20080618

Log

UTC

[18:02] <heno> #startmeeting
[18:02] <MootBot> Meeting started at 12:05. The chair is heno.
[18:02] <MootBot> Commands Available: [TOPIC], [IDEA], [ACTION], [AGREED], [LINK], [VOTE]
[18:02] <heno> \o/
[18:03] <heno> [TOPIC]: Hardy.1 SRU verification
[18:03] <MootBot> New Topic: : Hardy.1 SRU verification
[18:03] <sbeattie> hey
[18:04] <heno> Here is a nice tracker sbeattie and LaserJock have been working on http://people.ubuntu.com/~sbeattie/sru_todo.html
[18:05] <heno> some of the major packages like linux, openoffice and ltsp have been tested and seem good
[18:05] <heno> firefox 3.0 was also released today
[18:06] <heno> one area of concern is pulseaudio/alsa
[18:06] <heno> sbeattie: do you have a view on that
[18:06] <heno> ?
[18:06] <heno> it's an area where we need to appeal for more testing I think
[18:06] <sbeattie> heno: in terms of?
[18:07] <sbeattie> Yes, I believe it's an area where we should appeal for testing.
[18:07] <heno> sbeattie: what were the concerns expressed at last week's .1 meeting?
[18:07] <heno> or is there a log I can look at?
[18:08] <sbeattie> alsa/pulseaudio has a lot of problems as released, probably due to version mismatch between alsa's userspace and kernel.
[18:08] <heno> what are the concerns about the new versions?
[18:09] <heno> just that they are undertested?
[18:09] <davmor2> sbeattie: is that the cause of somethings playing in wrong speakers if two audio apps are running?
[18:09] <heno> and that it's a major change?
[18:09] <sbeattie> davmor2: that I don't know.
[18:10] <sbeattie> Yes, that it's a major upgrade for the userspace component, from 1.0.15 to 1.0.16.
[18:10] <pedro_> new version aka alsa 1.0.16 ?
[18:10] <pedro_> there's a resume about it here: https://bugs.edge.launchpad.net/ubuntu/+source/pulseaudio/+bug/221673/comments/16
[18:10] <ubottu> Launchpad bug 221673 in alsa-plugins "ALSA failing with PulseAudio in Hardy" [Undecided,Fix committed]
[18:10] <pedro_> seb128 talked with jordi about it and send us an IRC log which slangasek resume on that comment
[18:10] <pedro_> s/send/sent
[18:11] <heno> ok, thanks. that sheds some light
[18:12] <heno> has mgunes been around lately? This would be a good item to bring to the forums
[18:12] <slangasek> recently, there's an additional concern that the new alsa-lib causes regressions for xubuntu as well
[18:12] <slangasek> which bug I cannot currently find
[18:13] <cody-somerville> mmm
[18:13] <cody-somerville> I didn't see that bug either
[18:13] <sbeattie> the last comment in 221673 perhaps?
[18:13] <sbeattie> oh right, bug 240337
[18:13] <ubottu> Launchpad bug 240337 in alsa-lib "Upgrade of libasound2 to 1.0.16 on xubuntu 8.04 desktop  causes many applications to 'hang'" [Undecided,New] https://launchpad.net/bugs/240337
[18:14] <davmor2> http://www.mail-archive.com/ubuntu-bugs@lists.ubuntu.com/msg884888.html
[18:14] <MootBot> LINK received:  http://www.mail-archive.com/ubuntu-bugs@lists.ubuntu.com/msg884888.html
[18:14] <heno> so what is most needed here, general testing or better debugging information on know issues?
[18:14]  * heno suspects the answer is 'both' :)
[18:15] <sbeattie> I think that's right.
[18:15] <sbeattie> :-)
[18:15] <davmor2> or maybe a more standard way of testing for the upgrades?
[18:16] <heno> davmor2: yes we need that too. We could use your help on improving https://wiki.ubuntu.com/QATeam/PerformingSRUVerification :)
=== RoAk is now known as RoAkSoAx
[18:16]  * davmor2 looking
[18:16] <heno> to be as user friendly as the new test case pages
[18:17] <heno> Ok, I'll read some logs and bugs about know issues and appeal for wider testing
[18:17] <sbeattie> davmor2: did you mean in particular about testing alsa?
[18:17] <davmor2> heno: that's just scary :-/
[18:18] <heno> sbeattie, slangasek: anything else we should note about the remaining SRUs?
[18:18] <sbeattie> Because that would perhaps be useful, drawing up some specific things to test.
[18:18] <davmor2> sbeattie: no as a whole
[18:18] <slangasek> heno: nothing off the top of my head, no
[18:19] <sbeattie> other areas that we could use help verifying fixes are some of the wubi/installer related bugs.
[18:19] <heno> I see there are some installer/wubi/migration-assistant issues
[18:19] <davmor2> sbeattie: My point is it's okay to install -proposed but if you hardly use the apps that get updated you'll never know if there are any issues :)  Hope that makes sense
[18:19] <heno> ... :)
[18:19] <sbeattie> lupin, migration-assistant, casper.
[18:20] <davmor2> is there an updated version that can be dropped on a cd now rather than recompiling a version and I can test that then?
[18:20] <heno> davmor2: right, the bugs linked from http://people.ubuntu.com/~sbeattie/sru_todo.html should have test cases
[18:20] <heno> that brings us to CD testing :)
[18:20] <sbeattie> otherwise, the list of things needing verification has been pared down quite well, in part thanks to heno and stgraber for verifying openoffice and ltsp respectively.
[18:21] <sbeattie> yes, cd testing, when I started to do some installer verification testing last week, I found the alternatives cd was broken
[18:21] <heno> these are not built from proposed I guess http://cdimage.ubuntu.com/hardy/daily/current/
[18:21] <sbeattie> slangasek fixed it, but we need to some smoke testing of what's getting built.
[18:21] <heno> that might be useful in the future
[18:22] <sbeattie> There's some isos being built from proposed...
[18:22] <slangasek> heno: they are built from proposed
[18:22] <heno> ah, very cool
[18:23] <davmor2> slangasek: are these the iso that will be put forward as .1 or alpha 1 or both?
[18:23] <sbeattie> .1
[18:23] <slangasek> they're under /hardy/ - they have nothing to do with intrepid :)
[18:23] <heno> [TOPIC]: 8.04.1 ISO testing
[18:23] <MootBot> New Topic: : 8.04.1 ISO testing
[18:23] <davmor2> Ah right np's :)
[18:24] <heno> (just for the bookkeeping :) )
[18:24] <heno> davmor2: both need testing, and hardy.1 more strictly
[18:24] <davmor2> No probs heno.
[18:24] <heno> stgraber says we can list both on the tracker
[18:25] <davmor2> that's cool :)
[18:25] <sbeattie> slangasek: are the other *buntus getting built for .1 as well?
[18:25] <slangasek> yes
[18:25] <sbeattie> excellent
[18:25] <slangasek> but not until we fix a buglet introduced by removing packages from -proposed
[18:26] <sbeattie> oh, they're failing to build now?
[18:26] <stgraber> heno: it's indeed possible to do it, I must admit not having tested that for a long time :)
[18:27] <slangasek> actually, they /were/ failing to build, now they just need an archive fix; sorry, my explanation above was ill-informed and buggy :)
[18:27] <sbeattie> Heh, no problem, thanks for clarifying
[18:27] <heno> so we should do some light smoke testing of the current hardy.1 CDs and then do more full coverage as we get final-candidate images
[18:27] <sbeattie> agreed
[18:28] <davmor2> heno: when are the FC cds going to be around?
[18:28] <sbeattie> Is https://wiki.ubuntu.com/Testing/Isoscript still useful for keeping daily images current?
[18:28] <sbeattie> as far as anyone knows?
[18:28] <LaserJock> is there a list of all the updates/bugs fixed on the .1 ?
[18:28] <davmor2> sbeattie: it will be if you change directories
[18:29] <heno> indeed, we should study the changes that have gone in to see what to focus on
[18:29] <davmor2> or change hardy to intrepid
[18:29] <sbeattie> davmor2: intending for 8.04.1 images.
[18:29] <heno> we also talked about copying the test cases over to .../LTS/...
[18:29] <sbeattie> LaserJock: not yet, that's something I else I want/need to generate
[18:29] <heno> so we can keep them as they are while updating for Intrepid
[18:30] <heno> i.e. fork them
[18:30] <davmor2> sbeattie: yes change the path for where the script looks for the image
[18:30] <sbeattie> davmor2: thanks, good to know.
[18:30] <LaserJock> sbeattie: perhaps you can use pitti's script to extract out the bug #s for everything in hardy-updates?
[18:31] <davmor2> sbeattie: as long as there is still a current it should be fine :)
[18:31] <LaserJock> sbeattie: filtering out Universe that should give you quite a lot of info still
[18:31] <sbeattie> LaserJock: that's a possiiblity.
[18:31] <sbeattie> LaserJock: you interested and available to help out with that?
[18:31] <heno> That should only affect Ubuntu though as Kubuntu is not LTS FOR 8.04
[18:31] <sbeattie> ("no" is a perfectly valid answer)
[18:32] <LaserJock> sbeattie: some at least yeah. I'm interested in the answer for Universe as well
[18:32] <sbeattie> LaserJock: that's what I suspected
[18:32] <sbeattie> action: sbeattie (primarily) and laserjock to generate a list of fixes going in to 8.04.1
[18:32] <sbeattie> ACTION: sbeattie (primarily) and laserjock to generate a list of fixes going in to 8.04.1
[18:33]  * sbeattie hrm
[18:33] <cody-somerville> sbeattie, LaserJock: Make sure not to forget Xubuntu :)
[18:33] <heno> buggy bot
[18:33] <stgraber> I guess heno needs to do that, and it's [ACTION]
[18:33] <LaserJock> it doesn't like you. perhaps you need to feed it more bot snacks
[18:34] <LaserJock> cody-somerville: make sure to ping us about that, but yeah, we'll have to keep that in mind :/
[18:34] <heno> [ACTION]: sbeattie (primarily) and LaserJock to generate a list of fixes going in to 8.04.1
[18:34] <MootBot> ACTION received: : sbeattie (primarily) and LaserJock to generate a list of fixes going in to 8.04.1
[18:34] <sbeattie> heno what was that about kubuntu?
[18:34] <LaserJock> cody-somerville: maybe we can look for SRUs in packages in the Xubuntu Hardy seed?
[18:34] <cody-somerville> [ACTION] cody-somerville to liaison with LaserJock regarding fixes going in to 8.04.1 for Xubuntu specific packages.
[18:35]  * cody-somerville tried.
[18:35] <heno> sbeattie: because Kubuntu is staying on KDE 3.5 (+the 4.0 remix), Kubuntu isn't LTS for 8.04
[18:35] <heno> so we don't need to keep copies of the Kubuntu test cases around for 3/5 years
[18:35] <sbeattie> ah, got it.
[18:36] <heno> we are respinning the CDs now because of the ssl issues AFAIU
[18:36] <sbeattie> yes, branching the testcases would be good.
[18:36] <heno> which we don't expect to repeat :)
[18:36]  * slangasek nods
[18:36] <heno> [TOPIC]: Spec status
[18:36] <MootBot> New Topic: : Spec status
[18:37] <sbeattie> heno: one sec
[18:37] <heno> ok
[18:37] <sbeattie> slangasek: are you okay if we push for more widespread testing of the alsa update?
[18:37] <sbeattie> i.e. blog/forum announcements?
[18:38] <slangasek> I'm somewhat concerned that this would flood us with low-quality feedback while we're still trying to sort out the meaning of bugs like #240337
[18:38] <sbeattie> That one does seem to be missing hardware info
[18:40] <heno> too much testing of -proposed seem like a bit of a luxury problem
[18:40] <heno> we got a reasonable response after I blogged, but not a flood
[18:41] <sbeattie> slangasek: we could time it for a couple days out, to give Luke time to sort out that particular issue if possible.
[18:41] <persia> Luke doesn't have HW to replicate the issues with the bug
[18:43] <slangasek> persia: what hardware is that? the bug doesn't seem to say
[18:43] <heno> what HW is it, and can we get it to him?
[18:44] <persia> slangasek: Perhaps I'm confused then.  He was previously saying that he was having trouble replicating some of the pulse issues that might be related to the ALSA mismatch, as his HW worked.
[18:45] <persia> Nevermind.  I've read the bug backwards.  That's a 1.0.16 bug.  Sorry.
[18:45] <slangasek> ok :)
[18:46] <heno> ok, shall we move to specs?
[18:46] <sbeattie> that's fine.
[18:47] <heno> Thanks everyone for polishing off the specs
[18:47] <heno> they are now all in good shape and Approved
[18:47] <heno> (appart from a fw that are not started ...)
[18:48] <heno> and I need to re-read the mobile one
[18:48] <heno> cgregan: thanks for updating, looks good at a glance
[18:48] <cgregan> heno: I completed the updates we spoke about.
[18:48] <cgregan> :-)
[18:49] <cody-somerville> Is there anybody that can help me with this bug? It is rather important.
[18:49] <cody-somerville> https://bugs.edge.launchpad.net/ubuntu/+source/dbus/+bug/232364
[18:49] <ubottu> Launchpad bug 232364 in dbus "dbus-launch freezes for unknown reason at session start" [High,Confirmed]
[18:49] <heno> I also want to add one about automated CD testing and one about QAs role in advising on release status
[18:49] <cody-somerville> It should really be fixed for the point release if at all possible
[18:49] <cody-somerville> Oops, I guess I interjected there.
[18:50] <heno> cody-somerville: that's ok, can you bring it to #u-bugs or #u-testing afterwards?
[18:50]  * cody-somerville nods.
[18:51] <heno> bdmurray: I looked at the teams in LP as we talked about for the categories on the package info pages but found the mapping quite poor, so I wen't with a hand-crafted stucture loosely based on team interest
[18:52] <heno> we may need to refine that as we go, but we shouln't block on it
[18:52] <bdmurray> heno: I thought that we should improve the mapping in Launchpad
[18:53] <heno> bdmurray: right, wecan still go that route, but ATM it's quite inconsistent how teams relate to packages
[18:54] <bdmurray> okay, sounds good
[18:54] <heno> and if we make changes across the board it will affect the bug mail people get I guess
[18:55] <heno> I still like that approach but it was starting to look like a major blocker
[18:56] <heno> any other comments on specs? if not let's revisit them in 2 weeks
[18:56] <davmor2> maybe leave it till it's safe to play about with it and formalise a plan of attack and lay it out properly then?
[18:56] <LaserJock> heno: what spec were you just discussing?
[18:56] <heno> davmor2: I'd rather have the pages up sooner so we can get use from them
[18:57] <heno> even if the structure is not 'perfect'
[18:57] <heno> LaserJock: https://blueprints.launchpad.net/ubuntu/+spec/package-status-pages
[18:58] <davmor2> heno: sorry not well explained.  Keep things as they are now then when it is safe to swap stuff around re-think the layout then is what I mean :)
[18:58] <heno> davmor2: ok, then I agree :)
[18:58] <heno> any other topics for the meeting?
[18:58] <heno> (2 minute ones :) )
[18:59] <davmor2> one minute your clock is slow ;)
[18:59] <pedro_> not from me
[18:59] <LaserJock> heno: why would those status pages not be done by Launchpad, and for every package?
[19:00] <heno> LaserJock: we basically have those pages already for each package in LP ;)
[19:00] <LaserJock> virtually all the data is from Launchpad it looks like (except maybe popcon, but that probably should be in Launchpad as well ;-) )
[19:00] <heno> this will be more team focused, collecting packages together
[19:01] <LaserJock> heno: and wouldn't that go into a teams package report on Launchpad?
[19:01]  * LaserJock is playing a bit of devil's advocate, but is curious
[19:01] <heno> we are prototyping this now but the functionality should probably be implemented in LP later, I agree
[19:02] <LaserJock> ok
[19:02] <bdmurray> That's part of the motivation for having the assoication btwn teams and packages exist in lp
[19:02] <heno> LaserJock: I agree, we are just trying to move it along
[19:03] <davmor2> make assigning bugs easy too I guess :)
[19:03] <heno> bdmurray: right, I'd be happy to get a second opinion on the feasibility of doing that now
[19:03] <bdmurray> heno: I'll look into it with one of the teams
[19:04] <heno> bdmurray: I only had a brief look and don't know the dusty corners of LP as well as you do :)
[19:04] <heno> bdmurray: great, thanks
[19:04] <heno> ok, thanks all!
[19:04] <heno> #endmeeting

MeetingLogs/QATeam/20080618 (last edited 2008-08-06 17:00:23 by localhost)