15:00:19 #startmeeting manila 15:00:19 Meeting started Thu Dec 12 15:00:19 2024 UTC and is due to finish in 60 minutes. The chair is carloss. Information about MeetBot at http://wiki.debian.org/MeetBot. 15:00:19 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 15:00:19 The meeting name has been set to 'manila' 15:00:36 courtesy ping: dviroel vhari gouthamr carthaca msaravan pulluri ashrodri 15:00:40 hi 15:00:44 hi 15:02:36 Hi 15:02:50 o/ 15:03:19 hi 15:04:37 o/ 15:05:31 o/ hello everyone 15:05:47 thanks for joining, let's get started 15:05:52 our meeting agenda for today: 15:06:09 #link https://wiki.openstack.org/wiki/Manila/Meetings#Next_Meeting 15:06:22 let's start with our announcements 15:06:24 #topic Announcements 15:06:29 Schedule and Deadlines 15:06:38 #link https://releases.openstack.org/epoxy/schedule.html 15:06:43 next week is our spec proposal freeze 15:06:47 and the week of our bugsquash 15:07:01 as we agreed in the last meeting 15:08:57 still on schedule, I would like to propose cancelling two weekly meetings for this year, as I believe a lot of us will be on a break 15:09:20 the idea is to Cancel the Dec 26th and Jan 2nd upstream meetings 15:09:24 +1 15:09:34 +1 15:09:36 +1 15:10:09 awesome, thank you! a lot of hard work on this year, I believe we can catch that break and recharge our batteries for next year 15:10:44 the first weekly meeting of the next year will be on Jan 9th, which I will likely be around but I'll be still officially on PTO 15:10:52 but we can keep that meeting 15:11:18 and if I can't run it, I'll ask for some help 15:12:01 as for our bugsquash, I would like to propose reusing the slot of this weekly meeting 15:12:18 because we are already accounting for a meeting in that time frame and it would be easier to get everyone in the call 15:12:39 similar theme, but a bit different this time 15:12:58 I was chatting with vhari about our bug trends and it is still going up, but we have a lot of bugs that are in progress 15:13:07 so we would like to focus on doing reviews for this bugsquash 15:13:12 to changes that are in progress 15:13:16 s/changes/bugs 15:13:39 of course that will also require some follow-up on the authors side, but we believe it can be a good idea for us to close some bugs 15:15:31 we can start on Tuesday with a quick kick-off call and meet again on the time of this meeting and take the time to discuss things that may show up during reviews 15:16:09 if you prefer, we can start on Thursday too and keep focused on Friday and the beginning of the next week 15:16:30 initially I was going to suggest to have the first meeting on Wednesday but I will be mostly away next Wednesday 15:17:23 how does it sound? 15:17:30 +1 on Tuesday to Thursday 15:18:08 +1 either day works 15:19:12 msaravan gireesh jayaanand: how does that sound to you? 15:21:02 fine, we can plan it from January 15:21:21 fine for me 15:22:00 thanks, the idea is really to take some time to focus on reviews 15:22:01 we can meet coming weeks, for discussions and close on the bugs we wanted to work on. 15:22:11 it will be very nice if you can also participate :) 15:22:19 Sure, agree with your direction. We'll do what you suggest. 15:22:26 gireesh++ msaravan++ 15:22:40 gireesh++ msaravan++ thank you 15:23:00 I will share the details on the mailing list 15:23:18 that's all I had for $topic 15:23:36 is there an announcement you'd like to share? 15:26:38 #topic Share encryption updates (gouthamr/kpdev) 15:27:33 so this is a follow-up discussion to the one we had on the PTG 15:28:41 which mostly revolved on how we should proceed with the testing 15:30:59 we've discussed which first party drivers we could use for testing this feature and mentioned we'd check back on it 15:31:10 LVM and generic would be some options 15:32:09 so I would like to ask if you had a chance to take a look at it 15:34:45 alright, we can circle back in the future 15:35:19 We would definitely try netapp drivers when ONTAP implements barbican support 15:35:20 moving on 15:35:20 #topic Review focus 15:35:24 #undo 15:35:24 Removing item from minutes: #topic Review focus 15:35:46 ah, sure msaravan 15:35:48 actually can you undo so I could share some stuff for the minutes 15:35:57 yep 15:36:12 gouthamr: sure, please go ahead 15:36:24 I wasn’t ignoring this, sorry, just swamped.. I don’t recall thinking about LVM as an option here 15:38:07 Just generic; my thought was that we enable cinder/nova encryption via barbican through the generic driver - I believe, if I understood the spec correctly, we allow the end user to specify a secret or use barbican to create/store secret for a backend to retrieve and perform the encryption 15:39:19 it’s harder to implement something in the first party drivers where we are provided a secret… but, with the generic driver, cinder/nova have APIs to create encrypted volumes and mount them 15:40:46 are we going down this path; I suggested that we can chat at some point/brainstorm how to implement this… is there any interest to pursue this? 15:41:11 this would be a good solution for it 15:41:16 gouthamr: thanks 15:42:48 I'm not sure if kpdev is willing to continue with this implementation 15:43:06 it’s a bit frustrating, but recently at a TC meeting we were discussing how features are added to OpenStack; and we recalled that all project teams have a rule that any API feature has an implementation that is free and open source.. I think in the manila community this rule is not documented 15:43:06 msaravan: would you and the NetApp team willing to help with this too? 15:43:51 perhaps this gives the impression that features can just be implemented in one or two proprietary solutions.. 15:43:59 > and we recalled that all project teams have a rule that any API feature has an implementation that is free and open source.. I think in the manila community this rule is not documented 15:43:59 ++ 15:44:01 We'll work on integrating with netapp drivers for sure. Doing it for generic LVM, would be a overkill, as we are only #4 people working on cinder/manila 15:44:32 also, as it is not there in cinder, it would be totally a new code, and it'll take more and more time. 15:44:55 what is not there in cinder? 15:46:00 user defined key for LVM driver 15:46:42 they generate key in-house, and use it to encrypt all the volumes.. AFAIK. Please correct me if I am wrong. 15:47:46 the proposed implementation has two approaches 15:47:48 Its a block and they can encrypt.. for us its a NAS ..not sure how that can be achieved.. I am novice here.. 15:48:01 Need to explore that further 15:48:16 one with a encryption_ref and another with a share type with encryption specs 15:48:49 Okay, we can plan to meet next week if you have some time to discuss.. (30 mins to 1 hour) ? 15:49:12 you, kpdev, maurice, carlos, gireesh and I can connect 15:50:02 next Monday sounds okay to me 15:50:51 +1 15:51:27 maybe we can post the meeting bridge on #openstack-manila so if there are more interested parties they could join? 15:51:32 ack, lets chat on the #openstack-manila channel or a scheduled meetpad please.. 15:51:39 ++ 15:51:46 thank you 15:52:24 any other thoughts related to $topic? 15:54:20 #topic Open Discussion 15:54:30 couple of extra minutes for open discussion 15:55:01 please take a look at the specs, we seem very close to get them merged 15:55:12 they have been receiving some feedback so far 15:55:40 so thanks to everyone that has reviewed and please take a look if you didn't have a chance yet 15:57:36 alright, let's wrap up 15:57:59 thank you for joining 15:58:01 see you in #openstack-manila 15:58:05 #endmeeting