16:01:38 <jungleboyj> #startmeeting Cinder 16:01:39 <openstack> Meeting started Wed Jan 2 16:01:38 2019 UTC and is due to finish in 60 minutes. The chair is jungleboyj. Information about MeetBot at http://wiki.debian.org/MeetBot. 16:01:40 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 16:01:42 <openstack> The meeting name has been set to 'cinder' 16:01:51 <whoami-rajat> Hi 16:01:57 <jungleboyj> courtesy ping: jungleboyj diablo_rojo, diablo_rojo_phon, rajinir tbarron xyang xyang1 e0ne gouthamr thingee erlon tpsilva ganso patrickeast tommylikehu eharney geguileo smcginnis lhx_ lhx__ aspiers jgriffith moshele hwalsh felipemonteiro lpetrut lseki _alastor_ whoami-rajat yikun rosmaita enriquetaso 16:02:03 <enriquetaso> Hi o/ 16:02:03 <xyang> hi 16:02:12 <jungleboyj> @! 16:03:02 <mszwed> Hi 16:03:13 <woojay> Hello 16:04:25 <jungleboyj> Guessing we will have a small group today. 16:04:46 <jungleboyj> I think smcginnis_away is still on vacation today and I have a hard stop in about 40 minutes. 16:05:19 <jungleboyj> #topic announcements 16:05:29 <ganso> hello 16:05:47 <jungleboyj> I am hoping that everyone had a Merry Christmas and a Happy New Year. 16:06:40 <jungleboyj> Not a lot to announce this week other than to remind everyone that next week is Milestone 2 for Stien. 16:07:04 <jungleboyj> So, it is our driver and target driver freeze. 16:07:20 <jungleboyj> Which leads us to our first topic. 16:07:36 <jungleboyj> #topic Follow-up on driver review progress 16:07:54 <jungleboyj> I have 5 drivers in the meeting agenda. 16:08:20 <jungleboyj> woojay: I am seeing that you just added the Linstor driver. 16:08:37 <woojay> Yeah, just got the CI working w/ the tests. 16:09:35 <jungleboyj> Ok. Cool. I see that now. Ok, so we can take a look at that one. 16:10:04 <woojay> thank you. 16:10:21 <jungleboyj> woojay: Looks like good test results. Will you be pushing up other log files as well? 16:10:40 <mszwed> jungleboyj: regarding your question on gerrit - I will enable as many tempest tests as possible tomorrow 16:10:53 <jungleboyj> mszwed: Ok. Good. 16:11:04 <woojay> I can certainly do that. Still setting up a few things. 16:11:45 <mszwed> but, as I wrote before, not all tests are possible for SPDK, so i won't be able to enable 'all' 16:12:30 <jungleboyj> Ok. If you can get as many as possible enabled and then we will review the results. 16:13:00 <woojay> sure will. 16:13:40 <jungleboyj> I think that the Hedvig and Inspur are down to the point where they could be merged. The remaining issues are things that can be fixed after the driver merges. I will be working, however, to review these drivers today and tomorrow to ensure that is accurate. 16:13:40 <mszwed> Ok, thank you :) 16:14:09 <jungleboyj> CI for the RSD driver is failing. So, that will need to be investigated. 16:14:35 <jungleboyj> mszwed: Welcome. Can you remind me why not all tests can be run? 16:16:01 <mszwed> jungleboyj: some of them assume scsi interface, but SPDK uses nvme-of; some of the tests run 'volume manage' tests, but SPDK driver does not implement it 16:17:04 <jungleboyj> Oh, ok. But the driver is able to implement the required functionality? Create, Delete for volume and snapshot, extend, etc. 16:17:29 <mszwed> yes, everything that is required is there implemented 16:17:55 <jungleboyj> Ok. Good deal. Other drives don't have manage implemented and have those test cases disabled, so that is fine. 16:18:27 <mszwed> great :) 16:18:34 <jungleboyj> Anyone else have anything to say about the driver reviews? 16:19:40 <jungleboyj> Unfortunately we don't have much here in the way of core reviewers. 16:20:15 <jungleboyj> I will need to do a follow up note to the mailing list and hope they get attention first thing next week. I will take the first stab at it here. 16:20:34 <jungleboyj> No matter how many times we go through this we end up crunched at the end. :-( 16:20:56 <jungleboyj> If developers can do their best to be responsive to comments it would be appreciated. 16:21:18 <jungleboyj> Thanks for the efforts thus far to get CI working and to respond to comments. 16:21:47 <jungleboyj> #topic Horizon issue due to missing project_id in group-list 16:21:55 <jungleboyj> whoami-rajat: Your floor 16:23:37 <jungleboyj> I see your note in the bug and I think handling it as an API change makes sense Whether it is actually a bug or just a change ... Doesn't make a big difference. 16:24:00 <whoami-rajat> So horizon is implementing a new tab 'Groups' in their admin panel. As admin has access to list resources from all projects, project_id is a required response parameter. Recently they discovered that the group-list (detailed) doesn't return project_id in its response. I've proposed a spec and did the code for including that. 16:24:19 <whoami-rajat> just wanted to make sure is it the right approach or did i miss anything? 16:25:19 <jungleboyj> whoami-rajat: Thank you for triaging that. If it something that is needed to support enhanced functionality in Horizon I think it is appropriate to do it as an API change. 16:25:39 <jungleboyj> If the data is there but the API wasn't making it available, then that is the way to handle it. 16:25:50 <erlon> whoami-rajat, for me a spec seems a overkill 16:26:19 <whoami-rajat> jungleboyj: Thanks. the value existed in the db but just not returned. 16:26:22 <erlon> whoami-rajat, so, we already have the group-list API right? 16:26:38 <jungleboyj> erlon: Yeah, the spec is probably overkill but it should be a bick review. 16:27:05 <jungleboyj> erlon: For API changes I don't think we can over document. 16:27:21 <whoami-rajat> erlon: I thought a spec is mandatory for API changes. 16:27:43 <erlon> jungleboyj, if the api exist and is not returning something it should, its a bug ;) 16:27:50 <whoami-rajat> erlon: if its not required then i'm good with abandoning it. 16:28:22 <erlon> whoami-rajat, if is a new API at least is what it makes sense to me 16:28:45 <whoami-rajat> erlon: also i recently saw a spec for adding user_id in backup response. 16:29:05 <jungleboyj> whoami-rajat: Yeah, we have been doing those. 16:29:27 <jungleboyj> whoami-rajat: Like I said, we lean towards more documentation when it comes to changing/fixing APIs. 16:29:57 <erlon> but well, for the point of documenting there are some benefit 16:30:14 <jungleboyj> So, lets just look at the Spec, it is short, and get the patch in. We have our bases covered. There is a bug associated with it as well. 16:30:34 <jungleboyj> Best to get this fixed sooner than later for Horizon. 16:30:46 <whoami-rajat> jungleboyj: Agree. more doc is always better for me. 16:30:54 <jungleboyj> #link https://review.openstack.org/#/c/627541 16:31:06 <jungleboyj> #link https://review.openstack.org/#/c/627549 16:31:51 <whoami-rajat> jungleboyj: Thanks. 16:32:15 <jungleboyj> Ok, that is all we had on the agenda. 16:32:24 <jungleboyj> #topic open-discussion 16:33:29 <jungleboyj> Any other topics? 16:36:05 <jungleboyj> I am taking that as a no. 16:36:20 <jungleboyj> I do have one other reminder. 16:37:06 <jungleboyj> We now have the review priority option. Looks like Sean and I are using it. Would like to see other cores using it as well. So, if you feel that there is a patch that deserves attention. Please update the priority accordingly. 16:37:38 <jungleboyj> Ok. Since it is quiet in here I am going to wrap things up. 16:37:47 <jungleboyj> Happy New Year. Thanks for joining the meeting! 16:37:54 <jungleboyj> #endmeeting