14:00:35 <rafaelweingartner> #startmeeting cloudkitty 14:00:35 <opendevmeet> Meeting started Mon Jan 6 14:00:35 2025 UTC and is due to finish in 60 minutes. The chair is rafaelweingartner. Information about MeetBot at http://wiki.debian.org/MeetBot. 14:00:35 <opendevmeet> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 14:00:35 <opendevmeet> The meeting name has been set to 'cloudkitty' 14:00:41 <rafaelweingartner> Hello guys! 14:00:48 <priteau> Hello Rafael 14:00:50 <rafaelweingartner> Happy new year! 14:00:56 <priteau> Happy new year to you too! 14:01:06 <rafaelweingartner> I hope you guys had a good one 14:02:34 <rafaelweingartner> Roll count 14:02:35 <rafaelweingartner> \0 14:02:41 <MattCrees[m]> Hello! o/ 14:02:57 <priteau> \o 14:05:19 <rafaelweingartner> #topic secure RBAC changes 14:05:37 <rafaelweingartner> Let'ts star with this one, I am not sure what are the impacts on CloudKitty. 14:05:47 <rafaelweingartner> Are you guys aware of these requirements on other projects? 14:06:45 <priteau> I believe that gmann is going to propose some patches during the Epoxy cycle for CloudKitty to catch up with the goal 14:07:38 <rafaelweingartner> Ah I see 14:07:46 <rafaelweingartner> So we just need to monitor and review the patches then 14:09:18 <priteau> I am not finding where this was discussed 14:10:06 <rafaelweingartner> We have not discussed yet 14:10:17 <rafaelweingartner> you raised the topic, and we just added to the backlog 14:10:34 <priteau> I mean, where this was (briefly) discussed with gmann 14:11:11 <rafaelweingartner> ah I see 14:11:21 <rafaelweingartner> if you find it, I can then add to the etherpad as a reference 14:12:16 <priteau> I will send an email to restart the effort 14:12:55 <rafaelweingartner> ok, tjhanks! 14:15:26 <rafaelweingartner> Moving on. 14:15:29 <rafaelweingartner> #topic Target reviews 14:15:45 <rafaelweingartner> #link https://review.opendev.org/c/openstack/cloudkitty/+/921339 14:15:51 <rafaelweingartner> this one has a pipeline failing 14:15:58 <rafaelweingartner> should we try to fix it? Or ask for the author? 14:17:02 <priteau> The author has not been very active. We should probably pick it up. 14:17:11 <priteau> It may be a minor test issue. 14:17:26 <rafaelweingartner> do you guys have experience with prometheus and CloudKitty? 14:17:34 <rafaelweingartner> We have neer used it 14:18:40 <priteau> That's our usual configuration. 14:20:22 <priteau> Maybe one of my colleagues could look into it. 14:20:31 <rafaelweingartner> thanks, that would be awesome1 14:21:38 <rafaelweingartner> #link https://review.opendev.org/c/openstack/cloudkitty/+/936317 was merged 14:22:11 <rafaelweingartner> #link https://review.opendev.org/c/openstack/cloudkitty/+/933146 is an interesting one for you guys to review 14:22:22 <rafaelweingartner> it is a glitch that we found, and it is important to fix 14:25:16 <priteau> Wouldn't other workers pick up the unprocessed scopes? 14:25:49 <rafaelweingartner> depends 14:25:53 <rafaelweingartner> as the worker dies 14:26:05 <rafaelweingartner> there is a tendency, with time, that all of them die 14:26:13 <priteau> But I agree that killing the worker seems extreme 14:26:17 <rafaelweingartner> therefore, it reaches a point when cloudkitty processors will not process anything 14:26:32 <rafaelweingartner> because all worker processor were killed 14:26:42 <rafaelweingartner> it is not just the thread that dies, the whole process dies 14:26:59 <priteau> When running in Kolla, would the container engine restart the service? 14:27:11 <rafaelweingartner> no 14:27:18 <rafaelweingartner> it goes into a limbo 14:27:28 <rafaelweingartner> the python process, is there 14:27:37 <priteau> I see, these are threads 14:27:50 <priteau> It would only work if the head process was killed too 14:27:58 <rafaelweingartner> exactly 14:28:37 <priteau> My main comment is that we should have a bug report and associated release note 14:28:43 <rafaelweingartner> sure 14:28:51 <rafaelweingartner> can you ask Pedro that? 14:28:58 <rafaelweingartner> then, we can move on with that 14:30:07 <rafaelweingartner> Thanks for your help guys on these reviews! 14:32:24 <rafaelweingartner> and ast, but not least, we have this patch #link https://review.opendev.org/c/openstack/cloudkitty/+/876643 14:32:33 <rafaelweingartner> which has been in the making for a very long time 14:32:42 <rafaelweingartner> feedbacks there would be much appreciated 14:36:32 <priteau> It's been a long time I looked into it. I will need to revisit it before being able to discuss it again. 14:36:41 <rafaelweingartner> ok 14:36:44 <rafaelweingartner> no worries 14:36:51 <rafaelweingartner> we can discuss it deeper in the next meetings 14:37:57 <rafaelweingartner> Moving on 14:38:00 <rafaelweingartner> #topic Should we drop the "writer" code 14:38:04 <rafaelweingartner> I am not sure what is means 14:38:20 <priteau> Me neither, but I will try to explain :) 14:38:26 <rafaelweingartner> =) 14:39:17 <priteau> So last month we merged https://review.opendev.org/c/openstack/cloudkitty/+/936316 14:39:44 <priteau> I don't think I had seen it at the time, I only noticed it today while going through my email backlog 14:40:08 <rafaelweingartner> yes 14:40:12 <priteau> It looks like it is fixing code that doesn't work on Python 3 14:40:18 <rafaelweingartner> yes 14:40:28 <rafaelweingartner> but I do not see the link between that and the need to remove something 14:41:06 <priteau> I am getting there ;-) 14:41:22 <priteau> So, this would mean this code has been broken since the migration from Python 2 to Python 3 14:41:29 <priteau> That was several years ago 14:41:29 <rafaelweingartner> yes 14:41:44 <rafaelweingartner> I see, so it was not tested, and not used 14:41:45 <priteau> Which means, we are not testing it and no one is using it 14:41:48 <rafaelweingartner> for anybody else 14:41:54 <priteau> Or at least not reporting it 14:42:01 <rafaelweingartner> exactly 14:42:09 <priteau> I quickly looked into it but I have no idea what this code does 14:42:33 <priteau> Apparently it is something for the service to export reports, in different formats 14:42:43 <priteau> Including CSV 14:43:01 <rafaelweingartner> that is what I also guessed 14:43:03 <priteau> But it isn't the CSV that we can use through python-cloudkittyclient 14:43:13 <rafaelweingartner> I am also looking into it, but I do not clearly see what it is used for 14:43:24 <priteau> Because this uses API -> dataframes -> df-to-csv 14:43:53 <priteau> So my only guess is that it is some ancient code that can probably go 14:44:16 <rafaelweingartner> I see 14:44:27 <rafaelweingartner> I guess we should then query the mailing list 14:44:27 <priteau> We can look into it in more details to understand if there is any value here. But if not, I would propose to remove it entirely. 14:44:39 <rafaelweingartner> and if nobody respond, we can deprecate it, and then remove it 14:46:02 <priteau> It is mentioned here: https://docs.openstack.org/cloudkitty/latest/admin/install/install-source.html#install-the-services 14:46:06 <priteau> cloudkitty-writer: Reporting tool 14:47:05 <priteau> I'll look in the code history for clues 14:47:14 <rafaelweingartner> ok 14:47:31 <rafaelweingartner> I add some notes to etherpad 14:47:37 <rafaelweingartner> we can investigate this a bit further 14:49:17 <priteau> It is mentioned here too: https://docs.openstack.org/cloudkitty/latest/developer/roadmap.html#short-term-effort 14:49:36 <priteau> Add a v2 API endpoint allowing to generate reports - This will be a replacement for cloudkitty-writer. 14:49:47 <opendevreview> Merged openstack/cloudkitty master: Replace own hacking checks by built-on ones https://review.opendev.org/c/openstack/cloudkitty/+/936317 14:50:43 <rafaelweingartner> and I guess that is all, at least, that we have in the list here. 14:50:44 <priteau> It is listed as Implementation: Not started, but I suspect it may be https://review.opendev.org/c/openstack/cloudkitty/+/679636 14:50:52 <rafaelweingartner> Do you guys hae somethign else to add? 14:51:13 <priteau> I didn't realise we had this roadmap document. We should clean it up. 14:53:56 <priteau> Just adding that we may not all be available for the next meeting as we have a workshop that day. 14:54:15 <rafaelweingartner> ok 14:54:17 <rafaelweingartner> no worries 14:55:03 <rafaelweingartner> If that is all, I guess we can close then. 14:56:15 <priteau> Thanks for running the meeting 14:56:55 <rafaelweingartner> Welcome! 14:56:56 <rafaelweingartner> Thank you guys for participating. Have a nice week. 14:57:03 <rafaelweingartner> #endmeeting