20080109

Log

UTC

19:00   MootBot Meeting started at 19:00. The chair is heno.
19:00   MootBot Commands Available: [TOPIC], [IDEA], [ACTION], [AGREED], [LINK], [VOTE]
19:02   heno    [TOPIC] Things happening today: Kernel bug day and alpha 3 testing
19:02   MootBot New Topic:  Things happening today: Kernel bug day and alpha 3 testing
19:02   heno    just FYI really, no need for discussion
19:02   heno    unless someone has comments
19:03   bdmurray        Should I be testing the daily builds now?  If not when?
19:04   bdmurray        I can see holding Hug Days and testing daily builds at the same time as being challenging.
19:04   davmor2 lo
19:04   stgraber        pong
19:04   heno    bdmurray: yes, this will happen regularly
19:04   stgraber        sorry, at lug meeting
19:05   heno    we've had then on mondays sometimes close to releases
19:05   davmor2 bdmurray: got my vote testing is hectic enough normally
19:05   heno    should we consider that as a regular day?
19:06   davmor2 sorry heno for which?
19:07   heno    bug days on mondays
19:07   davmor2 right
19:07   heno    we can't really move testing
19:07   heno    as it's tied to thursday releases
19:07   davmor2 true :)
19:08   bdmurray        I think bug days on Mondays is problematic with holidays, timezones and weekends
19:08   pedro_  I'd like to propose Tuesday instead
19:08   pedro_  Monday is for me at least the busiest day of the week where you have to catch up tons of emails from the weekend
19:09   davmor2 Tuesdays sounds good as the majority of testing seems to happen on wednesday's
19:09   bdmurray        heno: did the distro team meeting used to be on Tuesdays?
19:09   heno    with bug day on tuesday the lists could be prepared monday evening
* heno can't remember
19:10   davmor2 sounding like a plan then :)#
19:10   heno    so I'm not sure why it's wednesdays
19:10   heno    it also clashes with this meeting :)
19:10   heno    each week
19:11   liw     so Tuesday sounds best?
19:11   heno    sounds good, bdmurray?
19:11   davmor2 probably because it keeps getting postponed until Wednesday :)
19:12   davmor2 testing that is :)
19:12   bdmurray        heno: that works for me but I'd like to check with the rest of the bugsquad and the marketing team
19:13   heno    bdmurray: yep, will you email the lists?
19:13   davmor2 makes sense no point if no-one else can do it :)
19:13   bdmurray        I think the first possible Tuesday would be the 29th then
19:13   bdmurray        heno: yes
19:14   heno    that's fine
19:14   heno    are we skipping it the week of the sprint?
19:15   bdmurray        That was my assumption but it was an assumption . . .
19:16   heno    [AGREED] We may move bug days to tuesdays to reduce conflicts
19:16   MootBot AGREED received:  We may move bug days to tuesdays to reduce conflicts
19:16   heno    bdmurray: we probably should skip it. we'll be quite busy
19:17   heno    [TOPIC] Following up on the target bug list with other teams
19:17   MootBot New Topic:  Following up on the target bug list with other teams
19:17   bdmurray        I'd be happy to help with the lists before leaving if someone wants to take that on
19:17   heno    I intended to attend the platform meeting today but missed it in the end :-/
19:18   heno    bdmurray: perhaps that should go to the bugsquad list as well
19:19   heno    we have several active triagers who are not here now
19:20   heno    desktop team meeting is tomorrow at 14.00 UTC, but I have a phone meeting then
19:21   heno    pedro_: are you on line by then, could you attend?
19:21   pedro_  let me check
19:21   heno    you might want to start attending desktop meetings anyway
19:21   heno    in the same way ogasawara attends kernel meetings
19:22   pedro_  heno: yep, ok i'll attend to it
19:22   heno    of course the gnome packagers are not on the desktop team, but on platform ...
19:22   heno    pedro_: thanks
19:23   heno    I'll propose the bug list as an agenda item there
19:23   heno    [TOPIC] Desktop testing status and a quick 'getting started with dogtail' intro
19:23   MootBot New Topic:  Desktop testing status and a quick 'getting started with dogtail' intro
19:24   heno    liw has written up https://wiki.ubuntu.com/Testing/Automation/DogtailTutorial
19:24   heno    so everyone can play along at home
19:24   heno    with automated desktop testing
19:25   bdmurray        Has anyone tested it on Hardy?
19:25   liw     the instructions on that page work on gutsy, at least for me; I haven't dared to try hardy yet, and all my attempts at trying them in hardy-inside-qemu result in a broken filesystem on the qemu image; however, I expect they'll work on hardy, too
19:25   heno    it failed for me on hardy
19:25   heno    in vbox
19:25   heno    haven't tried on real hardware
19:26   davmor2 I'll let you know latter I'm just rsyncing my bunch of images :)
19:26   liw     nothing in dogtail should be affected by real hardware vs emulation, but then again, that applies to most thigns...
19:26   heno    davmor2: that's great, thanks
19:27   davmor2 well it's what test machine were designed for :)
19:27   liw     anyway, a more general update: that sample script on the page is how far I've gotten; it's surprisingly much effort to get a dogtail script written, and the recorders in accerciser and dogtail are basically of no use at all, so it's all manual work, and trial-and-error, too
19:28   liw     writing the scripts is, in fact, so much effort that I'm not sure it's worth continuing on that path -- it'll take person-weeks of effort to get anything useful for hardy, and then it'll all probably break when the apps change for hardy+1, so they'll have to be updated/rewritten
19:28   heno    liw: I promised I would hook you up with the accerciser dev; this seems like a good time for that
19:28   liw     I found out today about a new tool called Strongwind (or found out about it again -- someone told me about it during UDS, I think), I'll have a quick look at that tomorrow
19:29   liw     strongwind's not packaged for debian or ubuntu, but that should not be a problem
19:29   bdmurray        I know who that someone is
19:29   bdmurray        What happens with the recorders?
19:30   liw     bdmurray, the recorders don't record anything except key presses, when typing into text areas, which are the least interesting part of the script
19:30   heno    I could swear we had accerciser recording native scripts in oslo
19:30   liw     it may be that I'm just stupid and can't figure out how to use the recorders, of course; if so, help is appreciated
19:30   liw     heno, I get the same results now as then, I think, but the results are not useful :(
19:30   heno    I failed at it too ladt week, FWIW
19:31   heno    ok
19:31   liw     that is, the recorder records that I type "hello, world", but fails to record into which widget I do that; it does record the app, though
19:31   heno    so, yeah if anyone wants to play with this and add eyeballs that'd be great
19:32   heno    liw: did you try macaroon?
19:32   heno    it likely has near-identical syntax
19:33   heno    (why he doesn't just always use dogtail scripts i don't know)
19:33   liw     heno, no, I should try that, too
19:34   heno    ok, I think we're done. liw we can talk more about this after the meeting
19:34   liw     I forget why I rejected it as an option last year, but given that dogtail isn't looking promising right now, I should look again at all the options
19:34   liw     heno, ack
19:34   heno    #endmeeting
19:34   MootBot Meeting finished at 19:34.

MeetingLogs/QATeam/20080109 (last edited 2008-08-06 17:00:08 by localhost)