From Fedora Project Wiki

< QA‎ | Meetings

Revision as of 21:21, 20 December 2010 by Jlaska (talk | contribs) (Updated minutes)
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)


People present (lines said)

  1. jlaska (108)
  2. adamw (33)
  3. wwoods (15)
  4. mkrizek (8)
  5. rbergeron (8)
  6. fenrus02 (3)
  7. Alam (2)

Unable to attend:

  1. Rhe (hopefully sleeping)


Previous meeting follow-up

  1. jlaska to ping jens+rhe and Alam for thoughts on a F15 l10n/i18n test day
  2. jlaska to merge User:Bruno#Mockup_for_QA_-_Tracking_bug_queries into QA:SOP_Blocker_Bug_Meeting
  3. Bodhi feedback patch from fcami (see infrastructure ticket#701 awaiting review
    • Still waiting on feedback from someone on the bodhi-devel side (lmacken likely)
  4. Check-in with Bruno on upstream investigation into bugzilla nested query search

Call for Test Days

Call for ideas/owners for Fedora 15 QA/Test_Days
Monitor list of approved features for ideas
Several events scheduled so far (GNOME 3, Xfce 4.8 and X Test Week)
Hurry and Alam scheduled another l10n/i18n test day (see
Adamw waiting on feedback for scheduling a systemd event
Next steps ...
  1. Contribute ideas to Talk:QA/Fedora_15_test_days
  2. Identify owners to organize and host test days

Requirements review for Fedora test case management

See fedora-qa ticket#152
Requirements review for Fedora test case management
Hurry drafting requirements at Rhe/tcms_requirements_proposal
Next steps ...
Hurry will continue refining requirements
All - please review and add any comments/concerns in Talk page

Critical Path test case development

See fedora-qa ticket#154
Still experimenting with some mock-ups/examples
Next steps ...
Adamw drafing wiki pages intended to document the process, expects updates shortly

AutoQA Update

clumens' anaconda_storage test has been reviewed and ready to land in git master branch
The patch representing second attempt for solving ticket #205 has landed into autoqa-devel. That means we will be able to send our test results as comments into Bodhi.
lmacken enabled us to use staging Bodhi instance for AutoQA development purposes (mainly he turned off email notifications). Kudos to him for that
mkrizek and I worked on ticket #241 [6] ("Add support for a staging server"). The patch is not posted yet (waiting for previous patch to land in master), but the code is available at mkrizek-staging branch
jskladan works on a different kind of a koji watcher. His work is available in the 'new_koji_watcher' branch. His work should enable us to hook up depcheck properly into our framework. Some discussion followed on how best to group -pending updates, the group agreed to discuss offline
wwoods still working on a blog post and code to handle ticket #248 (simultaneous depcheck test runs)
Team decided to serialize depcheck tests to avoid complexity relating to handling multiple depcheck tests running at the same time
Depcheck data - during busiest-ever month, we would need depcheck (on average) once every 33 minutes. Depcheck currently takes <= 1m to complete
mkrizek finished up autoqa staging support (pushed to mkrizek-staging)
mkrizek starting logging support to better capture output from watchers and 'autoqa'
jlaska posted updates to clumens anaconda branch - improved test, new compose-tree pungi test, and initial post-git-receive watcher to trigger events on git push
Next steps ...
  1. Wwoods posting blog article explaining the need for autoqa ticket#248
  2. Determine appropriate strategy for watcher and depcheck integration, and revise patchset as needed
  3. Merge clumens branch into master

Open discussion - <Your topic here>

Action items

IRC Transcript

jlaska #startmeeting Fedora QA Meeting 16:00
zodbot Meeting started Mon Dec 20 16:00:41 2010 UTC. The chair is jlaska. Information about MeetBot at 16:00
zodbot Useful Commands: #action #agreed #halp #info #idea #link #topic. 16:00
jlaska #meetingname fedora-qa 16:00
zodbot The meeting name has been set to 'fedora-qa' 16:00
jlaska #topic Gathering in the lobby .... 16:00
jlaska Hello all ... anyone lurking for likely our last meeting of 2010? 16:01
mkrizek jlaska: me! 16:01
jlaska mkrizek: hi there! Is it quiet there today? 16:02
mkrizek jlaska: it is, in the fedora qa box anyway 16:02
adamw yo 16:02
Alam hello 16:02
jlaska adamw: Alam Hi there! 16:05
jlaska sorry, I'm back ... just had a display hang 16:05
jlaska okay, anyone else? robatino, Viking-Ice, wwoods 16:05
* fenrus02 waves 16:05
jlaska fenrus02: hey! 16:05
jlaska okay, let's get started ... 16:06
jlaska #topic Previous meeting follow-up 16:06
jlaska #info jlaska to ping jens+rhe and Alam for thoughts on a F15 l10n/i18n test day 16:06
jlaska good news for me, rhe and Alam are already on top of this! 16:07
jlaska #info - F15 l10n/i18n test day planning 16:07
jlaska Alam: anything else you want to highlight or share on that subject? 16:07
adamw yeah, awesome job alam/rhe, thanks 16:08
Alam jlaska: we selected test day and on the way to plan test cases 16:08
jlaska Alam: that's great news ... sounds like you guys are really movin' 16:08
jlaska we'll touch on test days shortly, but feel free to raise any questions/concerns in the ticket or on :) 16:09
jlaska #info jlaska to merge User:Bruno#Mockup_for_QA_-_Tracking_bug_queries into QA:SOP_Blocker_Bug_Meeting 16:09
jlaska With the queries the Bruno provided, I did some minor re-org of the blocker meeting SOP page 16:09
jlaska #link 16:09
jlaska I'm not in love with my changes, but it at least gets those queries on the meeting radar for F15 16:10
adamw nice! 16:10
jlaska I also added excessive meetbot commands to the page 16:10
adamw simple and to the point and has links == good 16:10
jlaska (compliments to the Design team how to for example -- 16:11
jlaska adamw: thanks 16:11
jlaska last up ... 16:11
jlaska #info Bodhi feedback patch from fcami (see ticket#701 (infrastructure) awaiting review 16:11
jlaska I think fcami posted the updated patch last week ... so it's now on lmackens' radar for comment 16:12
jlaska not sure what else to track here 16:12
* wwoods lurks 16:12
jlaska wwoods: howdy :) 16:13
jlaska so, I guess we'll just stay tuned on that 16:13
jlaska alright ... quick recap ... 16:14
jlaska #topic Call for Test Days ... 16:14
* rbergeron lurks too ;) 16:14
jlaska rbergeron: hi there :) 16:14
jlaska so adamw and company have been going gang-busters on getting test days on the schedule 16:14
jlaska #link 16:14
jlaska Some nice additions this past week 16:14
jlaska including the l10n/i18n event discussed earlier (thanks Alam + Hurry) 16:15
jlaska #link - Consistent Network Device naming (pitched by shyamiyerdell) 16:15
jlaska adamw: any other updates or news to share on this topic? 16:16
adamw nope 16:16
adamw short and sweet, huh? :) 16:16
jlaska definitely! 16:16
jlaska oh ... did you hear back on systemd yet? 16:16
jlaska or are those folks already on break? 16:16
* rbergeron will just pipe in and say the cloud SIG has been talking about doing a test day - I will ping them this week and see when we might want to schedule that and put in a ticket 16:17
jlaska adamw: ^^^ ? 16:17
jlaska rbergeron: ooh, that'll be a fun new topic 16:18
adamw sorry, multitasking 16:18
adamw actually i have mgracik asking me stuff about systemd right now 16:18
adamw so i'll ask him 16:18
jlaska adamw: no distracted driving! :) 16:18
jlaska adamw: okay 16:18
rbergeron jlaska: indeed 16:19
jlaska so, the next topic was an update on the investigation bruno was doing with regards to query potential with upstream bugzilla 16:19
jlaska we can hold that off until he's available 16:19
jlaska but for those interested ... 16:19
jlaska #link 16:19
adamw general test day note 16:19
jlaska adamw: shoot 16:19
adamw remember, we only put thursdays in the calendar 16:19
jlaska ah yes, good reminder 16:20
adamw but if you really want to go in a week we have something else scheduled, you can go ahead and add a tuesday or wednessday 16:20
adamw we like the thursday 'cadence' (heh) but it's no big deal to double-dip a few weeks 16:20
jlaska #info We only put Thursdays in the test day calendar, but other days are available if really needed 16:20
jlaska and for somethings, double+triple dipping works out great (e.g. Xorg) 16:20
jlaska well, except for adamw's sanity 16:21
adamw right - if you want to run multiple related test days, it can work well to put them in a single week 16:21
rbergeron Is there a reason why only thursdays? Just easier to advertise/remember that way? 16:21
adamw as we're planning with i18n/l10n 16:21
adamw rbergeron: basically, yes, the idea is for it to be a regular event. thursday was picked because it was a nice quiet day, iirc. 16:21
* rbergeron just wonders if that is something interesting to put on the test days wiki page so people don't go creating their own stuff in the table. 16:21
adamw yeah, we could add a note there. 16:21
* rbergeron can do that real quicklike 16:21
adamw sure, just saying the above 16:22
rbergeron yup 16:22
* rbergeron goes a-typin' 16:22
jlaska clickity-clackity 16:22
jlaska When we try to host several unrelated test days in the same week, that can often be overload/confusing 16:22
jlaska and then if it's the same people responsible for hosting multiple events in one week ... that can be pretty taxing 16:23
jlaska as indicated by the hibernation adamw goes into after xorg week 16:23
jlaska alright, thanks for the updates adamw Alam and rbergeron 16:23
jlaska #topic Requirements review for Fedora test case management 16:24
adamw that's a vicious slander! 16:24
jlaska Just a reminder, Hurry is still collecting thoughts and developing requirements for a formal test case management tool 16:24
jlaska #link 16:24
* adamw breaks out the duellin' gloves 16:24
jlaska adamw: uh noes ... I think I just got slapped by your gauntlet! 16:25
jlaska New rule, never duel with a Canuck 16:25
jlaska Hurry and I have been discussing different approaches off-list, but most of hte thoughts are collected on the talk page 16:26
jlaska #link 16:26
jlaska it'll probably be easier to provide feedback to once finalized, but if there are any general ideas on how to organize thoughts etc... the Talk: page is your friend 16:26
jlaska so thanks Hurry for starting this process :) 16:27
jlaska #topic Critical Path test case development 16:27
jlaska adamw: you're on! 16:27
adamw uh, i'm working on it. =) 16:27
jlaska hehe ... should I send that to #info :) 16:28
adamw we thrashed out various issues in the ticket this week 16:28
adamw and i'm now drafting up wiki stuff which will document the whole process 16:28
adamw once i have those drafts in place i'll send them to the lists for review 16:28
jlaska yeah, that was some good discussion, thanks for managing that! 16:28
jlaska #info Adamw drafing wiki pages intended to document the process 16:28
adamw the biggest unsolved issue that i just have no idea how to tackle is how to express relationships between packages for test plans 16:28
adamw e.g. 'when yum is updated we should probably run the test plan for packagekit' 16:29
adamw that's really hard 16:29
jlaska hmm 16:29
adamw if anyone has bright ideas about that, go ahead and throw them at the ticket 16:29
adamw for now i am cheerfully ignoring it 16:29
jlaska well, that's another not tremendously difficult to manage thing with tcms ... so the future is promising 16:29
jlaska with only Categories to use, that is tough to manage with wiki 16:30
jlaska adamw: alrighty, thanks for the updates 16:30
jlaska #topic AutoQA Update(s) 16:30
jlaska note, we're missing kparal and jskladan this week 16:31
jlaska wwoods or mkrizek, any exciting updates on the autoqa front to share? 16:31
jlaska iirc, the big discussion last week was around the rework of the post-bodhi-update how/where the batching logic should live 16:31
jlaska I think wwoods and jskladan hashed out the details on list and over email 16:32
mkrizek I've got nothing:) 16:32
wwoods right - I think we finally sorted out some of the complexity relating to depcheck and handling multithreaded testing of updates 16:32
wwoods gordian knot style 16:32
jlaska hah! 16:32
wwoods we decided not to test updates multithreaded 16:32
jlaska problem, solution! 16:32
wwoods eventually we'll need locking/semaphore stuff so we can handle situations with multiple tests sharing data 16:33
* adamw notes intel and amd are both roadmapping futures where we have zillions of cores for everything...but hey. 16:33
adamw that's a long way off, i guess. 16:33
jlaska anyone with experience developing such stuff ... input would be appreciated! 16:34
jlaska mkrizek: how's the staging support coming along? 16:34
wwoods adamw: current data shows that, during our busiest-ever month, we would need depcheck (on average) once every 33 minutes 16:34
wwoods current average depcheck runtime: ~50-60sec 16:34
mkrizek jlaska: done 16:34
mkrizek jlaska: I will push it into mkrizek-staging 16:35
wwoods we're a long, long way from needing to care about multithreading *multiple depchecks for a single Fedora release* 16:35
adamw wwoods: oooh, nice numbers! 16:35
wwoods all other testing is still multithreaded. 16:35
jlaska mkrizek: ooh, nice, what's next for that? 16:35
jlaska #info Team decided to serialize depcheck tests to avoid complexity relating to handling multiple depcheck tests running at the same time 16:36
mkrizek jlaska: just starting coding logging suport, hopefully we will have /var/log/autoqa.log/different verbose modes soon 16:36
jlaska #info Depcheck data - during busiest-ever month, we would need depcheck (on average) once every 33 minutes. Depcheck currently takes <= 1m to complete 16:37
jlaska #info mkrizek finished up autoqa staging support (pushed to mkrizek-staging) 16:38
jlaska #info mkrizek starting logging support to better capture output from watchers and 'autoqa' 16:38
jlaska mkrizek: cool, nice ... that'll help avoid having to read 'autotest' mail to scan for failures 16:38
wwoods mkrizek: nice - using the python 'logging' module? 16:39
mkrizek wwoods: yes 16:39
* wwoods finally getting good at using 'logging' and kind of loves it now 16:39
jlaska Small update on the clumens branch. I made a more improvements to the test, drafted a compose-tree pungi test, and also added a preliminary post-git-receive watcher hook that triggers whenever a git push occurs for anaconda.git 16:40
mkrizek jlaska: and also helps debugging I would say 16:40
jlaska mkrizek: yes! 16:40
jlaska #info updates to clumens anaconda branch - improved test, new compose-tree pungi test, and initial post-git-receive watcher to trigger events on git push 16:40
jlaska alrighty ... anything else to report on the autoqa front? 16:40
wwoods don't think so 16:41
jlaska okay, open discussion time ... 16:42
jlaska #topic Open discussion - <your topic here> 16:42
jlaska anything not already discussed that folks would like to raise? 16:42
adamw where's my christmas present?! 16:43
jlaska adamw: I will send you a secular non-specific winter pleasantry 16:43
jlaska if no topics, I'll #endmeeting in 2 minutes 16:44
fenrus02 Happy Chrismakwanzika folks 16:44
jlaska indeed, happy Festivus to all 16:45
jlaska 1 min remaining until #endmeeting 16:45
wwoods for some reason my xmas goodies have been bacon-themed this year - I made two batches of maple-bacon chocolate chip cookies and four batches of maple baconated bourbon 16:45
wwoods I blame the meaty influence of the Hot Dog 16:45
wwoods uh I mean, praise. praise the meaty influence of the Hot Dog. 16:46
jlaska that was one delicious cookie! 16:46
fenrus02 wwoods, W00T! that is awesome with awesome sauce 16:46
jlaska okay all ... thanks for your time! 16:46
jlaska happy holidays :) 16:46
jlaska #endmeeting 16:46

Generated by 2.7 by Marius Gedminas - find it at!