox to start VM to run devstack
13:39:34 <ztang> let us try it tomorrow @RHCao
13:39:52 <RHcao> @zhiyuan i run the devstack right away and catch the error to shared with our members
13:40:35 <zhiyuan_> ok, but i think you can find the history log in /opt/stack/logs
13:40:50 <zhiyuan_> or /opt/stack/screen-logs
13:40:51 <zhiyuan_> not sure
13:40:53 <RHcao> ok
13:41:08 <ztang> OK, thank zhiyuan
13:42:52 <zhiyuan_> before the bug in devstack fixed, we need to change stack.sh a bit, as mentioned in tricircle readme
13:44:17 <zhiyuan_> one important point is that we need to keep code in each project compatible
13:44:47 <Yipei> @zhiyuan, are other VM/volume operations similar to this patch? #link https://review.openstack.org/#/c/265143/
13:44:51 <zhiyuan_> for example, update all the projects to the latest version
13:46:42 <zhiyuan_> delete operation is similar. but VM has other operations like shutdown, reboot, etc
13:47:22 <zhiyuan_> for these operations, you need to check the API format then register controller to handle such request
13:48:06 <ztang> let's continue to discuss CRUD for VM? RHcao seems interested in it.
13:48:38 <zhiyuan_> like this url format: /v2.1/​{tenant_id}​/servers/​{server_id}​/action
13:48:51 <RHcao> i am finding the error of log in devstack sorry for leave a while
13:49:33 <joehuang> to Yipei, quite similar
13:51:38 <zhiyuan_> if needed, i can submit a patch for VM start/stop to show an example
13:51:55 <Yipei> i think the code in this patch is helpful to us to implement other operations. but still need some comments and explanations. i think it is better to select some examples, we can study them offline
13:52:15 <RHcao> that is great and thank you so much
13:52:34 <longxiongqiu_> agree with Yipei
13:53:07 <RHcao> it will help us to start quickly
13:53:13 <joehuang> ok, let's try using this method to expand our bandwidth for code commit
13:53:40 <zhiyuan_> fine, the patch can be submitted tomorrow
13:53:41 <joehuang> #info using co-author way to expand commit bandwidth
13:54:21 <longxiongqiu_> thanks @zhiyuan_
13:54:24 <joehuang> #info one write small function as example, other follow this pattern to write new functions
13:54:51 <ztang> it sounds nice for us
13:55:21 <ztang> especially for longxiongqiu
13:55:23 <joehuang> ok, let's conclude the meeting, and discuss the devstack in #openstack-tricircle if needed.
13:55:36 <joehuang> yes
13:55:40 <joehuang> good to all
13:55:45 <joehuang> #endmeeting