14:01:49 #startmeeting cloudkitty 14:01:49 Meeting started Mon Dec 12 14:01:49 2022 UTC and is due to finish in 60 minutes. The chair is rafaelweingartner. Information about MeetBot at http://wiki.debian.org/MeetBot. 14:01:49 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 14:01:49 The meeting name has been set to 'cloudkitty' 14:01:53 Roll count 14:01:58 \O 14:02:00 o/ 14:02:28 Hi! 14:03:37 # Topic OpenSearch support 14:04:50 We already merged the notice, and Pierre is proposing a change in DevStack, which I guess was already merged. 14:04:54 #topic OpenSearch support 14:05:24 That's just step 0 for OpenSearch support really. The hard work is still to be done. 14:05:33 Which is compatibility with v2 14:05:52 I don't have time this week to look into it. Early next year. 14:06:27 I see 14:06:44 but the process has started. I mean, the notice, and the change in Devstack 14:07:40 Now, we are missing this item, "Crete the OpenSearch option in the Storage backend and deprecate elastic search", right? 14:08:02 Yes, create a new storage driver specifically for opensearch v2 14:08:10 Ok 14:08:17 Which, if I understand correctly, would be compatible with elasticsearch v8 too 14:08:39 But we probably won't be testing it 14:08:43 I see 14:09:43 Do we need anything else regarding OpenSearch? 14:10:52 Not right now, but we need to be aware that Kolla Ansible will release Zed this week, with OpenSearch v2 replacing Elasticsearch v7, which will break CloudKitty until we can add support. 14:11:06 Actually I should add a release note in Kolla Ansible itself about it. 14:11:17 So we will want to backport support to Zed once we have it. 14:11:31 Hmm 14:11:51 I see 14:12:26 thanks for that! 14:12:33 If you need my help, let me know 14:12:54 Thanks 14:13:05 Moving on then 14:13:09 #topic Monasca deprecation 14:13:17 #link https://review.opendev.org/c/openstack/cloudkitty/+/866140 14:13:24 we need to merge the patch with the notice 14:13:37 I've given my +2. 14:14:01 Ah, sorry, I missed that one! 14:14:10 mkarpiarz: can you do the honors? 14:14:25 Reviewed and merged. :) 14:14:32 thanks! 14:14:50 Moving on to the next topic 14:14:54 #topic add CloudKitty API reference docs 14:15:06 mkarpiarz: What is the status of that? 14:15:55 First we need to merge https://review.opendev.org/c/openstack/cloudkitty/+/867122 14:16:53 With API docs in api-ref/source we can then update a different repo and docs should be added to the landing page. 14:17:04 At least this is my understanding. :) 14:17:26 cool 14:17:30 let's see 14:17:41 I think it is good to merge. I gave my +2 there 14:18:49 Merged openstack/cloudkitty master: Add deprecation notice for Monasca https://review.opendev.org/c/openstack/cloudkitty/+/866140 14:18:57 Thanks! 14:19:24 In case you are curious, this is the repo I'm talking about: https://opendev.org/openstack/openstack-manuals/src/branch/master/www/project-data/latest.yaml#L452-L461 14:20:44 Looking now 14:20:49 I see, thanks 14:21:36 No change in generated docs, looks good. 14:22:36 Yes, we are simply moving API refs to a separate directory. 14:23:48 Like I said, it would be good to convert v1 refs to the format used by v2 refs but this is a completely separate task. 14:23:55 What's next? 14:24:41 My understanding is we need to add "has_api_ref: true" here: 14:24:43 https://opendev.org/openstack/openstack-manuals/src/branch/master/www/project-data/latest.yaml#L452-L461 14:25:23 "has_user_guide: true" is already there and this is why CloudKitty shows up here: https://docs.openstack.org/zed/user/ 14:25:58 cool 14:27:09 mkarpiarz: are you going to move on with that, and then we sync up next meeting? 14:27:49 Yes, this week I want to try to add our API refs to "openstack-manuals". 14:28:04 cool, thanks 14:28:07 The patch should merge in about 30 minutes 14:28:13 So, moving on. 14:28:16 #topic Target reviews 14:28:41 We have a new spec. I hava reviewed it, and it seems ok. Have you guys had time to review it? 14:28:52 #link https://review.opendev.org/c/openstack/cloudkitty-specs/+/866198 14:29:12 I am afraid I did not, sorry. 14:29:38 Me neither. :( 14:30:50 Ok, it is a long one to read in the meeting 14:31:05 but if you guys can try to check it today or tomorrow, it would be interesting for us 14:31:18 it is the first spec we see in quite some time 14:33:19 Do we know where the author is from? 14:33:53 yes, I know him 14:34:21 He is a colleague of mine 14:35:37 You guys are always in stealth mode :D 14:35:58 OK, I will add to my todo list to read it 14:36:04 =) 14:38:26 Ok, so moving on 14:38:30 I'll have a read too. Probably not tomorrow though. 14:38:35 we have the patch #link https://review.opendev.org/c/openstack/cloudkitty/+/862824 14:38:40 *not today 14:38:59 Have you guys had the opportunity to check it? 14:39:33 Please check out my last comment. 14:40:14 The feature seems to be working as expected and it adds description to each record. 14:40:27 There is the issue that we will duplicate the description to each record. We should at least warn the operator about this, it could bloat the database. 14:41:02 yes, it could 14:41:19 However, that already happens with two other fields 14:41:35 Can you suggest that there? Then, we have a story line for this topic 14:41:53 and, maybe, we can look for a better way of handling these metadata 14:43:09 Why can't we store this in SQL DB? Or even in metrics.yml? 14:43:27 The API can read the DB when it processes the response 14:43:27 that is something that Pedro proposed :) 14:43:44 but then we would need to change not just for the description, but also for other metadata, such as unit 14:44:24 There are things that are as they are, for historical reasons. We don't have to keep doing the same mistake ;-) 14:44:44 I could work on this improvement/optimization, but ideally it is something on top of those patches to separate the context, and to maybe apply it for other things, such as "metadata" 14:45:14 priteau: I agree with you 14:46:52 So, moving on. #link https://review.opendev.org/c/openstack/cloudkitty/+/861806 14:47:05 Mariusz has approved it, but still need one more 14:48:05 Also, this one #link https://review.opendev.org/c/openstack/cloudkitty/+/864269, It is provides an optimization for Gnocchi fecther, which would be a nice one to have 14:48:24 It is an optimization to control the memory foor print of CloudKitty processor when loading scopes to process 14:50:33 On the first you left a comment on the commit message. 14:51:54 What do you mean? 14:52:11 I see 14:52:16 that must have slipped through 14:52:32 This: https://review.opendev.org/c/openstack/cloudkitty/+/861806/2/COMMIT_MSG 14:52:39 Was that a proposal to rephrase? 14:52:52 I am pinging pedro now 14:52:55 Not very important, but anyway I need a bit more time to read through the code. 14:53:14 yes, the text was a bit hard to read. So, I tried helping him 14:55:46 I will review both patches, hopefully before the next meeting. 14:56:38 Should we discuss next meeting now? 14:57:11 sure 14:57:13 I guess so 14:57:17 we are running out of time 14:57:41 Are you guys going to be working on 26 of December? 14:57:57 Oh, a very good question! 14:58:07 Certainly not ;-) 14:58:14 Looks like I'm off. :) 14:58:18 I see 14:58:26 Pedro Henrique Pereira Martins proposed openstack/cloudkitty master: Allows multiple rating types for same metric for gnocchi https://review.opendev.org/c/openstack/cloudkitty/+/861806 14:58:28 I'll be back on January 3 14:58:28 so, we skip the next one then? 14:58:44 Yes, I think we should meet next on the 9th 14:59:01 ok 14:59:08 let's do that then 14:59:43 Sure 15:00:00 Do you guys have something else to add? 15:00:25 Merry Christmas :) 15:01:08 :D 15:01:59 I wish you guys a merry Christmas and Happy New Year. See you next year then =) 15:02:20 Thank you guys for participating. Have a nice week. 15:02:22 Thanks. Good bye guys. 15:02:36 Thanks and all the best to both of you. :) 15:02:44 It has been a pleasure to work this year with you guys. I hope we can keep the good work next year. 15:02:54 #endmeeting