/srv/irclogs.ubuntu.com/2008/10/14/#ubuntu-bugs.txt

wgrantjames_w: I sometimes wonder how people survive without multi-finger tapping...00:00
Awsoonnwgrant, do I need special hardware or will any touchpad work?00:01
wgrantAwsoonn: If your touchpad does scrolly things, it's the right kind00:01
Awsoonnit does inded, how do i enabe multitouch scrolling and tapping and such?00:02
wgrantAwsoonn: Multi-finger tapping needs to enabling; we have it on by default.00:02
wgrantBut it's very hard to do now.00:02
wgrantWhich is what I'm trying to fix.00:02
wgrantIt worked in Hardy.00:02
pedro_are you using a macbook?00:02
wgrantAnybody who feels like testing, even if they don't normally use those features: https://edge.launchpad.net/~wgrant/+archive00:02
Awsoonnif you ned testers, this is a fresh install I'm on right now00:02
wgrantpedro_: Me?00:02
pedro_wgrant: yes00:02
wgrantGrab the newer xfree86-driver-synaptics from there.00:03
wgrantpedro_: No.00:03
pedro_all this multi taping thing reminds me of a macbook00:03
wgrantpedro_: Why?00:03
pedro_you know where you don't have a second button for the menu and things like that00:03
wgrantMacbooks have two-finger scrolling on my default.00:03
wgrantAh.00:03
wgrantI rarely use buttons.00:03
Awsoonnwgrant, will I need to restart X?00:06
wgrantAwsoonn: You will.00:07
Awsoonnor enable it in a config file or anythign after i install your package?00:07
Awsoonnalrighty then. see you in a flash00:07
wgrantNo config required.00:07
Awsoonncool!00:10
wgrantIt works? What have you tried?00:11
Awsoonn2 finger tap works prety well00:11
Awsoonn3 finger worked once00:11
Awsoonnit usually just makes my mouse go crazy on teh screeen00:11
wgrant3 is hard to get right. You have to ensure the fingers hit around the same time with around the same pressure.00:11
Awsoonnok, I got it to work once more00:13
wgrantIt takes a bit of getting used to. But two-finger works reliably?00:14
Awsoonnyea,00:14
wgrantExcellent. Thanks.00:14
Awsoonnscrolling, hasn't worked yet for me here00:14
wgrantTwo-finger scrolling? Is it enabled?00:15
AwsoonnI don't know00:15
AwsoonnI dt enable anythign, so I guess not?00:15
Awsoonnlag is killing my typing here, sorry00:15
Awsoonnhow do I enable the scrolling function?00:15
wgrantxinput set-int-prop "Your Touchpad's Device Name Here" "Synaptics Two-Finger Scrolling" 8 1 000:16
wgrantYou can then scroll vertically just by dragging two fingers up or down anywhere on the touchpad.00:17
Awsoonnhow can I find my touchpad's name?00:17
wgrantxinput list00:17
Awsoonndoes it require a x-restart?00:18
wgrantNo.00:18
wgrantxinput list-props "Your Touchpad Here"00:19
wgrantThat will show the current settings.00:19
wgrantYou're looking for a "1 0" for Synaptics Two-Finger Scrolling.00:19
wgrantJust above the middle of that listing.00:20
hggdhwgrant, this is on your synaptics version, correct?00:20
wgranthggdh: Yes.00:20
Awsoonnok, I se it in there00:20
hggdhwithout restart, I still get the fetch failure on xinput list00:20
Awsoonnoh that is jsut too cool00:20
wgranthggdh: Yep... it's an amd64-specific bug that we're trying to work out.00:20
AwsoonnI had my fingers too close I guess00:20
AwsoonnI can boot this int 64-bit too if you need.00:21
wgrantAwsoonn: We can reproduce it, so that won't be necessary.00:21
wgrantBut thanks for testing!00:21
Awsoonnhey thanks for doing the hard work!00:22
Awsoonnmaybe you'll be the person I wanted to talk to :)00:22
Awsoonnrnoicly enough... I cant in here with a problem, on INtrepid, the volume buttons on my laptop cause the vlue to quickly go all teh way to max, and all teh way to low with one press. then the keyboard stops responding00:23
wgrant.... volume?00:23
wgrantDo you mean brightness?00:23
wgrantI've not seen that with volume before.00:23
wgrantDamn.00:23
Awsoonnif I were on a micro, I'd say that the interupt wasing being cleared after servicing, but I dont knwo how it works on a computer00:23
wgrantWhat kind of laptop is it?00:24
AwsoonnHP00:24
wgrantAnd can you recover by switching to a VT and back?00:24
AwsoonnI'll try :)00:24
wgrantAnd if you run xev and hit a volume key, do you see lots and lots of events?00:24
wgrantAnd if you turn off key repeat in System->Preferences->Keyboard, does it fix it?00:24
Awsoonnnot quite00:29
AwsoonnI have a slightly differant t set of snptoms00:29
Awsoonnswitchign to a VT and back fixes it00:29
Awsoonnbut, turnign off repeat doen'st fix it00:30
wgrantHmm.00:31
wgrantWhat does xev say?00:31
Awsoonnit makes it better maybe though, I don tget any mesages in xev, and the volume only goes up one click, but after that one click, the keyboard stops working and it can't go any farther up00:31
wgrantOK.00:31
wgrantHmm00:32
Awsoonninterestingly enough, i can't evn put my curso in the terminal after it borks00:32
Awsoonni'll paste bin you some xev00:32
Awsoonnhttp://pastebin.ubuntu.com/57192/00:33
Awsoonnif that means anything to you :) Ibegan to log the xev, presses volume up got one click, couldn't go any farther, then volume down, couldnt go any farther, did a few ups and downs after that for good measure, switched to a VT and back and paste bin'd it00:35
Awsoonnso you have the story to go with the log.00:35
wgrantI've so far been unable to catch the kernel guy I need...00:36
AwsoonnI don't even know where to start looking for that kind of code, otherwise I'd have looked myself00:36
wgrantExactly.00:37
Awsoonnwell, e-mail me or msg me if you need more help with it or anything00:37
wgrantWill do.00:39
wgrantThanks.00:39
Awsoonnaccually, do you think you could give me a pointer to where that code might lie?00:39
wgrantI have no clue.00:40
Awsoonnyou think it's in teh kernel? I was thinkign the same thing, but what module would tha tbe part of?00:40
wgrantA couple of us think it's in the kernel, yes.00:40
* wgrant knows the kernel not.00:41
Awsoonnand let em guess, a couple of the kernel guys think it's in X00:41
wgrantWe haven't managed to talk to them yet.00:41
Awsoonnright on, who is the guy you were hopign to track down?00:41
wgrantrtg00:41
Awsoonnawesome, a new name to me. :)00:41
crimsunwgrant: probably on more readily tomorrow US time, as today is a federal holiday00:42
Awsoonnrtg is Canadian?00:43
ograheh, no00:43
ograamerican00:43
wgrantcrimsun: Aha.00:43
* Awsoonn wonders how many people know it's Canadian thanks giving day00:43
* ogra does 00:44
* Awsoonn wonders is ogra is getting turkey tonight00:44
ogranah, i had pork, we dont celebrate thanksgiving in germany :)00:44
paulproteusPMT, Any luck with that btw?00:45
AwsoonnI recomend you start ThanksGiving in Germany... turkey s delicious afterall.00:45
PMTpaulproteus - trying.00:46
ograwell, i can have turkey without a reason ;)00:46
PMTit'll take awhile for me to figure out how to burn battery to trigger this00:46
greg-gPMT: glxdemo :)00:49
PMTworking on it00:51
PMTi'm using cpuburn and glxdemo00:51
PMTgotta travel to b-more00:53
PMTwill call back ca 11 PM sorry :(00:53
james_wPMT: please drop a note in the bug once you have tested, whether it works or not00:54
james_wwgrant: do you know what bug 282963 is a duplicate of?01:34
ubottuLaunchpad bug 282963 in gnome-power-manager "brightness keys act as if repeated many times, locks keyboard" [Undecided,New] https://launchpad.net/bugs/28296301:34
wgrantjames_w: It's a dupe of part of one of my bugs, but that turned out to in fact be two completely different bugs with similar symptoms. I'm not sure it's really a dupe of any.02:00
wgrantjames_w: Oh, I did file bug #261721. That's it.02:01
ubottuLaunchpad bug 261721 in gnome-power-manager "LCD brightness OSD steals keyboard focus" [Undecided,New] https://launchpad.net/bugs/26172102:01
james_wit sounded like the one you were describing02:01
wgrantIt is, yes.02:01
wgrantI'll comment with the current knowledge on my bug, and dupe the other one.02:01
james_wah, or is it two bugs and he also has them both?02:02
wgrantEhem.02:02
james_wthanks though02:02
wgrantIt's actually the same laptop.02:02
wgrantInteresting.02:02
=== erhesrhsrtb54vyh is now known as Elbrus
=== brian_ is now known as BrianFreytag
=== brian_ is now known as BrianFreytag
=== brian_ is now known as BrianFreytag
=== brian_ is now known as BrianFreytag
Awsoonnany reson why tring to install fglrx would want to remove ubuntu desktop as well as xorg?04:25
Awsoonn8.10 here btw04:25
Awsoonnjockey doesn't give me an option to install it interestingly enough as wel04:26
Hobbseebecause fglrx isn't compatible with the new X yet, iirc.04:26
Awsoonn:D awesome.04:27
Awsoonnso the solution is to wait then I suppose huh?04:27
Hobbseeyeah04:27
Hobbseeand hope that ati upgrades.04:27
Awsoonnthe open source drivers are pretty amazing imho, I've gone a full 2 days and didn't even realize I was using them :)04:28
Awsoonnin any case, thank you Hobbsee04:29
HobbseeAwsoonn: you're welcome04:29
=== brian_ is now known as BrianFreytag
dholbachgood morning06:11
TaimSo I am running up against a possible bug where my system hangs just after logging in.  I want to say it is either 1.  Network-Manager related, 2.  ATI xorg related or 3.  Something completely other.  The problem is that the system hangs so bad, sysreq magic and other fun things don't work.06:27
TaimI look at the usual suspect logs, but I don't see anything that really stands out to me.06:27
TaimI am also starting to wonder if it isn't a hardware issue as it only happens on a cold boot after being down for more than say 1 hour.06:28
TaimAfter that, a power cycle clears it up even if it's a hard reboot.06:28
TaimSo, uhmn....I don't want specific help.  Was hoping to get some idea of exactly how to traige where this should go.06:52
PMTmoop.06:56
PMTTaim - what wireless card are you using?06:56
TaimAtheros AR5001X+ (rev 01)06:56
PMTA-ha.06:56
PMTyou're connecting to a WPA network, aren't you.06:57
TaimSure am.06:57
PMTHigh-five. Welcome to ath5k land.06:57
Taimmnn..06:57
TaimShould have guessed.06:57
PMT:)06:58
PMTThis is hardly conclusive.06:58
PMTI'm just another user with the ath5k card and bugs with it doing the same thing.06:58
TaimTrue true.06:58
TaimSo, although it is yet something else to travel down, I am seeing only 1 bug on Atheros for anything 8.10.07:03
TaimIt also doesn't mean that this is truly my problem.07:03
PMThold please07:03
Taim<hold music> in my head.07:04
TaimRick Astly even in my head...that's bad.07:05
PMThttps://bugs.launchpad.net/ubuntu/+source/linux/+bug/26114207:06
ubottuLaunchpad bug 261142 in linux "ath5k driver causes Ubuntu to freeze" [Undecided,New]07:06
PMTbug 26925307:07
ubottuLaunchpad bug 269253 in linux "ath5k with AR5424 fails at startup or resume" [Undecided,New] https://launchpad.net/bugs/26925307:07
PMTbug 27650807:07
ubottuLaunchpad bug 276508 in network-manager "[intrepid] [Atheros AR5211] ath5k driver inoperative, wpa_supplicant never completes connection" [Undecided,Confirmed] https://launchpad.net/bugs/27650807:07
PMTalso bug 272185 (it's mislabeled as being intel-specific)07:07
ubottuLaunchpad bug 272185 in network-manager "[Intrepid] iwl3945 + iwl4965 -- network-manager will not connect to a WPA Enterprise network (disassociating by local choice (reason=3) )" [Undecided,In progress] https://launchpad.net/bugs/27218507:07
TaimHow are you searching on this?  I tried an advanced search with "Atheros" and "ath5k" as seperate queries.  To help me narrow down, I selected any of the intrepid and 8.10 milestones in the advanced search.07:09
* PMT shrugs.07:09
PMTI happen to have found these bugs by being subscribed to them. :)07:09
TaimAll I got was bug 26354307:10
ubottuLaunchpad bug 263543 in linux "ubuntu 8.10 beta boot failure - patch attached" [Medium,Fix committed] https://launchpad.net/bugs/26354307:10
PMTsearching with ath5k on ubuntu bugs works for me...07:11
PMTas a simple search07:11
PMThttp://xrl.us/otsmf07:11
TaimInteresting.07:12
TaimI was expecting more noise from previous versions.07:12
PMTath5k was just enabled in intrepid. :)07:12
PMTrather - it was just shipped in the kernel in a kernel version between hardy and intrepid.07:13
Taimhaha, so true...07:13
PMTso we get to be test dummies. :)07:13
TaimNever thought about that.07:13
PMTI happened to be following the driver's development, so that's why I knew that to start.07:14
TaimAh.  It's only a passing interest as it's only a backup machine for me.07:15
TaimI usually try to keep my wireless to Intel cards.07:16
=== bdmurray changed the topic of #ubuntu-bugs to: Hug Day! https://wiki.ubuntu.com/UbuntuBugDay/20081014 | Ubuntu BugSquad | http://wiki.ubuntu.com/BugSquad | Documentation: http://wiki.ubuntu.com/HelpingWithBugs | If you have been triaging bugs for a while, please apply to https://launchpad.net/~ubuntu-bugcontrol/ | Want to report a bug? Read https://help.ubuntu.com/community/ReportingBugs | User support (not related to triage) is in #ubuntu
PMTbdmurray - was this a not-subtle hint to go to #ubuntu?07:17
bdmurrayPMT: nope, its nearly the 14th in my tz07:18
PMToh, I didn't get a diff against the topic in my client, so I was guessing. :)07:18
PMTTaim - ah, it's my new laptop, so I have a slightly more active interest.07:20
PMTooh, james_w, testing now07:22
TaimPMT: I understand.07:22
PMTjames_w - tragically, the version with your patch still crashes.07:23
=== Rafik_ is now known as Rafik
=== ara_ is now known as ara
Tumiehi guys, i'm using an EEEPC with EEEbuntu,, but if a window is too big for my screen (800*480), and i click on something, it goes to the other (not viseble) side of the window,, i just can't click something..09:42
leoquantTumie try #ubuntu09:44
Tumie(this happens with xchat: server -> edit & amsn -> webcamsettings09:44
Tumieok,09:44
persiaTumie, Or try ubuntu-mobile (assuming you're willing to test intrepid)09:45
=== asac_ is now known as asac
=== pedro__ is now known as pedro_
seb128dholbach: so what are those harvest review buttons supposed to do?12:41
=== ara_ is now known as ara
persiaseb128, It just puts it in the "reviewed" category, in the hopes that other people don't have to review the same thing.12:46
seb128persia: so there is no way to hide items?12:46
persiaseb128, Not really.  Mark Reviewed is the closest to hiding.  Used to be that you couldn't even do that.12:47
* ogra hands seb128 a black edding marker12:48
persiaogra, The problem with using those on harvest is that it makes it hard to read the other applications.12:48
persiaseb128, I think Mark Reviewed takes it off the primary list.  If it doesn't, you could file a bug.12:49
ograt works as long as you dont scroll12:49
persiaogra, It breaks when I open a new application.12:49
ograso dont open it :P12:49
ograyour screen is trashed by then anyway, who would want to open other apps12:49
persiaheh12:50
seb128I was looking at harvest, it's useful but really noisy12:51
seb128lot of incorrect changes, closed bugs, duplicates, etc listed12:51
ograyou could use little black tape stries though, thats less harmfull12:52
ogra*stripes12:52
seb128hum, harvests seems to list all the fedora patches that are stored on their cvs12:53
seb128which means lot of patches which are deprecated and not used, they tend to clean the spec but let those there too12:53
persiaseb128, For the incorrect/dupes/etc. complain to bdmurray, who produces the LP input scripts.  For fedora, complain to dholbach, but I suspect the suggestion will be to mark all the reviewed patches reviewed, and then we only see the new ones (which could be useful).12:54
persiaYeah, fixing the fedora importer to parse the spec file is probably a good idea.12:54
seb128bug #24810312:56
ubottuLaunchpad bug 248103 in harvest "'mark reviewed' button does not appear to do anything" [Undecided,Confirmed] https://launchpad.net/bugs/24810312:56
seb128apparently the mark reviewed button not working is a known issue12:56
seb128the service used to list the nbs has outdated informations12:58
* ogra thinks probably mark doesnt simply have the time ...12:59
persiaFrustrating.  Mark Reviewed used to work.12:59
* ogra stops making silly jokes now 12:59
dholbachpersia: what seb128 said13:26
dholbachit's a harvest bug13:26
dholbachno idea, why it's failing13:26
dholbachI'll investigate13:26
seb128dholbach: thanks13:26
seb128dholbach: the fedora cvs listing old patches combined to the button non working makes it difficult to have clean lists13:26
dholbachseb128: I know13:27
seb128dholbach: I guess that the "list bugs has being closed upstream when other upstream tasks are closed but not the corresponding one" is a launchpad bug, right? you use the filter to list bugs which are closed upstream there?13:28
dholbachseb128: it's a LP bug and it's filed already13:28
dholbachseb128: which filter?13:28
dholbachseb128: that's a list that bdmurray generates - I'm not sure how much filtering and intelligence is added13:29
dholbachthe tasks mismatch is really annoying :/13:29
seb128dholbach: the show bugs which are resolved upstream launchpad search option13:30
seb128dholbach: anyway as you said that's a launchpad bug13:30
seb128dholbach: I guess the only bug on your side there is the review button not always working (and maybe the fedora list that should be filtered using the spec too)13:31
dholbachyeah, and I'm sorry not taking a look at it earlier13:31
dholbachI'm a bit busy with all kinds of things right now13:32
dholbachneed to head off for the berlin bug jam in a bit too13:32
seb128dholbach: everybody is busy no blame, I was just looking at cleaning the harvest desktop list to maybe start promoting it for contributors13:34
* dholbach hugs seb12813:34
* seb128 hugs dh13:40
james_wdebhelper?13:41
seb128james_w: no, dh<tab> to match dholbach who just left the channel ;-)13:42
* james_w hugs yada13:42
* Hobbsee boots james_w into the middle of next milenium13:43
seb128hey Hobbsee13:43
Hobbseehey seb128!13:43
Hobbseejames_w: should we trust any of your uploads, if you like yada?  :)13:43
james_wit's not the yada you should worry about13:44
Hobbseeit's the checkinstall?13:44
* james_w has a set of homebrewed erlang scripts to do packaging13:45
james_wdebian/control, what's that?13:45
persiajames_w, In case you weren't around, at one point discussion of yada was a kick-worthy offense.  Not sure if that's still being enforced.13:54
james_wheh :-)13:54
Hobbseepersia: it should be.   But i've already been dealing with nutcases today, and am getting lazy.13:54
persiaHobbsee, It used to be, although I've not seen anyone kicked for it since Dapper.13:55
Hobbseepersia: well, people got told off for kicking other people.  Or acting, really, at all.  So people try not to act as much as possible.13:55
Hobbseeat least in userland and such13:55
persiaInteresting.  Not sure that's best, but at least good to know.13:56
Hobbseepersia: yeah, i think many people hold your opinion, but, what can you do?13:56
=== LucidFox_ is now known as LucidFox
joumetaldoes bug 99505 have enough information to be confirmed?15:35
ubottuLaunchpad bug 99505 in gnome-power-manager "[apport] gnome-power-manager crashed with SIGSEGV in XInternAtom()" [Medium,Incomplete] https://launchpad.net/bugs/9950515:35
savvas_has anyone noticed a bug report about intrepid, when ejecting a cd/dvd rom, it inserts it back in?16:04
persiasavvas_, Yes.  I don't remember the number offhand, and it doesn't happen for me (my hardware isn't like that), but it's worth searching for it.16:06
savvas_I tried to search but couldn't find it :\ the package responsible seems to be "eject" package right?16:07
leoquantsavvas: https://bugs.launchpad.net/ubuntu/+source/linux/+bug/26401116:10
ubottuLaunchpad bug 264011 in linux "Intrepid: CD cannot be ejected" [Undecided,Invalid]16:10
leoquantejecting cd rom/dvd rom problems seems to be related to macĀ“s (in intrepid)16:13
savvas_not a mac, sorry :)16:14
savvas_This is my dvd drive:       product: DVD-RW  DVR-212       vendor: PIONEER16:16
savvas_seems to be my case, thanks leoquant:)16:17
leoquantnop16:18
savvas_hm.. so I should file a different bug report?16:18
leoquanti think so16:19
savvas_ah wait you're right16:21
savvas_my cd/dvds can be eject, but they are inserted right back in16:21
savvas_*ejected16:21
persiaI'm sure I saw someone discussing that a couple days ago, but if you can't find the bug, and file a new one, perhaps someone will mark it duplicate, or discover it is truly different.16:22
=== txwikinger2 is now known as txwikinger
savvas_which package should I use? eject or hal?16:25
savvas_or some other?16:26
bdmurrayTrying it in a tty w/o being logged into X might help determine that16:26
savvas_bdmurray: should I stop gdm and try from a tty?16:27
bdmurraysavvas_: logging out of X should be enough16:27
savvas_ah cool, let me save my work and I'll give it a shot16:28
savvas_be right back16:28
savvas0well um.. I used the eject command in a tty and it still ejects and re-inserts the dvdrom16:34
savvas0I'll try with the eject button on the device again16:35
bddebianBoo16:38
savvas0sorry for the delay, the dvdrom isejected and  inserted right back in, even while in tty17:01
bdmurraysavvas0: and that is when you press the button?17:03
savvas0yes bdmurray17:03
savvas0I've shut down the machine, booted and used tty117:05
savvas0another weird thing was that while in tty, I used ls -l /media/cdrom0 and the contents of the dvd media were "total 0"17:06
bdmurrayseb128: could you look at 279158?17:06
seb128bug #27915817:09
ubottuLaunchpad bug 279158 in ubuntu-meta "gnome-spell a redundant dependency of ubuntu-desktop" [Undecided,New] https://launchpad.net/bugs/27915817:09
seb128bdmurray: what about it?17:10
bdmurrayseb128: is gnome-spell really redundant and should those packages be updated?17:11
persiaThere's quite a few redundant dependencies in the seeds currently.  This late in the cycle doesn't seem the best time to try to hunt them down and fix them.17:11
seb128bdmurray: evolution doesn't use it nowadays, I'll update it in the next upload, not sure about either something else is using gnome-spell though17:11
seb128s/either/whether17:12
savvas0um.. another problem I have is that when I switch tabs in a program, like firefox, it doesn't show all the contents of the current tab17:23
savvas0I hope this weird behaviour is reported, because I really don't know how to phrase it or which package is in question, since I've noticed this on other programs as well17:24
TreenaksIs there any more I can do for bug 281825?18:34
ubottuLaunchpad bug 281825 in gzip "gzip, gunzip and zgrep manpages are missing" [Undecided,Confirmed] https://launchpad.net/bugs/28182518:34
chrisccoulsonTreenaks - you could subscribe 'ubuntu-main-sponsors', as you have attached a patch for the problem18:44
LaserJockbdrung: around?18:45
Treenakschrisccoulson: ok18:46
bdrungLaserJock: yes18:59
LaserJockbdrung: sweet19:01
LaserJockbdrung: looking at bug #27876419:01
ubottuLaunchpad bug 278764 in matplotlib "[intrepid] No output in ipython because because of wrong default backend" [Undecided,Confirmed] https://launchpad.net/bugs/27876419:01
LaserJockjust added a comment19:02
bdrungwe should switch the default backend to TkAgg or GtkAgg19:02
LaserJockbdrung: we need to set TkAgg as the default backend19:02
bdrungcurrently matplotlib pulls both in (tk and gtk)19:02
LaserJockbdrung: I had an idea though, that we should maybe create python-matplotlib-gtk and python-matplotlib-qt metapackages19:02
bdrungLaserJock: good idea.19:03
LaserJockeven if somebody has python-qt{3,4} installed there are deps on other gnomeish packages like python-glade219:03
bdrungmaybe add python-matplotlib-tk too19:04
LaserJockit's not just python-gtk2 vs python-qt419:04
LaserJockwell, I was thinking we'd keep tk and wx in a base python-matplotlib19:04
LaserJockso those are the base and then if you want the more DE-specific backends you can add those on19:05
LaserJockhmm, though wx is wxgtk19:05
bdrungLaserJock: currenlty${shlibs:Depends} pulls gtk libs19:05
bdrungwx should be part of mpl-gtk19:05
LaserJockright19:05
LaserJockhmm, it's slightly complicated19:06
LaserJockin terms of upgradability19:06
LaserJockwe'd probably want python-matplotlib to depend on mpl-gtk and mpl-qt and like mpl-base (that has tk and the rest of the deps)19:07
LaserJockor maybe it wouldn't need to dep on mpl-qt as that's never been default19:08
bdrungqt was never default19:08
LaserJockbut we shouldn't take away backends that people have already had :-)19:09
bdrungbut mpl-gtk would conflict with mpl-qt19:09
LaserJockwould it conflict?19:09
bdrungor how do you would handle /etc/matplotlibrc?19:10
LaserJockwell, I was thinking that mpl-base would always be installed and would have the default (TkAgg)19:10
LaserJockand then you'd want to put in a note somewhere that if you want GTK or Qt backends you need to install the packages19:11
bdrungmpl-gtk should have GtkAgg as default19:11
LaserJockbut you're right, the -gtk and -qt packages should set those as the default19:11
LaserJockbah, mpl has *too* much choice :-)19:12
bdrungyes :)19:12
persiaYou could use alternatives for the front-ends.19:12
LaserJockhmmmmm19:13
persiaAnd have the configuration point at the base alternative.19:13
LaserJockthat's a good idea19:13
LaserJockso we'd just have 3 different maplotlibrc files in /etc19:13
LaserJock-base, -gtk, -qt19:13
bdrungshould -qt use QtAgg or Qt4Agg as default?19:13
persiaWhen doing this, create the virtual mpl-frontend package, and have each frontend Provide: it, so you can add/remove later at leisure.19:14
persiaQt4Agg for intrepid.19:14
LaserJockyeah19:14
LaserJockwe can do python-qt4 | python-qt3 I think and default to Qt419:14
bdrungwe should contact the debian maintainer19:14
LaserJockok, that's Jaunty stuff though19:15
LaserJockfor Intrepid should we revert to TkAgg or GtkAgg?19:15
LaserJockI thought GtkAgg has some issues19:16
bdrungmom19:16
LaserJockI was thinking move moving python-tk into the | line so we'd have: python-tk | python-gtk2 | python-wxgtk2.8 | python-qt3 | python-qt419:17
persiaWait, is all of this in *one* binary package?19:18
LaserJockyes19:19
LaserJockthat's the current situation19:19
LaserJockand people are sometimes a bit upset that they have to download like 100MB just to install this plotting library19:19
persiaI'd be very tempted to split the frontends into separate binary packages, just so each would have the correct dependencies.19:19
persiaOh, yes, especially for a 100MB package.  That calls for splitting :)19:20
LaserJockthe package isn't 100MB, it's really small19:20
LaserJockit's the deps that kill us19:20
bdrunghave a look at bug #220137. if the current version fixes this we should use GtkAgg19:21
ubottuLaunchpad bug 220137 in matplotlib "python: matplotlib window does not show graph in interactive mode" [Undecided,Fix released] https://launchpad.net/bugs/22013719:21
LaserJockbdrung: it works with TkAgg here but not GTKAgg19:24
LaserJockI *thought* we were defaulting to TkAgg not Agg19:25
bdrungwith mpl 0.98.3?19:25
LaserJockthat's my bad for not checking that more thoroughly19:25
LaserJock0.98.3-3ubuntu219:25
bdrunggtkAgg was default a long time (if i read the changelog correct)19:25
bdrungok, then use TkAgg19:25
LaserJockit was yes, to give a default GUI backend19:26
LaserJockI thought Debian was using TkAgg by default (I swear that's what the Debian maintainer said)19:26
LaserJockbut it's actually Agg, which is not GUI19:26
LaserJocksince I already change the backend I didn't notice it19:26
LaserJockso I'll move python-tk to the front of the backend |19:27
LaserJockand people will have to just deal with installing some gnome libs for Intrepid19:27
LaserJockI think it's too late to split everything up and get alternatives set up, etc.19:27
LaserJockand we need to talk to Debian about that19:28
bdrungplease grab the debian changes 0.98.3-419:28
LaserJockyeah19:28
LaserJockI should have most of them in, but not all19:29
LaserJockI grabbed a bunch of the changes in -4 from SVN when I uploaded our -3ubuntu119:29
bdrungif you have the package ready i would like to have a look at it19:29
bdrungLaserJock: last comment in bug #278764: upstream should add an "auto" backend, which scans which backend is available19:35
ubottuLaunchpad bug 278764 in matplotlib "[intrepid] No output in ipython because because of wrong default backend" [Undecided,Confirmed] https://launchpad.net/bugs/27876419:35
LaserJockbdrung: I gotta run. I'll whip up a package and send it too you if you like19:41
bdrungLaserJock: yes19:42
bdrungor debdiff19:42
bdrungi found a typo. i will upload a fix to debians svn19:42
LaserJockbdrung: do you have commit access to the svn repo?19:43
bdrungyes19:44
bdrungsince some weeks19:44
LaserJockwow, sweet, was that easy to get?19:44
bdrungyes19:44
LaserJockcool19:44
bdrungmake an account on alioth.debian.org and request to join the python project19:45
LaserJockI gotta run, as I said, I'll send you an email today19:45
LaserJockor attach a debdiff to the bug, or something19:45
bucket529Need someone to change bug 282313 from 'Undecided' to 'Wishlist' - triaged packaging request19:47
ubottuLaunchpad bug 282313 in ubuntu "[needs packaging] Chandler" [Undecided,Confirmed] https://launchpad.net/bugs/28231319:47
bdrungbucket529: done19:55
bucket529bdrung: Thank you.19:55
bdmurraybucket529: Hi!  I've noticed you've assigned a few bugs to the grub package is that right?20:24
azimout_anyone here from bugcontrol?20:52
bdmurrayazimout_: I am20:54
azimout_hi, bill murray20:54
azimout_regarding this bug:20:54
azimout_https://bugs.launchpad.net/ubuntu/+source/linda/+bug/19156520:55
ubottuLaunchpad bug 191565 in linda "false warning when checking CLI packages" [Undecided,New]20:55
azimout_the package is not in the repositories of hardy and intrepid, not in debian, and generally abandoned20:55
azimout_what should we do about this bug?20:55
RAOFWontfix it, saying that it's been removed from current & development versions.20:56
RAOFSince that bug doesn't appear SRU worthy, Intrepid's the only place that bug applies, anyway.20:57
azimout_sorry, i meant brian murray. bill murray is the actor :-$20:57
azimout_that is correct20:59
raboofshould this be a bug against the kernel (which fails to suspend) or acpi-support (which fails to unmount the remote disk before suspending), or both?21:08
raboofsuspend fails when I try to suspend while a cifs partition that went offline is still mounted.21:08
chrisccoulsonraboof - your problem might be with pm-utils21:13
chrisccoulsonor the kernel21:13
raboofchrisccoulson: hmm, i'm not sure I'm using pm-utils21:15
raboofi'm calling /etc/acpi/sleep.sh (from acpi-support)21:15
raboofactually I don't quite see how pm-utils and acpi-support relate :)21:15
chrisccoulsonacpi-support is largely obsolete. with a normal configuration, pm-utils is used. Are you calling sleep.sh with the 'force' option?21:16
raboofchrisccoulson: no, only 'sleep'21:17
raboofso pm-utils replaces acpi-support nowadays?21:17
chrisccoulsonthis is a desktop right? with full gui? is gnome-power-manager running?21:18
chrisccoulsonif gnome-power-manager is running, then /etc/acpi/sleep.sh just exits without doing anything21:18
raboofit's a desktop with X, but gnome-power-manager is not running21:18
chrisccoulsonah, ok, so it will take care of the sleep. normally, the request to sleep from the desktop goes through HAL, which uses pm-utils21:19
raboofi'll see if pm-suspend works for me, too21:21
raboofok, seemed to work nicely. So the recommende way is to use pm-utils - can I safely remove acpi-support from the system?21:26
chrisccoulsonso, pm-suspend works successfully?21:28
chrisccoulsoni would keep acpi-support, as acpid uses it to handle acpi events from the kernel21:28
chrisccoulsononly remove acpi-support if you're not using acpid21:28
raboofi seem to be using that (or at least, it's running :) )21:29
raboofweird to have 2 so similar services in place though21:29
chrisccoulsonwell, there is a bit of overlap but they aren't that similar21:32
raboofwell, they both seem to have a sysvinit-like way of configure what happens on suspend/wakeup - though it looks like pm-utils' is a lot simpler21:33
chrisccoulsonyeah, i think you're right. but acpi-support contains scripts to handle a variety of ACPI events from the kernel (not just suspend/hibernate)21:34
chrisccoulsonwhilst everybodies minds are focused on gnome-power-manager, i have an issue which I think might be g-p-m (or it could be gnome-session). i wouldn't mind somebodies opinion before i open a bug report21:37
chrisccoulsonbasically, i have g-p-m set to ask me what to do when the power button is pressed (which I think is the default). In this case, the upstream logout session-dialog appears when i press the power button. If you don't respond to the dialog, it logs you out automatically after 60 seconds.21:38
chrisccoulsonthe problem is when multiple users are logged on to the system. if i press the power button, then the session dialog appears on everybodies desktops (including inactive users). 60 seconds later, all of the inactive users get logged out!21:38
chrisccoulsonfantastic!21:38
james_wnice21:42
chrisccoulsoncan you recreate that?21:42
james_wlet me try21:43
james_wI believe ted is going to upload a change so that it tries to call over dbus to request the shutdown dialog instead21:43
james_wI don't know if this will have an effect here21:44
james_wchrisccoulson: is the logout dialog global?21:44
james_wi.e. if the active user cancels no-one gets logged out?21:44
chrisccoulsononly the users who don't click get logged out21:44
chrisccoulsonthe logout dialog is just the gnome-session logout dialog21:45
chrisccoulsonteds fix to call the 'shutdown' dialog will make it worse, as that shuts the machine down if you don't respond within 60seconds21:45
chrisccoulsonso even if the active user cancels the dialog, the machine will still shut down without warning in 60s21:46
chrisccoulsonouch!21:46
james_wchrisccoulson: yeah, let's hope that won't happen21:46
chrisccoulsonhopefully. so, g-p-m shouldn't really call the session dialog unless the users session is the active one, should it?21:46
james_wI'm not sure it can know21:49
=== Toadstoo1 is now known as Toadstool
chrisccoulsonit must be able to though, because it knows not to automatically suspend the machine or turn off the monitor when a user is inactive21:51
chrisccoulsonunless that is enforced elsewhere?21:51
james_wwell, something should be able to21:54
james_wit uses consolekit to do that, so we probably need some more consolekit love here21:54
james_wwhen the user presses the power button pop up a dialog to all active sessions21:54
james_wif the user chooses to shutdown then ask to confirm if there are multiple users logged in?21:55
chrisccoulsonthat would be good.21:55
james_wdealing with a timed shutdown isn't going to be good though21:55
chrisccoulsonthe session dialog already pops up a policykit auth prompt if you try to shut down when multiple users are logged in21:55
RAOFdbus has the concept of active console; there are a bunch of policies which allow only the active console to frob the knob.21:57
chrisccoulsonRAOF: is that 'at_console'21:58
RAOFI think it might be.21:58
chrisccoulsondoesn't that open it for anyone logged in at a console (active and inactive)?21:59
james_wRAOF: that's deprecated now, replaced by consolekit22:00
chrisccoulsoni'll open a bug report against g-p-m for this problem anyway. i wonder if it was an issue in hardy?22:03
chrisccoulsonthere was no timeout on the session dialog in hardy though22:03
RAOFjames_w: Oh.  I thought it was talking to consolekit.  Oh, well.22:07
james_wwell, sorry, I think it is implemented via consolekit now22:08
james_wbut I believe 'at_console' itself is deprecated, and hoped to go away22:08
RAOFPoor at_console :)22:09
james_wchrisccoulson: yeah, I can reproduce22:09
chrisccoulsonthanks!22:09
james_wchrisccoulson: would you /join #ubuntu-desktop for a minute?22:11
chrisccoulsonnp22:11
=== txwikinger2 is now known as txwikinger
azimoutquestion: support for feisty ends in less than 5 days (19.10.2008)23:01
azimoutthere's still 131 bugs associated with it23:02
azimouthttps://launchpad.net/ubuntu/feisty/+bugs23:02
azimoutwhat are we going to do about that?23:02
greg-gazimout: the simple answer: it depends23:06
greg-git depends on the bug and if it is reproducible in any later version23:07
azimouti see. thanx23:10
charlie-tcaI got at least one in work, it's a memory leak against feisty but may carry over23:11
barmymattHi. I've provided some information on a bug and have now been asked to start the application from terminal (no problem) and to "attach my trace".  How do I create the trace that I'm being asked for?... or where should I look for more info? Cheers.23:17
greg-gcharlie-tca: yeah, best thing to do is to see if you can reproduce it with Intrepid.23:17
bdmurraybarmymatt: what bug?23:17
charlie-tcaWe're working on it23:18
barmymatt273159 - a problem with mythbuntu control centre23:18
bdmurraybug 27315923:18
ubottuLaunchpad bug 273159 in mythbuntu-control-centre "MCC Segmentation Fault" [Undecided,New] https://launchpad.net/bugs/27315923:18
barmymattthat's the one....23:19
barmymattso... how do I create the trace that I'm being asked for?23:20
bdmurraybarmymatt: it's not clear to me what he is looking for but you can find Mario (superm1) in #ubuntu-devel23:23
barmymattbdmurray: thanks, I'll find him.23:24
MTecknologyso - if a bug is marked Invalid, that means that it doesn't exist which means it could have been fixed or not, but nobody is on the other end to say that it is either way...23:39
MTecknologyis that right?23:39
bdmurrayBugs can also be Invalid because there wasn't enough information to determine the root cause so Fix Released is inappropriate.23:41
affluxhello23:41
affluxback from 5 days without internet, thanks to my isp, reading my bug mail. Just saw this mail from tdflanders: Hi there Pedro, Kjell Braden is not responding, possibly on a Holiday. I am reassigning all compiz.real bugs to you, since I do not know the developers of this package. I will also notify the launchpad team. This is obviously a very important bug. Cheers, Thomas.23:54
affluxThis guy obviously has too much free time.23:55
bdmurrayafflux: that's interesting23:55
affluxhe filled my inbox with ~40 mails, all containing about the same content but in different duplicates of the same bug.23:55

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