=== jono_ is now known as jono === asac_ is now known as asac === imlad|away is now known as imlad === doko__ is now known as doko === imlad is now known as imlad|away [15:57] hello [15:58] pitti: hey [15:59] morning [16:01] Wee [16:02] hi [16:02] #startmeeting [16:02] Meeting started at 10:02. The chair is slangasek. [16:02] Commands Available: [TOPIC], [IDEA], [ACTION], [AGREED], [LINK], [VOTE] [16:02] hello [16:02] [LINK] https://wiki.ubuntu.com/ReleaseTeam/Meeting/2009-09-25 [16:02] LINK received: https://wiki.ubuntu.com/ReleaseTeam/Meeting/2009-09-25 [16:03] in we go! [16:03] [TOPIC] action items [16:03] New Topic: action items [16:03] * cjwatson to *possibly* look at dbus on ia64 [16:03] I'm told this is resolved [16:03] yes [16:04] so, yay :) [16:04] turned out to be easy [16:04] (ScottK sends his regrets for this meeting, btw, work conflict) [16:04] agenda at https://wiki.ubuntu.com/ReleaseTeam/Meeting/2009-09-25, for those just arriving [16:04] * ogra is half way here [16:04] [TOPIC] QA Team [16:04] New Topic: QA Team [16:05] * Hardware testing [16:05] We have one new bugs [16:05] #435257 - Karmic takes forever to install on a Toshiba NB 200 [16:05] One solved bug: [16:05] #432497 - unable to install on Acer Aspire One because of broken SSD [16:05] And one outstanding: [16:05] #431970 - internal microphone not working on HP Mini Bixby and Toshiba NB100 [16:05] Any Qs on bugs? [16:05] I'm actually dismantling the Aspire One to replace the drive as we speak, so we should have results next week [16:06] bug 435257, bug 432497, bug 431970 [16:06] Launchpad bug 435257 in linux "Karmic takes forever to install on a Toshiba NB 200" [Undecided,Incomplete] https://launchpad.net/bugs/435257 [16:06] Launchpad bug 432497 in linux "Installing Karmic 20090917 on Acer Aspire One freezes on detecting filesystems" [Undecided,Incomplete] https://launchpad.net/bugs/432497 [16:06] Launchpad bug 431970 in alsa-driver "No sound recorded on internal microphone on Toshiba NB100" [Undecided,New] https://launchpad.net/bugs/431970 [16:06] cr3, all three need triage [16:06] mdz: will do [16:06] cr3: well, 435257 shouldn't have taken 2 days until the weekly release meeting to get targeted to karmic (I've just targeted it now) [16:06] part of the larger triage question, yes [16:07] ok, onto specs [16:07] * UbuntuSpec:karmic-qa-increase-apport-adoption [16:07] No news. [16:07] * UbuntuSpec:karmic-qa-extended-audio-testing [16:07] cr3 and fader to coordinate on testing a few laptops before integrating into the automated process. [16:07] * UbuntuSpec:karmic-qa-metrics-based-testing-phase-2 [16:07] On the client side, Phoronix has been integrated to run a few suites. On the server side, there were a few architectural problems but these are solved and now working on migration scripts. [16:08] Any Qs on specs? [16:08] cr3, no news on apport adoption? [16:08] didn't we make a major change in that area recently (the filebug redirect)? [16:08] we discussed that a lot in this week's desktop meeting [16:09] mdz: I thought that was done last week, my mistake. [16:09] right, we discussed the impact, and how to make it less strict [16:09] cr3, I think it was, but surely there has been some activity following that? [16:09] I would expect it to generate some discussion [16:09] so far the consensus was to not enforce it for ~ubuntu-bugcontrol, AFAIK [16:09] and for the QA team to be evaluating the results [16:10] mdz: I haven't received such information in time for this meeting, my apologies :( [16:10] https://wiki.ubuntu.com/DesktopTeam/Meeting/2009-09-22 "Changed bug reporting process", FTR [16:11] cr3, please take an action to ensure that the QA team collects and processes feedback about this change [16:11] mdz: and where shall I follow up with this information? [16:11] [ACTION] cr3 to ensure that the QA team collects and processes feedback about the filebug redirect change [16:11] ACTION received: cr3 to ensure that the QA team collects and processes feedback about the filebug redirect change [16:11] cr3, ubuntu-devel seems like an appropriate choice [16:11] mdz: will do, thanks [16:12] cr3: bug 432088 , FYI [16:12] Launchpad bug 432088 in malone "Disable +filebug redirection for ~ubuntu-bugcontrol" [High,Fix committed] https://launchpad.net/bugs/432088 [16:12] cr3, where is the latest hardware test report? [16:13] mdz: fader is on vacation, so I suspect his last report will date back to Wednesday. I'll try to find the url [16:13] we should have a report at every release meeting [16:14] even if fader happens to be away [16:14] http://people.ubuntu.com/~fader/hw-testing/current.html is the permanent "current" report, which includes tests as recent as the 23rd [16:14] LINK received: http://people.ubuntu.com/~fader/hw-testing/current.html is the permanent "current" report, which includes tests as recent as the 23rd === imlad|away is now known as imlad [16:14] slangasek: thanks [16:15] all the failures in that table are covered by the bugs mentioned earlier; there are a handful of "untested"s with documented reasons [16:16] we seem to still have some netbooks missing from the report [16:16] I've written to Marjo about that already, but we need to see it fixed for the next meeting [16:16] mdz: I'll double check the report reflects what we have in certification [16:16] which ones are missing? [16:17] the report is a bit misleading because it doesn't include all of the models we're supposed to be testing for 9.10 [16:17] slangasek, off the top of my head, the mini 10v [16:17] there's a full gap analysis in my email to marjo [16:17] ok [16:17] mdz: we don't have a mini 10 though, just a 9 [16:18] cr3, exactly! [16:18] it is a target which is not only untested, but we don't even have the hardware yet, i.e. it is way behind schedule [16:18] [ACTION] cr3, marjo to follow through on netbooks missing from the hw certification lab [16:18] ACTION received: cr3, marjo to follow through on netbooks missing from the hw certification lab [16:18] mdz: ah, I thought you meant we weren't reporting on the hardware we had on hand, I was scared for a bit [16:19] cr3, no, I've requested that we report on the goals for the release [16:19] well, not having the hardware that needs to be tested is also scary; but we have an action item for that now, so I think we can move on? [16:19] yes [16:19] anything else for QA? [16:20] we're missing test cases for eucalyptus [16:20] not from me [16:20] kirkland, cjwatson, et al have worked on this a little [16:20] I think help from the QA team would be appreciated [16:20] mdz: will this require integration work across systems, or can it be self contained in a single standalone system for testing? [16:20] cr3, it requires a minimum of two computers [16:21] would two VMs suffice? [16:21] pitti, two VMs can work for testing the installation portion only (you can't start VMs inside a VM yet) [16:21] but they need to be connected to a common virtual LAN [16:21] ah, of course [16:21] the setup is non-trivial in either case [16:21] mdz: I can do some work on that I'll have a chat with cjwatson and kirkland as too what they need [16:21] which is why we need a clear documented test plan [16:22] mdz: I'm afraid this will be difficult to automate for the karmic cycle [16:22] cr3, I am not looking for automation [16:22] we're not looking for automation, we're looking for documentation [16:22] excellent, just checkin [16:22] we need to create a test plan before we can even think about automation [16:22] davmor2, thank you [16:23] davmor2, do you have two computers you can use for this testing? [16:23] this use case will be good for discussion during the next uds for automation [16:23] 6 [16:23] davmor2, ok, please work with ttx on this [16:23] mdz: 6 but 4 on lan [16:23] mdz: np [16:23] [ACTION] davmor2 to work with ttx on creating a test plan for eucalyptus [16:23] ACTION received: davmor2 to work with ttx on creating a test plan for eucalyptus [16:23] he can point you to the existing documentation and forward test reports showing our current methodology [16:24] davmor2: thanks for stepping forward [16:24] nothing more from me for QA [16:24] anyone else? [16:25] [TOPIC] Desktop Team [16:25] New Topic: Desktop Team [16:25] pitti: hello [16:25] as usual, https://wiki.ubuntu.com/DesktopTeam/ReleaseStatus has all the details [16:25] This week we finally got the Karmic artwork from the Design team, and integrated the new icon theme, wallpaper, and usplash logo. [16:25] The session and messaging indicators landed a lot of urgent fixes and are in good shape for beta now. Empathy also got some bug fixing love; audio calls are now working reasonably stable, video is still brittle (currently being debugged). [16:25] We fixed 5 RC bugs, and got two new ones. [16:25] [LINK] https://wiki.ubuntu.com/DesktopTeam/ReleaseStatus [16:25] LINK received: https://wiki.ubuntu.com/DesktopTeam/ReleaseStatus [16:26] I just heared that we probably need to squeeze a new gdm artwork into beta, will chat with slangasek about that after meeting [16:26] ok [16:26] for final I'd like to finish the polkit-1 migration, so that we don't need to carry the old polkit stack in the default installation any more [16:26] what remains to be converted there? [16:27] this should be relatively easy to do, but doesn't quite count as "bug fixing" [16:27] slangasek: screen-resolutions-extra is one [16:27] (we can take that offline to #ubuntu-release after, if you wish) [16:27] the other is checkbox [16:27] checkbox doesn't "really" use polkit [16:27] I opened a bug and sub'ed cr3, I'd like to discuss it with him first [16:27] slangasek: yes, let's [16:27] too much detail for this meeting here [16:28] any questions? [16:29] oh, and of course the most critical bug EVER has been fixed: the nm-applet icon for ethernet which looked like a stray pacman ghost :-) [16:29] \o/ [16:30] hearts [16:30] Riddelll said that polkit-qt-1 won't make it, so we probalby have to live with bug 256245 [16:30] Launchpad bug 256245 in kpackagekit "Kubuntu GUI package manager does not warn if packages are unsigned" [High,Confirmed] https://launchpad.net/bugs/256245 [16:30] that's an o-o-b discussion, though ;) [16:30] slangasek: hearts> already fixed [16:30] ok [16:30] slangasek: alex did a followup upload which pretty much changes it back to "normal" [16:30] no hearts for us, sorry [16:31] pitti: could you go ahead and mark 256245 as 'wontfix' for karmic w/ explanation, then? [16:31] slangasek: will do [16:31] no sense in tracking it if we know it's blocked on other bits [16:32] any other questions from Desktop? [16:32] s/from/for/ [16:32] (done) [16:32] I need to check the upgrade script for Kubuntu some more [16:33] tseliot just fixed the intel-crash-on-logout bug which is great, although we now have another bug with KDM [16:33] otherwise Kubuntu is in decent shape [16:33] great [16:33] I'm still testing jockey [16:33] Apport works for me now [16:33] (in KNE) [16:33] jockey worked for me, once I gave it network [16:34] ok, cool [16:35] pitti: what's still beta wrt desktop-karmic-gnome-speech-replacement? [16:35] slangasek: it still needs a test plan/release note AFAIK [16:35] ok [16:35] similar for the other specs [16:35] it's by and large "testing", "test plan", etc. [16:36] seems like we'd want to have those for beta [16:36] are we likely to have them? [16:36] ok, will make sure that we will have them [16:37] thanks [16:37] [TOPIC] DX Team [16:37] New Topic: DX Team [16:37] ACTION: pitti to ensure that we have test plans/release ntoes for all specs [16:37] As usual, the status report is at: https://wiki.ubuntu.com/DesktopExperienceTeam/KarmicReleaseStatus [16:37] [ACTION] pitti to ensure that we have test plans/release notes for all specs [16:37] ACTION received: pitti to ensure that we have test plans/release notes for all specs [16:37] [LINK] https://wiki.ubuntu.com/DesktopExperienceTeam/KarmicReleaseStatus [16:37] LINK received: https://wiki.ubuntu.com/DesktopExperienceTeam/KarmicReleaseStatus [16:37] mostly bug fixes [16:37] of note [16:38] indicator-session gained a user session menu to switch to open new session; that was in fusa/jaunty though [16:38] * slangasek nods [16:38] also support for locking the session on suspend/user-switch [16:38] that may help with #407315 [16:39] xsplash was impacted by recent gdm patches for autologin [16:39] so we now have a nasty flicker; we've decided to work on that once the beta is out, to get a good baseline for a fina lset of fixes [16:40] the list of bugs on our radar is on the report [16:40] of note: #435522 - "xsplash timing out too early" [16:40] ^ especially on live system [16:40] #433719 - "indicator-applet crashed with SIGSEGV in g_str_hash()" [16:40] pitti: yep... [16:40] #430904 - Menu ordering gets "out of whack" [16:41] and a new unr crasher: #420478 - "netbook-launcher crashed with SIGSEGV in g_cclosure_marshal_VO..." [16:41] that njpatel is furiously working on [16:42] Thanks for the nice bug fix releases along the week BTW [16:42] questions on DX? [16:43] in Kubuntu: agatau fixed a bug in kmail MI and I had to work around konversation not compiling with MI support (agateau to find a proper fix), so we're in decent shape. [16:43] Riddelll: yes, thanks for pointing this out [16:44] davidbarth: please make sure that your release-targeted bugs get targeted to the release in launchpad (you can grab me or any other ubuntu dev to target them); that prevents us having two different lists that we're watching for "critical" bugs [16:44] KDM could do with upstart-ification at some point post-beta, nixternal has a script done already but needs keybuk and me to look over, but I believe our ksplashx stuff should work fine after that [16:44] slangasek: ok [16:45] Riddelll: post-beta, not pre-beta? [16:45] I was wondering that too [16:46] not being upstartified presents some ... significant challenges to the installer [16:46] although I must admit that the "Install Kubuntu" option appears to actually work in today's image [16:46] well I could upload it quickly today, but I've not had a chance to test it yet [16:46] also, it's quite an intrusive change which could use the testing in beta? [16:46] Riddelll: I can review it in the queue and poke Scott in person if I have doubts [16:47] and keybuk doesn't seem to be around to check with [16:47] Riddelll: so I would favor having it uploaded today [16:47] I guess he's travelling back from LinuxCon [16:47] ok I'll get testing on it now [16:47] AFAIK he's still /at/ plumbers :) [16:47] oh, ok; last plumbers was Tue-Thu [16:47] anyway [16:47] wrong, Wed-Fri [16:48] anything else for DX? [16:48] [TOPIC] Mobile Team [16:48] New Topic: Mobile Team [16:48] hey [16:48] here i am ! [16:48] hello [16:48] [link] https://wiki.ubuntu.com/MobileTeam/ReleaseStatus/Karmic [16:48] - Various bugs were fixed this week, good work was done to identify the armel openoffice breakage, workaround in progress [16:48] - work on moblin is ongoing, images build again [16:48] - UNR still in good shape even after lots of updates (we're making ourselves jobless here, should consider to add some artificial brekage :P ) [16:48] LINK received: https://wiki.ubuntu.com/MobileTeam/ReleaseStatus/Karmic [16:48] - closed spec on rootstock (offline-installer-gui) [16:48] - closed specs for marvell and freescale desktops [16:49] * ogra cant mention often enough how grateful we are for doko being with us on openoffice ! [16:50] yes, he's been hoovering up a lot of toolchain stuff this week [16:50] and keeping all armel HW in the company busy with testbuilds :) [16:50] and i mean *all* [16:51] is there an OOo workaround known now? [16:51] Yes [16:51] \o/ [16:51] binutils' change has been identified [16:51] the last workaround doko suggested was "ship a copy of old binutils with the OOo source package" :) [16:51] ah, much better :) [16:51] heh [16:51] I think we can revert it and rebuild oo.o [16:51] * pitti updates that on DesktopTeam/ReleaseStatus [16:51] * ogra guesses doko isnt happy having that sentence in the logs :P [16:52] Sadly, the binutils change is likely exposing a bug in another place [16:52] So we'll be hiding the real issue [16:52] ok; but we're at least closer [16:52] but oo.o will run on armel [16:52] no OOo b-d on binutils-source, nobody will notice the extra build time ;p [16:52] slangasek: yes; instead of knowing it's a toolchain change, we know now which part and which change [16:53] doko: gentoobuntu? [16:53] genTOOobuntu [16:53] :-) [16:53] haha [16:53] lool: I see bug #430277 marked as "discussion ongoing" in the status report [16:53] Launchpad bug 430277 in unr-meta "ubuntuone icon is colorful while the other panel icons are dark" [High,Invalid] https://launchpad.net/bugs/430277 [16:53] slangasek: So I escalated that bug via phone to design team on Monday evening [16:53] lool: what's holding up bringing this to a conclusion? [16:53] They wanted to chat with Mark about it and others [16:53] and I pinged them again over email today [16:53] "the theme will be improved" [16:54] presumably this issue now effects the desktop as well [16:54] discussion about *how* :) [16:54] will U1 get gray as well? [16:54] slangasek: We need to know whether we want coloured icons for everything notification area, or always black and white (runtime desaturation) or a miw (what we have) [16:54] everything else is now [16:54] lool: ok, have they been given a deadline for that to happen, so we're not trying to land these changes e.g., the week of RC? [16:54] pitti: Exactly the problem [16:54] pitti: many icons are visible only from time to time [16:55] such as tomboy, rhythmbox, bluetooth... [16:55] I feel that if it's a simple matter of changing to a thematically matching icon, we may as well get that in asap [16:55] Not really [16:55] but if it's about dynamically displaying different icons in different situations, then it's out of scope for Karmic [16:55] I think the proper fix is to either use colours or to runtime transform the icons [16:55] and we should stick with what he have and ship [16:55] I dont think we can live with all icons being b&w because it's too hard to distinguish them, and the mix is ugly [16:56] lool, right, my point is that it's too late to discussing "run time" fixes [16:56] rickspencer3: Well instead of patching a bunch of icons we'd be patching code [16:56] rickspencer3: Not sure why you feel it's harder [16:56] lool, my point exactly [16:56] because beta freeze was *yesterday* and it's time to stop making changes and fix crashers [16:56] Problem is that we dont have a finite list of icons [16:56] You can install any app [16:56] imnsho [16:57] rickspencer3: Ah you mean like not changing icon theme? :-) [16:57] patching icons would still present its own problems, in being able to make sure the right users get the right icons [16:57] I mean, for the desktop anyway, I feel it's too late to fix this bug, as there is no clear direction, and we need to channel resources to the Karmic end game [16:57] I think the easiest fix is to revert the notification area icons to colours in humanity [16:58] it's a clearly identified change and I think they have a colour version already [16:58] the easiest thing is to change nothing and live with the way it is for Karmic [16:58] wasn't really meaning to open a fresh discussion about this here, though; just trying to make sure we're on track for resolving this issue [16:58] since the bug has been open for some time now with no visible progress [16:59] This bug triggered my proposal to drop humanity in UNR which triggered the move to humanity in the desktop :-( [16:59] anyway, status quo is ugly but not breaking anything and design team has been pinged [16:59] hmm [16:59] slangasek, who gets to make the call in terms of it being "too late" ? [17:00] rickspencer3: the artwork deadline was yesterday; it's already late [17:01] so I propose we close the bug as "won't fix" and take it up in Lucid UDS [17:01] as regards "too late", any further changes are going to have to be discussed with the release team [17:02] rickspencer3: well, my understanding is that the current result looks half-cooked; I'm not keen to ship that way either [17:02] slangasek: I'll get back to you if I hear either way from the design team [17:02] slangasek: I can ask humanity upstream to prepare a colour revert if you like [17:02] I mean going back to colourful icons [17:02] lool: I think it makes sense to have that lined up [17:03] Ok [17:03] thanks lool - I presume the desktop will be expected to follow suit [17:03] Plus humanity folks are not busy with many more changes at this point any way [17:03] [ACTION] lool to ask humanity upstream to prepare a colour revert for bug #430277 [17:03] ACTION received: lool to ask humanity upstream to prepare a colour revert for bug #430277 [17:03] Launchpad bug 430277 in unr-meta "ubuntuone icon is colorful while the other panel icons are dark" [High,Invalid] https://launchpad.net/bugs/430277 [17:03] rickspencer3: we share the theme package now, so you will automatically [17:04] anything else on the mobile side? [17:04] we'll ping the doc team for screen shot updates when that comes [17:04] we're happy [17:04] pitti: better to warn them in advance that it may be coming, I think, so they don't waste time on screenshots they'll have to throw away? [17:04] slangasek: agreed [17:05] [TOPIC] Kernel Team [17:05] New Topic: Kernel Team [17:05] slangasek, ogra: thanks [17:05] pgraner: morning [17:05] apw: here? [17:06] slangasek, pgraner is on a plane now I think [17:06] ah [17:06] any kernel concerns that anyone else needs to bring up? [17:07] pgraner is on a plane, but said he would be online [17:07] a bunch ... for imx51, but nothing that wasnt mentioned in mobile already [17:07] bug list seems self-explanatory; including some that I talked through with the kernel team in person this week [17:07] slangasek, I wanted to make sure that linux-ec2 (source, binary, meta) get into main for beta [17:08] [ACTION] slangasek to get linux-ec2 into main for beta [17:08] ACTION received: slangasek to get linux-ec2 into main for beta [17:08] mdz: looks like the meta's the only thing not there currently [17:08] but yes, I'll follow through [17:08] slangasek, ok [17:08] it should be in the new uec seed, and so show up in component mismatches [17:08] ok [17:09] yes, it's in c-m [17:09] [TOPIC] Server Team [17:09] New Topic: Server Team [17:09] ttx: hi [17:10] I'm afraid ttx and I haven't synced up yet due to eucalyptus chaos, but I have a few agenda items here to add [17:10] ok, please go ahead [17:10] first the MIRs for ec2-images [17:10] o/ [17:10] it looks like several have been completed this week [17:10] mdz: I kinda updated https://wiki.ubuntu.com/ServerTeam/ReleaseStatus [17:11] unless I'm missing something, m2crypto is the only one left [17:11] [LINK] https://wiki.ubuntu.com/ServerTeam/ReleaseStatus [17:11] LINK received: https://wiki.ubuntu.com/ServerTeam/ReleaseStatus [17:11] cjwatson, ah, I'd missed that ec2-init was 'fix committed' (what does that mean for an MIR?) [17:11] it means approved [17:11] bug 434744 is the master bug for the fact that the uec images were not main-pure [17:11] but there are still unpromoted dependencies so I was holding off actually promoting it [17:11] Launchpad bug 434744 in ubuntu "ec2 and uec images contain packages not in main" [High,New] https://launchpad.net/bugs/434744 [17:12] kees requested a change in m2crypto [17:12] cjwatson, are there MIRs for the dependencies already? [17:12] i'd like to see the testsuite enabled, but beyond that, I'm ok with it [17:12] although said that he was fine with it going into main on the condition that the bug remains RC [17:12] or words to that effect [17:12] * kees nods [17:12] hmm, actually, ec2-init doesn't dep m2crypto [17:13] I thought it did for some reason [17:13] I assumed it was python-boto [17:13] the fixed ec2-init isn't uploaded yet [17:13] smoser, status of ec2-init? [17:13] the one in bzr head is okay [17:13] the one in the archive really shouldn't go anywhere near main [17:13] euca2ools dep: python-m2crypto [17:13] does ec2-init need a sponsor or something? [17:13] I don't know, need input from smoser [17:14] I tried to reach him a few times this week but have had trouble getting hold of him [17:14] I talked to smoser yesterday, but he didn't mention uploading/sponsoring [17:14] we just talked about the actual changes needed [17:14] soren: can you help here? [17:14] I've promoted m2crypto [17:15] in terms of review/sponsoring/testing ec2-init? [17:15] mdz, soren is good with my changes and is trying to incorporate and will get a new version in karmic [17:15] soren is away, I'll need someone else to pick that up [17:15] mdz: zul can do that. [17:15] ttx, ok, please confirm that he will do it today [17:15] mdz: he's been watching the MIRs to see them to completion; [17:16] ttx, can zul also take care of the test suite in m2crypto? [17:16] yeah I will do it today [17:16] [ACTION] zul to get new ec2-init sorted today [17:16] ACTION received: zul to get new ec2-init sorted today [17:16] cjwatson, odd, euca2ools was already in main [17:16] mdz: yes, I promoted it before noticing that it had an unpromoted dep [17:16] so with m2crypto promoted, and ec2-init getting promoted today, that should be the last of it [17:17] and decided it didn't matter so much since we knew we were going to do it anyway [17:17] remaining EC2 bugs are at [17:17] https://bugs.edge.launchpad.net/ubuntu/+bugs?field.tag=ec2-images [17:17] it's been whining at us in karmic_probs for a bit [17:18] one outstanding item I don't think we have a bug for on that list...switching vmbuilder to use seed tasks rather than the hardcoded package list for karmic [17:18] do we want to attempt that for beta? [17:18] I thought that was committed [17:18] branch? [17:18]