16:00:19 #startmeeting Horizon 16:00:20 Meeting started Tue Mar 25 16:00:19 2014 UTC and is due to finish in 60 minutes. The chair is david-lyle. Information about MeetBot at http://wiki.debian.org/MeetBot. 16:00:21 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 16:00:23 The meeting name has been set to 'horizon' 16:00:27 hello! 16:00:30 hi all! 16:00:31 Hello 16:00:32 hello o/ 16:00:33 Hello 16:00:47 hello! 16:00:52 hiya 16:00:52 hey 16:00:53 hi 16:00:59 Hello everyone 16:01:00 o/ 16:01:04 hi 16:02:01 We're in the middle of trying to wrap up RC1 for Icehouse. #link https://launchpad.net/horizon/+milestone/icehouse-rc1 16:02:25 We have 11 bugs left to merge 16:02:52 So we are looking good 16:03:04 david-lyle: Do we need to wait for them all to merge, or will there be a RC1 with the understanding than a RC2 is coming later too? 16:03:14 hello 16:03:15 We're going to target next Tuesday for RC1. 16:03:15 Is RC1 today? 16:03:24 Ooh, ok! 16:03:29 david-lyle: Thanks! 16:03:34 jpich: not all are required for RC1 16:04:09 There are only about 2-3 that I'd hold RC1 for, and the rest would target an RC2. 16:04:28 Understood 16:04:42 I believe there is one item that will require an RC2 and probably the translations 16:05:27 so please take the time to review the In Progress patches for RC1 16:05:56 Second general item is Summit Topics #link http://summit.openstack.org/ 16:05:59 david-lyle: Are there django openstack auth patches we should be keeping an eye on for the RC as well? 16:06:22 jpich: I'm treating that independently 16:06:55 My rationale is that we don't have any new features that require a django-openstack-auth upgrade 16:07:23 david-lyle: Ok, just wanted to make sure since they don't necessarily have the same visibility when they're not related to anything on the milestone page 16:07:25 makes sense! 16:07:26 so I'm not sure the value of requiring a new version of django-openstack-auth for Icehouse 16:07:36 * jpich thinking of the Keystone v3 issues 16:07:43 regarding openstack_auth, i see some translation update. I would like to update translations before its release. 16:08:21 so we should dig into this a little more 16:08:47 OpenStack dependency change freeze is EOD today 16:09:17 I'm pretty sure we'll need exceptions due to client updates :/ 16:09:33 if we want to require a newer version of django-openstack-auth, I need to release today and push the update through openstack/requirements 16:09:38 * jpich still hoping we can get password change with keystone v3 in time for Icehouse 16:10:22 I think dolphm is looking to release that change today, but supporting that would require a requirements change 16:10:57 The keystone client patch hasn't merged yet unfortunately 16:11:08 DepFreeze is new so I guess we'll see how it works out... 16:11:16 for openstack-auth, we can also say there's a newer version of the package, upgrade and your problems are fixed 16:11:44 jpich, if the patch makes it we'll figure out a way to utilize it 16:12:02 david-lyle: Cool, happy to know this! 16:12:05 Is it better to propose a translation import soon? All of them looks typo fix in translation. 16:12:42 for DepFreeze, the main motivation looks third party modules maintained outside of openstack projects. 16:12:44 o/ 16:12:58 hi 16:13:43 My understanding was not just outside, I think version rolls of the same package may be the exception 16:13:51 amotoki: That's my understanding of the rationale as well, but the emails about it make it sound larger 16:15:04 thanks. it seems better we think it applies to client libraries too anyway. 16:15:42 amotoki_: It can be troublesome for Horizon (e.g. that neutron client update with the new exceptions) but we'll figure it out :) 16:16:43 jpich: it is too short to address it :-( 16:17:06 amotoki_: Damn :( Ok, thank you for the update and for looking into it 16:20:27 ok, there can be exceptions to the freeze, but for openstack-auth, I don't think we need to use this 16:20:45 the other clients may cause us issues though 16:21:35 Ok 16:21:44 ok, so I'll release django-openstack-auth when we have translations, but not bump the required version 16:21:57 thanks. 16:22:14 np 16:22:49 are translations associated with the string freeze?.. or are these different translations? 16:22:52 circling back a bit http://summit.openstack.org/ 16:23:02 absubram: I think the translation topic will come back in a bit ;) 16:23:18 haha ok 16:23:28 we have 9 or 11 sessions proposed now, which is great 16:24:03 as I expect only 6-7 sessions we'll have to consolidate some of these 16:24:14 but if you have more topics please post them 16:24:19 I think people are still thinking up topics 16:24:29 david-lyle: can you recommend a cutoff date? 16:24:48 yes.. that;'d be good to know 16:24:53 I think it's only a few weeks before the summit 16:25:06 If there is a deadline it will become an emergency and I'll _have_ to think about it. 16:25:07 I don't have a firm date, as it's not set by me 16:25:10 Yes, previously it's been quite close to the summit, with particularly important sessions pre-approved earlier 16:25:19 oh I see 16:25:36 For HK summit, proposed deadline was around the release for most projects (nova, neutron...) though horizon did not. 16:26:06 Interesting, thanks amotoki 16:26:12 The summit sessions won't be finalized until after the PTL elections regardless, so the Juno PTL can be the ultimate decider of Juno sessions 16:26:19 lblanchard: last week you and jthopkin were mentioning something about hardware resource capacity management? Is there a session for this that you two have proposed? If so can I tag on to it? 16:26:21 So that would be around April 24th looks like 16:27:04 we've not had an embarassment of riches with topics in the past, so it has not been a large task 16:27:17 it is something we have been internally working on.. and personally I was thinking of it being more of an addition to the tuskar-ui.. 16:27:33 absubram: I think the plan is to cover this topic in the "Improve the Overview pages" session that jthompkins proposed…of course you are welcome to come! 16:27:44 awesome. thanks! 16:28:05 I did want to ask the the tuskar-ui folks which project they intended to submit to or both triple-o and horizon 16:28:12 I haven't seen any yet in either 16:28:48 jrist, jcoufal, jomara, tzumainn? 16:28:55 absubram: whether we talk through it in the Overview session or not, we should definitely talk about where it might fit best…even if we need to grab some space during another time. I think jcoufal would be interested too. 16:29:03 absubram: and julim 16:29:27 david-lyle: you mean where we're submitting sessions to? 16:29:30 david-lyle: unfortunately my time has been limited on Tuskar-UI as of late, so I'd refer to others 16:29:33 david-lyle: if we can get 1 session in Horizon, that would be awesome 16:29:51 for API etc it would definitely go to TripleO 16:29:59 oh definitely! sounds good.. thanks lblanchard 16:30:03 david-lyle, yep, what jcoufal said sounds right to me! 16:30:06 That sounds reasonable 16:30:14 +1 16:30:32 david-lyle: I'll work on it's detail 16:30:46 I don't think there is rush for the proposals, is it? 16:30:53 jcoufal: great thanks, I think it's an area of wide interest 16:31:07 no, we have another month or so 16:31:36 perfect 16:31:53 #topic Translation and String Freeze 16:32:33 so doug-fish had some questions and I'm hoping amotoki can provide some clarification 16:32:37 or jpich 16:32:48 amotoki_: I think it would be really helpful if the i18n team could update https://wiki.openstack.org/wiki/StringFreeze to be what the i18n team actually wishes to see 16:32:58 oh thanks - yes that is my concern 16:33:05 I know we don't exactly follow what's in the wiki 16:33:20 but its not clear to me exactly when we decide we should/shouldn't update an english user facing string 16:33:29 I think 95%+ of Horizon patches would be blocked if we followed what's in the wiki at the moment... 16:33:48 :-) 16:33:49 doesn't that cause lots of problems with translations? 16:34:15 doug-fish: most small updates does not matter from my experience. 16:34:46 - it doens't matter because the translation team can pick it up? 16:35:27 yes. if a change is small, transifex suggests old translations too. 16:35:44 amotok_i: I see a lot of typo patches being submitted to Horizon still. Is there a cut-off date when we should start blocking them till Juno? When should the "request exception from i18n team" process begin? Maybe that's a topic for the next i18n meeting 16:36:14 amotoki_ is the transifex process continuous? that is, if a havana patch went out today with a new user facing string, could it get translated somehow? 16:36:47 good question :) 16:37:01 doug-fish: we have resources both for icehouse and havana. They are maintained separately. 16:37:30 doug-fish: It could, as in we have havana "branches" in Transifex, but the po file updates aren't automatically updated on either side at this point (is my current understanding) 16:37:33 but syncing with transifex for havana is done manually. 16:38:16 manually by us in the community? 16:38:27 so our patch sets have updated po files as some point? 16:39:18 now we have jenkins job to upload PO file to transifex for master branch. 16:39:37 The job is kicked every time a patch is merged. 16:40:05 I did it manually previously :) 16:40:52 amotoki_: automation is great, huh? :-) will that job continue perpetually? so icehouse PO files could get sent to transifex during juno, or k? 16:40:53 I have a couple bugs in review that make string changes and that amotoki has seen and wished for other community feedback on as well.. and I don't want to step on any toes, if it needs to be pushed to Juno so be it, but wanted to make sure if it somehow did get added in RC1, it doesn't hurt the translation process 16:42:17 absubram: Are they small string changes or larger ones? 16:42:25 doug-fish: The job is targeted to master branch. Once the msater branch is opened to Juno, transifex resources will be synced with Juno. 16:42:46 so amotoki_, once we import the translations for Horizon, then should we, block further string changes? 16:42:57 doug-fish: a tricky point is after RC1 is shipped. I and daisy need to work on it. 16:42:58 jpich: I'd say small :)).. but one review does add two new strings 16:43:38 david-lyle: i hope so. RC1 cut is a possible deadline. 16:43:39 amotoki_: Thanks. I think I have a much better understanding of what we are doing now. 16:43:53 doug-fish: you're welcome. 16:44:30 jpich: absubram: The reason I -1'ed first is because it is a small feature addition with new strings. It is really a gray zone. 16:45:13 amotoki_: no worries.. that one got merged though this past weekend.. so thanks to you and david-lyle 16:46:05 amotoki_, RC1 seems reasonable 16:46:15 amotoki: this one is the Profile name and id addition in the network details.. anyway don't want to derail the discussion.. 16:46:47 btw, switching django from 1.5 to 1.6 introduces new strings to transaltions. Previously openstack/common strings are excluded. 16:47:02 amotoki_: I think updating that wiki page would help reviewers assess patches with a clearer understanding of what is helpful to the i18n and what might require exception approval. It sounds like "grey zone" may require asking the i18n list what they think? 16:47:46 jpich: sounds good. 16:48:14 amotoki_, is that just a path issue, seems like openstack/common should already be translated 16:48:35 amotoki_: Ok, thank you. I'll send an email to the i18n list with a summary of what was discussed here to get feedback 16:49:19 amotoki_, I see the lp bug now, lookign 16:49:36 david-lyle: basically it should be translated, but we have a number of unused openstack-common. 16:50:23 I noticed it because ~20 strings suddenly appeared. 16:50:36 so switched openstack/common/* to openstack fixes the issue? 16:50:51 s/switched/switching/ 16:51:06 david-lyle: yes, but it may be better to remove ignore option. 16:51:36 and retranslate the strings? or is oslo not including translations? 16:52:12 previously oslo is not translated in many languages (AFAIK no languages) 16:52:51 ah, then I agree, ignore may not be desirable 16:53:07 It seems a bit late to suddenly add many new strings to the translators workload though... 16:53:52 jpich: good point. openstack/common strings takes a time to translate without reading the code. 16:54:28 hmm..... how about keeping --ignore option for Icehouse. most strings are invisible to users. 16:54:34 let's go back to the original behavior for icehouse with amotoki_'s patch and fix properly in Juno 16:54:40 Sounds good to me, and we can revisit in Juno 16:54:45 +1 16:54:53 sold 16:54:55 :) 16:55:13 we're no worse off than we were before 16:55:25 any other translation questions issues? 16:55:30 amotoki: About the .mo file issues on the agenda, there is a bug open about it: https://bugs.launchpad.net/horizon/+bug/1196982 16:55:54 let's discuss on MO files in the bug report. 16:56:00 The concern about the first patch was that we wanted a way to regenerate them automatically, maybe you know how the other projects handle it? 16:56:04 Ok! 16:56:34 3 minutes of open discussion then? ;) 16:56:44 #topic Open Season 16:57:19 "open season" eh? :) 16:58:02 I am from the middle of the US, all phrases relate to farming, hunting or fishing, can't help it 16:58:09 haha 16:58:48 Actually I had a small item, we renamed the "Images & Snapshots" panel "Images" 16:58:56 but instances snapshots do live in there too 16:59:25 I think it'd be clearer to change back the verbose name (but no need to change the class names like the patch that changed this did) 16:59:53 I think an instance snapshot is a kind of images. 17:00:04 I can imagine some confusion as to where things go after clicking "Create snapshot" on the Instances panel 17:00:24 ah... 17:00:42 hi 17:00:44 for end-users I mean. I'm not sure if the "snapshot is an image" is clear 17:00:50 anyhow I'll open a bug and we'll see in reviews I guess 17:00:52 hi 17:00:59 time is over 17:01:01 Yup 17:01:04 oops indeed. 17:01:18 pballand: Error: Can't start another meeting, one is in progress. Use #endmeeting first. 17:01:18 Thanks everyone! Have a great week. 17:01:23 hi 17:01:28 thanks, have a good week everyone 17:01:29 #endmeeting