| === ion_ is now known as ion | ||
| Sarvatt_ | awesome! http://bazaar.launchpad.net/~pcjc2/notify-osd/fix_dropshadow/revision/428 | 01:17 |
|---|---|---|
| Sarvatt_ | notify-osd being screwed up with pixman since july in xorg-edgers fixed by that | 01:18 |
| Sarvatt_ | oh boy, don't tell me notify-osd needs copyright assignment... :) | 01:34 |
| * Sarvatt_ is confused as to how it can require copyright assignment to Canonical when it has source from OpenHanded in it (egg/) | 02:04 | |
| RAOF | New contributions get CA'd, I presume. | 02:05 |
| * ajmitch is surprised that such small patches can require copyright assignment | 02:12 | |
| === Bernardo is now known as Bernardo|Away | ||
| === Bernardo|Away is now known as Bernardo | ||
| === Bernardo is now known as Bernardo|Away | ||
| === Bernardo|Away is now known as Bernardo | ||
| === Bernardo is now known as Bernardo|Away | ||
| === Bernardo|Away is now known as Bernardo | ||
| === Bernardo is now known as Bernardo|Away | ||
| === Bernardo|Away is now known as Bernardo | ||
| === Bernardo is now known as Bernardo|Away | ||
| === Bernardo|Away is now known as Bernardo | ||
| === Bernardo is now known as Bernardo|Away | ||
| === Bernardo|Away is now known as Bernardo | ||
| === Bernardo is now known as Bernardo|Away | ||
| hyperair | Sarvatt: hmm did something break in mesa? i'm suddenly seeing a lot of damage bugs | 09:51 |
| hyperair | Sarvatt: okay, it seems that the latest mesa (on i965) doesn't like compiz++'s mipmaps. | 09:59 |
| hyperair | er okay, maybe not mesa. | 10:00 |
| hyperair | xxvi? | 10:00 |
| hyperair | =\ | 10:00 |
| mdeslaur | Could someone from the X team please give their blessing to LP: #607071, and I'll prepare a lucid SRU? | 13:56 |
| mdeslaur | bug 607071 | 13:56 |
| ubot4` | Launchpad bug 607071 in xorg-server (Ubuntu) (and 1 other project) "X server segfaults when using XTest (xvfb) (affects: 3) (heat: 45)" [Undecided,Confirmed] https://launchpad.net/bugs/607071 | 13:56 |
| mdeslaur | ROAF, bryceh, etc. ^ | 13:57 |
| Sarvatt | mdeslaur: big fat ack from me :) | 14:05 |
| jcristau | Sarvatt: would it make sense to upgrade to server-1.7-branch head as lucid sru? | 14:06 |
| Sarvatt | bryceh, RAOF: on that note, what do you think about upgrading xserver completely in lucid at some point? we could revert the xsfbs change | 14:06 |
| jcristau | or is that never going to be accepted? | 14:06 |
| Sarvatt | ^^ | 14:06 |
| mdeslaur | Sarvatt: great! Could you please add a "+1" to the bug? | 14:07 |
| Sarvatt | jcristau: yeah that's been bugging me, so many fixes we're missing. it's just the xsfbs abi changes making every driver need an update holding it back | 14:10 |
| Sarvatt | mdeslaur: done, thanks for that! | 14:10 |
| mdeslaur | Sarvatt: cool, thanks | 14:11 |
| jcristau | Sarvatt: merging the upstream fixes and leaving the packaging alone should be possible though? | 14:11 |
| * Sarvatt nods | 14:11 | |
| jcristau | ok, great :) | 14:12 |
| Sarvatt | hyperair: can't say I'm surprised, looks like there have been a lot of risky 965 composite speedup commits lately http://cgit.freedesktop.org/xorg/driver/xf86-video-intel/log/ | 14:32 |
| hyperair | Sarvatt: it seems that using mipmaps causes a whole bunch of textures to become black | 14:32 |
| Sarvatt | hyperair: btw did notification popups get fixed for you yesterday with the notify-osd upload? | 14:32 |
| hyperair | or maybe mipmaps broke | 14:33 |
| hyperair | Sarvatt: eh? i lowered my libpixmap or something version | 14:33 |
| hyperair | libpixman | 14:33 |
| Sarvatt | oh ya can upgrade it again now | 14:33 |
| hyperair | okay | 14:33 |
| Sarvatt | uploaded a fixed notify-osd to the PPA | 14:33 |
| hyperair | ah i see | 14:34 |
| hyperair | that's nice | 14:34 |
| hyperair | did you upload it to natty as well? natty's also got that libpixman version right? | 14:34 |
| Sarvatt | nope pixman doesn't build in natty | 14:35 |
| Sarvatt | doesn't pass the test suite at -O2 but does at -O1 :( | 14:35 |
| hyperair | =\ weird | 14:35 |
| hyperair | Sarvatt: what goes wrong in the test suite? | 14:36 |
| Sarvatt | guess I could force it to gcc-4.4 | 14:36 |
| hyperair | ah | 14:36 |
| hyperair | it's fixed. | 14:36 |
| Sarvatt | two of the tests fail, but it doesn't on debian's gcc-4.5 weirdly | 14:36 |
| hyperair | hm how weird | 14:37 |
| Sarvatt | ok now its uploaded for natty | 14:46 |
| greg-g | Sarvatt: so, I installed the PPA, restarted, and hooked up an external monitor. Didn't quite work: https://bugs.launchpad.net/ubuntu/+source/xserver-xorg-video-intel/+bug/682864 | 14:50 |
| ubot4` | Launchpad bug 682864 in xserver-xorg-video-intel (Ubuntu) "[GMA 4500MHD] Display freeze after VGA cable disconnected (affects: 1) (heat: 6)" [Undecided,New] | 14:50 |
| greg-g | Sarvatt: put simply: I had an output problem with the package in the PPA | 14:51 |
| Sarvatt | greg-g: did you upgrade to everything in that PPA or did you just force intel to upgrade? the libdrm packages were required too, just want to be sure | 14:51 |
| Sarvatt | greg-g: hmm, so the freeze went away but you got a new problem? | 14:52 |
| greg-g | Sarvatt: well, not sure about the freeze, didn't leave it installed long enough :) and dangit, no, I didn't upgrade the libdrm :/ | 14:53 |
| greg-g | I'll do that during lunch, I'm a little behind right now at work :) | 14:53 |
| greg-g | Sarvatt: thanks! | 14:53 |
| Sarvatt | greg-g: if you can please upgrade everything in that PPA (add it to your sources), you can install the ppa-purge package and sudo ppa-purge ppa:ubuntu-x-swat/x-updates to completely revert the changes and go back to stock packages after | 14:54 |
| greg-g | oh, nice, didn't know that | 14:54 |
| greg-g | that's handy | 14:54 |
| Sarvatt | it's very possible you do get that new regression with the packages in there, the actual fix for what I believe your bug is is a minor change and that's a backport of the natty versions which have a huge amount of changes. I'll try backporting just the fix now | 14:55 |
| * greg-g obviously didn't read the full description on the PPA page :/ | 14:55 | |
| greg-g | Sarvatt: I went ahead and took the time to upgrade everything in that PPA and restart, same issue | 15:01 |
| greg-g | I took a photo if you want to see it. The screenshot I took this time didn't show the issue. | 15:02 |
| Sarvatt | greg-g: ok thanks, will have a new package for you to try in a bit | 15:02 |
| greg-g | Sarvatt: rock. I'm going to be away from IRC (hopefully) for the next few hours, just fyi. Just ping me and I'll see it. | 15:03 |
| greg-g | Sarvatt: this is the top right hand corner of my external monitor, showing the issue: http://www.flickr.com/photos/grggrssmr/5238275134/ | 15:04 |
| Sarvatt | sheesh, why did I put off such an easy backport thinking it was touching refactored code when it was just source file names changed? :) | 15:07 |
| Sarvatt | hyperair: baah notify-osd fails to build on natty too | 15:07 |
| hyperair | Sarvatt: lol. wait, was it the binutils-gold issue? | 15:08 |
| Sarvatt | yeah needs -lX11 | 15:08 |
| hyperair | ah | 15:09 |
| Sarvatt | err, can't work out where to add it, notify_osd_LIBS isn't it | 15:13 |
| Sarvatt | greg-g: ok updated package building here if ya could purge x-updates and try it when you get a chance - https://launchpad.net/~sarvatt/+archive/gold | 15:19 |
| Sarvatt | ah hah https://wiki.ubuntu.com/RobSavoye/GoldFixes/notify-osd | 15:29 |
| hyperair | Sarvatt: http://people.ubuntu.com/~hyperair/compiz-crash.log <-- could you takea look at this log? | 15:39 |
| hyperair | Sarvatt: smspillaz tells me that XGetWindowProperty should never hang, but it hung there | 15:39 |
| Sarvatt | hyperair: just curious, whats your libxcb version? | 15:41 |
| Sarvatt | didn't you do a screwed up natty upgrade not disabling edgers? | 15:42 |
| hyperair | *** 1.6+git20100703.75ff427d-0ubuntu0sarvatt2 0 | 15:42 |
| Sarvatt | yeeeep | 15:42 |
| hyperair | Sarvatt: i did a screwed up maverick upgrade not disabling edgers. | 15:42 |
| hyperair | and i'm still on maverick. | 15:42 |
| Sarvatt | oh you're on maverick now? | 15:42 |
| hyperair | yep | 15:42 |
| hyperair | and compiz is self-compiled | 15:42 |
| hyperair | there's a nice build_compiz++ script done by soreau | 15:43 |
| hyperair | and it stops beeping annoyingly after two rounds of hacking it | 15:43 |
| Sarvatt | yeah I used that script for a long time on maverick | 15:46 |
| hyperair | ah | 15:46 |
| hyperair | Sarvatt: so anyway, i think everything's using maverick xorg-edgers | 15:47 |
| hyperair | Sarvatt: and this particular hang happens when closing seahorse's password prompt | 15:47 |
| hyperair | i.e. pressing enter with a valid password | 15:48 |
| Sarvatt | hmm | 15:48 |
| Sarvatt | hyperair: have you checked for anything in ~/.xsession-errors when it happens? | 15:51 |
| hyperair | Sarvatt: nay. | 15:51 |
| hyperair | Sarvatt: i could check the next time | 15:51 |
| hyperair | i don't think compiz logs things into .xsession-errors though | 15:52 |
| Sarvatt | was thinking about seahorse spewing an error there | 15:54 |
| Sarvatt | hyperair: can you install libxcb1-dbg and do a echo "0x00007ff411ee22ba" | eu-addr2line -e /usr/lib/debug/usr/lib/libxcb.so.1.1.0 ? | 16:18 |
| hyperair | Sarvatt: lol i've never managed to get addr2line to work before | 16:26 |
| hyperair | always returns ??:0 | 16:26 |
| hyperair | well let's see.. | 16:26 |
| hyperair | yeah it's ??:0 | 16:27 |
| Sarvatt | hyperair: can you downgrade to maverick libX11 and see if it still happens? | 16:31 |
| hyperair | alright | 16:31 |
| Sarvatt | whats the version on your libX11 btw? | 16:33 |
| Sarvatt | launchpad isn't loading the PPA page dangit :) | 16:33 |
| Sarvatt | newer than 0806? | 16:33 |
| * Sarvatt squints at http://cgit.freedesktop.org/xorg/lib/libX11/commit/?id=4b8ff7db39f2fe7ef12968d462aaf3f9054b6c18 | 16:35 | |
| Sarvatt | hmm > A simple test program to reproduce the issue is appended below: just > call XGetWindowProperty on a non-existent window. | 16:35 |
| Sarvatt | 2:1.3.4+git20100720.554da76e-0ubuntu0sarvatt3 | 16:37 |
| Sarvatt | hyperair: this is a long shot but it looks like it might be related, uploading a new libX11 to the PPA | 16:38 |
| Sarvatt | uploaded | 16:44 |
| Sarvatt | hyperair: maverick->natty will be screwed up if you dont disable edgers btw, I stopped updating a lot of things in natty because the PPA was getting too crazy. lucid->maverick has pretty much the same thing in both so its ok | 17:11 |
| hyperair | Sarvatt: hmm what do you mean PPA getting too crazy? | 17:11 |
| Sarvatt | like 200 packages | 17:12 |
| Sarvatt | stuff in experimental is really up to date now and we synced it all so dont need git for a lot of those | 17:13 |
| Sarvatt | darnit, I *know* I've seen an bug report about the twinview regression in https://bugs.launchpad.net/bugs/680811 somewhere but can't find it | 17:16 |
| Sarvatt | its probably working as intended now and wasnt before :) | 17:21 |
| * Sarvatt digs for the documentation on twinview | 17:23 | |
| Sarvatt | looks like its working as intended to me, twinview makes multiple displays appear as a single screen to X so maximize would maximize to both screens.. | 17:29 |
| Sarvatt | "When TwinView is enabled, the NVIDIA driver provides its own version of the Xinerama extension to inform window managers of where the TwinView display devices are. Without that information, window managers would only know about the size of the entire combined desktop. Some window managers ignore this information and treat the screen as one big desktop anyway. It sounds like the version of Compiz you have is one such window manager." hmm | 17:30 |
| jcristau | maximize is a wm op | 17:31 |
| jcristau | if the wm is xinerama-aware, it should do the right thing on twinview, like it does on randr | 17:31 |
| === yofel_ is now known as yofel | ||
| === Bernardo|Away is now known as Bernardo | ||
| === Bernardo is now known as Bernardo|Away | ||
| bryceh | hrm, bug reporters have been busy this weekend - http://www.bryceharrington.org/X/Reports/ubuntu-x-swat/totals-natty-workqueue.svg | 18:32 |
| Sarvatt | I see our friend vesafb in there :) | 18:39 |
| Sarvatt | got rid of a few there | 18:40 |
| bryceh | great | 18:41 |
| Sarvatt | https://bugs.launchpad.net/ubuntu/+source/xserver-xorg-video-intel/+bug/685767 is http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=600405 | 18:42 |
| ubot4 | Launchpad bug 685767 in xserver-xorg-video-intel (Ubuntu) "Dim LCD Screen After Blank Screensaver Activated (affects: 1) (heat: 6)" [Undecided,New] | 18:42 |
| Sarvatt | looks like http://cgit.freedesktop.org/xorg/driver/xf86-video-intel/commit/?id=33c08882c0d551afb28baef643279901dcc65fa9 is the fix | 18:44 |
| jcristau | right | 18:47 |
| Sarvatt | which package should this be, gnome-session, gnome-panel? https://bugs.launchpad.net/ubuntu/+source/xorg-server/+bug/685555 | 18:48 |
| ubot4 | Launchpad bug 685555 in xorg-server (Ubuntu) "window list, show desktop, rubbish bin, workspace selector - all crash on login (affects: 1) (heat: 4)" [Undecided,New] | 18:48 |
| Sarvatt | gnome-panel sounds right | 18:48 |
| Sarvatt | doing the bugs as they come in is great, almost back down to 0 :) | 18:49 |
| bryceh | yeah :-) | 18:49 |
| Sarvatt | https://bugs.launchpad.net/ubuntu/+source/xserver-xorg-video-intel/+bug/680135 bugs like that really suck though | 18:49 |
| ubot4 | Launchpad bug 680135 in xserver-xorg-video-intel (Ubuntu) "natty Intel Corporation Mobile 945GM/GMS, 943/940GML bad performance (affects: 1) (heat: 503)" [Undecided,New] | 18:49 |
| Sarvatt | oops sorry, "bugs" | 18:50 |
| jcristau | reassign to kde | 18:50 |
| bryceh | yeah could be kwin specific | 18:50 |
| Sarvatt | nvidia-graphics-driver bugs condensed to 1 | 18:51 |
| Sarvatt | looks like people installing nvidia-current on a livecd with no persistant storage expecting it to work | 18:53 |
| Sarvatt | darn, dpkgterminallog is empty, gonna have to boot the livecd and see whats happening | 18:54 |
| jcristau | how can that ever work? wouldn't the livecd have loaded nouveau? | 18:54 |
| Sarvatt | these people do have persistent storage enabled, I'm guessing they dont have enough of it to do the actual install (needs like 2gb worth to do last I checked) | 18:55 |
| jcristau | k | 18:55 |
| === Bernardo|Away is now known as Bernardo | ||
| Sarvatt | package nvidia-current 260.19.21-0ubuntu1 failed to install/upgrade: subprocess installed post-installation script returned error exit status 1 -- yeah I get that if I use a 1gb USB stick with the excess set to persistent storage because there isnt enough space, looked at my old bug | 18:57 |
| Sarvatt | bug surprise, fglrx is the same as nvidia :) | 18:58 |
| Sarvatt | bryceh: ok down to 5 bugs | 18:58 |
| Sarvatt | we could move the nvidia/fglrx ones over somewhere else, I'm sure its the not enough persistent storage problem | 18:59 |
| bryceh | Sarvatt, awesome | 18:59 |
| bryceh | Sarvatt, on 685017 do you see any crash related evidence? It's looking bare to me | 18:59 |
| Sarvatt | nope, I just know from personal experience that's a problem, ran into that the first time back in jaunty | 19:00 |
| bryceh | Sarvatt, think it's really an -nvidia bug, or dkms? | 19:01 |
| Sarvatt | maybe it shouldn't be offered if there isn't enough space, but i'm not sure where that could be detected. thats why I was saying maybe we could move them all to another package | 19:02 |
| Sarvatt | maybe wishlist on jockey? | 19:02 |
| Sarvatt | I bet I could punt at least 50 bugs to that bug if we did that | 19:02 |
| bryceh | ah, it's an insufficient space problem? | 19:02 |
| Sarvatt | yeah | 19:02 |
| Sarvatt | it's the weird way aufs works from what I saw, things like take up 2x the amount of space that they really take up | 19:03 |
| bryceh | http://launchpadlibrarian.net/60071510/Df.txt | 19:03 |
| === Bernardo is now known as Bernardo|Away | ||
| bryceh | alright, let's see how we can improve this | 19:04 |
| Sarvatt | i'm not sure how to interpret that, hmm | 19:04 |
| bryceh | this feels like a type of bug that's going to generate lots of bug reports | 19:06 |
| Sarvatt | bryceh: if you look at all nvidia-graphics-drivers or fglrx-installer bugs the majority are the same as this one | 19:06 |
| bryceh | for the workflow of - if unity is notifying people they don't have 3D and so they try to install it via jockey, and this happens | 19:06 |
| bryceh | hmm | 19:07 |
| Sarvatt | https://bugs.launchpad.net/ubuntu/+source/nvidia-graphics-drivers | 19:07 |
| Sarvatt | start from last page | 19:07 |
| bryceh | Sarvatt, is there already one with a detailed write up we can dupe them all to? | 19:07 |
| Sarvatt | subprocess installed post-installation script returned error exit status 1 is almost always installing on a livecd | 19:07 |
| Sarvatt | I dont know, lets see | 19:08 |
| bryceh | ah, yeah all the post-installtion errors | 19:08 |
| Sarvatt | thats how it errors out when out of space | 19:08 |
| Sarvatt | https://bugs.launchpad.net/ubuntu/+source/nvidia-graphics-drivers/+bugs?field.searchtext=subprocess+installed+post-installation+script+returned+error+exit+status+1&orderby=-importance&search=Search&field.status%3Alist=NEW&field.status%3Alist=INCOMPLETE_WITH_RESPONSE&field.status%3Alist=INCOMPLETE_WITHOUT_RESPONSE&field.status%3Alist=CONFIRMED&field.status%3Alist=TRIAGED&field.status%3Alist=INPROGRESS&field.status%3Alist=FIXCOMMITTED&field.assigne | 19:08 |
| Sarvatt | e=&field.bug_reporter=&field.omit_dupes=on&field.has_patch=&field.has_no_package= | 19:08 |
| Sarvatt | heh | 19:08 |
| bryceh | OMFG | 19:09 |
| bryceh | that's a lot | 19:09 |
| bryceh | so... | 19:10 |
| bryceh | jockey could check for available space before installing and issue a warning | 19:10 |
| Sarvatt | 66 nvidia-current, 45 fglrx-installer, 3 nvidia-graphics-drivers-96, 26 nvidia-graphics-drivers-173 | 19:10 |
| bryceh | or it could indicate that installing the driver on the livecd just won't work | 19:10 |
| Sarvatt | something like a warning that a minimum of XGB free space is needed if launched on a livecd would be nice | 19:11 |
| Sarvatt | the loopback stuff on the livecd is odd, 500mb free space advertised isn't enough to install the packages needed for nvidia-current build it and rebuild the initrd and all that | 19:12 |
| Sarvatt | 500MB is more than enough I mean but the way it works doesn't let it work, it duplicates stuff on the persistent storage drive and needs a ton more space | 19:14 |
| bryceh | ok not spotting an obvious dupe of this filed against jockey | 19:14 |
| bryceh | (but didn't look through all the bug reports) | 19:14 |
| bryceh | shall I go ahead and turn one of these into a jockey "warn on out of space" bug? | 19:15 |
| Sarvatt | I know I've duped a ton of bugs about this problem before when I filed one back in jaunty or karmic, wonder where that went | 19:15 |
| Sarvatt | ohh I bet it's under nvidia-graphics-drivers-180 | 19:15 |
| Sarvatt | 15 more under that | 19:16 |
| Sarvatt | bryceh: yeah that sounds like a plan | 19:17 |
| Sarvatt | it's almost like it duplicates any space used between the ram drive and the persistent storage device if you have one so things take 2x the space | 19:18 |
| bryceh | so it installs -nvidia twice? | 19:21 |
| bryceh | or it is reporting double the space actually used? | 19:21 |
| bryceh | or it is reporting double the space actually available? | 19:21 |
| Sarvatt | making a liveusb now to see if I can work it out | 19:24 |
| bryceh | Sarvatt, how does 685017 look so far? | 19:25 |
| Sarvatt | looks good to me | 19:26 |
| Sarvatt | bah can only find 1gb usb sticks lying around | 19:27 |
| bryceh | just use up 800M of space | 19:27 |
| bryceh | put a few mesa tarballs on it ;-) | 19:28 |
| Sarvatt | dont get enough free space to install nvidia on 1gb | 19:29 |
| bryceh | oh | 19:32 |
| bryceh | wow | 19:32 |
| bryceh | seriously? | 19:32 |
| Sarvatt | you only get like 250mb persistent storage on 1gb.. | 19:33 |
| bryceh | huh | 19:33 |
| bryceh | Sarvatt, can you fill in the 'steps to reproduce' section on bug 685017 ? | 19:34 |
| ubot4 | Launchpad bug 685017 in nvidia-graphics-drivers (Ubuntu Natty) (and 1 other project) "Jockey should warn on insufficient disk space in LiveCD environment, else get error "package nvidia-current 260.19.21-0ubuntu1 failed to install/upgrade: subprocess installed post-installation script returned error exit status 1" (affects: 5) (dups: 4) (heat: 36)" [Critical,Triaged] https://launchpad.net/bugs/685017 | 19:34 |
| bryceh | for now, just put in what you suspect are steps to reproduce | 19:34 |
| Sarvatt | bryceh: the livecd takes up like 750mb of it :) | 19:34 |
| bryceh | oh right, I was thinking that was free storage aside from the livecd | 19:35 |
| Sarvatt | got the livecd downloading now, found a 2gb stick | 19:36 |
| Sarvatt | i'll try with different amounts of persistent and see how much is needed | 19:37 |
| === Bernardo|Away is now known as Bernardo | ||
| bryceh | ok I think the bug report is good to go | 19:44 |
| bryceh | Sarvatt, feel free to make any corrections/changes/comments to that bug you think would clarify it better | 19:44 |
| Sarvatt | doing some livecd tests now, will look after | 19:48 |
| === Bernardo is now known as Bernardo|Away | ||
| Sarvatt | bryceh: we've got a new issue | 20:21 |
| Sarvatt | this is a squashfs problem | 20:21 |
| Sarvatt | there's an lzma problem killing initramfs-tools, then the nvidia-current setup dies | 20:23 |
| Sarvatt | http://paste.ubuntu.com/540414/ | 20:23 |
| Sarvatt | so the natty crashes aren't related to the no free space thing | 20:26 |
| Sarvatt | ugh my VT's are all kinds of messed up with 2.6.37-8, plymouth stuff is on VT1, resolution is wacky on all the others | 20:28 |
| Sarvatt | I filed an automatic bug against module-init-tools that it offered, modprobe crashed with SIGBUS | 20:35 |
| jcristau | sounds like kernel screwup | 20:38 |
| Sarvatt | nouveau actually works on my NVc3 now though \o/ | 20:40 |
| Sarvatt | well on the kernel side, fbdev in X | 20:40 |
| bryceh | Sarvatt, weird | 20:43 |
| bryceh | Sarvatt, so you think this is what caused the current bug? or do you still think disk space is a factor? | 20:43 |
| Sarvatt | https://bugs.launchpad.net/ubuntu/+source/module-init-tools/+bug/686145 go figure everything optimized out | 20:44 |
| ubot4 | Launchpad bug 686145 in module-init-tools (Ubuntu) "modprobe crashed with signal 7 (affects: 1) (heat: 10)" [Medium,New] | 20:44 |
| Sarvatt | definitely not disk space related in natty, there's something else | 20:44 |
| Sarvatt | https://bugs.launchpad.net/ubuntu/+source/nvidia-graphics-drivers/+bug/686147 | 20:46 |
| ubot4 | Launchpad bug 686147 in nvidia-graphics-drivers (Ubuntu) "package nvidia-current 260.19.21-0ubuntu1 failed to install/upgrade: subprocess installed post-installation script returned error exit status 1 (affects: 1) (heat: 6)" [Undecided,New] | 20:46 |
| bjsnider | wait a minute. it fails to install why? | 20:48 |
| Sarvatt | Setting up nvidia-current (260.19.21-0ubuntu1) ... | 20:50 |
| Sarvatt | update-initramfs: deferring update (trigger activated) | 20:50 |
| Sarvatt | lzma: Encoder error: -2147467259 | 20:50 |
| Sarvatt | dpkg: error processing nvidia-current (--configure): | 20:50 |
| bjsnider | well, update-initramfs is run by the postinst script | 20:52 |
| Sarvatt | looks like that error means lzma didn't have enough space to create the archive | 20:52 |
| Sarvatt | leading back to squashfs being screwed? :) | 20:52 |
| Sarvatt | [ 276.442208] SQUASHFS error: zlib_inflate error, data probably corrupt | 20:52 |
| Sarvatt | [ 276.442216] SQUASHFS error: squashfs_read_data failed to read block 0x2796fdc1 | 20:52 |
| bjsnider | it's not specifically an nvidia-current problem though right? | 20:52 |
| Sarvatt | nope | 20:52 |
| Sarvatt | wonder if it only happens with persistent storage | 20:56 |
| Sarvatt | I guess in a way it is the same old out of space error, its just being triggered by a bug somewhere else making it think it has no space (the lzma errors) | 21:08 |
| cnd | Sarvatt, I saw all the input module breakage this weekend :) | 21:14 |
| cnd | will you be uploading xorg-server 1.9.99 to xorg-edgers now too? | 21:14 |
| Sarvatt | cnd: yeah just waiting for the last big merges to go through | 21:15 |
| cnd | Sarvatt, cool :) | 21:15 |
| Sarvatt | tomorrow hopefully | 21:15 |
| cnd | awesome | 21:18 |
| cnd | once that's done I can do multitouch development on xserver master | 21:19 |
| cnd | instead of the constant back/forward-porting between 1.9 and master | 21:19 |
| jcristau | you don't build your drivers? | 21:22 |
| bjsnider | Sarvatt, there's now a changelog for the .21 blob. fixed 3 issues | 21:25 |
| bjsnider | or are we up to 26 now | 21:26 |
| cnd | jcristau, was your question posed to me? | 21:27 |
| bjsnider | oh right, it's 26 | 21:27 |
| jcristau | yes | 21:27 |
| cnd | jcristau, the 1.9 server I've been keeping in ppa:utouch-team/xorg-unstable is 1.9+all input changes from master | 21:27 |
| cnd | so it's compatible with the 1.10 drivers | 21:27 |
| Sarvatt | oh boy | 21:28 |
| Sarvatt | -nr changed to -background none? | 21:28 |
| cnd | though I've only been pushing master versions of evdev and synaptics to the ppa | 21:28 |
| * Sarvatt has to patch gdm/kdm/etc now then | 21:28 | |
| Sarvatt | i dont think i'll be using xorg-edgers with xserver 1.10 anytime soon with all of the input stuff, synaptics 1.3.0 is practically unusable here with the crappy acceleration | 21:31 |
| Sarvatt | drop 06_dont_trap_access_to_timer_and_keyboard.diff upstream | 21:32 |
| Sarvatt | drop 16-xaa-fbcomposite-fix-negative-size.diff fails | 21:32 |
| Sarvatt | drop 121_only_switch_vt_when_active.diff fails | 21:32 |
| Sarvatt | drop 189_xserver_1.5.0_bg_none_root.patch upstream | 21:32 |
| Sarvatt | drop 190_cache-xkbcomp_output_for_fast_start_up.patch fails | 21:32 |
| Sarvatt | drop 197_xvfb-randr.patch fails | 21:32 |
| Sarvatt | drop 203_gestures-extension.patch fails | 21:32 |
| Sarvatt | (thats whats needed for master) | 21:32 |
| Sarvatt | 16-xaa-fbcomposite-fix-negative-size.diff: obsolete more like it, http://cgit.freedesktop.org/xorg/xserver/commit/?id=6118346d64e3c2fbe1fe2f041ea773dd2a3c0438 | 21:37 |
| Sarvatt | cnd: how do you want to do all the input stuff in there? | 21:38 |
| Sarvatt | or do you want to do it separate to start? | 21:42 |
| Sarvatt | like i'll just get everything built and ya can use that as a base for the patches and we can move it over once its worked out, or we could do separate git branches for all this somewhere | 21:43 |
| Sarvatt | bjsnider: where do you see the changelog at? | 21:51 |
| Sarvatt | oh sorry missed that ya thought the .21 one was .26 | 21:52 |
| bjsnider | i still can't find a changelog for 26, which makes me a bit nervous | 21:53 |
| Sarvatt | only problem I've had with it was glxgears causing the NVRM error and a 30 second or so freeze, happened one time in the past week | 21:54 |
| bjsnider | what if it's supposed to say "transmitted all data to the government for archiving"? | 21:54 |
| bjsnider | or "added spambot rootkit" | 21:56 |
| jcristau | bjsnider: they're already doing that, they just don't say it in the changelog | 21:57 |
| Sarvatt | yeah I assume that kinda crap from any blob :) | 21:57 |
| bryceh | Sarvatt, for bug #685017 did you think that was a dupe of 686145 ? | 22:16 |
| ubot4 | Launchpad bug 685017 in nvidia-graphics-drivers (Ubuntu Natty) (and 3 other projects) "Jockey should warn on insufficient disk space in LiveCD environment, else get error "package nvidia-current 260.19.21-0ubuntu1 failed to install/upgrade: subprocess installed post-installation script returned error exit status 1" (affects: 5) (dups: 4) (heat: 36)" [Critical,Triaged] https://launchpad.net/bugs/685017 | 22:17 |
| Sarvatt | bryceh: nah, 685017 is still valid, what's happening is the postinst scripts cant run because of a filesystem error somewhere making it think it has no space | 22:20 |
| bryceh | Sarvatt, would the make.log have relevant details? | 22:20 |
| bryceh | or would it be in a dpkg log of some sort? | 22:21 |
| Sarvatt | I described it as best as I could in https://bugs.launchpad.net/bugs/686147 still unsure where to find the actual bug on the natty livecd | 22:22 |
| ubot4 | Launchpad bug 686147 in nvidia-graphics-drivers (Ubuntu) "package nvidia-current 260.19.21-0ubuntu1 failed to install/upgrade: subprocess installed post-installation script returned error exit status 1 (affects: 1) (heat: 6)" [Undecided,New] | 22:22 |
| Sarvatt | it wasnt in any of my logs | 22:22 |
| Sarvatt | i got the info manually installing nvidia-current | 22:22 |
| Sarvatt | lzma: Encoder error: -2147467259 means no space to unpack the archive | 22:23 |
| bryceh | Sarvatt, ok, where is that seen? | 22:23 |
| Sarvatt | Setting up initramfs-tools (0.98.1ubuntu7) ... | 22:24 |
| Sarvatt | update-initramfs: deferring update (trigger activated) | 22:24 |
| Sarvatt | lzma: Encoder error: -2147467259 | 22:24 |
| Sarvatt | dpkg: error processing initramfs-tools (--configure): | 22:24 |
| Sarvatt | subprocess installed post-installation script returned error exit status 1 | 22:24 |
| Sarvatt | after sudo apt-get install nvidia-current from a terminal | 22:24 |
| Sarvatt | i can't find that error in any of the logs | 22:24 |
| bryceh | ok | 22:24 |
| Sarvatt | on my own system where I got the problem, looked through /var/log/ | 22:24 |
| Sarvatt | I mean there is still the problem where it errors out because of insufficient space, but there's another bug most likely in the kernel on the natty livecd causing it to trigger every time | 22:29 |
| bryceh | ew | 22:30 |
| bryceh | ok thanks, updating bug report | 22:30 |
| RAOF | That's pretty hostile error reporting! | 22:30 |
| Azelphur | bryceh did you get my backtrace earlier? | 22:31 |
| bryceh | heya RAOF | 22:31 |
| Azelphur | bryceh http://dl.dropbox.com/u/3832397/misc/gdb-Xorg.txt.tar.gz same bug as yesterday with X crashing in arecord, this time I have debug symbols, I was also on the -proposed repo | 22:31 |
| bryceh | Azelphur, not that I recall | 22:31 |
| bryceh | Azelphur, ah, ok busy with this other bug at the moment, I'll look in a bit | 22:32 |
| Sarvatt | bryceh: https://bugs.launchpad.net/ubuntu/+source/initramfs-tools/+bugs?start=150 | 22:34 |
| Sarvatt | those are the same bug, both nvidia-current and initramfs-tools fail the same way on the natty livecd here | 22:34 |
| bryceh | Sarvatt, https://bugs.launchpad.net/ubuntu/+source/initramfs-tools/+bug/562312/comments/7 | 22:35 |
| ubot4 | Launchpad bug 562312 in initramfs-tools (Ubuntu) "package initramfs-tools 0.92bubuntu71 [modified: usr/sbin/update-initramfs] failed to install/upgrade: lzma: Encoder error: -2147467259 (affects: 7) (dups: 5) (heat: 56)" [Undecided,Triaged] | 22:35 |
| Sarvatt | \o/ | 22:35 |
| Sarvatt | yay, progress | 22:36 |
| bryceh | alright, enough on that bug for now... left the OR with a couple tasks... will check back later | 22:36 |
| Sarvatt | what a maze, so usb-creator-gtk needs a bug to make it leave free space now | 22:38 |
| * Sarvatt gives up for now and has dinner too | 22:39 | |
| bryceh | Sarvatt, ?? | 22:39 |
| bryceh | Sarvatt, reference? | 22:39 |
| Sarvatt | you just linked it! | 22:39 |
| bryceh | oh | 22:39 |
| bryceh | I thought you meant, "We are required to file bugs to request additional disk space be provided" | 22:39 |
| Sarvatt | the disk creator is using all of the free space for the persistent storage if you slide it all the way and it needs space on the actual fat32 filesystem to store the kernel to do upgrades | 22:39 |
| bryceh | ah | 22:41 |
| Sarvatt | i know I always just max out the slider every time | 22:42 |
| bryceh | Azelphur, btw it'd be helpful in the future if you post logs without compressing them... then I have to download and manually unzip and stuff | 22:43 |
| bryceh | Azelphur, also, what is your bug report #? | 22:43 |
| Azelphur | bryceh don't have one yet, I wasn't sure it was a bug | 22:44 |
| Azelphur | bryceh I compressed it because it's big, it's like 20MB | 22:44 |
| Azelphur | I guess I should get a report up for it | 22:45 |
| bryceh | #12 is the same call as #7, so #7 -> #12 is an infinite loop | 22:46 |
| bryceh | Azelphur, file it with ubuntu-bug. If possible, repro the issue, reboot, and then file | 22:46 |
| bryceh | that way it'll include some stuff like Xorg.0.log.log from the failure | 22:47 |
| bryceh | not a big deal if that's inconvenient, I think this trace has the info needed to troubleshoot | 22:47 |
| Azelphur | bryceh will do, fyi there's nothing at all in Xorg.*, I figure it crashes before the writing to log happens | 22:47 |
| bryceh | I think you're right | 22:47 |
| bryceh | my guess is that it gets stuck in a loop, fills some sort of crash handler buffer and then the crash handler crashes, or something analogous to that | 22:48 |
| bryceh | Azelphur, also when you file the bug report, be verbose in explaining steps to reproduce the problem, please | 22:49 |
| Azelphur | it happens completely at random unfortunately | 22:49 |
| Azelphur | I've seen it happen on multiple occasions with nobody anywhere near any input devices | 22:49 |
| bryceh | Azelphur, then explain in detail how you set up the machine | 22:50 |
| Azelphur | will do :) | 22:50 |
| bryceh | i.e. you need to explain why the record extension is turned on, and what is requiring it | 22:50 |
| Azelphur | that'd be x11vnc I guess | 22:50 |
| bryceh | Azelphur, think of it like a high school chemistry experiment, where randomly the solution turns blue... explain how you set things up, what things you poured in, how hot you made the bunsen burner, etc. ;-) | 22:51 |
| Azelphur | yup :) | 22:51 |
| bryceh | btw, never use the word "random" in a bug report, it just means you didn't think about things enough ;-) | 22:52 |
| Azelphur | good tip :) | 22:52 |
| RAOF | It is possible to have bugs with random (or, rather hidden) triggers, but they're few and far between. :) | 22:53 |
| Azelphur | although this happens all by itself, as I say it'll happen when nobody has had any interaction with the machine for the past 30 minutes with nobody anywhere near any input devices | 22:53 |
| bryceh | hmm, poking through the source code it's not totally obvious what's going on here | 22:55 |
| bryceh | but I still think the record extension is listening to events that it's generating from its listens | 22:55 |
| bryceh | Azelphur, that can be a strong clue too | 22:56 |
| bryceh | for example... | 22:56 |
| bryceh | if it *always* happens when no one is near it, then that suggests maybe looking at screensaver or power saving relationships | 22:57 |
| Azelphur | nope, it happens when people are using it too, and there's no screensaver :) | 22:57 |
| Azelphur | (this is a Ubuntu machine powering the TV in the front room, MythTV \o/) | 22:57 |
| bryceh | cron jobs? | 22:57 |
| bryceh | mythtv starting recording something, or starting process to remove commercials or something? | 22:58 |
| Azelphur | I have a couple, but none of them do anything with X. But I guess it could somehow have effect | 22:58 |
| Azelphur | maybe | 22:58 |
| bryceh | running low on memory? | 22:58 |
| Azelphur | nah always got plenty of memory, 2GB total 1.5GB in cache | 22:59 |
| bryceh | like raof said, there *are* some bugs which have truly random triggers, but the fact that this is occurring within a callback tree strongly suggests to me that there is *some* process that triggers it | 22:59 |
| Azelphur | yea | 22:59 |
| Azelphur | I'll start logging times and see if I can come up with anything. | 23:00 |
| bryceh | Azelphur, great, yeah look for any patterns at all for when it occurs or does not occur, even wacky things | 23:00 |
| bryceh | once there was a bug (not against X) which broke the printer driver, which only occurred on Thursdays | 23:02 |
| bryceh | so every week bunches of people would report "me too!" then the next day on Friday everyone would chime in "Oh, it's fixed now"... only until the next week | 23:02 |
| RAOF | That's totally awesome :) | 23:03 |
| Azelphur | haha | 23:04 |
| Azelphur | bryceh not as weird as my weirdest bug report, My best one was "Using volume control on keyboard changes gnome theme" | 23:04 |
| bryceh | lol | 23:04 |
| Azelphur | and it actually got fixed too, it was a bug. | 23:04 |
| Azelphur | (not on the computer we're debugging) I use separate X screens, and it was causing the OSD for the volume control to flip out, so gnome-settings-daemon crashed, which caused the gnome theme to change \o/ | 23:05 |
| RAOF | Yeah. That was my guess as to the cause :) | 23:05 |
| bryceh | heh | 23:05 |
| cnd | bryceh: I hope you're feeling better than me | 23:28 |
| cnd | I think I've come down with a cold... | 23:28 |
| bryceh | cnd, uh oh... I'm fine myself | 23:29 |
| * bryceh breaks out the echinecea | 23:29 | |
| cnd | heh | 23:29 |
| cnd | feels like just a head cold, for what it's worth :) | 23:30 |
| bryceh | cnd, bummer :-/ | 23:32 |
| bryceh | no biggie if I get it, but hopefully ogasawara escapes it | 23:32 |
| cnd | yeah | 23:33 |
| bryceh | cnd, air travel is hazardous to your health | 23:46 |
| cnd | heh, the hazards of working at Canonical... | 23:47 |
| cnd | even though this trip wasn't for Canonical :) | 23:47 |
| Sarvatt | dang, sprint is only a month away already | 23:47 |
| Sarvatt | we get to get felt up at the airport 6 times a year :) | 23:48 |
| cnd | bryceh: angela liked the program today | 23:48 |
| cnd | they give you $50 for every 30 days you bike to work :) | 23:48 |
| cnd | it's at the top of her list so far | 23:49 |
| cnd | bryceh, RAOF: btw, I've got a patch against maverick xserver and evdev cooking to fix crashes during gestures | 23:51 |
| cnd | so that's your heads up :) | 23:51 |
| RAOF | Ta :) | 23:51 |
| cnd | RAOF, oh yeah, when I'm on the west coast, our time lines up much better :) | 23:52 |
| RAOF | c'mon xserver. Build! Faster! MUSH! | 23:52 |
| RAOF | :) | 23:53 |
| bryceh | cnd, :-) | 23:58 |
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!
