16:00:11 #startmeeting Fedora QA Meeting 16:00:11 Meeting started Mon Jan 11 16:00:11 2010 UTC. The chair is kparal. Information about MeetBot at http://wiki.debian.org/MeetBot. 16:00:11 Useful Commands: #action #agreed #halp #info #idea #link #topic. 16:00:17 #meetingname qa 16:00:17 The meeting name has been set to 'qa' 16:00:23 #topic gathering 16:00:36 ok, welcome all, let's start 16:00:37 * maxamillion is here ... but about to walk over to a meeting, taking laptop 16:00:39 who's here? 16:01:08 * jlaska lurking ... but likely unresponsive :( 16:01:24 * tk009_ here 16:01:39 * kparal poking wwoods 16:03:17 no one really seems to be about today 16:03:50 ok, seems that's all for today 16:03:52 ok, back 16:04:15 I have an unfortunate message for you - jlaska is out, so I will lead the meeting 16:04:18 * Viking-Ice jumps in.. 16:04:29 that means it will be short and poorly moderated :) 16:04:44 ack! pokes! 16:04:47 I hoped to have wwoods help, but he's somewhere out too 16:04:49 ah 16:04:55 came to save me, good! 16:04:57 heh! 16:05:28 alright, jlaska and adamw are not here. the same is probably for lili and rhe 16:05:33 let's go on 16:05:34 sorry, was distracted fixing up my mail filters 16:05:51 #topic Previous meeting follow-up 16:05:54 #topic Previous meeting follow-up 16:06:06 * Viking-Ice investigating adding an LXDE test day to the F-13 test day schedule 16:06:23 any updates on that Viking-Ice? 16:06:42 * kparal still learning the meetbot tags 16:06:47 #info Viking-Ice investigating adding an LXDE test day to the F-13 test day schedule 16:07:27 * jlaska encourages excessive use of # info :) 16:07:43 ok, let's give some time to Viking-Ice 16:07:48 #info maxamillion investigating scheduling a XFCE test day to F-13 test day schedule 16:08:07 should I talk or wait for Viking-Ice to finish? 16:08:13 go ahead 16:08:26 he didn't start yet :) 16:09:13 ok, so I spawned a conversation with cwickert and nirik and we've hashed out some good stuff as far as test planning and how to improve the test-day, we've also started as xfce@lists.fedoraproject.org because we want to start getting more of a community built around the Xfce SIG/Spin 16:09:40 nice 16:09:55 we discussed the possibility of pushing forward with Xfce 4.8 and the testing/QA implications that would bring up and considered submitting it as a feature 16:10:16 sorry I to finish some stuff here @ work 16:10:23 but after some review and some discussion with upstream there is potential for the release to slip and we don't feel there would be enough time to test thoroughly 16:10:23 Viking-Ice: no problem 16:10:51 maxamillion: I talked to Jannis, the release manager today and he said he thinks they will be on time 16:11:05 but we do plan to be working closely with upstream in pushing 4.8 to rawhide as soon as possible after the Fedora 13 branch 16:11:09 maxamillion: sounds like a prudent decision 16:11:13 cwickert: oh? 16:11:30 cwickert: should we bring that back up on xfce@lists.fp.o for a topic of conversation? 16:11:38 ok 16:11:47 spoke to cwickert and he agreed on having an lxde test day would be a great idea next task write lxde test and find a date for the day 16:11:51 for the interested, here is the detailles schedule: https://fedoraproject.org/wiki/Features/Xfce48#Detailed_Description 16:12:04 Viking-Ice: just a second to let maxamillion finish, thanks 16:12:10 cwickert: oh! awesome, thanks for the info and the link 16:12:22 kparal: can you #link that wiki page please? 16:12:27 sure 16:12:30 #link https://fedoraproject.org/wiki/Features/Xfce48#Detailed_Description 16:12:34 #link https://fedoraproject.org/wiki/Features/Xfce48#Detailed_Description 16:12:43 kparal: thank you :) 16:12:46 feel free to correct my moderation :) 16:12:57 maxamillion: should I give you another action point for the next week? 16:12:58 lol 16:13:25 * nirik notes no need for link, it should just pick up anything that looks like a URI. 16:13:27 kparal: uhmmm... yeah, I think that would be a good idea. Label it something along the lines of "Xfce 4.8 update" 16:13:38 nirik: good to know 16:14:11 #action maxamillion bring back conversation about Xfce 4.8 updato on xfce@lists.fp.o and inform about conclusion 16:14:28 alright, anything to add or let Viking-Ice talk? 16:14:42 that's more or less all I've got for now, I still need to update a wiki page on the Xfce test day stuff, but that's about all I have to report 16:14:57 ok, thanks 16:15:05 anytime :) 16:15:07 Viking-Ice: are you ready now? 16:15:13 Ah yeas spoke to cwickert and he agreed on having an lxde test day would be a great idea next #action write lxde tests and find a date for the day 16:15:16 #info Viking-Ice investigating adding an LXDE test day to the F-13 test day schedule 16:15:46 Viking-Ice: assign that action item to you I suppose? 16:15:46 It would be good to have an lxde mailing list ( since it's twice as popular as XFCE ) 16:15:50 kparal yes 16:15:55 especially the new login manager LXDE will need to be tested 16:15:56 #action Viking-Ice write lxde tests and find a date for the day 16:16:10 * wwoods senses a growing rivalry 16:16:35 wwoods: between Xfce and LXDE? nope, it's getting better in fact 16:16:37 wwoods: not yet, but we might need to spawn one ;) 16:16:59 so who can we talk about new mailing list for lxde? 16:17:07 there were some problems in the past, but we figured them out and work together now 16:17:09 *talk to 16:17:10 but yeah ... we actually have a pretty good working relationship, similar ideals but very different approach and solution 16:17:27 nirik: can you set up lxde@lists.fedoraproject.org? 16:17:28 kparal: file a infrastructure ticket. 16:17:28 jds2001 for mailing list 16:17:35 and add me as admin? 16:17:56 yeah, I can at some point here... or if you file a ticket someone else could get to it first. ;) 16:18:01 I suppose so if you are up for it.. 16:18:13 nirik: anyone can file a ticket? 16:18:19 yes. 16:18:23 anyone with a FAS account 16:18:25 #action cwickert to file an infrastruture ticket for the lxde mailing list 16:18:35 (oh it seems like a *friendly* rivarly. the best kind!) 16:18:45 https://fedorahosted.org/fedora-infrastructure/ 16:18:50 alright. meetbot registers the action items even for non-moderators, right? 16:19:03 not sure 16:19:05 no 16:19:09 #action cwickert to file an infrastruture ticket for the lxde mailing list 16:19:19 ok 16:19:21 wwoods: lol, yeah .... I was honestly a bit surprised to see the popularity of lxde .... I wasn't aware, but good for them! Its a great project :) 16:19:50 people want a smaller footprint it makes sense 16:20:00 goes in circle I suppose.. 16:20:09 yup, and with netbooks being so popular its even more prevelent 16:20:31 perfect, anything else on mind on this topic? 16:20:58 nothing from me.. 16:21:10 ok, thanks Viking-Ice 16:21:29 now we have here a few action items for which the people are not available right now 16:21:41 * adamw and rhe to discuss ways to add install testing as a QA activity 16:21:55 * maxamillion throws #info at kparal 16:21:57 :) 16:21:58 so I guess I will just reschedule them for next meeting 16:22:11 #info adamw and rhe to discuss ways to add install testing as a QA activity 16:22:15 #action adamw and rhe to discuss ways to add install testing as a QA activity 16:22:28 * kparal catches, thanks maxamillion 16:22:36 #info jlaska will reach out to mclasen for thoughts on F-13 fit'n'finish test days 16:22:41 kparal: also we can ask adamw/rhe (whenever they are around) to put a post on the planet if they have anything new and shiny to talk about 16:22:57 wwoods: good idea 16:22:59 kparal: anytime ;) 16:23:13 here is the link to the LXDE mailing list ticket: https://fedorahosted.org/fedora-infrastructure/ticket/1921 16:23:22 #link https://fedorahosted.org/fedora-infrastructure/ticket/1921 16:23:28 thanks 16:23:40 jlaska or mclasen here? 16:23:54 seems not, let's reschedule 16:24:01 #action jlaska will reach out to mclasen for thoughts on F-13 fit'n'finish test days 16:24:12 kparal: I updated my status on the wiki . 16:24:17 just a bit late for updates sorry :( 16:24:32 jlaska: oh sorry, didn't refresh 16:24:48 so update from jlaska: Ideas posted to Talk:QA/Fedora_13_test_days. A main fit'n'finish track is likely not in plan for F-13. 16:24:54 kparal: no worries ... I was late to post :( 16:25:03 #link https://fedoraproject.org/wiki/Talk:QA/Fedora_13_test_days 16:25:32 so it seems there will be no main fit'n'finish track for F13 16:25:58 another item 16:26:03 #info jlaska to reach out ot fedora-{test,devel}-list for trading reviews 16:26:14 update from jlaska: Request sent, see https://www.redhat.com/archives/fedora-devel-list/2010-January/msg00127.html 16:26:22 #link https://www.redhat.com/archives/fedora-devel-list/2010-January/msg00127.html 16:26:42 kparal: I'm still looking for some help there ... so if anyone is interested in the first of many package reviews related to autoqa ... this is a nice way to help the project 16:27:07 jlaska: I think pmuller offered some package review, am I right? 16:27:26 oh yeah, so all the lists have moved to lists.fedoraproject.org 16:27:34 so: http://lists.fedoraproject.org/pipermail/devel/2010-January/043856.html 16:27:40 ah 16:27:41 same message, just in the new location 16:27:47 #link http://lists.fedoraproject.org/pipermail/devel/2010-January/043856.html 16:27:54 thanks wwoods 16:28:12 kparal: oh I musted have missed any comments from pmuller 16:28:19 kparal: is this for beakerlib review or for autotest-client? 16:28:37 jlaska: I heard something on the conference call we had about beakerlib 16:28:44 jlaska: but I may be wrong 16:28:50 kparal: oh right ... you're right ... that should be posted soon 16:29:36 ok, if anyone interested in package reviewing, you know where to go 16:29:42 last action item 16:29:46 #info jlaska to reach out to beland for guidance/ideas on how to document the process (or point to existing documentation) for how bugs are noted (common_bugs, release notes, install guide etc...) How to determine which bugs land in which place 16:30:01 no action taken on that yet, so just reschedule 16:30:06 #action jlaska to reach out to beland for guidance/ideas on how to document the process (or point to existing documentation) for how bugs are noted (common_bugs, release notes, install guide etc...) How to determine which bugs land in which place 16:30:23 and now for something completely different 16:30:28 #topic AutoQA project update 16:30:50 wwoods: any news on your front? 16:31:02 (save me from typing for some time :)) 16:31:11 heh! let's see 16:31:35 mostly I've been: 1) filing/updating tickets for future work with respect to installation testing 16:32:00 we have a new(ish) milestone: https://fedorahosted.org/autoqa/milestone/Automate%20installation%20test%20plan 16:32:18 #topic AutoQA project update - install automation 16:32:25 #link https://fedorahosted.org/autoqa/milestone/Automate%20installation%20test%20plan 16:32:28 the tickets (mostly) outline the work that will need to be done to lay groundwork for automating the installation test plan 16:32:45 there's some improvements needed in our little autoqa virtguest library 16:32:53 #info wwoods updated tickets regarding install automation 16:32:54 and some other setup code and things 16:33:25 mostly that was from some discussions at FUDCon and the documentation sprint just before the new year 16:33:41 I don't expect that we'll begin work on that stuff *immediately* 16:34:08 the autoqa depcheck test(s) are coming first 16:34:14 that's: 16:34:15 https://fedorahosted.org/autoqa/milestone/autoqa%20depcheck 16:34:24 #topic AutoQA project update - Deps/conflicts prevention 16:34:33 # link https://fedorahosted.org/autoqa/milestone/autoqa%20depcheck 16:34:37 #link https://fedorahosted.org/autoqa/milestone/autoqa%20depcheck 16:34:45 this requires a new hook - post-bodhi-update - which should trigger when someone files a new update request in bodhi 16:34:58 I've been working with lmacken to figure out how to query for that info 16:35:06 (and fix some bugs in the bodhi API) 16:35:23 #info wwoods working with lmacken to figure out how to query bodhi 16:35:24 I'll be hacking on the API today, and hopefully have a new release into staging today/tomorrow 16:35:33 lmacken: awesome, thank you 16:35:38 very nice 16:35:47 ninja status 16:35:50 :P 16:36:14 so that milestone is missing the important ticket - actually write the depcheck test 16:36:19 I'll file that in a bit 16:36:26 :) 16:36:45 ok, thank you for the info wwoods 16:36:49 wwoods: so we'll probably want bodhi to prevent an updates push if there is an autoqa depcheck job in progress, or if some updates failed it, I assume? 16:36:56 so that's what I'm working on at the moment - that and working with kparal on rpmguard, which was just merged into the master branch of autoqa 16:37:15 lmacken: each update will be checked individually - we'll want some way to flag updates which fail the test 16:37:22 and/or prevent them from being pushed 16:37:41 wwoods: gotcha. will autoQA have a JSON API, or somethign? 16:37:45 I assume the simple implementation would just be to add negative karma from autoqa 16:37:48 or will bodhi have to listen to the AMQP bus? 16:38:03 ah, having autoqa add a comment to bodhi would be great 16:38:03 wwoods: is satisfactory just not to push it automatically, or you want the owner to prevent push until some QA approves or issue is fixed? 16:38:07 that would make things much easier 16:38:10 well autoqa right now is all about testing, not so much with the reporting right? 16:38:27 we still need to have a test results tracking software? 16:38:35 right - currently autoqa doesn't really have anything to query 16:38:50 in the future it'll be emitting AMQP messages on test completion, I'm sure 16:39:17 but that's the Glorious Future and we're stuck in the grim present 16:39:21 and our test results software would be listening on the bus to get it's updates 16:39:45 so: initial simple implementation would just be to have the test give karma and a useful message 16:40:00 #info Glorious Future will be better than grim present 16:40:16 near-future we may want a special way for autoqa to mark an update as bad, or to give autoqa power to retract a request 16:40:22 kparal: ha 16:40:28 :) 16:41:02 lmacken: so give some thought to a good way to implement that and we'll talk more about it once there's actually a working hook/test 16:41:23 wwoods: sounds good 16:41:35 on the koji side we have a vague idea how to implement it - have new builds land in, say, dist-blah-candidate-untested 16:41:54 and autoqa will have the power to move packages that pass their tests to dist-blah-candidate 16:42:24 (or something that listens to autoqa output) 16:42:26 right 16:42:28 hmm... we could also move tested builds to dist-f13-updates-tested, or something, and only mash autoqa-tested tags 16:42:33 from a bodhi perspective 16:42:47 lmacken: interesting - we'll talk more about this later, I'm sure 16:42:54 yeah, I'll churn on it 16:43:09 ok, thank you guys 16:43:20 well, now I will say a few words about my stuff 16:43:30 #topic AutoQA project update - rpmguard package comparison 16:43:44 so, there are some updates on rpmguard 16:43:58 first of all it's finally merged in master, thanks to wwoods 16:44:24 so it's closer to be really working then ever 16:44:25 \o/ 16:44:38 #info rpmguard merged in master 16:44:49 also, I have created a small blogpost about recent updates 16:44:51 #link http://kparal.wordpress.com/2010/01/11/rpmguard-a-wiki-page-and-a-new-output-format/ 16:45:31 I have created a separate wiki page for rpmguard, so it should be easier now to collect all basic data - what, where, why, when 16:45:33 :) 16:45:36 #link https://fedorahosted.org/autoqa/wiki/rpmguard 16:45:54 there's even some origin of a documentation 16:46:07 I have also changed the output format to be more rpmlint-like 16:46:49 the small task ahead is to start sending output to autoqa-results 16:46:51 * jlaska a fan of the updated output format :) 16:47:16 and the big task ahead is to publish results in some other format then 50 emails/day 16:47:21 heh 16:47:36 which I have started to discuss in autoqa-devel today 16:48:01 ok, that would be about rpmguard 16:48:11 #topic AutoQA project update - packaging/deployment 16:48:20 there is some update from jlaska about packaging autoqa 16:48:27 let's copy n paste it here 16:48:41 autotest-client - awaiting review - RHBZ #548522 16:48:54 autoqa - packaged (see autoqa-0.3-2.fc12.src.rpm), will submit once autotest-client is accepted 16:49:00 gwt-* - investigating scope of packaging needed 16:49:05 autotest - packaged (see autotest-0.11.0-4.el5.src.rpm, will submit for review after gwt* accepted 16:49:31 so, the things are moving forward, good job, jlaska :) 16:49:45 packaging gwt's deps is going to be a nightmare 16:49:51 yes 16:49:58 because each dep will spawn another 2 or 3 deps 16:50:03 blarg 16:50:03 I'll be soliciting help on that task 16:50:04 if not more 16:50:07 let's employ ghostbusters 16:50:11 is it worth trying to do a sprint or something? 16:50:19 wwoods: yeah ... that was one suggestion 16:50:34 * wwoods thinks of it as more of a siege 16:50:36 wwoods: once I close our scoping all the packaging changes ... that would be one good idea to tackle it 16:50:55 #info jlaska considering a sprint about packaging gwt 16:51:37 sorry, I have been disconnected 16:51:51 jlaska: did anyone review your autotest-client package yet? 16:52:07 maxamillion: there was an updated comment this morning ... I'll be checking that shortly 16:53:01 * jlaska notes ... meeting conflict ended 16:53:02 jlaska: ok, let me know because if not then I should be able to get a formal review in this evening (*should*) 16:53:20 maxamillion: that's much appreciated, I'll let you know 16:53:34 jlaska: no problem, sounds good 16:53:56 ok, seems like we go to open discussion 16:54:00 #topic Open discussion 16:54:16 anyone have a topic he/she wants to discuss? 16:56:16 ok, thank you everyone, see you on next meeting 16:56:22 it seems I didn't even made such a mess as a chairman as I supposed :) 16:56:26 #endmeeting