09:00:37 <hiro-kobayashi> #startmeeting blazar 09:00:37 <openstack> Meeting started Tue Feb 28 09:00:37 2017 UTC and is due to finish in 60 minutes. The chair is hiro-kobayashi. Information about MeetBot at http://wiki.debian.org/MeetBot. 09:00:38 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 09:00:41 <openstack> The meeting name has been set to 'blazar' 09:00:52 <hiro-kobayashi> #chair priteau 09:00:52 <openstack> Current chairs: hiro-kobayashi priteau 09:01:01 <hiro-kobayashi> #topic Rollcall 09:01:07 <hiro-kobayashi> o/ 09:01:09 <priteau> o/ 09:01:15 <tejaswi> o/ 09:01:20 <bertys_> hi all 09:01:28 <hiro-kobayashi> masa will join later, maybe in 15 minutes 09:02:00 <hiro-kobayashi> Is r-mibu? 09:02:41 <hiro-kobayashi> OK lets start a meeting 09:02:53 <hiro-kobayashi> #topic Action items from the last meeting 09:03:10 <hiro-kobayashi> There is not a list of action items, but actions in my mind are: 09:03:19 <hiro-kobayashi> #1 Vote Blueprints 09:03:38 <hiro-kobayashi> #2 Work on bugs for v.0.2.0 09:03:46 <hiro-kobayashi> Anything else? 09:04:34 <hiro-kobayashi> About #1, it looks most of us have voted. 09:04:45 <hiro-kobayashi> Here is the result: https://etherpad.openstack.org/p/Blazar_status_2016 09:05:41 <priteau> Blueprints highlighted in bold font are the ones with the most votes 09:05:51 <priteau> We have 5 of them 09:05:59 <hiro-kobayashi> yes 09:06:40 <hiro-kobayashi> And blueprints with at least 1 vote are [mid] 09:06:53 <hiro-kobayashi> And blueprints with no vote are [low] 09:07:12 <hiro-kobayashi> These results have been reflected to the Launchpad 09:07:23 <hiro-kobayashi> #link https://blueprints.launchpad.net/blazar 09:07:52 <priteau> Thank you hiro-kobayashi for doing this 09:08:48 <hiro-kobayashi> We should detail them, and assign someone after releasing v0.2.0 09:09:39 <hiro-kobayashi> Now we should focus on v.0.2.0 09:09:45 <hiro-kobayashi> hi masahito 09:09:48 <hiro-kobayashi> #chair masahito 09:09:49 <openstack> Current chairs: hiro-kobayashi masahito priteau 09:09:52 <masahito> hi, all. sorry late. 09:10:05 <hiro-kobayashi> masahito: Now we are checking action items 09:10:19 <hiro-kobayashi> #1 Vote Blueprints 09:10:29 <hiro-kobayashi> #2 Work on bugs for v.0.2.0 09:10:48 <hiro-kobayashi> #1 finished 09:10:49 <masahito> hiro-kobayashi: got it. I'll check eavesdrop for past discussion. 09:11:15 <hiro-kobayashi> let's move on #2 09:11:47 <hiro-kobayashi> Here is the status of v0.2.0: https://launchpad.net/blazar/+milestone/0.2.0 09:12:42 <hiro-kobayashi> Do you have anything to discuss here? 09:13:04 <hiro-kobayashi> #topic v0.2.0 status 09:13:16 <priteau> I opened a bug yesterday about ClimateFilter issues 09:13:33 <hiro-kobayashi> priteau: thanks for that 09:13:41 <priteau> is it fixed by https://review.openstack.org/#/c/438796/ 09:14:11 <hiro-kobayashi> Yes, I confirmed it works in my environment 09:14:32 <priteau> Great, thanks 09:14:40 <priteau> Should this be fix released? https://bugs.launchpad.net/bugs/1657826 09:14:40 <openstack> Launchpad bug 1657826 in Blazar "blazarclient doesn't work with keystone v3 API" [Medium,Fix committed] - Assigned to Masahito Muroi (muroi-masahito) 09:14:55 <priteau> or do we want until we have python-blazarclient on pypi? 09:15:42 <masahito> priteau: you hit the bug? 09:16:00 <priteau> masahito: 1657826? not since you merged the patch 09:16:02 <masahito> In my local, blazarclient works well with keystone v3 09:16:12 <priteau> I am talking about the launchpad status of the bug 09:16:22 <priteau> it is "Fix committed" rather than "Fix released" 09:16:31 <masahito> oh, I see. 09:16:37 <hiro-kobayashi> Yes, i think we can mark it fix-released 09:17:15 <masahito> right, it should be fix committed. 09:17:33 <hiro-kobayashi> not fix-released? 09:18:01 <hiro-kobayashi> Other completed pathces are "Fix Released" 09:18:04 <masahito> sorry, I was confused. 09:18:39 <priteau> I have changed its status 09:18:51 <hiro-kobayashi> Thanks! 09:18:54 <masahito> IMO, until we have pypi package it's better to use fix released b/c we only have master codes in git. 09:19:01 <priteau> +1 09:19:15 <priteau> master branch is effectively a continuous release 09:19:31 <hiro-kobayashi> +1 09:19:35 <priteau> Thanks masahito for reviewing my patch for https://bugs.launchpad.net/blazar/+bug/1663204. I will reply to your comments 09:19:35 <openstack> Launchpad bug 1663204 in Blazar "Non-admin users cannot create reservations with default policy rules" [High,In progress] - Assigned to Pierre Riteau (priteau) 09:20:00 <masahito> priteau: first of all, thanks the patch. 09:20:34 <masahito> My comments is more focusing on coding itself. Overall it looks good. 09:20:56 <hiro-kobayashi> Thanks priteau and masahito 09:21:43 <hiro-kobayashi> And thanks for https://review.openstack.org/#/c/434775/ 09:22:04 <hiro-kobayashi> I want to discuss Documentation 09:22:33 <hiro-kobayashi> Should we publish docs on readthedocs? 09:22:53 <masahito> If it works on readthedocs. 09:23:06 <priteau> I am not sure if we can publish several sets of docs 09:23:13 <priteau> We would need both release notes and documentation 09:23:41 <hiro-kobayashi> I tried and confirmed "docs" works with readthedocs 09:23:55 <hiro-kobayashi> I can push the patch for that 09:24:10 <priteau> What kind of patch is needed? 09:24:19 <hiro-kobayashi> But I don't know how to integrate readthedocs with releasenotes 09:24:24 <hiro-kobayashi> with reno 09:25:06 <hiro-kobayashi> priteau: trivial changes. Adds sphinx requirements into requirements.txt 09:25:12 <masahito> IIRC, readthedocs only works with pure markdown or rst file 09:25:44 <hiro-kobayashi> And add some paths into docs/source/conf.py 09:26:04 <hiro-kobayashi> masahito: Yes I think so 09:26:31 <masahito> ok, then if it works, we can point out it. if not, we can add how to generate the doc on README.md. how do you think? 09:26:33 <hiro-kobayashi> So we should make another version of release notes for readthedocs 09:26:43 <priteau> Another solution is, once we have the final release notes, we copy/paste them to the wiki 09:27:30 <masahito> +1 for priteau. it looks nice. 09:27:46 <hiro-kobayashi> +1 too 09:28:33 <hiro-kobayashi> Then I will push the patch for publish "doc" in readthedocs 09:29:04 <hiro-kobayashi> and After the release, we will copy the release note into "doc" repo 09:29:26 <hiro-kobayashi> Does it make sense? 09:29:46 <priteau> what would be the doc repo? 09:30:07 <hiro-kobayashi> doc directory in the blazar repo 09:30:31 <hiro-kobayashi> #link https://github.com/openstack/blazar/tree/master/doc/source 09:31:02 <hiro-kobayashi> Because readthedocs publishes documents in doc directory 09:31:34 <hiro-kobayashi> Here is a sample: http://test-blazar.readthedocs.io/en/latest/introduction.html 09:31:39 <masahito> I think we don't need to keep copy of releasenotes in docs repo. 09:31:59 <priteau> I agree with masahito 09:32:10 <priteau> Or maybe just an index page, like ironic does: https://github.com/openstack/ironic/blob/master/doc/source/releasenotes/index.rst 09:32:15 <priteau> pointing to the wiki 09:32:17 <masahito> just has a link to wiki in blazar repo in github. 09:32:35 <priteau> otherwise that's a lot of duplicated content 09:33:14 <hiro-kobayashi> OK, then I will add a point to the wiki 09:34:36 <hiro-kobayashi> After we publish docs in readthedocs, I think we should remove duplicated wiki pages. How do you think? 09:35:26 <priteau> or mark them as outdated and link to readthedocs? 09:35:43 <hiro-kobayashi> priteau: its better. 09:35:53 <priteau> keeps the history 09:36:13 <hiro-kobayashi> Make sense! Then I will do that. 09:36:26 <masahito> hiro-kobayashi: thanks the volunteer. 09:36:43 <hiro-kobayashi> That's all for documents, thanks! 09:36:56 <hiro-kobayashi> Oh, not all, sorry 09:37:14 <hiro-kobayashi> tejaswi: How's your patch? 09:37:31 <hiro-kobayashi> https://review.openstack.org/#/c/436557/ 09:37:35 <tejaswi> hiro-kobayashi: I've updated v1 API 09:37:53 <tejaswi> I've replaced VM reservation with host reservation 09:38:14 <hiro-kobayashi> Thank you very much! 09:38:42 <tejaswi> should we update v2 API ? 09:38:42 <hiro-kobayashi> the document would be a great help 09:39:12 <hiro-kobayashi> tejaswi No. Because v2 API docs are auto generated as far as I know 09:39:14 <priteau> I believe v2 API is autogenerated 09:39:21 <tejaswi> ok 09:39:31 <priteau> although I still have things I want to verify in it 09:40:19 <tejaswi> In my devstack setup, the API to update host is throwing 500 error. Are you also facing the same issue ? 09:40:39 <priteau> I haven't tried update-host 09:40:51 <priteau> and tempest doesn't check it 09:40:51 <hiro-kobayashi> Me, neither 09:40:55 <tejaswi> ok. 09:41:01 <masahito> me too. 09:41:34 <hiro-kobayashi> Then that should be reported as a bug 09:41:39 <priteau> tejaswi: can you open a bug? 09:41:49 <tejaswi> ok 09:42:01 <hiro-kobayashi> tejaswi Thanks! 09:42:41 <tejaswi> The command to update host "climate host-update --extra abc=1 1" does not throw error, however if we use REST API, it throws 500 error. Ideally, the CLI and REST responses should be the same 09:42:42 <hiro-kobayashi> Anything else about on-reviewing patches? 09:43:19 <priteau> Anyone would have some time to rebase and refresh the patch for checking physical reservation min/max parameters? 09:43:55 <masahito> priteau: which one? 09:43:56 <hiro-kobayashi> priteau: Do you have a link for that? 09:44:56 <hiro-kobayashi> tejaswi: That's should be investigated. Could you describe about it on Launchpad bug? 09:45:09 <tejaswi> ok 09:45:45 <hiro-kobayashi> #action tejaswi to open bug about 'host update' 09:45:47 <priteau> the bug is https://bugs.launchpad.net/blazar/+bug/1404190 09:45:47 <openstack> Launchpad bug 1404190 in Blazar "No validation for --physical-reservation arguments of lease-create command" [High,Confirmed] - Assigned to Pierre Riteau (priteau) 09:46:06 <priteau> client patch https://review.openstack.org/#/c/143044/ 09:46:34 <masahito> I've never checked it. 09:46:57 <hiro-kobayashi> OK, I'll do that 09:47:02 <priteau> and server bug: https://bugs.launchpad.net/blazar/+bug/1400642 server patch: https://review.openstack.org/#/c/140306/ 09:47:02 <openstack> Launchpad bug 1400642 in Blazar "Internal Server Error while creating lease with physical reservation" [Undecided,In progress] - Assigned to chintha govardhan (chintha-govardhan) 09:47:19 <masahito> But from release side, I don't think it blocks the release 09:47:49 <hiro-kobayashi> masahito: +1 09:48:00 <hiro-kobayashi> Because it's not critical 09:48:09 <hiro-kobayashi> Hi GeraldK 09:48:28 <priteau> It would be good if it was fixed in it though. It throws an exception in the middle of lease creation and leaves it half-created 09:49:13 <masahito> Does it make sense it backport after release 0.2.0.? 09:49:32 <masahito> If we want to make blazar and blazarclient perfect, meaning no bugs, we can't release. 09:49:41 <priteau> :-) 09:49:46 <priteau> Yes we can backport of course 09:50:16 <masahito> I'm planning to cut 0.2.0 release once we merge the five patches. 09:50:22 <GeraldK> hi, sorry, I arrived very late... 09:50:32 <masahito> https://review.openstack.org/436557 09:50:37 <masahito> https://review.openstack.org/#/c/436690/ 09:50:41 <masahito> https://review.openstack.org/#/c/438293/ 09:50:45 <masahito> https://review.openstack.org/#/c/437004/ 09:50:50 <masahito> https://review.openstack.org/#/c/434775/ 09:50:57 <masahito> https://review.openstack.org/#/c/434777/ 09:51:32 <masahito> does it make sense? 09:52:05 <priteau> 6 patches? 09:52:31 <masahito> oh. '6 patches' is right. 09:52:50 <GeraldK> two of the patches have status "-1" 09:53:00 <hiro-kobayashi> And I want to merge readthedocs related patch which we have discussed. How do you think? 09:53:19 <GeraldK> the others seems to be okay more or less 09:53:31 <masahito> hiro-kobayashi: Could you add it into https://review.openstack.org/#/c/434775/ ? 09:53:58 <hiro-kobayashi> masahito: yes i can 09:54:01 <masahito> GeraldK: Of course, we need to solve the -1 :) 09:54:33 <GeraldK> hiro-kobayashi: OPNFV opnfvdocs project has also started using ReadTheDocs, so if you have any questions on it, we could also reach out to them 09:54:56 <masahito> GeraldK: after resolving the -1s I'll adds the docs. 09:55:03 <hiro-kobayashi> GeraldK: Thanks! That's a great help 09:55:15 <priteau> Should we talk about deprecation policy before the meeting ends? 09:55:31 <GeraldK> masahito: sure, I know. I just wanted to say it seems we have only 2 out of 6 where some more work is needed. 09:55:37 <hiro-kobayashi> priteau: yes we should 09:56:09 <hiro-kobayashi> #topic Deprecation rules 09:56:11 <masahito> GeraldK: i see. 09:56:52 <priteau> My personal preference is to remove code as soon as possible so there is less to maintain, especially while we have very few users 09:57:09 <hiro-kobayashi> priteau: +1 09:57:14 <masahito> +1 09:57:20 <GeraldK> +1 09:57:42 <priteau> masahito: but I thought you proposed to keep climate code until after P release? 09:57:46 <masahito> If it doesn't hit your environment, I think we can do :-) 09:58:04 <masahito> I thought it could hit your environment. 09:58:12 <priteau> nah don't worry at all about it 09:58:18 <masahito> :-) 09:58:37 <priteau> we will continue running our own fork for now since we have some Chameleon-specific patches 09:58:55 <priteau> and rebase them on Ocata when we upgrade 09:59:31 <priteau> If we follow OpenStack rules I think we can remove deprecated code 3 months after release with the deprecation note 09:59:47 <priteau> #link https://governance.openstack.org/tc/reference/tags/assert_follows-standard-deprecation.html 10:00:09 <hiro-kobayashi> priteau: +1. We should follow it 10:00:43 <masahito> ok. 10:01:01 <hiro-kobayashi> Unfortunately we have no more time 10:01:19 <masahito> then mark all climate namespace is deprecated in O. and will remove it in May or somedays. 10:01:32 <hiro-kobayashi> +1 10:01:44 <priteau> Probably in June 10:02:25 <hiro-kobayashi> Let's go to #openstack-blazar if AOB 10:02:30 <priteau> So that the 3 months are March-April-May 10:02:36 <priteau> All right, thanks everyone 10:02:44 <masahito> priteau: all right. 10:02:45 <hiro-kobayashi> Thanks! 10:02:53 <hiro-kobayashi> #endmeeting