=== Melodist is now known as Symphonist === Symphonist is now known as Rhapsodist === Rhapsodist is now known as Melodist [00:26] You know there's something terrible going on when the crash notifying application notifies itself crashed. [00:27] haha yes [00:28] * penguin42 worries that someone has just tagged my bugreport that I added a patch to with the patch tag even though the patch and the description both have BIG hairy warnings in [00:40] Daekdroom: I had a lot of things crashing the other day until I installed python-gobject [01:02] penguin42: which bug? [01:02] yofel: bug 605686 [01:02] Launchpad bug 605686 in linux (Ubuntu) "[maverick] nointremap needed - Blocked an interrupt request due to source-id verificiation failure" [Undecided,New] https://launchpad.net/bugs/605686 [01:03] yofel: I mean it's correct it does have a patch on, and it works - but it does need someone who has a clue about the code they're working with - it sounded a promising idea and seems to work, but heck modern interrupt handling is black magic [01:04] penguin42: oh that, that's brians 'search for patches in ubuntu bugs' script that automatically ads the patch tag and subscibes either the reviewers team or in this case jFo [01:04] ah ok [01:04] it should obvious to anyone who reviews it :-) [01:05] don't worry, JFo will know what to do with it === jenkins is now known as ubuntujenkins === ubuntujenkins is now known as jenkins === jenkins is now known as jenkins_ === jenkins_ is now known as jenkins [01:21] yofel: So I'm trying to remember the story from yesterdays classrom stuff - what happens if a bug is fixed in a version is in debian but we're passed the date when we pull stuff from debian? [01:22] if we're not at final freeze yet you could file a sync/merge request [01:24] yofel - see bug 604087 [01:24] Launchpad bug 604087 in iscsitarget (Ubuntu) "iscsitarget fails with can't create a target 2 0" [Undecided,Confirmed] https://launchpad.net/bugs/604087 [01:26] hmm, and disk read performance is a LOT better on maverick in kvm [01:38] hmm, time for bed === yofel_ is now known as yofel [03:58] Hello all. Probably a common question...In A2 how do i get my install to default to KDE instead of the netbook remix interface? [04:00] you can choose at login [04:00] what WM you want to use [04:00] you need to install KDE though [04:01] * holstein doesnt know what A2 is [04:01] you might want to try #ubuntu-mobile [04:01] i am using KDE. A2=alpha2 [04:01] AH [04:02] yeah, im not sure then [04:02] i had NBR going [04:02] but ubuntu is also doing the same thing on another box. seems like they made the default UI the netbook interface. Not sure why, but it looks like hell on a large display. [04:02] and i installed other WM's [04:02] and just chose at loing [04:02] login* [04:02] rsouthard: then its not at the location i suggested? [04:03] holstein: its not a window manager, its the desktop display, [04:03] i can choose at login but when i choose gnome or kde it still defaults to the netbook ui === nhandler_ is now known as nhandler [04:04] probably some setting somewhere. checked out several files in /etc/ with grep and nothing matched a rather broad regex. [04:04] whatever you kids are calling them these days ;) [04:04] rsouthard: look at "system settings > Workspace" [04:04] i will check it [04:05] not on maverick... yet ;) but I am on kde 4.5 and have a (greyed out option as i dont ahve the netbook interface installed) to change workspace type there [04:06] Sweet. That was it. Thx James:147. Much appreciated. [04:06] didnt know that option was there. must be new to 4.9.x [04:06] * 4.4.8.x [04:08] rsouthard: its new... i mean moved in 4.5 at some point... before RC1 i think.... but it was there in 4.4.x, just in a wieard place [04:09] and not usable unless you had plasma-netbook and plasma-desktop install which was almost noone [05:39] https://blueprints.launchpad.net/ubuntu/+spec/foundations-m-uefi-support [05:39] interesting... my EFI breaks quite horribly if I use grub-efi-amd64 to boot a 32-bit kernel. [05:59] nouveau / Xorg broken ? [06:23] Once the UEFI images come out, I'll reinstall my Maverick partition to 64-bit. [12:29] anyone else sometimes is not able to leftklick? [12:29] * penguin42 hasn't had any leftclick problems === om26er_ is now known as om26er === om26er_ is now known as om26er [15:04] Is it just me or is sound muted/low each time I login? [15:05] yep [15:05] I have no sound on boot [15:05] its been like that's almost all cycl [15:06] BUGabundo_remote: And it's just a matter of unmuting it and turning it up and it seems happy? [15:06] me too [15:06] crimsun_ said he was going to workon that [15:06] but I haven't seen much of him around, so I assume work busy [15:06] penguin42: yes [15:06] penguin42, that's a papercut bug still [15:06] is there a bugid? [15:07] no [15:07] I never file audio bugs, unless told so [15:07] Why not? O.o [15:07] most of them are very HW specific [15:07] BUGabundo_remote: This feels more generic I guess [15:07] and if crimsun_ is not around, its no use to file them [15:07] penguin42: if it does affect a larger number of users, yes [15:08] I hadn't that perception [15:08] the flickering pointer/metacity problem is more of an issue for me [15:08] no idea what that is [15:08] I haven't had any problems with metacity. [15:08] I just have GTK bug [15:08] Compiz, on the other hand, keeps crashing. [15:08] Daekdroom: I get a flickering cursor [15:09] where it makes *everyrthing* painfully SLOWWWWWWWWWWWWWWWWWWWWWW [15:09] had to give up on FreeNX, cause it was unusable [15:09] Daekdroom: And in .xsession-errors Window manager warning: Got a request to focus 0x2401051 (dg@major: ) with a timestamp of 0. This shouldn't happen! [15:09] compiz is fine for me [15:09] BUGabundo_remote: Do you have any droppings in .xsession-errors? [15:09] I have a RAM usage problem that I still have to report. [15:09] xD [15:10] my xsessions seems pretty clear [15:10] As a matter of fact, I'll look for it in Launchpad. [15:11] anyone managed to get f-spot to import ? [15:12] BUGabundo_remote, you should report audio bugs, most regular users don't know how to do it properly [15:12] joaopinto: LOL [15:12] joaopinto: search for "open" audio bugs I have filed [15:13] and you will get why I stop doing it, without being asked by an audio team member [15:13] I once filed a pulseaudio bug that lasted through two versions. [15:13] pff [15:13] It got "magically" fixed. The bug I think it's still open xD [15:13] is that all? [15:13] I have bugs open from 2006 [15:13] lol [15:13] EOL [15:14] heck, I filed a dupe of an OOo bug from 2003 [15:14] So, I found bug 569273 [15:14] Launchpad bug 569273 in indicator-application (Ubuntu Lucid) "memory leak in gnome-power-manager on lucid" [Undecided,Confirmed] https://launchpad.net/bugs/569273 [15:14] too bad I have no clue if my memory leak is caused by power manager. [15:15] Daekdroom: What symptoms are you seeing? [15:15] !valgrind Daekdroom [15:15] !valgrind | Daekdroom [15:15] Sometimes after boot, I'll see RAM usage as high as 700MB without any program open.. [15:15] When it's usually down to 200-ish [15:15] Killing X doesn't solve it. [15:16] I'm trying to spot a pattern so I can file a bug. [15:16] Daekdroom, you can see the memory usage on processes, unless it's a driver leak you can identify the leaking app [15:16] joaopinto, and if I can't see anything using too much memory, should I assume it's the kernel? [15:17] grr, kvm hates me [15:17] ubottu, botsnack [15:17] Yum! Err, I mean, APT! [15:17] hmm I've seen another bug like that, but I'll need to find it [15:17] it has that mouse pointer integration turned on here, just that it doesn't work out of the box, I have to let it catch and release the mouse focus like a dozen times until I can use the mouse in the VM [15:18] yofel: Not seen that, although I've found removing the vmware mouse driver can help [15:18] hm, don't have vmware installed, only virtualbox [15:19] maybe even that's too much [15:19] why dont i have permission to see bug 602091? [15:19] Bug 602091 on http://launchpad.net/bugs/602091 is private [15:19] ah [15:20] Daekdroom: You might want to look at bug 598849 [15:20] Launchpad bug 598849 in dpkg (Ubuntu) "After reboot after update, alot more memory is used" [Undecided,Confirmed] https://launchpad.net/bugs/598849 [15:22] Daekdroom: Although, to me I feel that what's really happening here is that ureadahead or something else is happening on 1st boot and getting a lot of memory cached, not sure if it's actually in use [15:26] zakscott: now you can see bug 602091 [15:26] Launchpad bug 602091 in aptdaemon (Ubuntu) "aptd crashed with AttributeError in _emit_acquire_item()" [Undecided,New] https://launchpad.net/bugs/602091 [15:27] penguin42: here it is in use, but I also suspect ureadahead (in use as: drop_caches doesn't have any effect, memory stays used) [15:27] Daekdroom: ^ [15:28] thanks yofel [15:28] and for me it gets up to like 900MiB on x86 and ~1.3GiB on x86_64 [15:28] yofel: An interesting thing to try would be to write something that actually tried to use lots of memory (not jus tmalloc it) and see if it pushed it out [15:29] I confirmed it but won't post anything on the bug given how I haven't spotted a dpkg pattern. [15:29] well, maybe we should just go ahead and reassign that to ureadahead, but I'm not sure there, every time I get it I have an unusual long fsck run in my bootchart too [15:30] hold on hold on! [15:30] I have to check something.. [15:30] yofel: Yeh it's either kernel or something 1st boot related, I doubt it's actually dpkg [15:30] maybe plymouth hogs memory due to fsck messages? [15:31] no idea [15:31] yofel: You should be able to see that in a ps though [15:31] Going to check if it's related to booting after using NTFS through Windows 7 xD [15:31] i just removed bootchart entirely [15:31] no ntfs here.. [15:32] well, it can't be bootchart, or it would happen every time, not just once [15:35] yofel: Well, if the boot takes longer for some reason.... [15:36] yeah, but 6-800 MiB difference is too much for bootchart.. unless it has some heavy leak [15:38] hmm im getting OoM error on boot also [15:38] "[ 2.844791] Out of memory: kill process 309 (plymouthd) score 39 or a child" [15:39] "[ 2.848100] ureadahead invoked oom-killer: gfp_mask=0xd0, order=0, oom_adj=0" [15:40] "[ 2.872798] Out of memory: kill process 308 (ureadahead) score 37 or a child" [15:45] I wonder if you're all seeing this one: https://bugs.launchpad.net/ubuntu/+source/ureadahead/+bug/501715 [15:45] Launchpad bug 501715 in ureadahead (Ubuntu Maverick) "Kernel trace buffer should be cleared and size restored after profiling" [High,Triaged] [15:45] right before the plymouthd oom it says i have 0 swap space [15:46] Given how I have no clue what ureadahead is.. [15:46] Anyway, I doubt it's dpkg. [15:46] Daekdroom: I think the idea is to record the files normally used during boot, so that on future boots it can bring all those files back in right at the beginning [15:46] It struck me in 4 ubuntu boots in a row. [15:47] 2 were following a windows 7 reboot, 2 were following a ubuntu reboot. [15:47] Daekdroom: But to do that it means on boots after an update it has to do a profile to see what needs to be loaded [15:50] penguin42: seems to be that bug [15:51] should I let pm-utils remove pm-utils-powersave-policy? [15:51] I let it do it. I think pm-utils-powersave-policy needs a rebuild [15:51] * yofel goes patching maverick ureadahead [16:00] * BUGabundo_remote piggy backs on yofel [16:01] ^^ [16:03] I wonder when they're going to give the sound indicator proper Last/Pause/Next buttons :| [16:10] Daekdroom, mine has the new button? [16:11] ZykoticK9, so does mine, but it's awfully done. [16:11] C'mon, it doesn't even have a Play button if you pause the song :| [16:11] It's just a static image. [16:11] Daekdroom, i'm not talking about the > buttons there where there origionally - they are proper buttons now [16:12] ZykoticK9, uh.. are they? [16:12] Let me find a screenshot of how they look in here. [16:13] Daekdroom, i'm in my 10.04 install right now so can't screenshot for you unless i reboot - but I'll see if yours are the same if you can post quickly - i'm leaving in a very little while [16:14] ZykoticK9, I don't even need to post a screenshot of my own, I found it [16:14] ZykoticK9, http://andregondim.eti.br/wp-content/uploads/2010/07/sommavericknovo1.png [16:14] Daekdroom, oh sorry - ya that's what my button look like as well [16:14] wait, you guys have a sound applet? :( [16:14] zakscott, ubuntu default's? [16:14] *ubuntu's default [16:15] never had it, and its not under "add to panel" [16:15] zakscott, it's part of indicator applet [16:15] zakscott, that's the new volume control in 10.10 [16:16] ZykoticK9: New in 10.10? I still have something that looks the same as 10.04 [16:16] ahh gotcha, i actually removed that before.. my mistake [16:16] penguin42, open rhythmbox and play a song [16:16] ah, I don't use rhythmbox [16:16] penguin42, mine looks like the screenshot Daekdroom posted [16:16] Banshee doesn't integrate with it. [16:16] same [16:16] Exaile? Lol, not even with indicator applet [16:16] what music player do you use penguin42? [16:18] ZykoticK9, anyway, I'd find it awesome if it used the same buttons Rhythmbox do, so it could blend well with any theme. [16:19] zakscott: Exaile [16:20] good find, ill have to try it out later [16:20] It looks a lot like Amarok [16:21] * Daekdroom didn't like Exaile [16:21] Daekdroom: It's a bit basic [16:22] Exactly. [16:22] Daekdroom: It does work though and it's quite hackable which is nice [16:23] ok, ureadahead patched, built, installed, all updates installed -> me goes rebooting... [16:27] Last day of https://wiki.ubuntu.com/UbuntuDeveloperWeek starting in 34 minutes in #ubuntu-classroom [16:27] libdns64 became deprecated, I guess that means I should remove it [16:40] yay, ureadahead patch seems to work :D [16:44] if someone else wants to try it: https://edge.launchpad.net/~yofel/+archive/test-ppa/+sourcepub/1238189/+listing-archive-extra [16:44] yofel: They're using a pretty evil trick - IMHO that tracing mechanism was intended for debugging where the person doing it might sanely choose a buffer size and know they are using some ram while they do the debug or performance analysis [17:43] i can't seem to find the maverick daily alternate install cd [17:44] it isnt here http://cdimage.ubuntu.com/daily/current/ [17:44] any ideas? [17:44] There are no images today [17:44] wait some days, to many bugs yet [17:44] We have hope they may be back tomorrow [17:45] how can i just get the latest one that is available? [17:45] Try the daily-live/current [17:45] It is from the 14th [17:45] is there a kde/text equivalent of update-manager-core? Installing half of gnome isnt really what I want to do... [17:46] actually i can just use the alpha [17:46] (want to try out btrfs) [17:47] jussi_: Yeh if you look on the maverick-alpha-2 thing it tells you how to do a server update [18:04] bbl [18:04] anyone else notice nvidia isnt working with 2.6.35-7-generic #12? [18:07] gnomefreak: it was working fine here, what about -8 ? [18:08] problems ? :) [18:08] yofel: not sure yet. updateing kernel now [18:09] s/updateing/updating [18:11] what was the bug exactly gnomefreak ? [18:11] cause I'm having issues also atm [18:11] dupondje: ? nvidia doesnt work (non-free drivers) [18:11] not sure what more info i can give that would explain it anymore [18:12] error ? [18:12] cause nouveau here is also broken ... [18:12] dupondje: no error doesnt run. nouveau works here (im using it now) [18:13] weird :) [18:13] i cant get a TTY so i had no choice. i prefer TTY to low graphics [18:13] https://bugs.launchpad.net/ubuntu/+source/libdrm/+bug/606244 => this is what I have [18:13] Launchpad bug 606244 in libdrm (Ubuntu) "X doesn't find a screen and is not starting due a race condition" [Undecided,New] [18:13] maybe -8 will fix the issues [18:14] running -8 here [18:15] even ps aux seem to lock here now :s [18:15] can't even reboot [18:15] dupondje: you cant get a terminal? [18:15] I can, but reboot doesn't reboot :s its locked [18:16] hint sudo reboot will reboot your system [18:16] ah [18:16] push the little botton on the tower? [18:16] well yea have to :( [18:17] s/botton/button my typing sucks badly today [18:17] * gnomefreak prtty much been gone for 3ish weeks well 2 or 3 [18:18] how was your vacation? [18:20] ok back in track :) [18:20] dupondje: reboot fixed you? [18:21] well fixed your system at least [18:21] rm /var/lib/ureadahead/pack [18:21] sudo reboot [18:21] :) [18:22] ah [18:22] sudo reboot is a command? O.o [18:22] * Daekdroom has always used sudo shutdown -r now [18:22] sudo reboot is a command [18:22] one shuts sytem down the other reboots system ;) [18:23] shutdown -r now reboots. [18:23] * gnomefreak be back need a smoke [18:23] oh [18:23] sudo reboot now works too [18:24] gnomefreak: also have plymouthd crash ? [18:24] Actually, shutdown -P does a full shutdown, shutdown -r does a reboot [18:24] dupondje, do you also have high memory usage problems involving ureadahead [18:24] ? [18:32] dupondje: none of my splashes work and yes plymouth is crashing see bug 598035 for plymouth and bug 551746 for gnome-splash [18:32] Launchpad bug 598035 in plymouth (Ubuntu) "Plymouth does not work" [Undecided,Incomplete] https://launchpad.net/bugs/598035 [18:32] Launchpad bug 551746 in gnome-art (Ubuntu) "Failing to using a gnome splash image" [Undecided,New] https://launchpad.net/bugs/551746 [18:33] LP is slow :( [18:33] gnomefreak: my plymoutd crashes here: https://bugs.launchpad.net/ubuntu/+source/linux/+bug/606393 [18:33] :) [18:33] Launchpad bug 606393 in linux (Ubuntu) "BUG: unable to handle kernel NULL pointer dereference at 00000000000003c0" [Undecided,New] [18:34] no i dont get that [18:35] * gnomefreak needs to add a boot option but i dont recall how to do that. anyone help :/ [18:36] vi /boot/grub/grub.cfg ? [18:39] dupondje: will see if i can [18:40] dupondje: im not seeing a place for it. [18:41] gnomefreak: edit /etc/default/grub ? [18:42] or just add it at boot time? [18:45] yofel: i guess i can add 'plymouth:debug=file:/var/log/plymouth-debug.log' anywhere? problem i dont see anything about plymouth in the file [18:46] depends where you need to add that, if plymouth reads that from the kernel command line then add it to quiet splash [18:47] yofel: see steve's last comment on bug 598035 he doesnt say anything other than add it to boot options [18:47] Launchpad bug 598035 in plymouth (Ubuntu) "Plymouth does not work" [Undecided,Incomplete] https://launchpad.net/bugs/598035 [18:48] I would say add it to the kernel command line [18:49] GRUB_CMDLINE_LINUX_DEFAULT="quiet splash"? [18:49] yes [18:49] Anyone seen any suggestions on the thing about having to remove the grub graphics parameters? [18:50] separate it using , or just add another "....." [18:51] "quiet splash plymouth:debug..." [18:52] yofel: ok thanks i will add it [18:58] be back these updates are going to take forever and its like watching paint dry [19:16] penguin42: what do you mean with "grub graphics parameter"? [19:35] geser: I need to remove the load_video and set gfxpayload=keep [19:37] penguin42: isn't setting gfxpayload=text enough in your case? for that 'echo "GRUB_GFXPAYLOAD_LINUX=text" >> /etc/grub/default' is enough to disable [19:38] geser: Yes it probably is [19:39] if you regenerate the grub config afterwards (update-grub) you should have "set gfxpayload=text" in your grub config [19:39] did you file a bug that gfxpayload=keep doesn't work for you [19:41] geser: No, I thought it was a known - I certainly saw a comment somewhere about it [19:42] penguin42: ATI graphic card? [19:42] yeh, it's probably bug 605614 [19:42] Launchpad bug 605614 in linux (Ubuntu) "[ATI] GPU lockup with gfxpayload=keep" [Undecided,New] https://launchpad.net/bugs/605614 [19:42] yep [19:48] well nvidia works now but still no luck with plymouth log [21:26] Cookies [21:29] So... RGBA Window Decorations... Will GNOME/Ubuntu ever get them? KDE, OS X, and Windows 7 have lovely smooth window edges. It would be nice if Mutter (dubbed Metacity 3) was planning to implement it. I don't think that it's "feature creep". It seems like a fundamental to a modern desktop experience now, especially considering that the default GNOME theme, Clearlooks, and Ubuntu's default, feature rounded corners. [21:33] I've been hearing about the client-side decorations, and how they're a terrible thing. I'm not specifically referring to that at all, however. I'm talking about a proper window decorator that knows that an alpha channel is :-P === luis__lopez is now known as luis_lopez [22:29] damn I love quad core to build kernels :) [22:30] dupondje: Yeh they're nice :-)) Now if only there were about 50 zillion less modules [22:34] * penguin42 gently wonders how to push an xfwm patch I submitted nearly a year ago [22:41] does anyone else see 'You have old & broken userspace please consider updating mesa" [22:42] where do you see that ? [22:42] penguin42, I read in #radeon earlier today that it should be ignored and that it's leftovers from old code. [22:42] dupondje: dmesg [22:42] Daekdroom: Ah ok [22:43] (not that my 3d is happy, but it's the 1st time I've run without edgers for a while - so it might be normal) [22:43] No edgers? Blasphemy :O [22:43] Daekdroom: Yeh, I thought I should try maverick without it first [22:44] I did, but back then it still had 7.8.1 Mesa :| [22:44] dupondje: your 3rd and 4th most likely are idling most of the time since software is always playing catch up with hardware [22:44] I can't live with that. [22:45] Volkodav: well gcc is quite good @ 4 threads :) [22:45] Daekdroom: I've got a couple of bugs though at the moment; the 1st is a constant stream of errors in metacity logs about focus and timestamps; but I've tried doing xfwm --replace and X is still getting battered at 50% cpu for no apparent reason [22:46] dupondje: I normally give it a few more than 4 on a quad core; I've got DEB_BUILD_OPTIONS=parallel=10 in my .bashrc [22:46] mm :) should try [22:46] anyway its build now [22:46] dupondje: half the time it does not know when to kick the other 2 - that's for sure [22:48] dupondje: Just make sure you have plenty of ram and /tmp [22:48] dupondje: multithreading facility should be programmed to be able to use the four cores at a time [22:48] Very few software products can do that currently [22:49] gcc included [22:49] Volkodav: make -j manages very nicely :-)) [22:49] brb :) [22:50] penguin42: Did you trace the cores usage ? [22:51] Volkodav: Just letting top show my % free on each one [22:51] what's the status of btrfs in maverick? [22:51] Volkodav: make -j on a large package can easily use 4 cores most of the time [22:51] virtuald: +1 [22:52] aha, deferred [22:52] penguin42: when that happens and ....say your first 2 CPU's get to 20-30% usage what are the other 2 doing ? [22:53] Volkodav: It's unusual for that to happen; it would be more normal to find all 4 doing 10-15% - I'm talking about getting all 4 to 100% [22:54] (of course it depends how long how you're averaging over) [22:54] I find that hard to believe penguin42 [22:54] Volkodav: Have you got a quad core to try it on? [22:54] yes I do [22:55] I do think that 4 seperate gcc processes should each use a core full time, well, nearly [22:55] Volkodav: OK, grab yourself a big package to build, say a kernel - and set DEB_BUILD_OPTIONS=parallel=8 and build the package [22:55] yofel: Yeh [22:55] and the usage is core 1 - 60-70 5 core 2 -50-60 core 3 -20-15 core 4 - 10-5 [22:56] I have a dual core and I never have more than 2 threads on a multithread program because it makes everything irresponsive xD [22:56] define big penguin42 [22:56] yofel: And that's actually why I run with more than 4, you end up with some processes waiting for disk etc - so give the OS 8 processes to schedule and it might be able to keep 4 cores busy [22:56] Volkodav: Kernel will do [22:56] well, depends on how mch memory gcc needs and on the cache/memory speeds too... [22:56] yofel hi long time no see [22:57] hey [22:57] how's that test on SSD ? TRIM works on mine I found out [22:57] Daekdroom: that provesmy point [22:58] Volkodav: I've got a standard git copy of the kernel source; did make defconfig and then just done make -j 10 at the top - I'm seeing less than 1% idle on all the cores [22:58] Volkodav, as a matter of fact, it doesn't. [22:58] well, it works on mine too, well somehow. Hdparm says that the first few sectors are kept with data, but all other sectors do end up with zeroes [22:58] what version of hdparm yofel ? [22:59] Volkodav: I've got 8GB of RAM that helps mind you, so it's managing a make -j 10 without swapping [22:59] and there we are - built [22:59] And some SSD's need 4-6 hours idling before the TRIM kicks in [22:59] Volkodav: well, whatever we have in maverick, I didn't check this for over 2 weeks now though [22:59] penguin42: most people got rid of swap these days [23:00] Volkodav: Yeh but I'm still one of the weird guys who reckons swap lets you lose the RAM better [23:00] on modern hardware anyway [23:00] what I noticed when compiling though, is that sometimes the CPU will be at 100%, but not at full speed, I guess it doesn't get the data from memory fast enough [23:00] yofel: What do you mean by full speed? [23:01] what is your memory yofel ? [23:01] yofel, powermanagement still clocks it down? [23:01] and I am one of the rare persons that has no problem to fill up 6GiB RAM and needs swap :P [23:01] Daekdroom: could be, didn't worry about it much [23:02] yofel: If it's waiting for RAM it'll show as the CPU being busy, I don't think it will clock down for that - it'll only be clocking down if it'seither too hot or if the kernel doesn't have enough to schedule [23:02] hm, well, it did it for *some* reason, it wasn't too hot though [23:03] yofel: Then it's waiting for {disk, swap, something} [23:03] or you just aren't trying hard enough to keep it busy :-) [23:03] it wasn't disk, that's for sure [23:04] yofel - what is your swapiness set to ? [23:04] 1m48s for a make bzImage (with -j 10) [23:04] (and that's with X eating 50% of one cpu for no good reason) [23:05] cat /proc/sys/vm/swappiness [23:05] Volkodav: on my SSD system 0, others 60 [23:05] I don't really get what effect that has so I don't change it [23:06] same here on SSD and I bit into 1 mb from 4 gig over a week of pretty heavy usage [23:06] it does make a difference on my box at least [23:23] Hey there [23:23] I've just download the alpha 2 live dvd but I got troubles when trying to install it on my computer. The partitioner just allow me to install it on my usb device (used for the boot) and not on my hard disk [23:25] Sephorah: Odd, how is the hard drive connected - normal SATA? [23:26] Yup [23:26] Sephorah: Can you get the output of dmesg somewhere we can see it - e.g. a pastebin and possibly the contents of /proc/partitions? [23:27] When I choose "Default" after the boot, I got a console with initramfs so I have to choose "Install" from the menu but the problem I just told occurs [23:27] I'm currently using Windows [23:27] I used a 16GB usb key and Unetbootin [23:28] ah, I've only ever done it off a CD - if it's dropping you to the initramfs on the install image then there is something wrong with that install image [23:29] hmm actually [23:29] Do you know any other way to boot the maverick-dvd-i386.iso on my usb key under windows? [23:29] yes [23:30] how is your usb formatted ? [23:30] FAT32 [23:31] do the full install on it with linux fs ext3 or 4 [23:31] and it will work [23:32] I'm gonna try the CD image before [23:32] makes sense too [23:32] because it's less than 4GB [23:32] and I heard that 4GB+ images can cause trouble [23:32] true [23:32] after a reboot my gnome is freezing after login screen [23:32]