Wednesday, 2017-06-07

*** diablo_rojo has quit IRC00:56
*** diablo_rojo has joined #storyboard02:54
*** gomarivera has joined #storyboard02:58
*** gomarivera has quit IRC03:03
*** jamesmcarthur has joined #storyboard03:04
*** jamesmcarthur has quit IRC03:08
*** jamesmcarthur has joined #storyboard03:17
*** jamesmcarthur has quit IRC03:19
*** jamesmcarthur has joined #storyboard03:20
*** jamesmcarthur has quit IRC04:22
*** diablo_rojo has quit IRC04:33
*** jamesmcarthur has joined #storyboard06:23
*** jamesmcarthur has quit IRC06:27
*** gomarivera has joined #storyboard08:59
*** gomarivera has quit IRC09:03
*** jamesmcarthur has joined #storyboard09:23
*** jamesmcarthur has quit IRC09:27
*** dongfeng has joined #storyboard09:42
*** dongfeng has quit IRC10:26
*** jani has joined #storyboard11:15
*** jamesmcarthur has joined #storyboard11:24
*** jamesmcarthur has quit IRC11:29
janisorry if this is offtopic but i sort of thought you guys are maintainers of git-review ?13:34
janiand follow up if the answeris yes - Is there any way to prevent setting of topic from local branch name ?13:35
*** gomarivera has joined #storyboard13:59
persiajani: I believe the answer is "no", but I may be mistaken14:09
persiaBut I believe the answer is `git review -t ${topic} ...` or `git review --topic=${topic} ...`, based it git-review(1).14:09
persiaErr, respecting the order of the questions, that is.14:10
*** diablo_rojo has joined #storyboard14:21
*** jamesmcarthur has joined #storyboard14:22
janiyeah, i was sort of expecting that.  thanks.14:35
janibackstory for this was that someone setup some work for our jenkins that will trigger "topic" builds if topic is set and most who work on branches, submit a single change to gerrit. and because the jenkins scripting was checking for presence of GERRIT_TOPIC variable, ofcourse all the changes where topic builds (because branchname == topic)14:40
janiso i was just wondering if there's a sane way to actually enforce that topic is only set if/when specifically requested ..14:41
janican work around this..14:41
fungijani: you want the #openstack-infra channel, but anyway yes that's something i want to do in the next git-review release (which is also waaay overdue)14:53
funginow that gerrit lets you set/change review topics without having to create a new patchset, the topic default munging in git-review is obsolete14:54
fungiand actively harmful in my opinion14:54
fungijani: https://review.openstack.org/32653914:55
fungidiablo_rojo: if you're around, mind submitting a patch to add yourself as the phase 1 assignee at https://git.openstack.org/cgit/openstack-infra/infra-specs/tree/specs/task-tracker.rst#n21814:57
fungidiablo_rojo: and then remove the entry for that spec from the help wanted section of the index at https://git.openstack.org/cgit/openstack-infra/infra-specs/tree/doc/source/index.rst#n7714:58
*** gomarivera has quit IRC15:08
*** gomarivera has joined #storyboard15:57
*** diablo_rojo has quit IRC16:28
*** diablo_rojo has joined #storyboard17:19
*** jamesmcarthur has quit IRC17:56
*** jamesmcarthur has joined #storyboard18:11
*** jamesmcarthur has quit IRC18:13
*** jamesmcarthur has joined #storyboard18:15
*** jamesmcarthur has quit IRC18:15
*** jamesmcarthur has joined #storyboard18:16
*** jamesmcarthur has quit IRC18:16
*** jamesmcarthur has joined #storyboard18:16
*** jamesmcarthur has quit IRC18:17
*** jamesmcarthur has joined #storyboard18:17
*** jamesmcarthur has quit IRC18:18
*** jamesmcarthur has joined #storyboard18:18
*** jamesmcarthur has quit IRC18:18
*** jamesmcarthur has joined #storyboard18:19
*** jamesmcarthur has quit IRC18:19
*** jamesmcarthur has joined #storyboard18:19
*** jamesmcarthur has quit IRC18:20
*** jamesmcarthur has joined #storyboard18:20
*** jamesmcarthur has quit IRC18:21
*** jamesmcarthur has joined #storyboard18:21
*** jamesmcarthur has quit IRC18:21
*** jamesmcarthur has joined #storyboard18:23
*** jamesmcarthur has quit IRC18:23
*** jamesmcarthur has joined #storyboard18:24
*** jamesmcarthur has quit IRC18:25
*** jamesmcarthur has joined #storyboard18:25
*** jamesmcarthur has quit IRC18:25
*** jamesmcarthur has joined #storyboard18:26
*** jamesmcarthur has quit IRC18:26
*** jamesmcarthur has joined #storyboard18:27
*** jamesmcarthur has quit IRC18:27
*** jamesmcarthur has joined #storyboard18:28
*** jamesmcarthur has quit IRC18:28
*** jamesmcarthur has joined #storyboard18:29
*** jamesmcarthur has quit IRC18:29
*** jamesmcarthur has joined #storyboard18:30
*** jamesmcarthur has quit IRC18:31
*** jamesmcarthur has joined #storyboard18:31
zara_the_lemur__storyboard meeting in 27 minutes! It'll be a short one!18:32
*** jamesmcarthur has quit IRC18:32
*** jamesmcarthur has joined #storyboard18:33
*** jamesmcarthur has quit IRC18:33
*** jamesmcarthur has joined #storyboard18:34
*** jamesmcarthur has quit IRC18:34
*** jamesmcarthur has joined #storyboard18:35
*** jamesmcarthur has quit IRC18:36
*** jamesmcarthur has joined #storyboard18:36
*** jamesmcarthur has quit IRC18:36
*** jamesmcarthur has joined #storyboard18:37
*** jamesmcarthur has quit IRC18:38
*** jamesmcarthur has joined #storyboard18:38
*** jamesmcarthur has quit IRC18:43
*** jamesmcarthur has joined #storyboard18:53
zara_the_lemur__storyboard meeting in 1m!18:58
zara_the_lemur__I had so many windows open, haha19:37
persiaBother.  Counted hours wrong.20:00
persiaAfter reading backscroll, I wanted to mention that an interloper (other than I)  joined in the watcher discussion, talking about wanting a project-wide common interface to see what was happening with milestones.20:01
persiaSomething like the automated worklists with e.g. pike-3 that were discussed a few days ago, but as a common model for *every* openstack project.20:02
persiaI suspect it can just be openstack-specific best practice documentation to make something like that happen (although the worklist owner should probably be someone formally important), but I suspect that might be a wider need (especially for folk who want to view *other* project's status)20:02
-openstackstatus- NOTICE: The Gerrit service on review.openstack.org is being restarted now to clear some excessive connection counts while we debug the intermittent request failures reported over the past few minutes20:07
*** jamesmcarthur has quit IRC20:11
*** jamesmcarthur has joined #storyboard20:13
zara_the_lemur__as in, to see what things had been fixed in which releases, retrospectively?20:24
zara_the_lemur__I would guess that most people interested in that information would be interested in a cluster of projects at once, though that is just a guess20:25
persiaMore, which things are targeted at a given milestone.20:28
persiaBased on the earlier discussion with jdandrea and fungi, I think the right answer is to suggest everyone tag things appropriately, and then have an automatic worklist that shows everything, whether complete or not.20:29
persiaAlong with smaller automatic worklists that show things for individual projects.20:29
persiaThe trick is having the display work right, so that all the right information is on the page.20:29
persiaMight end up needing a new (small) client, or maybe the worklist view would work.20:29
* persia isn't quite sure20:29
fungiyeah, i'm surprised watcher is relying on that in lp. most projects stopped creating series and targeting bugs becaus ethe release managers decided it was too much work20:30
persiaThe use case that was mentioned in the meeting earlier today was when someone caught something interesting in a meeting (or from a mailing list) by accident, and wanted to get a quick overview of current activity in a project (outside their normal cluster).20:30
fungithat's why it was never included in the original specifications for the sb design (along with blueprints, for similar reasons)20:31
persiafungi: It wasn't watcher: it was an interloper who wanted to be able to see what watcher was doing.20:31
fungioh, gotcha20:31
zara_the_lemur__my understanding of sb milestones is that they're retrospective, so you can say what was done in one but not what will be done in one20:31
zara_the_lemur__(though my memory is somewhat hazy)20:32
zara_the_lemur__(iirc you can't create a milestone in the future)20:32
persiaThe impression I had was that watcher's big feature-wish was blueprint dependencies.  For that, I thought a text file in a spec repo to be processed by dot or graphviz would probably be the simplest way (in terms of actual development required vs. just documentation).20:32
fungiwell, the release management tools are likely going to use task references in commit messages to leave comments on the corresponding stories the first time a release is tagged including the commit20:33
persiazara_the_lemur__: You can also target a task to a milestone, even a future milestone.20:33
fungiwhich would vover the retrospective part20:33
persiaMore importantly (from the perspective of my hypothetical model of the interloper) blueprints can be targeted, so one can get an idea of what is happening there.20:33
persia+1 on using the release management tools for retrospectives (and encouraging appropriate metadata tagging in commit messages to support this)20:34
fungiright, i agree story tags should work fine. we can come up with standard ones in openstack land we encourage projects to use20:34
persiazara_the_lemur__: I think your memory of SB milestones is correct, which is why I fiddle with my patch to remove them entirely every once in a while.  LP milestones are a little different.20:34
persiafungi: I suspect the release team's nomenclature should work (e.g. "pike-3")20:35
* fungi nods20:35
* zara_the_lemur__ isn't sure she's right about sb milestones, has checked and notes they have expiration dates, and once they have expired you can't map tasks to them; beyond that would need to spend more time refreshing20:40
* zara_the_lemur__ is super sleepy today20:40
persiaI think the current SB milestone implementation isn't well-specified or well-implemented, and is best ignored for now :)20:40
persiaBut don't let me speculation keep you up late20:40
* persia should probably go do something else anyway20:41
zara_the_lemur__hah, I just didn't want to give the impression I'd researched it thoroughly recently20:41
zara_the_lemur__'nobody knows what they are, but whatever they are, they probably aren't what people want' probably covers it20:42
SotK+120:49
SotKI think our milestones should go away to be optionally replaced if needed20:49
* persia will set aside some time to rebase the patch-in-process this weekend, and submit even if he can't get the migrations to work right this time21:00
*** betherly_ has joined #storyboard21:24
*** jamesmcarthur has quit IRC21:29
*** betherly has quit IRC21:29
*** mcnanci_ has quit IRC21:29
*** gary_perkins has quit IRC21:29
*** betherly_ is now known as betherly21:29
*** jamesmcarthur has joined #storyboard21:30
*** gary_perkins has joined #storyboard21:30
*** mcnanci has joined #storyboard21:32
*** jamesmcarthur has quit IRC21:50
*** gomarivera has quit IRC22:53
*** gomarivera has joined #storyboard22:54
*** gomarivera has quit IRC22:58
*** openstack has joined #storyboard23:14
*** gomarivera has joined #storyboard23:30
*** gomarivera has quit IRC23:34

Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!