09:01:11 #startmeeting Smaug 09:01:11 Meeting started Tue Jun 21 09:01:11 2016 UTC and is due to finish in 60 minutes. The chair is yuval. Information about MeetBot at http://wiki.debian.org/MeetBot. 09:01:12 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 09:01:14 The meeting name has been set to 'smaug' 09:01:17 Hello everybody 09:01:21 hola! 09:01:26 hello 09:01:36 hello yuval 09:01:37 hi 09:01:45 hi 09:01:47 Saggi is out today, Yuval has agreed to take over 09:01:59 Hi 09:02:26 #topic Smaug Big-Tent 09:02:42 Well, as you all know, Smaug was admitted as a Big-Tent 09:03:00 Congratulations, and thank you all for you hard work 09:03:11 :). 09:03:11 incredible work everyone!!!! 09:03:25 Congratulations 09:03:44 Congratulations 09:04:05 That being said, we still have more on our path 09:04:12 #topic Testing 09:04:33 zhangshuai did very good job on fullstack tests 09:04:59 last week we managed to make the fullstack tests run on the jenkins check 09:05:15 good news 09:05:19 thanks for approval 09:05:21 I think we should pay more attention to list checkpoints patch. It have block several fullstack tests. 09:05:30 zhangshuai: thanks 09:05:51 zhangshuai good job. 09:05:53 please make sure you write fullstack tests for new features, besides the normal unit tests 09:05:56 I think that we should disable the quota on the fullstack devstack , it will reach the limit soon when we add more test 09:06:17 chenying__: yes, I'll get to that in a couple of minutes 09:06:47 anyone would like to add something regarding the tests? 09:06:57 Hi eran Long time no see. 09:07:04 hi how are you ? 09:07:18 I am fine :) 09:07:22 #topic Patches and Reviews 09:07:40 As chenying__ has pointed out, the 'list_checkpoint' patch requires reviews 09:07:53 can you please share the link 09:07:59 Also, zengchen patches to the operation engine require more review 09:08:01 gampel: sure 09:08:09 luobing is updating this patch to fix fullstack error. 09:08:22 #link https://review.openstack.org/310710 09:08:28 yuval:totally agree. 09:08:37 yuval:thanks 09:08:51 https://review.openstack.org/#/q/project:openstack/smaug-dashboard+status:open 09:08:53 yuval: Yes I think so. 09:09:07 So please dedicate some of your time for reviewing them 09:09:12 also dashboard has a lots of patches which need be reviewed 09:09:20 xiangxinyong: thanks for pointing that out 09:09:35 yuval: you are welcome 09:09:52 Also, if you think that some patches are high priority, add them to the Trello and mark them with a red label 09:10:07 ok 09:10:09 310710 is in merge conflict 09:10:46 Make sure you don't submit patches with DOS endlines (\r\n) - only UNIX endlines (\n) 09:11:17 Anything else regarding patches and reviews? 09:11:29 yuval:will you merge your new design? 09:12:02 zengchen: yes, after the fullstack patches will be merged 09:12:32 yuval:ok. good! 09:12:57 #topic Welcome oshidoshi 09:13:06 oh, that's a nice topic 09:13:09 * oshidoshi blushing 09:13:09 I would also like to welcome oshidoshi! 09:13:14 welcome 09:13:17 welcone 09:13:17 oshidoshi: welcome 09:13:19 thanks guys 09:13:41 oshidoshi: could you introduce yourself? 09:13:43 :) 09:13:48 he is also known as Eshed 09:13:58 :) 09:14:01 :) 09:14:02 okay, Oshidoshi was a joke during the OpenStack Tokyo summit 09:14:02 welcome 09:14:05 gampel: oh.. 09:14:12 now I carry this cross on my back... hmm.. IRC 09:14:30 do you like this name? 09:14:34 :) 09:14:48 well... what can I say.. a name's a name 09:15:04 :) 09:15:07 good name! 09:15:20 anyways, let's move on to the next topic..? 09:15:23 #topic Launchpad Bugs 09:15:33 I started to review the launchpad bugs 09:15:42 some of them are very old, and might be irrelevant 09:15:57 please take a moment to check your old bugs 09:16:04 and close/update them 09:16:20 yuval:got it. 09:16:23 also, if you have old patches which are no longer relevant, abandon them 09:17:02 I've been doing a little review on the dashboard patches, looks like people are not looking there enough... 09:17:15 is that your way of saying "we don't like javascript"? :-D 09:17:48 https://review.openstack.org/#/c/310023/ 09:17:57 how about this patch? 09:18:13 anyone works on it 09:18:49 Now I recall that we said maybe to hold it a bit until the new design is merged 09:19:09 But the network protectable needs to enter 09:19:24 oshidoshi: agree with you. but we love javascript. :) 09:19:34 https://review.openstack.org/305522 09:19:56 chenhuayi: is here ? 09:20:15 looks no update for a long time 09:21:21 I think that the network protection plugin is very impotent and what will drive other to join , it is a major differentiation 09:21:57 As I konw He have finished the neutron plugin. 09:22:24 chenying__: seems like it is not in the gerrit 09:22:36 chenying__: there is only an old verion of it 09:22:40 @gampel, I have finished debug the neutron plugin. now working on the patch. 09:22:52 chenhuayi: great 09:23:08 chenhuayi: thx excellent, :-D 09:23:19 Ok, lets move on 09:23:26 #topic Documentation 09:23:35 there will be three patchs. 09:23:57 Before the big tent submission, we did some documentation effort 09:24:22 After the merge of the protection plugin new design 09:24:34 We'll need a guide for writing protection plugins 09:24:52 Meanwhile, we can create guides for protectable plugins and bank plugins 09:25:14 yuval: +1 09:25:32 Anyone would like to take writing such guide (bank, protectable) ? 09:26:01 it is not high priority, imo 09:26:21 You can add these tasks to trllo. 09:26:35 Ok, I'll add them to the Trello 09:27:33 https://trello.com/b/Sudr4fKT/smaug 09:28:28 Ok 09:28:39 So think about it 09:28:48 "Your browser was unable to load all of Trello's resources." :( 09:29:02 It shouldn't be much work and it is low priority. I'll add to the trello, assign to yourself if you take it 09:29:13 #topic Open Discussion 09:29:27 Anything else you want to talk about? 09:29:30 yuval: yeah 09:29:49 I want to present the app backup solution here 09:30:04 I have a question, when we use keystone client to fetch the service's endpoint, such as nova, cinder, if the keystone don't retrun the endpoint in the token because of configuration, how to solve the problem? 09:30:38 zhonghua-lee: sure 09:30:51 zengchen: what do you mean? 09:30:51 anyone consider how Smaug to support the application backup? e.g. DB, such as Oracle 09:31:12 and others apps, such as sharepoint.... 09:31:41 zhonghua-lee: that's a tough question... I think from Smaug's point of view, we need to see that the APIs can accommodate 3rd party backup solutions for the above 09:31:43 what I know, some backup software has his own method 09:32:07 specifically about Sharepoint, it's a little problematic, since it's a Windows-based software 09:32:11 zengchen: that is a good question. What if the protectable just doesn't list the resources? 09:32:25 :I mean, how to get the endpoint of other service in the smaug. we know keystone may not return the catalog in the token. 09:32:25 also some of the backup software suppoort his policy. 09:33:03 application backup, some backup software maybe have its own scheduler/policy. 09:33:05 zehngchen: okay, here's a thought... 09:33:10 oshidoshi: I mean do we consider to pass the policy to the plun-in directly? 09:33:34 what if we add some validation on the response, e.g. if there's a VM instance, we expect also a Volume and Image protectables... 09:34:02 so, if we don't get them from Keystone, we can send some warning to the user, like "VM without Volume" or "VM without Image", etc. 09:34:59 zhonghua-lee: I'm not sure I follow what you propose... maybe take it offline, or discuss over some BP... 09:35:26 oshidoshi: ok. 09:35:43 Anything else? 09:35:58 who's coming to OpenStack Day Beijing? 09:36:09 oshidoshi: in the smaug we need create other service's client, so we should get the endpoint first. 09:36:23 oshidoshi: I think I will be there. 09:37:14 Anyone else planning to come there? 09:37:17 Yuval? 09:37:17 Me 09:37:22 :) 09:37:47 yuval: see you... 09:37:51 chenying__: are you coming? 09:38:10 yuval: yes I will 09:38:13 oshidoshi:maybe we should solve the problem, not just seed warning. 09:38:42 zengchen: if we can solve the problem by ourselves, fine. but what if the user did not set it up correctly? 09:39:42 oshidoshi:yes, that's another good way. but first we should solve it. 09:40:32 zengchen: okay, we can discuss this after the meeting, I need to get better grasp on the scenario 09:40:55 oshidoshi: ok, thank you. 09:41:02 Anything else? 09:41:10 nope 09:41:16 zengchen: you're welcome 09:41:29 Thank you all :) 09:41:34 #endmeeting