09:02:02 <chenying> #startmeeting karbor 09:02:03 <openstack> Meeting started Tue Aug 29 09:02:02 2017 UTC and is due to finish in 60 minutes. The chair is chenying. Information about MeetBot at http://wiki.debian.org/MeetBot. 09:02:05 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 09:02:07 <openstack> The meeting name has been set to 'karbor' 09:02:52 <zhonghua> hi 09:02:52 <chenying> hi 09:03:00 <jiaopengju> hi guys 09:03:14 <chenying> waiting 2 minutes for people to join 09:03:37 <yuval> hey 09:04:08 <chenying> hi yuval jiaopengju zhonghua 09:05:00 <chenying> let's start 09:05:21 <chenying> #topic Pike release (tomorrow) 09:06:19 <chenying> so yuval It seems that there are not any patches that need be meged to pike branch. 09:06:42 <yuval> yes, seems like we are ready for the pike release tomorrow 09:07:04 <chenying> What's about the patches about network plugins that jiaopengju work on? 09:07:51 <yuval> well, they are important, but they are not yet tested or reviewed 09:07:52 <chenying> Will we merge them to pike branch int the future? 09:08:02 <jiaopengju> chenying: I will try to split the patch and re-test it soon. 09:08:03 <yuval> we can later backport to pike and make a minor release 09:08:14 <chenying> sound good. 09:08:14 <yuval> chenying: we can 09:09:02 <chenying> So do you have any question about this topic Pike release? 09:09:47 <chenying> yuval: you do great job on karbor project release. 09:10:14 <chenying> ping gengchc_ 09:10:18 <yuval> chenying: thanks :) 09:10:20 <chenying> welcome. 09:10:25 <gengchc_> hello 09:11:01 <chenying> next topic 09:11:06 <zhonghua> celebrate the new release 09:11:32 <chenying> #topic Virtual PTG 09:12:26 <chenying> I have listed the queens goals for karbor in this link : https://ethercalc.openstack.org/karbor-queens 09:13:28 <yuval> The PTG is in Denver, CO, September 11 - 15 09:13:56 <yuval> We have decided to do a virtual PTG 09:14:52 <yuval> that means deciding on topics and hosting the discussion over voice/video 09:15:33 <zhonghua> video? how to join it? 09:15:50 <yuval> we need to find a suitable platform 09:17:28 <zhonghua> how about zoom? 09:17:48 <chenying> sound good. 09:18:51 <yuval> not familiar with it 09:18:51 <zhonghua> AFAIK, k8s community used it as the meeting tool 09:20:04 <zhonghua> another is skype. 09:20:58 <chenying> One problom, not all voice/video can be accessed from the pc in my office. 09:22:09 <jiaopengju> chenying: how about wechat group :) 09:22:57 <jiaopengju> tacker team use it sometimes 09:23:20 <yuval> I'm ok with zoom 09:23:50 <chenying> So we can use zoom. 09:25:13 <yuval> Alright, we just need to decide on sessions time and topics 09:25:22 <chenying> IMO, 11 - 15 five days may be too long for karbor PTG? :D two or three days? 09:26:37 <yuval> The first three days (11-13) are for more major projects (nova, cinder, neutron) while the next two days (14-15) are for our class of projects 09:27:28 <chenying> so (14-15)? 09:28:51 <chenying> #info Karbor Virtual PTG September 14 - 15 09:30:06 <yuval> yes 09:30:20 <chenying> the detail about the session and spec can be added here in this link. https://etherpad.openstack.org/p/karbor-queens-planning 09:31:38 <chenying> Are there any quesion about this topic? 09:31:57 <yuval> nope 09:33:27 <chenying> #topic spec: Adding more protection parameters retention period 09:33:42 <chenying> ping gengchc_ 09:35:47 <chenying> he is offline. So we can discuss the questions about this spec on karbor IRC channel. 09:35:54 <chenying> next one 09:36:15 <chenying> #topic Open Discussion 09:36:34 <chenying> any quesions want to discuss? 09:37:02 <yuval> nothing from me :) 09:37:24 <chenying> yuval As I mentioned in the IRC channel about the fullstack for pod proection. 09:38:25 <chenying> What's I want to see, the fullstack for pod proection will not be introduced to karbor. I will add doc about how to run the k8s cluster on the openstack environment(devstack) and how to protect pod with karbor in future. 09:38:25 <yuval> chenying: ok..? 09:38:27 <gengchc_> Can we discuss spec: Adding more protection parameters retention period? 09:38:33 <chenying> what your oppion about it 09:38:51 <yuval> chenying: I need to read more about it 09:39:16 <yuval> chenying: not sure I understand what's the exact problem 09:39:56 <chenying> OK. If you have any quesion, we can discuss in irc channel. 09:40:05 <chenying> ping gengchc_ 09:40:40 <chenying> about the spec: Adding more protection parameters retention period. Do you have any quesiton about it? 09:41:59 <chenying> As yuval's comment about this spec, we thin these parameters about retention period could not be add to the plan API. 09:42:10 <chenying> S/thin/think 09:42:53 <chenying> ping gengchc_ 09:43:34 <gengchc_> I see your suggestion,we can not place max_backups parameter in a plan.Can we add one karbor object that used to keep max_backups parameters? 09:45:41 <chenying> The retention period are the parameters about the scheduled policy. yuval have explain the reason. 09:46:37 <chenying> We can consider add these parameters about retention period to the scheduled_operation API. 09:46:59 <chenying> yuval What's your oppion about it? 09:47:02 <gengchc_> ok 09:47:46 <chenying> ping yuval 09:48:19 <gengchc_> chenying:can we add these parameters to ScheduledOperation? 09:48:29 <yuval> chenying: gengchc_: these are two different approaches: 1. stating the retention policy as a 'state', 2. stating a scheduled operation to perform operations in order to achieve that state 09:49:19 <gengchc_> chenying:ok 09:49:49 <yuval> chenying: gengchc_: my suggestion is: as we have a 'protect' operation in the Operation Engine service, create a new operation which wakes up, manages the retention policy 09:50:13 <yuval> chenying: gengchc_: scheduled operation can save extra info in them, including max_backups and retention_period 09:50:20 <yuval> chenying: gengchc_: (for example) 09:51:12 <chenying> yuval I agree that adding max_backups and retention_period to the extra info of scheduled operation API. 09:51:24 <yuval> chenying: gengchc_: I believe we can avoid adding these fields into the data model and API of karbor 09:53:39 <gengchc_> chenying:ok.I avoid adding these fields to these fields into the data model and API of karbor, if we use the state to solve the problem. 09:53:49 <chenying> yuval: If we add these the extra info of scheduled operation API. do we need define a schema about it (extra info field.) not add new field to data module. 09:54:01 <chenying> If we add these to 09:55:18 <yuval> not sure I understand 09:55:50 <chenying> adding max_backups and retention_period to the extra info of scheduled operation API. 09:56:38 <yuval> what about it? I think we should refrain from doing so 09:56:44 <chenying> Do we need define the schema about this extra info filed. What parameters can be supported 09:58:55 <yuval> chenying: no, just create a new operation type (like protect) and have it parse the 'operation_definition' param 09:59:11 <yuval> chenying: a new operation type in the Operation Engine, not the Protection Service 09:59:23 <yuval> https://developer.openstack.org/api-ref/data-protection-orchestration/v1/index.html#create-scheduled-operation 09:59:28 <chenying> OK. I see. 09:59:41 <yuval> time's up 10:00:44 <chenying> so we can discuss it in karbor IRC channel. 10:01:11 <gengchc_> different protection objects, we may need to set the max_backups to different values. 10:01:41 <yuval> hey, time is up 10:01:44 <chenying> #endmeeting