QA/Meetings/20100712

From FedoraProject

Jump to: navigation, search

Contents

Attendees

People present (lines said):

  • jlaska (133)
  • wwoods (121)
  • adamw (67)
  • j_dulaney (66)
  • kparal (29)
  • Viking-Ice (25)
  • jskladan (11)
  • tertl3 (11)
  • dafrito (6)
  • vaschenb (6)
  • fedbot (4)
  • zodbot (3)
  • Southern_Gentlem (1)

Regrets:

Agenda

Previous meeting follow-up

  1. jlaska to cleanup (or remove) the Critical Path Packages#Background section so that it provides _some_ value
  2. j_dulaney to draft a combined proventesters / joinproventesters wiki page for list review
  3. wwoods to evaluate nss-softokn dependency problem for proper 'depcheck' coverage

Pre-Alpha Rawhide Acceptance Test Plan #1

Owner - User:jlaska
Summary
Per task#4 on the fedora-qa schedule, a rawhide acceptance test run was scheduled for 2010-07-09.
See rel-eng ticket#3842 , images built on Jul 12, 2010.
Next steps ...
Testing underway, results posted to Test_Results:Fedora_14_Pre-Alpha_Rawhide_Acceptance_Test_1
Test summary will be sent to test@lists.fedoraproject.org

Fedora 14 Install Matrix Review

Owner - User:rhe
Summary
Updated Fedora 14 install test plan template and test matrix available
The proposed pages will be used to test Fedora 14
Next steps ...
Rhe has requested review and make suggestions/corrections as needed. Feedback encouraged!

Proventester Update

Owner - User:Maxamillion, User:Adamwill
Summary
Adam Miller announced that the proventesters wiki page (QA/JoinProvenTesters) is no longer a draft.
There are several TRAC requests to join the proventesters group.
User:Adamwill drafted proventester instructions and sent to the list for review.
Several draft documents were out for review: User:Adamwill/Draft_proventesters, User:Dafrito/Proven_tester, User:Jdulaney/Proven_Tester
Next steps
Choose a draft wiki page
Discuss generating recurring critpath+proventester metrics

AutoQA PUATP

Owner - User:wwoods
Summary
In order to accomplish the QA:Package_Update_Acceptance_Test_Plan, wwoods proposed prioritizing the following autoqa milestones
depcheck
multi-hook test support
Finalize test base class patchset
Finalize label patchset
Next steps...
  1. Start picking up tickets in the above milestones

Open discussion - <Your topic here>

Reporting bugs upstream vs fedora bugzilla

Owner - User:Johannbg
Summary
Asked for QA to make a decision on a debate from devel@ regarding whether bugs should be submitted against Fedora, or against the upstream project.
Expressed concerns for time wasted by bug reports and maintainers if some bugs are not tracked in Fedora, but upstream
Next steps...
Not enough time for discussion, and no consensus reached.
Continue discussion on list

Upcoming QA events

  • 2010-07-08 - Pre-Alpha Rawhide Acceptance Test Plan #1
  • 2010-07-15 - Pre-Alpha Rawhide Acceptance Test Plan #2
  • 2010-07-22 - Pre-Alpha Rawhide Acceptance Test Plan #3
  • 2010-07-23 - Alpha Blocker Meeting (f14alpha) #1
  • 2010-07-30 - Alpha Blocker Meeting (f14alpha) #2

Action items

  1. adamw to talk with lmacken about generating recurring email reports on proventester

IRC Transcript

jlaska #startmeeting Fedora QA Meeting 15:00
zodbot Meeting started Mon Jul 12 15:00:19 2010 UTC. The chair is jlaska. Information about MeetBot at http://wiki.debian.org/MeetBot. 15:00
zodbot Useful Commands: #action #agreed #halp #info #idea #link #topic. 15:00
jlaska #meetingname fedora-qa 15:00
zodbot The meeting name has been set to 'fedora-qa' 15:00
jlaska #topic Gathering ... 15:00
* jlaska waits a few minutes for folks to arrive 15:00
adamw mor-diddly-orning 15:00
* jskladan tips the hat 15:01
* kparal joins up 15:01
* dafrito waves 15:01
* tertl3 arrives 15:01
* j_dulaney killes a Kligon 15:01
j_dulaney kills 15:01
* j_dulaney decides to sleep as soon as this is done 15:02
kparal what an attendence :) 15:02
kparal nice 15:02
j_dulaney better than last week 15:02
* wwoods here 15:02
jlaska howdy all 15:02
* vaschenb newbie is nervous :-) 15:02
* Southern_Gentlem 15:02
j_dulaney vaschenb: why? 15:03
kparal he don't know what tortures to expect :) 15:03
vaschenb j_dulaney: it's my first time :-D 15:03
j_dulaney Read and learn 15:03
jskladan he's still waiting for the hazing to start :) 15:03
j_dulaney Ah 15:03
adamw alright, time to start thinning the herd! 15:03
wwoods oh boy a new guy! do we have the dump truck full of manure ready or what 15:03
* adamw revs up the chainsaw 15:03
jlaska wwoods: haha 15:03
wwoods I mean er uh. WELCOME 15:03
adamw oh, that's right, wwoods - manure before chainsaw. where ARE my manners 15:04
jskladan  :-D 15:04
kparal lol 15:04
j_dulaney tertl3 is also new 15:04
* jskladan lolz 15:04
tertl3 yes 15:04
tertl3 I am a noob 15:04
jlaska alrighty, I think we have critical mass ... let's get started 15:04
vaschenb tertl3: welcome into virgin group :-) 15:04
jlaska #topic Previous meeting follow-up 15:05
jlaska #info jlaska to cleanup (or remove) the Critical Path Packages#Background section so that it provides _some_ value 15:05
jlaska nothing sexy, but I added some wording to http://fedoraproject.org/wiki/Critical_Path_Packages#Background 15:05
adamw that looks a lot better. 15:06
jlaska Unless there is rioting in the streets, let's get this off my list :) 15:06
dafrito no riots here ;) 15:06
jlaska dafrito and adamw are the wiki-experts ... suggestions always encouraged 15:06
jlaska alrighty ... thanks 15:06
jlaska this next one really has several threads tracking it ... 15:07
jlaska #info j_dulaney to draft a combined proventesters / joinproventesters wiki page for list review 15:07
jlaska we have a few drafts ... shall we save the updates on this for the proventester update 15:07
adamw sure, that'd make sense to me 15:07
j_dulaney Indee 15:07
jlaska alrighty ... 15:07
j_dulaney indeed 15:07
jlaska last one ... 15:08
jlaska #info wwoods to evaluate nss-softokn dependency problem for proper 'depcheck' coverage 15:08
jlaska and from what I can tell, this was done already 15:08
jlaska #link http://lists.fedoraproject.org/pipermail/devel/2010-July/138478.html 15:08
wwoods right - we still want to write up a test case for this, when we start writing depcheck test cases 15:08
jlaska roger, I think you've got that captured already in trac too 15:09
wwoods but I think that description of the problem will be sufficient to allow that, and theoreticallly depcheck should catch that, if it runs mash 15:09
wwoods (which it will) 15:09
adamw cool 15:09
jlaska alrighty ... that's all I had on my list from last week 15:09
jlaska anything I missed? 15:09
jlaska okay ... diving into the agenda 15:10
jlaska 2 brief updates ... 15:10
jlaska #topic Pre-Alpha Rawhide Acceptance Test Plan #1 15:10
jlaska < insert thunder and lightning > 15:10
* j_dulaney ducks and covers 15:11
jlaska it has started ... we hit our first of 3 Fedora 14 pre-alpha acceptance test milestones last Thursday 15:11
jlaska these are still very much experimental test runs ... intended primarily to run through the rawhide acceptance test plan 15:11
jlaska updated anaconda+pykickstart were built last week, and install images were built by rel-eng over the weekend 15:12
jlaska Testing is underway ... and I'll send a test summary to the list in the next day or so 15:12
j_dulaney available in the nightly builds? 15:12
jlaska no ... nightly installable rawhide images are not provided 15:13
jlaska this milestone is intended to help scrub custom-built rawhide install images before F-14 Alpha ... to help shake out early bugs 15:13
jlaska #info jlaska to send test summary to test@l.fp.org once complete 15:14
jlaska #link https://fedoraproject.org/wiki/Test_Results:Fedora_14_Pre-Alpha_Rawhide_Acceptance_Test_1 15:14
adamw the nightly composes are running, though. and i think they're rawhide. 15:14
jlaska yes indeed, those are just package repos. So as always, you can update to rawhide using the documented procedures 15:14
jlaska https://fedoraproject.org/wiki/Releases/Rawhide#Installing_Rawhide 15:14
jlaska Okay, next up ... 15:15
jlaska #topic Fedora 14 Install Matrix Review 15:15
adamw er, what? i mean the nightly live images. 15:15
jlaska Hurry (rhe) has been putting the finishing touches on the much improved install test matrix 15:15
adamw oh well. 15:15
jlaska adamw: ah I see, yes you are right :) 15:15
jlaska adamw: too much nightly content to keep track of 15:15
adamw fwiw, i don't particular recommend rawhide right now, for GNOME anyway. it's so broken i'm in XFCE. 15:16
adamw just as a note =) 15:16
jlaska #info Hurry has an updated F14 install test plan out for review at - https://fedoraproject.org/wiki/QA:Fedora_14_Install_Test_Plan 15:16
jlaska adamw: good to note, thanks 15:16
* j_dulaney couldn't get it to boot last time he tested it 15:16
jlaska #info Hurry has an updated F14 install test matrix out for review as well - https://fedoraproject.org/wiki/QA:Fedora_14_Install_Results_Template 15:16
adamw i've been following the trac tickets, looks like nice stuff 15:16
jlaska yeah, should be some nice improvements 15:17
jlaska easier to see which tests/results impact the different use cases (cd, dvd, boot.iso etc...) 15:17
jlaska Hurry welcomes feedback on those 2 wiki documents. Unless anything drastic, we'll be using those to track install testing for F-14 15:18
jlaska okay ... next topic ... 15:18
jlaska #topic Proventester Update 15:18
jlaska quite a bit of wiki love over the last week 15:19
jlaska who wants to take this one? 15:19
dafrito there ended up being three drafts, and I think we settled on adamw's 15:20
adamw drafts: 15:20
adamw https://fedoraproject.org/wiki/User:Adamwill/Draft_proventesters <--- adamw 15:20
adamw https://fedoraproject.org/wiki/User:Dafrito/Proven_tester <--- dafrito 15:20
adamw https://fedoraproject.org/wiki/User:Jdulaney/Proven_Tester <--- j_dulaney 15:21
j_dulaney I believe adamw's to be the best 15:21
adamw dulaney's came first, then dafrito's, then mine. so far, dafrito and I say we prefer mine, mike c says he likes it too, no-one else has said much. =) 15:22
adamw I also like how none of us can agree on the page name =) 15:22
j_dulaney LOL 15:22
jlaska does this replace QA/JoinProvenTesters and Proven_tester? 15:22
dafrito hehe 15:22
j_dulaney jlaska: indeed 15:22
j_dulaney adamw's seems to be the clearest of the lot to me. 15:23
j_dulaney My version is a kluge combining what was really two drafts. 15:23
jlaska yeah, I like adding back the sections for specific types of feedback ... but like I said in email, that might be specific to the odd way my brain works :) 15:23
jlaska so, should we consider this topic _voted_ on then? 15:24
j_dulaney jlaska: I concur 15:24
j_dulaney however you spell it 15:24
adamw if so i'll go ahead and move mine into the live namespace, and edit one of the pages to be a redirect to it 15:24
adamw j_dulaney: got it in one 15:24
dafrito yeah, my version was off-topic, I'd go with adam's 15:25
jlaska +1 15:25
j_dulaney I'll go ahead and axe mine 15:25
jlaska btw ... good work all around folks, nice to see this process flow with proposed drafts 15:25
dafrito Did we want to rename the page to proventesters as well? 15:26
adamw nah, i'll keep the Proven_tester name 15:26
jlaska let's see how many different page names we can come up with :) 15:26
adamw it's best to avoid page renames whenever possible 15:26
adamw mediawiki gets lost after one level of redirects 15:26
kparal really? doh 15:27
adamw i'll put this draft into the Proven_tester name and edit JoinProvenTesters to be a redirect to it 15:27
adamw kparal: yeah. try creating a page, then rename it twice, then go to the URL of the first name 15:27
j_dulaney adamw: I'd say go ahead 15:27
jlaska the websites team has a script they run periodically to remove all those double/triple redirects 15:27
jlaska but still, it can get confusing 15:28
jlaska #info team agreed to make User:Adamwill/Draft_proventesters the official Proven_tester page 15:29
jlaska Alright, I'm not aware of anything else we need to be tracking for Proventesters 15:29
jlaska perhaps the proventester mentor requests? 15:29
j_dulaney I was going to ask, how do y'all thik it's going since activation? 15:30
adamw well, it's hard to tell 15:30
adamw that's one thing i wanted to mention 15:30
jlaska that brings up a good follow-up topic 15:30
adamw it'd be nice to be able to track proven tester feedback 15:30
adamw after all, we all know how easy metrics are! (sorry, injoke for the bugzappers crowd) 15:30
jlaska lmacken: has some scripts he used to generate metrics ... we could investigate using those scripts for a weekly test summary? 15:31
jlaska http://lists.fedoraproject.org/pipermail/devel/2010-June/137413.html 15:31
adamw i think there's already a report on bodhi feedback sent to -devel periodically; i've been meaning to email the author to ask if it can be adjusted to report only on feedback from proventesters, and only for critpath packages 15:31
j_dulaney There seems to be a lot of 0 karma 15:32
jlaska I'm seeing 11 untested F13 critpath updates - https://admin.fedoraproject.org/updates/critpath?release=F13&untested=True 15:33
j_dulaney In several cases, it seems that the 0 karma was given for packages that are core to Fedora, but people are unfamiliar with them 15:33
jlaska Shall we keep this on the list for a future meeting? 15:34
j_dulaney It seems that folks aren't looking at what a package does 15:34
* j_dulaney is guilty of that 15:34
jlaska adamw: do you have a link to one of those sample reports? Perhaps someone is interested in following-up on that topic? 15:34
adamw i can see a couple of packages in that report which are somewhat trickyt 15:35
wwoods might be good to ask the package maintainers for some info about how to do basic package verification/acceptance 15:35
wwoods *somebody*'s gotta have a clue 15:35
adamw for koji and mash, clearly we need someone in releng to be active as a proventester 15:35
jlaska e.g. mash, koji 15:35
adamw iscsi-utils is very hardware-specific 15:35
j_dulaney Indeed 15:36
adamw as is wacom 15:36
wwoods adamw: not really - you can have both iscsi endpoints be all-software, as I understand it 15:36
wwoods wacom, yes 15:36
adamw wwoods: okay, knowing-what-the-hell-you're-doing specific =) 15:36
j_dulaney what is wacom? 15:36
wwoods one could argue that all package testing has that specific requirement. 15:36
adamw j_dulaney: tablets 15:36
j_dulaney Ah 15:36
adamw j_dulaney: (the ones artists use) 15:36
jlaska so do we have 2 topics here ... 15:36
wwoods wacom tablets are far less rare 15:36
j_dulaney I'm a moron then 15:36
j_dulaney I have one 15:36
wwoods you can get one for like $40 15:36
jlaska 1) generating proventester metrics 15:36
adamw wwoods: sure, you know what I mean - knowing-what-the-hell-you're-doing-with-something-somewhat-obscure 15:37
j_dulaney I didn't think about using it for testing 15:37
jlaska 2) providing some package-specific wiki test guidelines 15:37
jlaska  ? 15:37
adamw j_dulaney: okay, then problem solved - plug it in, check you can scribble on it at every boot, and +1 the wacom updates 15:37
j_dulaney adamw: roger 15:37
adamw jlaska: i can't find the karma reports i'm thinking of, any more :/ 15:38
adamw jlaska: i'm fairly sure it wasn't just an opium dream though... 15:38
jlaska adamw: did you see the lmackenreport I linked to above? 15:38
adamw yeah, that's not what i was thinking of, though close 15:38
jlaska okay 15:38
adamw i guess it's not too important, obviously luke's the guy to ask 15:38
* adamw strictly follows his rule to never touch a drop of opium till after 2pm 15:39
jlaska okay ... anyone want to take this up with Luke, see how perhaps we can generate proventester reports on a recurring basis? 15:39
adamw i will 15:39
jlaska everyone take 1 step back ... except adamw! 15:39
jlaska hehe 15:40
jlaska thanks adamw 15:40
adamw one other from the list; openldap is there because it can be involved in login, i think? 15:40
jlaska #action adamw to talk with lmacken about generating recurring email reports on proventester 15:40
adamw so we kinda need someone with an LDAP auth system to +1 those updates 15:40
j_dulaney I wish there was a list somewhere of what the critpath updats do 15:41
jlaska I think sudo requires it 15:41
adamw jlaska: oh, hmm...didn't know that 15:41
adamw j_dulaney: how do you mean, exactly? what the update changed? 15:41
jlaska okay, so anything else we want to track on this topic for next week? 15:41
adamw hum 15:42
adamw i think the mentoring process is going fine so far 15:42
j_dulaney adamw: What the library does, so we know if we can test it or not 15:42
adamw j_dulaney: ah, so just what jlaska mentioned above 15:42
adamw j_dulaney: a couple of things you can do - look at the RPM description (rpm -q packagename) and also what requires the package (use repoquery for that) 15:42
adamw that often gives you a good idea of what it's for 15:43
jlaska # repoquery -q --whatrequires $foo 15:43
wwoods ITYM rpm -qi packagename 15:43
j_dulaney @ping 15:44
j_dulaney @ping 15:44
fedbot pong 15:44
j_dulaney @ping 15:44
fedbot pong 15:44
fedbot pong 15:44
jlaska that was odd 15:44
j_dulaney Ugh 15:45
j_dulaney @ping 15:45
fedbot j_dulaney: You've given me 3 commands within the last minute; I'm now ignoring you for 5 minutes. 15:45
jlaska alright, I'd like to leave at least 15 minutes for the last topic 15:45
wwoods j_dulaney: please don't do that in the channel, esp. in the middle of a meeting 15:45
wwoods j_dulaney: consider privmsgs for talking to the bot 15:45
jlaska anything else to track for next week? 15:45
j_dulaney I apoligize, my connection went wacky 15:45
jlaska alrighty ... we'll follow-up on the metrics topic next week then 15:46
jlaska thanks folks 15:46
jlaska next up ... 15:46
adamw wwoods: rpm -qi gives you the package description. repoquery --whatrequires tells you what requires it (well, you also have to do it for everything it provides). 15:46
wwoods (adamw: right you said "look at the RPM description (rpm -q packagename)" - rpm -q isn't gonna give you the description.) 15:47
jlaska #topic AutoQA Package Update Acceptance Test Plan 15:47
jlaska wwoods: apologies, I was hoping to leave you more time to discuss today 15:47
adamw wwoods: oh yes, thanks :) 15:48
wwoods jlaska: s'ok 15:48
wwoods so right - we've got this excellent acceptance plan for package updates 15:48
wwoods https://fedoraproject.org/wiki/QA:Package_Update_Acceptance_Test_Plan 15:48
wwoods and we've been working on automating as much of it as possible 15:49
kparal (preferably everything) 15:49
wwoods kparal: indeed! 15:49
wwoods one of the key things adhering to this plan *should* accomplish 15:49
wwoods is preventing us ever having broken deps in the repos (yay) 15:49
kparal big yay! 15:50
wwoods among other various important things that will be helpful for packager/tester sanity. 15:50
wwoods so. we've had this plan kicking around a long time, I've been messing with the depcheck test for months, etc. 15:50
wwoods and now we've finally got autotest/autoqa running in the Fedora infrstructure 15:51
wwoods AND the Fedora 14 release cycle is starting up soon 15:51
wwoods so I feel like now's a good time to try to get as much done on the PUATP as possible. 15:51
wwoods I broke things up into a few milestones: 15:51
wwoods https://fedorahosted.org/autoqa/milestone/Package%20Update%20Acceptance%20Test%20Plan 15:51
wwoods https://fedorahosted.org/autoqa/milestone/Package%20Update%20Acceptance%20Test%20Plan%20-%20package%20sanity%20tests 15:52
wwoods https://fedorahosted.org/autoqa/milestone/Package%20Update%20Acceptance%20Test%20Plan%20-%20depcheck 15:52
wwoods blerg, long URLS 15:52
wwoods but you can just check out the roadmap - https://fedorahosted.org/autoqa/roadmap 15:52
* kparal has installed %20->space converter into his eyes 15:52
* adamw needs some o' those 15:53
* j_dulaney notes he could get involved here. 15:53
wwoods kparal: you pointed out that we had a goal-setting meeting a while back and decided that resultdb was basically the Top Of The List for stuff we wanted to get done 15:53
wwoods which is still true for the larger roadmap, I think 15:54
kparal yea, I remember it somewhat like that :) 15:54
wwoods right - that's definitely an incredibly important part of the AutoQA roadmap 15:54
wwoods but I think that can go on in parallel with automating PUATP 15:55
wwoods and there are a lot of people interested in seeing PUATP automated as soon as possible 15:55
kparal I think the important part is not having everyone work on 5 different tasks. but we can shift priorities if needed, that's no problem 15:55
wwoods kparal: agreed 15:55
jlaska +1 15:55
j_dulaney indeed 15:55
wwoods so I'm suggesting that - as much as is possible - we should shift work to PUATP-related tasks for a while 15:55
kparal alright, so depcheck testing is the #1 now? 15:56
adamw j_dulaney: for a start, subscribe to autoqa-devel list 15:56
adamw j_dulaney: https://fedorahosted.org/mailman/listinfo/autoqa-devel 15:56
j_dulaney adamw: roger 15:56
wwoods I keep getting approached by board members / desktop team members / etc asking about depcheck and friends 15:56
wwoods I know FESCo is interested 15:56
wwoods and we've put a good deal of work into PUATP already 15:56
wwoods and I'd love to have it functional - it'll make us look real good if it works the way we want it to 15:57
kparal wwoods: ok, so I believe there's enough work even for me, right? :) 15:57
wwoods heh 15:57
wwoods kparal: yeah I think there's plenty to go around! 15:57
wwoods heh 15:57
kparal should we all work on it? 15:57
wwoods In as far as you can find tasks/tickets to take, definitely 15:58
kparal alright then 15:58
wwoods I'd like to get it moved far enough forward that we have at least a couple of the mandatory PUATP tests running for new updates 15:58
wwoods I'm not suggesting NO WORKING ON ANYTHING UNTIL THIS IS COMPLETE 15:59
wwoods that's obviously sill 15:59
wwoods err, silly 15:59
* kparal is fine with taking it as the main point of focus 15:59
* j_dulaney will slave to get it done 15:59
wwoods but let's seriously all try to take bits and pieces and get this thinger done 15:59
jlaska wwoods: you mentioned a sprint idea in your autoqa-devel mail, are the type of remaining tasks things that might lend well to a day of hacking on this stuff? 16:00
j_dulaney that would be cool 16:00
jskladan ok, from my point of view - i'd like to see the "common base class for tests" in master (at least smth like that) at least before we start with the "multiple hook tests" 16:00
* kparal will need some help with getting started on depcheck, maybe sprint could help 16:00
wwoods jlaska: yeah I think so! the package sanity tests, for instance - those tickets could use a day's worth of work 16:00
wwoods jskladan: absolutely agreed - I'm going to review that patch right after the meeting 16:01
jskladan other than that, i don't see any obstacle on the road (yet :) ) 16:01
wwoods and the same for kparal's label patch 16:01
wwoods because we need that for ticket #156 in the package sanity tests 16:01
wwoods and for proper virt stuff 16:01
kparal I think its agreed, we will start working on depcheck and its requirements 16:02
wwoods now we're kind of over time for the meeting, but we can divide up a few obvious tickets right now 16:02
jskladan well, sadly i need to go home after the meeting, but feel free to potentialy bug me via email and i'll respond during my night time, or we can check the code via telephone tomorrow or smth like that, if you should find it too confusing/not ok/whatever 16:02
wwoods or - yeah, I don't want to keep the Brno guys (it's beer-o-clock there!) 16:03
* j_dulaney wants something to start on 16:03
wwoods so if we're agreed that depcheck / PUATP is the thing to do 16:03
* jskladan wwoods reads minds! :) 16:03
wwoods either feel free to take tickets in trac 16:03
wwoods or let's discuss who should/can take tickets on autoqa-devel 16:04
wwoods j_dulaney: definitely recommend subscribing to that list 16:04
j_dulaney wwoods: already have 16:04
wwoods https://fedorahosted.org/mailman/listinfo/autoqa-devel 16:04
kparal wwoods: I'll try to explore it a little (probably ask you some stuff) and then start working on some of its tickets 16:04
wwoods excellent 16:04
wwoods yeah I'm happy to answer questions or provide assistance on any of it 16:04
wwoods if anyone wants to take the HelloWorld test, that's probably a pretty easy ticket for anyone with a bit of python skill 16:05
kparal maybe a good start for vaschenb ;) 16:05
wwoods kparal: I was thinking that too, actually 16:05
wwoods that's: https://fedorahosted.org/autoqa/ticket/195 16:05
vaschenb huh :-) 16:06
wwoods also: if there's tickets that seem invalid / unneeded for finishing PUATP, either move them around or bring it up on the list or in #fedora-qa 16:06
jlaska wwoods: what's the finish line for us ... closing out all three of the milestones you linked above? 16:06
wwoods yeah - I'd like to see all three of those milestones complete before, let's say, F14 Beta 16:07
wwoods That's October 15, according to http://poelstra.fedorapeople.org/schedules/f-14/f-14-quality-tasks.html 16:08
wwoods Actually, I think we could probably shoot for F14 Alpha 16:08
wwoods (September 7) 16:08
vaschenb kparal: ok, tomorrow I'll do what you want, now I've to go destroy my body... 16:08
jlaska perhaps even just the depcheck milestone by F14-Alpha 16:08
wwoods once we hit Alpha we're going to want to put more work into release testing / installer automation 16:09
jlaska and the remainder by F14-Beta (bonus points of course for wrapping this up early too) 16:09
wwoods but ideally this should all be up and running when F-14 is released 16:09
wwoods so F14 can be one of the smoothest releases we've ever had 16:09
wwoods and then we all get well-earned QA Victory Beers 16:09
jlaska heh, I like those 16:09
wwoods jlaska: that sounds like a reasonable schedule 16:10
wwoods anyone object / further discussion on timelines? 16:10
* vaschenb is out, idling here for complete log of meeting... 16:10
jlaska vaschenb: cya 16:10
kparal vaschenb: log will be available on wiki pages and in ML 16:11
* wwoods takes that as "no objections", updates milestone due dates 16:11
jlaska wwoods: thanks! 16:11
jlaska I better go steal all the easy tickets first! (just kidding) 16:11
* j_dulaney wants the easy stuff 16:12
j_dulaney The idea being I start off easy 16:12
wwoods kparal: I might bug you for rpmfluff info and/or help writing depcheck test cases (ticket #202) 16:12
kparal wwoods: yes, sure 16:12
wwoods the pst tickets need review - some of them are finished/obsolete, like #139: "pst: Find a way how to test packages coming to updates-testing" 16:12
wwoods that's the post-bodhi-update hook, yay 16:12
kparal wwoods: we were actually thinking to ask vaschenb to use rpmfluff to create a few broken packages for us. but we will see 16:13
wwoods can we schedule a pst ticket sprint day sometime this week? wednesday maybe? 16:13
wwoods kparal: ooh, also a good idea 16:13
wwoods or maybe j_dulaney? rpmfluff is a fun little tool 16:13
wwoods is rpmfluff C or python? I forget 16:13
kparal wwoods: I would wait with the pst stuff 16:13
kparal wwoods: rpmfluff is python 16:13
j_dulaney I'd like to look into it 16:13
wwoods kparal: fair 'nuff 16:14
kparal wwoods: pst stuff can wait until we have depcheck working. pst stuff wasn't really worked on recently, because psplicha had some other work. but we will surely come back to it 16:14
wwoods okay, let's revisit pst status in next week's meeting 16:15
wwoods first priorities are: multi-hook testing, label / base test class patches 16:15
wwoods and the depcheck test 16:15
* Viking-Ice joins in Better late then never ;) 16:16
wwoods someone should review output of rpmlint / rpmguard on the -results list 16:16
wwoods err. I mean. do we have a policy about which rpmlint/rpmguard test results we actually want to consider errors/failures 16:16
wwoods or should someone review the current output we've been getting and write such a policy? 16:17
adamw writing one would be good 16:17
wwoods we probably need to run that past the packaging folks 16:17
adamw i don't believe there is such a policy 16:17
kparal I would do one thing after another :) 16:17
wwoods adamw: ah - well the PUATP will require one 16:17
jskladan will, i do not want to be mean or something, but looks like we're heading back to the "all at once" again :)) 16:17
wwoods jskladan: heh, maybe you're right 16:17
jlaska if the output from those tests is more than just informational, yeah, we'd need to work that path. But perhaps that's a future task? 16:18
adamw wwoods: there's https://fedoraproject.org/wiki/Packaging/Guidelines#rpmlint 16:18
adamw wwoods: that's about the most 'official' thing that currently exists, afaik 16:18
wwoods so yeah, depcheck only this week, we can do a status checkin on pst and rpm{lint,guard} policy next week sometime 16:18
jlaska wwoods: the priority you outlined earlier seems pretty focused to me: multi-hook testing, label/base test class patches and depcheck 16:18
jskladan jlaska: +1 16:19
wwoods jlaska: right, just want to be sure we're accounting for all the PUATP pieces in the medium-term 16:19
jlaska that's 2 milestones ... and then working those 2 patches into master, right? 16:19
wwoods yup! 16:19
jlaska wwoods: ah, true true 16:19
jlaska okay, I'll include a link to each of those milestones in the minutes 16:20
wwoods and the other PUATP pieces.. we just reviewed them and kparal and jskladan have gently reminded me that they can wait 'til later 16:20
wwoods heh 16:20
* j_dulaney apologizes for bad connection 16:20
jlaska anything others can do to help with those 2 patchsets out for review? 16:20
j_dulaney wwoods: I don't know if you got my request to email me info on rpmfluff? 16:21
wwoods jlaska: not sure. documenting changes / new capabilities perhaps? 16:21
wwoods we'll probably need to update the wiki pages on writing new tests, for instance 16:21
j_dulaney wwoods: and that I'd like in on that? 16:21
jlaska wwoods: okay ... I'll jump on kparal's label thread tomorrow 16:21
jlaska j_dulaney: google is your friend ... https://fedorahosted.org/rpmfluff/ 16:22
wwoods j_dulaney: didn't see that, but I'll definitely try to keep you in the loop 16:22
j_dulaney thanks jlaska, wwoods 16:23
jlaska any other thoughts/concerns before open floor? 16:23
kparal not from kparal 16:23
j_dulaney Chan 'eil idir as j_dulaney. 16:24
jlaska okay, thanks gang. wwoods I'll try to capture this in the minutes, but kick me if I miss anything 16:25
wwoods jlaska: with gusto 16:25
jlaska not too much gusto :) 16:25
* wwoods brought the extra-heavy boots 16:25
jlaska #topic Open discussion - <your topic here> 16:25
jlaska wwoods: hah ... not the steal tipped kodiac work boot 16:25
* j_dulaney has nothing 16:26
jlaska okay gang, we've run over today ... that's my fault for not leaving enough time 16:26
Viking-Ice I asked for one thing to add to the list 16:26
Viking-Ice upstream bugzillas vs our bugzilla should reporters be directed to upstream bugzilla tracks system and what not 16:27
Viking-Ice Where does QA stand on this issue 16:27
adamw i don't think we have time to discuss that properly now. 16:27
tertl3 hi 16:27
tertl3 this is a long meeting 16:27
Viking-Ice adamw: why not 16:27
wwoods tertl3: they're not usually this long, heh 16:27
jlaska tertl3: not usually this long, thanks for sticking around :) 16:27
tertl3 not problem, I was playing TF2 16:28
Viking-Ice we need to settle this once and for all 16:28
tertl3 i play it in a windpw though 16:28
adamw Viking-Ice: we're already 30 minutes over time =) 16:28
* j_dulaney is about to pass out 16:28
tertl3 j_dulaney, dont drink the Nyquil! 16:28
wwoods Viking-Ice: I'd say that's more a bugzappers question but my gut feeling is that plain ol' bug-reporting users should be reporting stuff to RHBZ 16:28
j_dulaney LOL 16:28
tertl3 j_dulaney, i'm just kidding 16:28
Viking-Ice wwoods: this affects all QA mostly reporters thou and perhaps triagers 16:29
tertl3 i like the tylenol cold and cough 16:29
* jskladan needs to go, bb gang! 16:29
wwoods but that our talented and well-trained bugzapper ninja squadrons should feel free to copy/move bugs upstream as they see fit 16:29
jlaska jskladan: take care 16:29
j_dulaney peace 16:29
adamw wwoods: practically speaking that ain't going to solve the problem, as bugzappers coverage hovers solidly around the 2% of components mark... 16:29
Viking-Ice wwoods: is it not the maintainers responsibility do play that role not the triagers one 16:29
tertl3 I havent booted into fedora in a few days 16:30
jlaska Viking-Ice: I'm not sure I understand what the "issue" is 16:30
jlaska there are clear instructions on when to file upstream, vs when to file in Fedora? 16:30
Viking-Ice jlaska;: reporters being directed to upstream bugzillas 16:30
j_dulaney jlaska: I've not seen such 16:30
jlaska they are or aren't being reported to upstream? 16:30
kparal I personally report to upstream whenever possible, whenever I'm sure it's not Fedora specific 16:31
Viking-Ice maintainers close bug file this stuff upstreasm 16:31
Viking-Ice upstream 16:31
jlaska is that good, bad? 16:31
jlaska what's your stance/recommendation? 16:31
Viking-Ice when in fact they should act as the bridge between components the maintain and upstreasm 16:31
Viking-Ice upstreasm 16:31
jlaska ah 16:31
Viking-Ice frack 16:31
Viking-Ice upstream! 16:32
Viking-Ice my stance is simple all bugs should be reported to our bugzilla 16:32
wwoods wait is this about that thread on the devel list? are maintainers closing bugs in rhbz and saying "file this upstream instead" or something? 16:32
Viking-Ice and maintainers should act as the bridge between upstream bugzilla and our bugzilla for their component 16:32
wwoods Viking-Ice: yeah historically that's the policy, and anything that needs upstream attention either a) the upstream is alert and pays attention to our bugzilla, or b) the maintainer or some diligent tester pushes things upstream when needed 16:33
Viking-Ice wwoods: reoccurring thread happens every release sometimes often in the release cycle 16:33
wwoods general policy is still "fedora reports go to rhbz" 16:33
wwoods don't think we've changed anything about that 16:33
j_dulaney Ugh, can't type fast enough in current state. 16:34
Viking-Ice wwood: Well some maintainers ignore or simple close bugs wontfix file upstream 16:34
j_dulaney File problem with maintainers? 16:35
Viking-Ice waste of everybody's time reporters triagers and maintainers mean if this is going to be the procedure then simply remove the components from our bugzilla 16:35
wwoods so this is about that mail thread. 16:35
wwoods that thread gets a big tl;dr frmo me 16:35
jlaska Viking-Ice: I think that's a bit extremist 16:35
Viking-Ice wwoods: yup it's "why the WONTFIX?" this time 16:35
Viking-Ice jlaska: not really 16:36
jlaska alright, so we're not going to reach any conclusions on this topic in the meeting 16:36
j_dulaney indeed 16:36
wwoods worth thinking about, maybe revisit in other meetings 16:36
jlaska unless others want to weigh in, let's take this up at a future meeting, or on list 16:36
j_dulaney listsounds good, too drawn out for meeting 16:37
jlaska Viking-Ice: alright, if we have something that needs review of finalizing for next week ... I'll add it to the list 16:38
jlaska otherwise ... we can continue the debate/discussion on the list 16:38
Viking-Ice we need to settle this or atleast provide QA stance on the topic 16:38
Viking-Ice put it on next meetings agenda 16:38
Viking-Ice gather feedback from the list this week 16:38
jlaska I'm missing the exigency, but am open to learning 16:39
jlaska okay folks ... thanks for your time today 16:39
jlaska as always, I'll send minutes to the list/wiki 16:39
jlaska #endmeeting 16:39

Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!