15:00:13 #startmeeting Fedora QA Meeting 15:00:13 Meeting started Mon Jun 6 15:00:13 2011 UTC. The chair is jlaska. Information about MeetBot at http://wiki.debian.org/MeetBot. 15:00:13 Useful Commands: #action #agreed #halp #info #idea #link #topic. 15:00:17 #meetingname fedora-qa 15:00:17 The meeting name has been set to 'fedora-qa' 15:00:20 #topic Roll Call 15:00:37 * Viking-Ice here.. 15:00:41 yo 15:00:41 tgr__: Hi there ... we're just doing roll call atm 15:00:48 Viking-Ice: adamw: hey hey 15:00:50 * tflink is here 15:01:00 * athmane is there 15:01:15 well, i'm here :) 15:01:21 * jlaska greets tflink & athmane 15:01:29 * jsmith lurks 15:01:50 * j_dulaney waves 15:01:52 * kparal here 15:01:58 hi j_dulaney && kparal 15:02:21 * vhumpa says hi 15:02:39 helloooo 15:02:39 * rbergeron takes a seat 15:02:42 jlaska, vhumpa, kparal 15:02:55 okay, let's get started 15:03:08 we don't have a complicated/lengthy agenda today ... just checking in on a few recurring topics 15:03:17 as always, feel free to raise topics during open-discussion 15:03:37 and thanks to vhumpa j_dulaney and adamw for #chair'ing last week 15:03:38 .bacon 15:03:38 I love bacon, you love bacon, WE ALL LOVE BACON! 15:03:50 jlaska: was fun 15:03:53 Indeed 15:04:09 So, I'm skipping the 'previous meeting follow-up' topic today ... since that's really covered by the agenda 15:04:23 * jlaska queues adamw first .. 15:04:27 #topic Release Criteria Updates 15:04:32 well, vhumpa too really 15:04:51 What's the word on the "too-similar menu names" proposal that went out last week? 15:05:16 jlaska: I started a discussion on test + desktops mailing lists 15:05:16 * adamw defers to vhumpa 15:05:37 People are supportive of the idea that *something* needs to be done with the issue 15:05:45 #info vhumpa started a discussion on test + desktop lists last week 15:06:39 vhumpa: ideas spread from modifying the app launcher to make sure that they would differentiate the apps with same names properly - to just renaming some of the problematic apps 15:07:09 I guess depending on the solution ... a different group of people would need to make the changes? 15:07:21 For example of this issue: You all know terminal/terminal etc. 15:07:22 Indeed 15:07:39 jlaska: Yes, the first one, simply, is upstream 15:07:53 thus not something I think we can do very quickly 15:08:24 Should gnome-shell offer e.g. popups for the app icons, that would present one set of means how to deal with the issue 15:08:35 how does QA fit into this discussion as in is this not something all the *DE should take care of among themselves ? 15:08:44 But, I am not sure how reasonable it would be to push that through 15:08:45 right,viking 15:08:59 the fix isn't our problem exactly 15:09:00 Viking-Ice, +1 15:09:01 Viking-Ice: yeah ... I'm just seeing this as QA bringing this to appropriate desktop attn 15:09:14 That brings me to what FEdora can do... 15:09:27 what we're concerned with is whether this should be a release requirement 15:09:40 Indeed 15:09:57 It's a problem of a few apps really... So what we can do is merely to rename some of them, in their desktop files 15:09:57 I'm not seeing this as an release requirement 15:10:03 the QE connection: 15:10:13 enforce that it is done with a requirement 15:10:41 if we tests each desktop separately this issue will not raise, afaik 15:10:58 athmane: true 15:11:05 athmane: partially 15:11:07 The biggest issue I see is apps within the same desktop 15:11:16 there are issues even withing single desktop 15:11:17 what's the ideal outcome for this topic? ... upstream acknowledgement of the issue? 15:11:35 I suppose so 15:11:38 ... formalizing tests and applicable criteria? 15:11:58 (depending on whether it's accepted or not) 15:12:03 Meaning.. upstream acknowledgment is a "pony" perhaps 15:12:06 we should probably add it to the desktop menus test case 15:12:22 I'm thinking that a good outcome would simply be that what the user sees is different names for different apps 15:12:33 I agree with Adam on this one. 15:12:36 I don't see any feedback from anyone upstream on this topic ... have they weighed in on this yet? 15:12:52 jlaska: nope 15:13:04 maybe we should fill bugs on upstream tracking app ? 15:13:05 The actual names don't necesarily have to be different for the actual app, just the menu choices 15:13:16 are these apps that any of the *DE ship by default or is this something that is mixed apps between *DE ? 15:13:36 Viking-Ice: Default 15:13:55 For instance, within just Gnome: Softare Update and Software Updates 15:14:08 so we have a plan for testing 15:14:11 Some are issue in deafult, some become an issue when you have multiple environments installed 15:14:13 What's the next step? Should we focus on trying to get feedback/input from upstream on this topic? 15:14:19 anyway this sounds to me just something that the relevant *DE maintainers need to take care of not something related to QA per se 15:14:31 then the question is, should it be a release criterion, i.e., should we require same name situations to be resolved for release 15:14:36 adamw: should we move forward with testing and criteria without feedback from GNOME? 15:14:40 The QA angle would be enforcement 15:14:44 I would concentrate in choosing menu names around Fedora desktops to minimize this issue... Upstream should come later 15:14:57 adamw: I think so 15:15:00 jlaska: feedback would be good, i guess 15:15:08 adamw: seems like it should be required to me 15:15:16 how can we create criteria and tests without their input? 15:15:24 * jlaska might be missing something though 15:15:57 The problem lies in how Fedora names apps, which I am not sure how connected is to upstream really 15:15:58 err ... I'd want to avoid creating tests and criteria that GNOME isn't interested in honoring/fixing etc... 15:16:13 vhumpa: good point ... it really depends on the implementation 15:16:16 jlaska: well, addressing it upstream is only one approach 15:16:30 yes, I see now, gotcha 15:16:37 i can see, for instance, that if upstream GNOME decide they don't care, we would decide Fedora still does care 15:16:45 right, that makes sense 15:16:50 adamw: yes 15:16:56 so when will we know which of those routes to take? 15:16:57 sounds like a ( test ) candidate for fit and finish 15:16:59 +1 15:17:46 I guess we can conclude that if we don't get GNOME feedback, then it's up to Fedora to decide? 15:17:53 So the question is 1) Do we ask upstream to help with this 2) We just rename a few menu items in Fedora 15:18:04 * j_dulaney goes with 2 15:18:10 +1 15:18:14 both 15:18:15 Easiest solution 15:18:17 * jsmith goes with 1) 15:18:19 both 15:18:20 we'll likely go with #2 ... but I'd like to give #1 another attempt 15:18:21 vhumpa: i think that's kind of up to the devs to decide really 15:18:47 i think we may be going round in circles at this point? 15:18:54 yup ... let's wrap up on this topic 15:18:54 1) definitely too - but facing reality that it would be a more long term solution but for later 15:19:07 True 15:19:10 anyone want to approach GNOME with this topic this week? 15:19:14 if not ... I'll take it 15:19:38 or any other #action items ... feel free to grab 15:19:45 If I know how to approach them, I will 15:20:17 vhumpa: okay, thank you 15:20:29 anything else to cover on this before next week? 15:20:34 i'll sync up with you on that 15:20:40 is this only relevant to Gnome or is this problem present in all *DE we ship ? 15:20:41 Meaning: we'll be aproaching for modifing the launcher, right? You don't just mean renaming apps on Upstream side 15:20:57 Unsure on that 15:21:01 #action vhumpa/adamw - reach out to GNOME for opinions on presenting duplicate application names in overview 15:21:04 Viking-Ice: I'm not sure about within other DEs 15:21:21 Viking-Ice: KDE solves it iirc, but it's not specific to a single DE 15:21:44 While we are here, anything else on release criteria? 15:21:59 adamw: any other notable criteria changes to highlight? 15:22:30 er, i think i did some 15:22:36 heh 15:22:40 but i think we may have covered them last week 15:22:43 okay 15:22:50 oh, the 'release-blocking desktops' thing may have been this week 15:23:01 That was last 15:23:08 okay. then, i think nothing new. 15:23:12 (sorry, it's been a busy week.) 15:23:19 okay, then moving on 15:23:32 I'm switching the next two topics so we don't keep tgr__ waiting too long 15:23:40 #topic IPv6 Test Day 15:23:46 #link https://fedoraproject.org/wiki/QA/Test_Day:2011-06-08_IPv6 15:24:06 #info World IPv6 Day is happening on June 8, along with a Fedora IPv6 test day 15:24:22 so this is just intended as a check-in for Test Day preparedness 15:24:57 * j_dulaney has the network in his house setup for IPv6 already 15:25:14 i haven't checked in on this for a few days i'm afraid 15:25:17 since my last email shot 15:25:25 anyone know of any recent developments? 15:25:25 looks like we have 2 test cases linked .. .and one in need of a test case 15:25:40 tgr__: any updates/concerns on your end with regards to test day prep? 15:25:43 #link http://fedoraproject.org/wiki/QA:Testcase_NetworkManager_ipv6 15:25:47 #link 15:25:50 #link http://fedoraproject.org/wiki/QA:Testcase_NFS_ipv6 15:26:14 #info test case needed for ipv6 printing 15:26:29 Red Hat will announce the event in a blog and refer to the Fedora test and ask for participation 15:26:41 tgr__: do you know when that is going out? 15:27:01 * Viking-Ice throws in --> http://ipv6eyechart.ripe.net/ <-- for interested parties reading the meeting logs.. 15:27:03 tomorrow if everything goes right 15:27:10 awesome 15:27:18 the press blog? 15:27:20 I've written a howto which covers enabling ipv6 on RHEL/fedora 15:27:21 today would be a good day to be blogging about this for everyone else 15:27:35 using configuration file/NetworkManager 15:27:59 and instructions how to verify if a website has been reached via IPv6 or not 15:28:10 this alone should get us some feedback on any basic issues 15:28:27 tgr_: You want to link to that for info? 15:28:36 Or is it not online, yet? 15:28:48 * j_dulaney notes that he'll put something in his blog as well. 15:28:49 it's not online yet, we are having ISP troubles 15:29:04 the wiki links to rawhide live images ... I assume we just want F15 live images for this? 15:29:21 if we can't get it up within red hat I will provide the info via the fedora wiki 15:29:22 jlaska: +1 15:29:23 yeah, good catch, let's fix that 15:29:26 * jlaska fixes 15:30:37 i'm working on getting www.fedoraproject.org listed as participant on isoc.org 15:30:46 anyone want to volunteer to send an event reminder to test-announce@ ? 15:30:47 it's currently only listed as IPv6 enabled website 15:30:55 tgr__: nice! 15:31:12 jlaska: i can do it 15:31:24 adamw: thank you 15:32:03 Anyone object if I move the different setup procedures out into unique wiki pages? 15:32:07 just to clean up the main page a little? 15:32:13 or is that not really needed 15:32:16 jlaska: Good idea 15:32:18 i think that's a good idea 15:32:31 okay, I'll make a minor adjustment after the meeting 15:32:37 yeah sounds great 15:32:53 i will add instructions how to do setup if isp provides native ipv6 connectivity 15:32:56 there's some boilerplate text still in there too which we should remove 15:33:08 +1 15:33:11 adamw: like the test results stuff? 15:33:13 "Provide a list of test areas or test cases that you'd like contributors to execute. For other examples, see Category:Test_Cases. " 15:33:15 and yes 15:33:21 though we need to set up a proper table for that 15:33:29 okay ... I'll include that in my wiki cleanup 15:33:33 thanks 15:33:42 #action adamw - send test-announce@ for IPv6 test day 15:33:53 #action jlaska - test day wiki cleanup (remove boilerplate) 15:34:10 #action tgr__ - provide wiki instructions for native ipv6 connectivity 15:34:20 feel free to grab any #action's that I missed 15:35:25 tgr__: thanks for joining today ... anything else you want to cover before we move on? 15:35:40 jlaska: i think i'm done, thanks 15:35:48 tgr__: great, thank you! 15:35:54 #topic AutoQA Updates 15:36:08 * kparal goes on stage 15:36:19 it's time for a regular autoqa check-in! 15:36:24 I have only one update today 15:36:30 * jlaska sees he has plenty of unread autoqa-devel mails to catch up on 15:36:47 and that is the announcement of 'pretty patch' that was just posted into autoqa-devel 15:36:49 #link https://fedorahosted.org/pipermail/autoqa-devel/2011-June/002344.html 15:37:07 woah, that knocks out quite a few tickets :) 15:37:22 this patch should allow us to create pretty html logs 15:37:22 * j_dulaney was just looking at that 15:37:35 yay! ... 15:37:36 #link http://kparal.fedorapeople.org/autoqa/upgradepath2.html 15:37:36 Shiny 15:37:40 they should be more concise and readable than the previous logs 15:38:13 ooooooh pretty 15:38:19 More shiny 15:38:22 Thus the name :) 15:38:26 * j_dulaney likes shiny 15:38:27 now people will break their packages just to see the failure 15:38:38 now we need to review the patch and merge into master. but the most of the work should be done already 15:38:45 adamw: let's hope not :) 15:38:46 #link http://kparal.fedorapeople.org/autoqa/depcheck.html 15:39:13 that's awesome stuff 15:39:22 I think I can understand depcheck output now :) 15:39:23 Wow, the shiny just keeps piling up 15:39:33 hopefully I can stop bugging tflink for help with that! 15:39:37 this patch should be the core of 0.5.0 release, hopefully to come really soon 15:39:51 together with email reduction patch from tflink 15:40:01 How's that one going? 15:40:16 pretty much done, other than a little bit more of cleanup and testing 15:40:30 * j_dulaney keeps getting distracted; cheerleaders 15:40:40 I'm planning to send out a patch email to autoqa-devel today 15:40:59 tflink: great 15:41:06 tflink: nice! 15:41:27 tflink: I still haven't heard back yet on the email notification for all passed results 15:41:27 +1 15:41:49 jlaska: I'll make sure that the configuration works so that we can change it later 15:41:54 without changing code 15:41:58 jlaska: I have talked to some developers and they liked the idea of not getting bothered when everything works fine 15:41:59 tflink: but I now see my lucky ping recipient online ... so I'll see if we can get some more info 15:42:41 kparal: okay, good to know ... sounds like this will be tunable (without patching) based on how tflink is implementing 15:43:27 well, and that was the big announcement of today. I have no further updates 15:43:31 #info tflink finalizing test result email reduction patchset - expecting patch out for review later today 15:43:52 With help from lmr, I've been packaging what will become autotest-0.13.0 15:44:09 finding a few bugs here and there (nothing major), but so far it's working okay 15:44:35 note, those changes are in the autoqa fedora-15-testing repo ... so make sure you *arent* using that when you are testing for the next autoqa release 15:44:35 Need to run, bye for now everybody. 15:44:38 vhumpa: cya! 15:45:14 #info Packaging for soon-to-be-released autotest-0.13.0 almost complete 15:45:39 kparal: tflink: anything else to cover on AutoQA? I guess it depends on patch review for when we'll start the packaging machine for autoqa-0.5.0 ? 15:46:05 jlaska: no. yes. :) 15:46:30 nothing I can think of. review and testing for this week, yes 15:46:38 heh, okay ... thanks for the autoqa updates all 15:46:52 #topic Open Discussion - 15:47:08 * rbergeron raises her hand 15:47:10 I've got just a quick status update if there are no other open discussion topics 15:47:14 rbergeron: what's up? 15:47:21 jlaska: go first, i have a few minor things 15:47:26 s/I've got/I have/ 15:47:45 #topic Open Discussion - fedora-qa F15 TRAC tickets 15:48:07 I'm doing some TRAC ticket maintenance to prepare for the retrospective tickets 15:48:09 * Viking-Ice points out we need to start looking at potential features being introduced and if we need to cover that ( grub2 and btrfs pop up to my mind ) 15:48:11 #link https://fedorahosted.org/fedora-qa/query?status=new&status=assigned&status=reopened&milestone=Fedora+15 15:48:28 Viking-Ice: +1 15:48:37 Viking-Ice: good point ... I believe grub2 is in TRAC already, we'll need something for btrfs I believe 15:48:56 there are still some open tickets in the 'Fedora 15' TRAC milestone ... I closed out all the completed test events already 15:49:15 and I'll likely start annoying ticket owners to find the most suitable outcome for any remaining tickets 15:49:21 prepare to be annoyed! 15:49:52 #chair rbergeron 15:49:52 Current chairs: jlaska rbergeron 15:49:55 rbergeron: #topic away 15:50:00 Woot. thanks 15:50:04 jlaska: Since I'm already using btrfs, I can start on test cases for it 15:50:04 #topic Schedule 15:50:12 #link https://fedoraproject.org/wiki/Releases/16/Schedule 15:50:21 #info schedule is posted, feedback is welcome, please. 15:50:41 rbergeron: is there any *easy* way to diff the F15 and F16 schedules? 15:50:51 I know you guys have a retrospective; if there are things to be converted into schedule changes, let me know. 15:50:59 jlaska: ahahahahaha. 15:51:08 okay, I'll be bugging you about any schedule topics that come out of the retrospective 15:51:15 I assume you mean the "original" schedule vs. schedule as it turned out? 15:51:19 rbergeron: I should clarify ... a human-readable diff :) 15:51:26 I haven't added anything *new* for you guys. 15:51:33 I'd speculate that the easiest way would be this: 15:51:38 are any of the windows noticeably different? 15:52:09 http://rbergero.fedorapeople.org/schedules/f-15/f-15-quality-tasks.html vs. http://rbergero.fedorapeople.org/schedules/f-16/f-16-quality-tasks.html 15:52:17 adamw: they shouldn't be. 15:52:32 Other than - the dates showing in *that* F15 schedule are the slipped dates. 15:52:39 you might want to compose any previous schedules as in what they where and how they turned out to be 15:52:39 ah, okay 15:52:57 This schedule is more or less as the original f15 schedule was. 15:53:16 Just, 6 months later. 15:54:01 I'll try to come up with more focused feedback after finishing the retrospective ... but the branch timing and Alpha still lend to slippage 15:54:08 I don't have any great ideas at the moment 15:54:14 will try to process 15:54:33 * rbergeron nods 15:54:47 happy to have a brainstorming meeting with you / whoever else is interested. 15:54:51 (or should be there, lol) 15:54:52 * jlaska notes ... there isn't an option to "Create a new month" in the schedule 15:55:12 * Viking-Ice is not foreseeing any slips this release cycle.. 15:55:30 any other schedule q's/comments? 15:55:59 #topic Fixing features 15:56:06 I'll keep this brief: 15:56:43 Basically some folks have mentioned that the feature process is perhaps not quite as robust as it could be, or could account for "different types of features" better (aka: marketing-ish features vs. stuff that is going to break the universe features) 15:56:54 https://fedoraproject.org/wiki/Fixing_features <--- your feedback is welcome. 15:57:12 #info if you have any thoughts on the good, the bad, the ugly in the feature process, feel free to add your commentary to wiki page. 15:57:17 And that's all on that. 15:57:24 * jlaska queues for reading 15:57:25 * rbergeron looks around before continuing 15:57:51 features aren't mandadory process afaik .. 15:58:26 yes 15:58:47 well, i think that depends. and I think that's part of the problem. 15:58:55 but not going to open pandora's box at the moment. :) 15:59:04 Just wnated to give a heads-up to that, if you're interested. 15:59:07 yes, save that for #pandora :) 15:59:12 #topic Cloud stuff 15:59:35 You may notice that there are a boatload of cloud features for f16, we're already talking about test-day stuff, possibly breaking into two test days. 15:59:45 ke4qqq posted something to the cloud list for anyone who might be interested in helping us work that stuff out. 15:59:49 http://lists.fedoraproject.org/pipermail/cloud/2011-June/000632.html 16:00:06 (basically, we haz lots of stuff, we should start planning now rather than last second like last time.) 16:00:06 awesome 16:00:23 because there is going to be a lotta stuff. :) 16:00:23 it's definitely a good idea to hash out a clear test day topic 16:00:31 'cloud test day' is pretty vague, so splitting it like last time is good\ 16:00:40 Are there going to be any clouds setup specifically for us to test on? 16:00:40 we can always make room for more test days, so don't worry about having too many 16:00:45 yeah, I think that worked pretty well 16:00:47 wasn't too vague 16:00:49 yeah, and we can group them by different types of cloud apps. 16:00:56 cloud test week :) 16:01:12 j_dulaney: unknown. that's part of what we need to solve ahead of time, so we can get that kind of thing set up for folks without it being a nightmare. 16:01:17 anyway ... will have to see how the features fall out 16:01:18 cloud test week sounds like a good way to proceed 16:01:22 s/fall/pan/ 16:01:33 +1 16:01:40 But: would appreciate any feedback if you're on the cloud list. :) 16:01:43 Viking-Ice: I know you like the test week idea ... assuming we have a series of clear topics, that might work 16:01:48 rbergeron: okay 16:02:03 And would like to invite folks to come to a meeting maybe in 2 weeks or so, but will update on that next week. :) 16:02:07 That's it. :) 16:02:17 Sweet 16:02:25 * j_dulaney is getting hungry 16:02:51 #topic Open Discussion - Last call for topics 16:03:05 jlaska, cloud test week which would cover Aeolus,CloudFS,CloudStactk, Sheepdog testing.. 16:03:06 * jlaska sets the fuse for 2 minutes 16:03:40 Viking-Ice: yeah, could be ... will see what comes out of that thread 16:04:05 1 minute until #endmeeting ... 16:04:14 * j_dulaney wanders off in search of food and to start thinking about btrfs test case 16:04:19 Peace, y'all 16:04:21 Viking-Ice: yeah, and all the other ones I know of in the pipeline but aren't posted yet (openstack, pacemaker-cloud, $others) 16:04:24 cya j_dulaney 16:04:35 30 seconds until #endmeeting ... 16:05:02 Thanks everyone for your time today!! ... I'll follow-up with minutes to the list 16:05:06 #endmeeting