Re: [gentoo-user] May GMN Tips and Tricks
Alan McKinnon writes: This month has been a treasure trove of such things here on gentoo-user. Oh my, an I have some 6500 unread e-mails... that's hard to catch up. But I'll have a look into this month then :) Wonko
[gentoo-user] Only 4 of 8 GB usable
Hi there! So I installed another 4 GiB RAM into a Gentoo amd64 system that had 4 GiB already. But it still sees only 4 GiB, not 8 GiB: leela ~ # uname -a Linux leela 3.6.11-gentoo #3 SMP Mon Feb 4 15:37:48 CET 2013 x86_64 AMD A6-3500 APU with Radeon(tm) HD Graphics AuthenticAMD GNU/Linux leela ~ # free -m total used free sharedbuffers cached Mem: 3688 3269419 0108 1050 -/+ buffers/cache: 2110 1577 Swap: 2047 54 1993 Huh? Any idea why this is? The BIOS shows the full 8GiB, and lshw finds it. dmidecode shows that 8G should work: leela ~ # dmidecode -t 16 # dmidecode 2.11 SMBIOS 2.7 present. Handle 0x0008, DMI type 16, 23 bytes Physical Memory Array Location: System Board Or Motherboard Use: System Memory Error Correction Type: None Maximum Capacity: 8 GB Error Information Handle: Not Provided Number Of Devices: 2 In case this helps, I uploaded the outputs of dmesg [1], lshw -c memory [2] and full dmidecode output [3]. The dmesg output is somewhat weird though, it has several 'vmalloc: allocation failure: 0 bytes' entries. I suspected those were causing the problem, but I found that I needed to activate CONFIG_KALLSYMS=y, and they are gone. But still only 4 GiB RAM. The system is using an old kernel right now, so I cannot get the current dmesg, sorry for this. Probably related: Since I inserted this 2nd RAM module, wakeup from hibernate-ram does no longer work. Does this ring any bells? I'm out of ideas. Except than pulling out the 4 GB, or trying another mainboard. [1] http://www.wonkology.org/tmp/lshw.txt [2] http://www.wonkology.org/tmp/dmesg.txt [3] http://www.wonkology.org/tmp/dmidecode.txt Wonko
Re: [gentoo-user] Qt blocking @world update
Alan McKinnon writes: On 05/11/2013 15:37, Alex Schuster wrote: [kde-misc/fsrunner pulls in QT 4.8.4] This does not make any sense, does it? Actually, it does make sense, in a weird kind of way kid3 and fsrunner are not part of KDE proper (i.e. they are not shipped in the huge KDE tarballs). So they may be inconsistent with the main release due to no QA checks beyond what the dev does. And I doubt the gentoo KDE team checks such packages before updating ebuilds. But what exactly is it that pulls in the older Qt? I would use this approach: Remove from world every KDE package that is not in kde-base (quickpkg first to make restores easier), then update world and do a depclean. Chances are very good it will complete cleanly. Well, I was at this point already, after excluding fsrunner and kid3. Then emerge all those KDE packages back in using the -t option to emerge and see what is causing issues. emerge fsrunner would happily just install fsrunner, but emerge -Dpu fsrunner again wants to downgrade Qt to 4.8.4. The same goes for kid3. No big deal, I don't really need those. I think the odds are very good you will find an out-of-sync package that directly DEPENDS on some old version of Qt (or something equally silly). That package might even already be in the emerge output, but buried in the voluminous output portage gives these days But emerge -uD @world no longer complains. Of course I have other problems now... but I will start a new thread for that. Alex
Re: [gentoo-user] Qt blocking @world update
Alan McKinnon writes: Excuse the top-posting; if I try inter-post between all those blockers you'll never find what I reply :-) I would, but for everyone else it's a mess. E-mail with line breaks is not suited for this kind of output. First I recommend to sync your tree again, just in case you got yours between two Qt commits and things are not consistent anymore. Ah, I have this problem for weeks now, but did not care enough. And did not have the time for this. You seem to have at least two things happening: python-exec qt To deal with the first, try remove python-exec and re-merge it (but quickpkg a backup first) quickpkg python-exec emerge -avC python-exec emerge -av1 python-exec This is untested so I don't know if it will bork. If it does, you have a quickpkg that you can untar and get things back. Thanks, a simple upgrade just worked, as I wrote in the reply to Walt. Onto Qt: I've had similar things over the years and it always made little sense. Eventually I removed all references to Qt from world, sets in use and USE then let portage figure out what to do. Rationale: Qt is a basic toolkit that stuff uses, so stuff should decide what it needs and not me. I want the stuff and if that requires Qt then just let portage give me what is required. This will show all references to Qt to consider: grep -ir /qt /etc/portage/ /var/lib/portage/world* /etc/portage/sets.portage/qt-split:x11-libs/qt-assistant /etc/portage/sets.portage/qt-split:x11-libs/qt-core /etc/portage/sets.portage/qt-split:x11-libs/qt-dbus /etc/portage/sets.portage/qt-split:x11-libs/qt-demo /etc/portage/sets.portage/qt-split:x11-libs/qt-gui /etc/portage/sets.portage/qt-split:x11-libs/qt-opengl /etc/portage/sets.portage/qt-split:x11-libs/qt-phonon /etc/portage/sets.portage/qt-split:x11-libs/qt-qt3support /etc/portage/sets.portage/qt-split:x11-libs/qt-script /etc/portage/sets.portage/qt-split:x11-libs/qt-sql /etc/portage/sets.portage/qt-split:x11-libs/qt-svg /etc/portage/sets.portage/qt-split:x11-libs/qt-test /etc/portage/sets.portage/qt-split:x11-libs/qt-webkit /etc/portage/sets.portage/qt-split:x11-libs/qt-xmlpatterns /etc/portage/package.use/misc:dev-qt/qt-creator doc examples perforce qtscript /etc/portage/package.use/misc:dev-qt/qt-meta:3 doc mysql /etc/portage/package.use/misc:dev-qt/qthelp doc /etc/portage/package.use/misc:dev-qt/qtsql mysql /etc/portage/package.use/misc:=dev-qt/qtgui-4.8.4-r1 gtkstyle /etc/portage/package.use/misc:=dev-qt/qtsql-4.8.4mysql /etc/portage/package.use/misc:=dev-qt/qtgui-4.8.4-r1 gtkstyle /etc/portage/package.use/misc:=dev-qt/qtgui-4.8.4-r1 gtkstyle /etc/portage/package.use/misc:=dev-qt/qtsql-4.8.4mysql /etc/portage/package.use/misc:=dev-qt/qtgui-4.8.4-r1 gtkstyle /var/lib/portage/world:dev-qt/qt-meta:3 In your case, I see portage wants to downgrade several Qt packages due to fsrunner, but there's nothing in that ebuild or the kde4-base eclass it inherits, which leads me to believe you might have a config setting somewhere that wants to exclude latest Qt somehow. I commented them all out, I still get the error about multiple Qt versions. Portage and the tree by itself isn't doing it, here's my output: $ emerge -pvt fsrunner These are the packages that would be merged, in reverse order: Calculating dependencies... done! [ebuild N ] kde-misc/fsrunner-0.7.5:4 USE=(-aqua) -debug 18 kB Total: 1 package (1 new), Size of downloads: 18 kB Same here, except that it's emerged already. OK, I have no clue how to further debug this. But what I did is: for (( i=500; i 0; i-=20 )) do emerge -DautvNj $( head -n $i /var/lib/portage/world ) done This failed until $i was 260, so I tried a little more, and removed media-sound/kid3 from @world. Along with fsrunner of course. Now, it's building. This does not make any sense, does it? Wonko
Re: [gentoo-user] Qt blocking @world update
Walter Dnes writes: On Sat, Nov 02, 2013 at 11:02:27PM +0100, Alex Schuster wrote * One or more packages are either masked or have missing dependencies: * * dev-lang/python-exec:=[python_targets_python2_7(-),python_targets_python3_2(-),-python_single_target_python2_6(-),-python_single_target_python2_7(-),-python_single_target_python3_2(-),-python_single_target_python3_3(-)] pulled in by: * (dev-java/java-config-2.2.0::gentoo, installed) [snipping LOTS of similar output again] Let's start at the top, as the python errors may cascade and cause other errors. From that output, it looks like you do not have any version of python_single_targetX_Y enabled. That could be your problem right there. What python settings do you have in make.conf? I have 3 lines. PYTHON_SINGLE_TARGET=python2_7 PYTHON_TARGETS=python2_7 USE_PYTHON=2.7 I only have PYTHON_TARGETS=python2_7, but commented. Don't remember when or why I did this. Bit it is set per default, emerge --info gives: PYTHON_SINGLE_TARGET=python2_7 PYTHON_TARGETS=python2_7 python3_2 USE_PYTHON is unset AFAIK PYTHON_TARGETS defines for which versions of python you build packages. And in case a package only allows to be built for a single version of portage, this is set with PYTHON_SINGLE_TARGET. But what does USE_PYTHON do? Is it documented anywhere? I don't find it in the man pages of make.conf, portage or emerge. Also what do you get when you type eselect python list? In my case I get [i660][waltdnes][~] eselect python list Available Python interpreters: [1] python2.7 * [2] python3.2 weird ~ # eselect python list Available Python interpreters: [1] python2.6 [2] python2.7 * [3] python3.2 [4] python3.3 So, all looks fine to me. USE_PYTHON=2.7 emerge --resume gives the same error. But then I upgraded python-exec. This went without problems, and now the python errors are gone, and I only get this: * One or more packages are either masked or have missing dependencies: * * =dev-libs/icu-3.8.1-r1:0/51.1= pulled in by: * (net-libs/webkit-gtk-1.10.2-r300::gentoo, installed) * * ~app-emulation/emul-linux-x86-medialibs-20130224 pulled in by: * (app-emulation/emul-linux-x86-qtlibs-20130224::gentoo, installed) Hmm. I have icu-51.2 installed, and emerge -u icu would happily upgrade to 51.2-r1. But with -deep I get app-emulation/emul-linux-x86 blockers (xlibs, baselibs, opengl). Well, I'd say let's skip this. The failed emerge -e was days ago, the portage tree has changed, my interest is not so much continuing this failed emerge, but being able to update @world again. Thanks for responding Walt! Alex
[gentoo-user] Qt blocking @world update
Hi there! My @world update did not go well. It was much worse some while ago, so I just did an emerge -e @world, after manually removing stuff from /var/lib/portage/world until I got no complaints any more. I had to remove kde-misc/publictransport and kde-misc/plasma-emergelog for that. After most was done, it stopped after one package failed to build, and was unable to resume due to blockers. emerge --resume gives this: weird portage # emerge -aj --resume These are the packages that would be merged, in order: Calculating dependencies... done! * Invalid resume list: * * (u'ebuild', u'/', u'sys-apps/lshw-02.17b', u'merge') * (u'ebuild', u'/', u'net-print/foomatic-filters-4.0.17-r1', u'merge') [snipping some dozen lines] * (u'ebuild', u'/', u'media-video/kmplayer-0.11.3d-r1', u'merge') * (u'ebuild', u'/', u'media-libs/phonon-vlc-0.6.2', u'merge') * * One or more packages are either masked or have missing dependencies: * * dev-lang/python-exec:=[python_targets_python2_7(-),python_targets_python3_2(-),-python_single_target_python2_6(-),-python_single_target_python2_7(-),-python_single_target_python3_2(-),-python_single_target_python3_3(-)] pulled in by: * (dev-java/java-config-2.2.0::gentoo, installed) [snipping LOTS of similar output again] * * dev-lang/python-exec:=[python_targets_python2_7(-),python_targets_python3_2(-),-python_single_target_python2_6(-),-python_single_target_python2_7(-),-python_single_target_python3_2(-),-python_single_target_python3_3(-)] pulled in by: * (dev-libs/libpeas-1.8.1::gentoo, installed) * * dev-lang/python-exec:=[python_targets_python2_7(-),python_targets_python3_2(-),-python_single_target_python2_6(-),-python_single_target_python2_7(-),-python_single_target_python3_2(-),-python_single_target_python3_3(-)] pulled in by: * (dev-python/pygobject-3.8.3::gentoo, installed) * * =dev-libs/icu-3.8.1-r1:0/51.1= pulled in by: * (net-libs/webkit-gtk-1.10.2-r300::gentoo, installed) * * ~app-emulation/emul-linux-x86-medialibs-20130224 pulled in by: * (app-emulation/emul-linux-x86-qtlibs-20130224::gentoo, installed) * * The resume list contains packages that are either masked or have * unsatisfied dependencies. Please restart/continue the operation * manually, or use --skipfirst to skip the first package in the list and * any other packages that may be masked or have missing dependencies. Wow, I don't even... anyway, I thought emerge -DuN @world might work now, but it does not, again due to Qt problems. And those I do not understand: Total: 178 packages (148 upgrades, 9 downgrades, 12 new, 1 in new slot, 8 reinstalls, 4 uninstalls), Size of downloads: 349,914 kB Fetch Restriction: 1 package Conflict: 18 blocks !!! Multiple package instances within a single package slot have been pulled !!! into the dependency graph, resulting in a slot conflict: dev-qt/qtgui:4 (dev-qt/qtgui-4.8.5-r1::gentoo, installed) pulled in by =dev-qt/qtgui-4.8.5:4[accessibility,dbus(+)] required by (kde-base/libkworkspace-4.11.2::gentoo, installed) ~dev-qt/qtgui-4.8.5[aqua=,debug=,egl=,qt3support=] required by (dev-qt/qtopengl-4.8.5::gentoo, installed) (and 283 more with the same problems) (dev-qt/qtgui-4.8.4-r1::gentoo, ebuild scheduled for merge) pulled in by =dev-qt/qtgui-4.7.4:4[accessibility,dbus] required by (kde-misc/fsrunner-0.7.5::kde, installed) =dev-qt/qtgui-4.7.4:4[accessibility,dbus] required by (media-sound/kid3-2.2.1::kde, installed) ~dev-qt/qtgui-4.8.4[accessibility=,aqua=,debug=,qt3support] required by (dev-qt/qt3support-4.8.4::gentoo, ebuild scheduled for merge) (and 1 more with the same problems) dev-qt/qtcore:4 (dev-qt/qtcore-4.8.4-r5::gentoo, ebuild scheduled for merge) pulled in by ~dev-qt/qtcore-4.8.4[aqua=,debug=] required by (dev-qt/qtscript-4.8.4::gentoo, ebuild scheduled for merge) (and 4 more with the same problem) (dev-qt/qtcore-4.8.5::gentoo, installed) pulled in by ~dev-qt/qtcore-4.8.5[aqua=,debug=,qt3support=] required by (dev-qt/qtopengl-4.8.5::gentoo, installed) =dev-qt/qtcore-4.8.5:4 required by (app-office/akonadi-server-1.10.3::gentoo, installed) (and 12 more with the same problems) dev-qt/qtscript:4 (dev-qt/qtscript-4.8.5::gentoo, installed) pulled in by ~dev-qt/qtscript-4.8.5[aqua=,debug=] required by (dev-qt/designer-4.8.5::gentoo, installed) (and 2 more with the same problem) (dev-qt/qtscript-4.8.4::gentoo, ebuild scheduled for merge) pulled in by ~dev-qt/qtscript-4.8.4[aqua=,debug=] required by (dev-qt/qtgui-4.8.4-r1::gentoo, ebuild scheduled for merge) dev-qt/qtdbus:4 (dev-qt/qtdbus-4.8.4::gentoo, ebuild scheduled for merge) pulled in by ~dev-qt/qtdbus-4.8.4[aqua=,debug=] required by (dev-qt/qtgui-4.8.4-r1::gentoo, ebuild scheduled for merge) (dev-qt/qtdbus-4.8.5::gentoo, installed) pulled in by =dev-qt/qtdbus-4.8.5:4 required by (app-office/akonadi-server-1.10.3::gentoo, installed)
Re: [gentoo-user] Multiple package instances within a single package slot
Helmut Jarausch writes: You are not alone, Alex, please see https://bugs.gentoo.org/show_bug.cgi?id=486438 Thanks! Alan McKinnon writes: On 05/10/2013 20:30, Alex Schuster wrote: Neil Bothwick writes: And whatever package I try to update, emerge wants to remerge libreoffice. Happens with all the packages I tried, which are adobe-flash, python, zsh, xterm. If portage believes LO needs to be rebuilt, it will try to do so whichever packages you are emerging, just let it happen. I already did that, twice, when updating portage and chromium. emerge still wants to remerge libreoffice whenever I upgrade a package. With a suitable --jobs setting, portage will emerge all the other packages before LO is done. No, at least with portage, the installation happened after libreoffice was installed. weird ~ # emerge -1uaj portage These are the packages that would be merged, in order: Calculating dependencies... done! [ebuild r U ] app-office/libreoffice-4.1.2.2-r1 [4.1.2.2] [ebuild U ] sys-apps/portage-2.2.7 [2.2.0_alpha188] [...] Emerging (1 of 2) app-office/libreoffice-4.1.2.2-r1 Emerging (2 of 2) sys-apps/portage-2.2.7 Installing (1 of 2) app-office/libreoffice-4.1.2.2-r1 Installing (2 of 2) sys-apps/portage-2.2.7 Jobs: 2 of 2 completeLoad avg: 3.06, 2.76, 2.63 Now that you have upgraded portage, does this behaviour (always wanting to rebuild LO) still happen? Yes. As I wrote, I upgraded portage (and LO), then claws (excluding LO this time), then chromium (along with LO). No more packages yet, but every upgrade would remerge LO. Downgrading to portage 2.2.6 does not help. Neil Bothwick writes: On Sat, 5 Oct 2013 20:30:36 +0200, Alex Schuster wrote: If portage believes LO needs to be rebuilt, it will try to do so whichever packages you are emerging, just let it happen. I already did that, twice, when updating portage and chromium. emerge still wants to remerge libreoffice whenever I upgrade a package. As a workround, you could remove LO from your world file, do your updates but don't depclean, then put it back with emerge -n libreoffice It may be quicker than trying to track down the cause. I don't worry about this. I will add --exclude app-office/libreoffice to EMERGE_DEFAULTS_OPTS, and wait for the bug to be fixed by someone. Alex
Re: [gentoo-user] Multiple package instances within a single package slot
Alan McKinnon writes: On 04/10/2013 17:40, Alex Schuster wrote: Well. Sort of. Emerge also wanted to re-merge libreoffice, I have no idea why. The same happened yesterday when I upgraded portage. Whatever :) This time, I used --exclude app-office/libreoffice to avoid this. probably a poppler or icu or java update, or any one of the many things libreoffice uses that changes ABI at the drop of a hat. Recent portage with subslot support triggers a libreoffice rebuild when that happens, it is seldom an error. You can leave it out of the world emerge to speed things up, but revdep-rebuild is probably going to also find it and want to do the same No, there is something wrong here. When I updated portage, it also remerged libreoffice. Upgrading claws-mail updated dev-libs/libdbusmenu and dev-libs/libindicate, and wanted to remerge libreoffice, which I avoided. Next, I updated chromium, which also remerged libreoffice. Again. And whatever package I try to update, emerge wants to remerge libreoffice. Happens with all the packages I tried, which are adobe-flash, python, zsh, xterm. Alex
Re: [gentoo-user] Multiple package instances within a single package slot
Neil Bothwick writes: On Sat, 5 Oct 2013 17:59:52 +0200, Alex Schuster wrote: No, there is something wrong here. When I updated portage, it also remerged libreoffice. Upgrading claws-mail updated dev-libs/libdbusmenu and dev-libs/libindicate, and wanted to remerge libreoffice, which I avoided. Next, I updated chromium, which also remerged libreoffice. Again. And whatever package I try to update, emerge wants to remerge libreoffice. Happens with all the packages I tried, which are adobe-flash, python, zsh, xterm. If portage believes LO needs to be rebuilt, it will try to do so whichever packages you are emerging, just let it happen. I already did that, twice, when updating portage and chromium. emerge still wants to remerge libreoffice whenever I upgrade a package. With a suitable --jobs setting, portage will emerge all the other packages before LO is done. No, at least with portage, the installation happened after libreoffice was installed. weird ~ # emerge -1uaj portage These are the packages that would be merged, in order: Calculating dependencies... done! [ebuild r U ] app-office/libreoffice-4.1.2.2-r1 [4.1.2.2] [ebuild U ] sys-apps/portage-2.2.7 [2.2.0_alpha188] [...] Emerging (1 of 2) app-office/libreoffice-4.1.2.2-r1 Emerging (2 of 2) sys-apps/portage-2.2.7 Installing (1 of 2) app-office/libreoffice-4.1.2.2-r1 Installing (2 of 2) sys-apps/portage-2.2.7 Jobs: 2 of 2 completeLoad avg: 3.06, 2.76, 2.63 Alex
[gentoo-user] Multiple package instances within a single package slot
Hi there! Some may remember me from posting here often. But since a year, I have a new life, and much less time for sitting at my computer. Sigh. And my beloved Gentoo got a little outdated. So, a @world update does not work. I thought I give emerge -e @world a try, this should sort out the problems, but this also does not go well. I don't want to bother you with the whole lot of output emerge gives me, and just ask a specific question at the moment. I get the 'Multiple package instances within a single package slot have been pulled into the dependency graph, resulting in a slot conflict' message, and several affected packages. One example is claws: mail-client/claws-mail:0 (mail-client/claws-mail-3.9.0-r1::gentoo, ebuild scheduled for merge) pulled in by ~mail-client/claws-mail-3.9.0 required by (mail-client/claws-mail-address_keeper-1.0.7::gentoo, ebuild scheduled for merge) (mail-client/claws-mail-3.9.2::gentoo, ebuild scheduled for merge) pulled in by (no parents that aren't satisfied by other packages in this slot) Looking at the ebuild, I see that claws-mail-address_keeper rdepends on claws-mail-3.9.0. But being on ~amd86, 3.9.2 would be current. I can solve this by masking versions greater than 3.9.0. Two questions: Why can't portage deal with this itself, and simply install the highest version that fulfills all requirements? And how do I notice an update to claws-mail-address_keeper that would allow a newer version of claws-mail? Other than remembering those masks and go through them once in a while? Similar problems happen with sys-boot/syslinux, pulled in by sys-boot/unetbootin, media-sound/jack-audio-connection-kit, pulled in by app-emulation/emul-linux-x86-soundlibs, and all dev-qt packages, where I did not yet figure out what to do. I am running portage 2.2.7. Alex
Re: [gentoo-user] Multiple package instances within a single package slot
Kerin Millar writes: On 04/10/2013 11:50, Alex Schuster wrote: [...] (mail-client/claws-mail-3.9.0-r1::gentoo, ebuild scheduled for merge) pulled in by ~mail-client/claws-mail-3.9.0 required by (mail-client/claws-mail-address_keeper-1.0.7::gentoo, ebuild scheduled for merge) (mail-client/claws-mail-3.9.2::gentoo, ebuild scheduled for merge) pulled in by (no parents that aren't satisfied by other packages in this slot) Looking at the ebuild, I see that claws-mail-address_keeper rdepends on claws-mail-3.9.0. But being on ~amd86, 3.9.2 would be current. I can solve this by masking versions greater than 3.9.0. Two questions: Why can't portage deal with this itself, and simply install the highest version that fulfills all requirements? Your use of --emptytree makes it slightly harder to determine from the above output, because the conflict messages will not correctly distinguish merged (installed) packages from those that are yet to be merged. I get some more errors without --emptytree (media-libs/x264, dev-libs/icu, dev-libs/boost, app-text/poppler, dev-util/boost-build, dev-lang/ocaml, x11-base/xorg-server), so I gave -e a try. Do you have mail-client/claws-mail-address_keeper in your world file? Sure. If so, that would mandate its installation as part of the @world set (no if or buts). In turn, that would exhibit a hard dependency on claws-mail-3.9.0, which obviously cannot co-exist with 3.9.2, even if you have unmasked it. Right. Try removing the entry from the world file if it's there, then seeing whether the conflict is handled any differently. I guess this would install 3.9.2, as there's no reason not to do this. And how do I notice an update to claws-mail-address_keeper that would allow a newer version of claws-mail? Other than remembering those masks and go through them once in a while? As of the 3.9.1 ebuild, there is a comment above the collection of blocks that states: Plugins are all integrated or dropped since 3.9.1 Further, from the 3.9.1 release notes: All plugins previously packaged as 'Extra Plugins' are now contained within the Claws Mail package. Thus, it's possible that the address_keeper plugin has been folded into the core. In turn, that would explain why it must block the plugin as a separate package. Good catch! Thanks, also to Neil. I unmerged this plugin, and claws updates just fine. Well. Sort of. Emerge also wanted to re-merge libreoffice, I have no idea why. The same happened yesterday when I upgraded portage. Whatever :) This time, I used --exclude app-office/libreoffice to avoid this. Alex
Re: [gentoo-user] Removing pulseaudio
Walter Dnes wrote: On Wed, Apr 24, 2013 at 12:12:56AM +0200, Michael Hampicke wrote Im not a gnome user as of yet, but I can tell you that the day is coming (Gnome 3.8 I believe) when gnome will not work without PA, so you will have to install it if you want newer Gnome. That's true, gnome3.8 will require you to install pulseaudio-2 From a logic chapter in a highschool math text, the contrapositive version of this is that removing pulseaudio will require removing gnome. I don't use it much, but I have Gnome installed, so I can play around with it if I like. Whenever PulseAudio gets updated, I manually rename /usr/bin/pulseaudio. I was never able to configure it, despite some help from this list in the past, I think my problem is that my internal sound card has two devices, and the HDMI one is default. For ALSA I was able to switch them, with PulseAudio I had no success. Sound behaviour is very erratic, and killing the pulseaudio process (or not enabling it to start at all) seems to help. Although it still happens that Amarok or Flash do not play sound, even though the test sound works fine in the Phonon setup. Quite annoying, but these days I have no time for that any more :-( Alex
Re: [gentoo-user] Emerge problems
meino.cra...@gmx.de writes: FIXED! The problem seemed to be *~-file in package.use left from my last vim session...sigh Huh? I once filed a request that *.bck files should be ignored, because NEdit creates such files per default, and was told that they already ignore those. https://bugs.gentoo.org/show_bug.cgi?id=346075 Alex
Re: [gentoo-user] Restart frozen X
Helmut Jarausch writes: On 02/27/2013 04:18:51 AM, Joseph wrote: I can login to the system over ssh. I've tried to restart/zap xdm it doesn't help. What else can I do to restart X? You can use the SysRq Key (I use CtrlAltPrtSc) and then RE R is okay, this resets the keyboard, and often helped me in the past to get my keyboard back. But E would kills all processes except init. Use K instead, this only kills processes on the current virtual console. Alex
Re: [gentoo-user] No space left on device ?
Florian Philipp writes: tmpfs uses as much memory as necessary and nothing more. In theory, it doesn't hurt to add all your memory to it as tmpfs will start to swap when you run out of memory. However, it is usually a better idea to unmount the tmpfs and use a regular file system whenever you need more space. As Volker noted, it is probably best to use 2GB tmpfs and when you emerge libreoffice, (and maybe firefox and co.) to switch back to using a regular fs. You could also expand tmpfs so that it can eat all memory not used by your applications under normal circumstances. In order to avoid manual intervention when building large packages, I do it that way: In /etc/portage/package.env I have entries like these: app-emulation/virtualboxsafecflags.conf j1.conf app-office/libreoffice notmpfs.conf j1.conf dev-java/icedteanotmpfs.conf dev-lang/R j1.conf games-fps/alienarenanotmpfs.conf games-fps/worldofpadman notmpfs.conf kde-base/kdmj1.conf kde-base/plasma-workspace j1.conf kde-base/systemsettings j1.conf mail-client/thunderbird notmpfs.conf media-sound/amarok debug.conf ~net-mail/dovecot-2.1.15j1.conf net-misc/nx j1.conf sys-boot/grub grub.conf www-client/firefox notmpfs.conf Which means that for those packages the .conf scripts in /etc/portage/env.d/ are sourced. j1.conf has the line 'MAKEOPTS=-j1' in it, so those packages are not being compiled in parallel. I happen to have problems with many packages due to my MAKEOPTS being '--jobs --lod 5', somehow this make much more trouble than MAKEOPTS=-somelarge number. notmpfs.conf has 'PORTAGE_TMPDIR=/var/portage/tmp', while my normal PORTAGE_TMPDIR is /var/portage/tmpfs. It is 4G in size, still this is not enough for many packages. Firefox and Thunrbird are fine with the size, but they tend to be compiled both at once, and then it is not enough. safecflags.conf is: CFLAGS=-pipe -march=amdfam10 -O2 CXXFLAGS=$CFLAGS debug.conf: CFLAGS=$CFLAGS -O2 -ggdb CXXFLAGS=$CFLAGS FEATURES=-buildpkg splitdebug And grub.conf is 'export DONT_MOUNT_BOOT=blabla', this avoids Grub messing around with my /boot directory. Isn't portage just cool? Wonko signature.asc Description: PGP signature
Re: [gentoo-user] Re: udev-197: what to do -- S0LVED
Stefan G. Weichinger writes: # cat /proc/version Linux version 3.6.11-gentoo # zgrep -i devtm /proc/config.gz CONFIG_DEVTMPFS=y CONFIG_DEVTMPFS_MOUNT=y # mount | grep tmpfs udev on /dev type devtmpfs (rw,nosuid,relatime,size=10240k,nr_inodes=493463,mode=755) tmpfs on /run type tmpfs (rw,nosuid,nodev,relatime,mode=755) shm on /dev/shm type tmpfs (rw,nosuid,nodev,noexec,relatime) cgroup_root on /sys/fs/cgroup type tmpfs (rw,nosuid,nodev,noexec,relatime,size=10240k,mode=755) I should edit /etc/fstab, I assume: # grep tmpfs /etc/fstab # glibc 2.2 and above expects tmpfs to be mounted at /dev/shm for # (tmpfs is a dynamically expandable/shrinkable ramdisk, and will shm /dev/shmtmpfs nodev,nosuid,noexec 0 0 I still have this line in my fstab on one host... Same mistake as I mentioned a few days before ... the syntax seems to have changed to: tmpfs /dev/shmtmpfs nodev,nosuid,noexec 0 0 Right? ... but I don't have it at all on another. /dev/shm is mounted just fine though. CONFIG_DEVTMPFS_MOUNT seems to be responsible for that, although the help text says that it does not work when using an initramfs, which I do: CONFIG_DEVTMPFS_MOUNT: This will instruct the kernel to automatically mount the devtmpfs filesystem at /dev, directly after the kernel has mounted the root filesystem. The behavior can be overridden with the commandline parameter: devtmpfs.mount=0|1. This option does not affect initramfs based booting, here the devtmpfs filesystem always needs to be mounted manually after the roots is mounted. With this option enabled, it allows to bring up a system in rescue mode with init=/bin/sh, even when the /dev directory on the rootfs is completely empty. Alex
Re: [gentoo-user] udev-191 bit me. Insufficient ptys
Alan McKinnon writes: On Sat, 2 Feb 2013 16:21:10 +0100 Alex Schuster wo...@wonkology.org wrote: Michael Mol writes: [system does not boot after UDEV upgrade] Ran into the same problem, with my sister's PC. Which I had updated from remote, so I did not see the elogs. I do not think it is correct behaviour to continue building udev although the system wouldn't boot with that kernel option missing. I would expect the udev ebuild to check the running kernel for that option, and refuse to build until it has it set. Or until building is forced by some USE flag or an environment variable. Had these things not been handled better in the past? There's a furious debate going on in -dev about this very thing, and the bottom line is that your statements above are way too simplistic. - there is no guarantee that /proc/config.gz represents the kernel the binary will actually run on (this emerge might well be the last process you ever run on that kernel) - there is no guarantee that /usr/src/linux corresponds to anything at all (it's a symlink and can point to anything, even invalid stuff) - there is no guarantee that the build host will run the code (think build farms, crossdev etc, so every available config cannot possibly be valid) - and a couple more Sure, all this is not guaranteed. But IF there is a /proc/config.gz and a /usr/src/linux/.config without the DEVTMPFS entry, it is quite probable that the system will not boot. And I think a single line 'DEVTMPFS is not set in this kernel. Udev will not run.' along many others is not enough. Basically, the only thing left for the ebuild devs is to notify the user with the important information. That's okay with my PC I am sitting at. But on my sister's PC, I just logged in and started a world update, not monitoring the process all the time. She turned the thing off before I was able to read the elog, and she was surprised when it did not boot the next day. How should I have known what would happen? The question is not whether to halt the build or not (that cannot and will not be done) but how to do the communication: - news item There is one, from 2013-01-23, ending with 'Apologies if this news came too late for you.' Okay, if that one came a little earlier, I would have been fine. - elog - README - some arb notice on a web site somewhere . This is gentoo, the distro that does not hold your hand and gives you every opportunity to keep both pieces. This is a good example of such. I'm using Gentoo for 10 years now, and this is the first time such a thing has happened to me. Normally, the devs do quite a good job informing people about such changes that need to be dealt with, but this time I was not pleased. But I'll stop complaining. This incident just seems a little odd to me, unusual for Gentoo. Alex
Re: [gentoo-user] udev-191 bit me. Insufficient ptys
Michael Mol writes: So, I botched the upgrade to udev-191. I thought I'd followed the steps, but I apparently only covered them for one machine, not both. [...] Udev also complained about DEVTMPFS not being enabled in the kernel.[2] I couldn't get into X, but I could log in via getty and a plain old vt, so I enabled it, rebuilt the kernel, installed it and rebooted...and now that's presumably covered. Ran into the same problem, with my sister's PC. Which I had updated from remote, so I did not see the elogs. I do not think it is correct behaviour to continue building udev although the system wouldn't boot with that kernel option missing. I would expect the udev ebuild to check the running kernel for that option, and refuse to build until it has it set. Or until building is forced by some USE flag or an environment variable. Had these things not been handled better in the past? Alex
Re: [gentoo-user] continue an installation
Willie WY Wong writes: Suppose that I tried to emerge a package, and the compilation phase went through without problems, but it got stopped in the installation phase. Is there a way to (after I fixed the problem) to tell portage to install the (now all already compiled binaries sitting in /var/tmp/portage) directly without having to redo the compiling phase? Case in point: I just tried to update dev-lib/boost to 1.52. The compilation went without a hitch, but the installation died because of file collision against (I think) boost-1.49.0-r1000. Now that the colliding files are no longer there, is there a way to tell portage to go ahead an install boost-1.52 from the compiled sources in /var/tmp/portage ? FEATURES=keepwork emerge -1ua boost If you also want to avoid collisions: FEATURES=keepwork -collision-protect -protect-owned emerge -1ua boost Wonko
Re: [gentoo-user] python 3.1 removed
James wrote: Python 2.7 is my default setting. I also had python 3.1 and 3.2 both installed. I read about how I should get rid of 3.1 and force those apps that need/want python 3 to use python 3.2. (makes sense but I did not fully research it). So I did these steps: emerge -C python:3.1 eselect python update --python2 eselect python update --python3 eselect python list Available Python interpreters: [1] python2.7 * [2] python3.2 python-updater (cleaned up) All is OK? Yes. Miss anything? Nope. Foolish? No, why? Wonko
Re: [gentoo-user] SSD configuration
Luis Gustavo Vilela de Oliveira writes: Well, with 8Gb RAM, i recommend use tmpfs on PORTAGE_TMPDIR, just while u are compiling anything. Or even with 6Gb too. I have 16 GB, with 8GB for $PORTAGE_TMPDIR on tmpfs. There were issues with some packages having not enough space, so I have this in /etc/portage/package.env: app-office/libreoffice notmpfs.conf dev-java/icedteanotmpfs.conf games-fps/alienarenanotmpfs.conf games-fps/worldofpadman notmpfs.conf games-sports/vdrift notmpfs.conf mail-client/thunderbird notmpfs.conf www-client/firefox notmpfs.conf /etc/portage/env.d/notmpfs.conf has this entry, changing PORTAGE_TMPDIR to real HDD space: PORTAGE_TMPDIR=/var/portage/tmp Most of these packages compile with 8 GB of space, but not with parallel merges, like when Thunderbird and Firefox are both being built at the same time. Alex
Re: [gentoo-user] SSD configuration
Jacques Montier writes: I bought a 250Go SSD M4 Crucial , read (of course) Gentoo documentation and installed the drive on my desktop pc (Asus MB, Intel ie7 and 6Go RAM). 1- Everything seems to work perfectly, but i would like to know if my configuration is ok or could be optimized. /tmp and /var/log are on tmpfs Like Volker said. Yikes! Or is that just a typo and you meant /var/tmp? Still, I would prefer to have that on the HDD. /boot, / and /var are on SSD (sda), swap, /home, /usr/portage, /var/tmp and /var/log on a 1To SATA HDD (sdb) I would put the portage tree on the SDD. Wonko
Re: [gentoo-user] gentoo install problem
刘焕杰 writes: Hi guys, I try to install Gentoo this morning. I follow the instructions in the official website. But after I reboot, it appears like below: this is (none). unknown_domain Gentoo Linux 3.5.7 (none) login: Put the host name in /etc/conf.d/hostname, and the fully qualified domainname in /etc/hosts, like 127.0.0.1 localhost myhost.mydomain. And I can log in as root, but it says it is a read-only file system. I can't modify any file. Any messages about this while booting? Like having an unclean file system, but /sbin/fsck.ext3 from sys-fs/e2fspropgs missing? You can halt the output by pressing Ctrl-S, and enable it again with Ctrl-Q. Before I reboot, I can't umount /mnt/gentoo and /mnt/gentoo/dev, it says those devices are busy. But /mnt/gentoo/proc and /mnt/gen- too/boot umount succeed. As Cr0k said. Wonko
Re: [gentoo-user] genlop and tab completion acts weird
Dale writes:
I been noticing something weird. If I try to use tab completion with
the genlop command, I get things like this:
root@fireball / # genlop -t -f /var/-su: /etc/make.globals: No such file
or directory
hp-toolbox.lock ^C
Similar here, with missing /etc/make.conf. It's working when I create it,
as symlink to /etc/portage/make.conf. I do have /etc/make.globals, it
points to ../usr/share/portage/config/make.globals.
The problem lies in /usr/share/bash-completion/genlop,
where /etc/make.{conf,globals} are being referenced.
https://bugs.gentoo.org/show_bug.cgi?id=439234
Wonko
Re: [gentoo-user] emerge --sync
Silvio Siefke writes: i try to install a Gentoo Vserver by Hosteurope. Im have take the last stage archive, because the vserver Archiv is old i think. When i want run emerge --sync it gives only this message: [...] ERROR: out of memory in flist_expand [receiver] rsync error: error allocating core memory buffers (code 22) at util.c(117) [receiver=3.0.9] rsync: connection unexpectedly closed (2861 bytes received so far) [generator] rsync error: error in rsync protocol data stream (code 12) at io.c(605) [generator=3.0.9] Retrying... Can me someone tell what is it? As it says, you're out of memory. It seems you are low on RAM, what does free -m say? Maybe you need to add some swap space? Wonko
Re: [gentoo-user] emerge --sync
Alan McKinnon writes: On Mon, 15 Oct 2012 21:03:59 +0200 Silvio Siefke siefke_lis...@web.de wrote: On Mon, 15 Oct 2012 19:50:07 +0200 Alex Schuster wo...@wonkology.org wrote: As it says, you're out of memory. It seems you are low on RAM, what does free -m say? Maybe you need to add some swap space? lvps5-35-240-192 / # free -m total used free sharedbuffers cached Mem:164600 11 164589 0 0 0 -/+ buffers/cache: 11 164589 Swap:0 0 0 You have 164M of RAM, that is not enough. Packages like gcc and glibc free -m outputs megabytes, so this would mean he has 164 G of RAM, with only 11 M being used... something is wrong here. Not sure what. Wonko
Re: [gentoo-user] Understand Portage not
Silvio Siefke writes: i try to build freecad from source, in Portage is mask. I try to build the requirements, but i understand really not what portage me say with this message. !!! Multiple package instances within a single package slot have been pulled !!! into the dependency graph, resulting in a slot conflict: The complete message found here http://nopaste.info/c7ead120ee.html I think attaching that is considered to be more appropriate, as we don't know whether nopaste.info will be available for as long as the mail archives. I don't really understand the problem, as the multiple package instances seem to have the same version. But maybe it helps to add --tree to the options, this shows which stuff gets pulled in why. And maybe adding --update --deep may sort things out. I also had success sometimes with doing what the output says, adding --backtrack=30. Wonko
Re: [gentoo-user] virtualbox - serial port
J. Roeleveld writes: Joseph syscon...@gmail.com wrote: ls -l /dev/ttyS0 crw-rw 1 root tty 4, 64 Sep 17 20:56 /dev/ttyS0 Is the above correct permission? Those are default permissions. However those normally won't give a normal user access. You can change the permissions of that file/device to enable your user to have access. I am typing this on my mobile and can't quickly tell you how to do that on a permanent basis. But for a quick change you can use 'chown' to change the owner to your own user. What about 'gpasswd -a user tty' to add the tty group to the user? Needs a re-login to make use of the changes. Wonko
Re: [gentoo-user] Re: dev-util/valgrind-3.7.0-r4 and glibc
Nikos Chantziaras writes:
What the other posters said, except that you shouldn't add splitdebug
in your make.conf. If you do that, it will affect all packages.
What you do instead is put this text into
/etc/portage/env/sys-libs/glibc (yes, it must be a text file, not a
directory):
CFLAGS=${CFLAGS} -g
CXXFLAGS=${CXXFLAGS} -g
FEATURES=${FEATURES} splitdebug
Or put sys-libs/glibc splitdebug.conf in /etc/portage/package.env, and
FEATURES=splitdebug in /etc/portage/env/splitdebug.conf. The CFLAGS
change should not be necessary. And I also think that you can simply use
FEATURES=... instead of FEATURES=$FEATURES ... and hope someone will
correct me if not.
Wonko
Re: [gentoo-user] Gentoo is the best linux distro
Alan McKinnon writes: On Mon, 10 Sep 2012 17:33:05 -0700 Mark Knecht markkne...@gmail.com wrote: Like Paul and many others I've never looked back. I'm no power user, and contrary to a lot of the press out there I don't think you need to be to use this distro. That's actually quite perceptive and correct. You don't need to be a genius wise-ass to use Gentoo. You just need to have some brain-smarts and a willingness to look after your own stuff yourself. It takes some more time though to maintain it, compared to the other distros. And the installation is much more complicated of course. But unless you need very basic stuff only, it pays off later I think. When you get into trouble, there are decent howtos that usually do not simply explain _what_ to do, but _why_. When you installed your own Gentoo, you already know a lot about Linux. And where to look in case of problems. Other distros often hide what's going on deeper, and that's nice when all works, but when not, you're screwed. It's also much more fun to actually _solve_ problems on Gentoo, than just googling how some other Ubuntu user 'solved' his problem by trying various commands that you do not understand what they do, but that might also work in your case. Or not. Wonko
Re: [gentoo-user] Gentoo is the best linux distro
Michael Mol writes: On Wed, Sep 12, 2012 at 9:33 AM, Neil Bothwick n...@digimed.co.uk mailto:n...@digimed.co.uk wrote: Instead we get, try USE=-* :P Try MAKEOPTS='-j1' Which in fact often helps... especially for me, I am using MAKEOPTS=-j --load=4, and I often experience build problems that are not reproducible with a fixed number of jobs, regardless how large. Turn off distcc revdep-rebuild And emerge -e world perl-cleaner --all python-updater lafilefixer --justfixit. Wonko
Re: [gentoo-user] My PC died. What should I try?
Peter Humphrey writes: On Tuesday 28 August 2012 21:57:43 Alex Schuster wrote: I wrote: Well, all I can do now is to get a new board and see if things will be okay then. [...] So I had to wait. And when it became available, I wondered if it might be the processor instead that has the problem, so I let the PC shop diagnose CPU and board. This took until today, and they confirmed it was the board indeed, not the CPU. Let me get this straight. The shop ran tests and concluded that the motherboard was faulty, not the CPU? Yes. Fine, I bought the board ...it having been tested and found faulty! Well, obviously not the defective board I already owned, but a new one of the same type. Yes. Defects happen, and because one specific board suddenly has a problem after working fine for half a year, I do not assume that all of these boards will likely fail. And it seems to be the only board having the features I want, at least in the price range of about 100€. Most have two memory banks only, so I would either have to use only 8GB out of 16 GB, or buy new RAM. And I want on-board graphics, I do not want to buy an extra graphics adapter that needs power or has a noisy fan. There were NVidia boards I think, but I prefer Radeon, that finally seems to work just fine, after having lots of trouble in the past with both NVidia and an older Radeon system. Wonko
Re: [gentoo-user] My PC died. What should I try?
Dale writes: Alan McKinnon wrote: Rule #1 in dealing with odd weird strange computer faults is ALWAYS test with another PSU of at least twice the capacity you think you need. +1 I always start with the P/S. Well, unless I see something else unrelated letting the smoke out. Even then tho, a bad P/S can cause the smoke to get out of something else too. It's good advice all the way around. Why not let the computer shop test the P/S? If it blows up something of theirs, it's bad. ;-) I would have preferred to give them the whole PC, but I cannot carry that around easily when going to work by bus and tram, so I could drop it of the store when I leave work in the evening. It was easier to just carry mainboard and CPU in a small bag. Well, not really true, I gave them the hardware on Friday, and on Saturday I could have used the car to transport the PC, but I was somewhat busy that day, and just didn't think about the PSU frying the board. And I had hoped that they would test the board right when I was there on Friday, so I could leave with the new one. Or with the new CPU, if that had turned out to be defective. Wonko
Re: [gentoo-user] My PC died. What should I try?
Volker Armin Hemmann writes: Am Dienstag, 28. August 2012, 22:57:43 schrieb Alex Schuster: This sucks. Is it a faulty board again? Is something (the PSU?) killing the board once I turn the thing on? What will happen when I have the next board and try again? Argh. so - instead of changing the PSU, the obvious culprit, you got a new board AND USED THE SAME PSU? YEAH :) Thinking about this now, yes, it would have made sense to test with another PSU first. But it wasn't so obvious to me, I simply thought I had bad luck with a bad board, that died. Happens. I am just saying - one faulty PSU fried three of my boards. Enermax... will never buy again. So - instead of changing the PSU, the obvious culprit, you let it fry another board, and then... yet another one? Just saying :) I once had the opposite problem, a mainboard seemed to kill PSUs. That was weird. The fans spin, so not all hope is lost. Keyboard, ps/2? usb? It's a PS/2 keyboard. But before you do anything else, change the PSU. I tried another one this morning, same problems. I guess the board is fried. So I'll order another one, and this time use another PSU. Wow, they say it will take 2-3 weeks. So I'll see if there's another board that will fulfill my needs... and there is. Radeon 3000 instead of 4250, and I remember having big trouble with my last Radeon 3250 system... and no eSATA which I probably wouldn't miss anyway, but it also has no PATA at all. I can (and have to) live with this it seems, but it's somewhat inconvenient. Wonko
Re: [gentoo-user] My PC died. What should I try?
I wrote: Well, all I can do now is to get a new board and see if things will be okay then. This took longer than expected. The board I wanted (the same I already have) was not available, I had to order it. Strange, there is only one that has the features I want - AMD3+ chipset, four memory banks, USB 3, and on-board graphics. So I had to wait. And when it became available, I wondered if it might be the processor instead that has the problem, so I let the PC shop diagnose CPU and board. This took until today, and they confirmed it was the board indeed, not the CPU. Fine, I bought the board, installed it in the PC, and guess what - it doesn't work. On the first boot I saw some BIOS status messages, hard drives and such, but the keyboard did not react, and then it did not boot, I got a black screen only. And on subsequent tries, with everything (2 ISDN cards, 4 hard drives) except for the DVD drive removed, the screen does not even turn on. All fans spin, and the DVD-ROM tray opens when I press the eject button. That's all. No keyboard LEDs. This sucks. Is it a faulty board again? Is something (the PSU?) killing the board once I turn the thing on? What will happen when I have the next board and try again? Argh. Wonko
Re: [gentoo-user] SSD performance tweaking
Frank Steinmetzger writes: On Thu, Aug 23, 2012 at 12:15:20PM +0200, Alex Schuster wrote: The size of an erasable block of SSDs is even larger, usually 512K, it would be best to align to that, too. A partition offset of 512K or 1M would avoid this. Unless the filesystem knows this and starts bigger files at those 512 k boundaries (so really only one erase cycle is needed for files =512 k), isn't this fairly superfluous? Yes, I think it is. When you search for SSD alignment, you read about this alignment all the time, even on the German Wikipedia, and many resources say that this can have a big impact on performance. But I could not find a real explanation at all. Besides that, it's not so easy to do the alignment, at least when using LVM. I read that LVM adds 192K header information, so even if you align the partition start to an erasable block size of 512K, the actual content is not aligned. See [*] for information how to overcome this. That is, if you believe the alignment to erasable blocks is important, personally I do not know what to think now. It wouldn't hurt, so why not apply it, but it seems like snake oil to me now. Wonko http://tytso.livejournal.com/2009/02/20/
Re: [gentoo-user] SSD performance tweaking
Volker Armin Hemmann writes: Am Sonntag, 26. August 2012, 13:41:09 schrieb Alex Schuster: Frank Steinmetzger writes: Unless the filesystem knows this and starts bigger files at those 512 k boundaries (so really only one erase cycle is needed for files =512 k), isn't this fairly superfluous? Yes, I think it is. When you search for SSD alignment, you read about this alignment all the time, even on the German Wikipedia, and many resources say that this can have a big impact on performance. But I could not find a real explanation at all. Besides that, it's not so easy to do the alignment, at least when using LVM. I read that LVM adds 192K header information, so even if you align the partition start to an erasable block size of 512K, the actual content is not aligned. See [*] for information how to overcome this. That is, if you believe the alignment to erasable blocks is important, personally I do not know what to think now. It wouldn't hurt, so why not apply it, but it seems like snake oil to me now. Wonko http://tytso.livejournal.com/2009/02/20/ because erasing is slow. You can not overwrite data on a ssd. you have to erase first, then reprogramm. Also, erasing shortens lifetime. Yes, I know that. But why exactly does it help to align a partition to the erasable block size? I don't get it. Why isn't it sufficient to align to the usual 4K block size, so that a block never spans over two erasable blocks? Wonko
Re: [gentoo-user] SSD performance tweaking
Am 26.08.2012 16:21, schrieb Volker Armin Hemmann: Am Sonntag, 26. August 2012, 14:49:08 schrieb Alex Schuster: Volker Armin Hemmann writes: Am Sonntag, 26. August 2012, 13:41:09 schrieb Alex Schuster: Yes, I know that. But why exactly does it help to align a partition to the erasable block size? I don't get it. Why isn't it sufficient to align to the usual 4K block size, so that a block never spans over two erasable blocks? well, for one, there are lots of ssd which have 8k pages. Not 4k. Whatever. Then align to 8K instead. But what does this have to do with the erasable page size? Wonko
Re: [gentoo-user] NFS mounts and uid/gid/user names
meino.cra...@gmx.de writes: to not to stress the SD-card of my single board computer too much I mounted a directory of my PC via NFS at my single board computer, so that compilations and other task which need to be done while installing will access the hd and not the SD-card. (The singleboard computer is a Cortex A8/OMAP based one: 32 bit. The PC is 64 bit AMD based.) So far so nice...everything works fine: I can see the directory on both ends. In /etc/exports on the PC I entered this: /tmp/NFS 192.168.178.25(async,rw,no_subtree_check) When setting chmod 700 /tmp/NFS, chown root:root /tmp/NFS on the server side (PC) I cannot write to the directory as root on the client side (single board computer). On both sides root is 0:0. When setting chmod 777 /tmp/NFS on the server side, I am able to write at the client side to the that directory, bit listing that files shows me that they become owned by nobody:nobody which is nobody:x:65534:65534:nobody:/nonexistent:/bin/sh on the server side. How can I acchieve, that files written on one side remain the same uid/gid assignment on the other side? Add 'root_no_squash' to your options in /etc/exports. Have a nice weekend! Will do! Wonko
Re: [gentoo-user] SSD performance tweaking
Mark Knecht writes: I'm currently just using a single large partition ext3. I didn't do anything special in fdisk so the partition might not be aligned as best it could be. I don't know. See if the partition's starting block is 63 as it used to be in the past. In this case the alignment is wrong, as SSDs have 4K (or even 8K) sectors consisting of 8 (or 16) 512 byte blocks. The starting block should be divisible by 8 (or 16) because of the large sector size, if not, a file system sector spans over two drive sectors, and both heed to be accessed when reading a file system sector. The size of an erasable block of SSDs is even larger, usually 512K, it would be best to align to that, too. A partition offset of 512K or 1M would avoid this. Wonko
Re: [gentoo-user] Invalid module format
Tamer Higazi writes: I did what you say, now the magic issue comes, the kernel drivers ARE BUILT for this kernel, here the modinfo output: tamer@office ~ $ sudo modinfo /lib/modules/3.3.8-gentoo/kernel/drivers/net/wan/wanpipe.ko +filename: /lib/modules/3.3.8-gentoo/kernel/drivers/net/wan/wanpipe.ko license:GPL description:Sangoma WANPIPE: WAN Multi-Protocol Driver author: Nenad Corbic ncor...@sangoma.com depends:sdladrv,wanrouter vermagic: 3.3.8-gentoo SMP mod_unload modversions Does modinfo say exactly the same for another, regular kernel module you compiled when building the kernel? I am running the 3.3.8 SMP kernel, and I don't know why he doesn't load the modules. This is what droves me crazy about it any other ideas?! Did you use the same compiler version for building the kernel and for wanpipe? Maybe enabling CONFIG_MODVERSIONS in your kernel will help (Enable loadable module support -- Module versioning support), but I doubt it. Wonko
Re: [gentoo-user] My PC died. What should I try?
Dale spent two cents: Just my two cents here. Problems like this are usually the power supply. Could it be the mobo, yes it could but the power supply is more likely, usually cheaper to replace and easier to. I had a friends puter that was acting weird, random reboots and such, it was the power supply. A bad power supply can cause all sorts of weird problems. Indeed. Well, so can bad capacitors or a hair crack on a motherboard, but those are rare I tink. If you can, unplug everything including the CD/DVD drive. No hard drives either. Just play with the BIOS. Basically, don't try to boot anything, just look at the BIOS itself. If it acts weird, start with the power supply. If you have to, go to a local place and pick up a cheap power supply. I got three from a friend that once were mine, and I know that at least one of them is definitely working. But the effect was the same. Random problems are hard to fix sometimes. You just have to swap things until you find the bad part. I would put the odds at 80% that it is the power supply tho. I hoped so, as I do not have board or CPU to swap. While at it, do you know what brand and the wattage of your power supply? It could be that someone on here as experience with that particular brand or even that exact model. I could look it up, but then, it's not new, and was one of the few parts that survived a major hardware failure half a year ago. Maybe it got damaged a little aready then. It seemed to work fine, so I kept using it. These things are not cheap, as I tend to buy quality ones that are silend and efficient. I'll get a new board tomorrow, and hope I will have all back working soon. I'm very used to my desktop PC. I have a notebook that is way faster, but it's new and I don't have all my stuff on it yet. Oh, and it runs Windows 7... I'm not sure yet if I will a Gentoo VM, or if I will install Gentoo natively and run Windows in the VM. The best would be the option to have both, I think I read an article on how this could be accomplished. With Gentoo it's not much of a problem, I did that already, but Windows will need some tweaking. And I do not have much time for this these days. Wonko
[gentoo-user] My PC died. What should I try?
Hi there! Two days ago, my PC suddenly died, after working fine for half a year. I used myrtcwake as usual to suspend to RAM, and it woke up in the morning. But after two minutes, the screen went blank and nothing, even SysRq, gave a reaction. I tried booting a couple of times again, and sometimes it did not even reach KDM. Now, I cannot even run Grub (from my USB stick) any more, I only see a GRUB string at the top right, then nothing happens. Booting with SystemRescueCD also freezes sometimes. If not, I can make it freeze after seconds by running 'memtester'. Booting good old memtest86 ran for an hour and only found one error, then I aborted, removed three of my four memory modules (4GB each), and tried different ones in the first bank. Memtest86 again did not find much errors, but froze once. Running memtester after booting from SystemrescueCD again makes the thing freeze in seconds. It once also froze while being in the BIOs setup. What could be the problem? CPU, board, or even the PSU? I do not think it has to do with bad memory. I removed most of the other stuff (hard drives, PCI cards). I have no similar hardware so I cannot simply exchange things, the question is what to buy and try. How would you proceed? The fan is still working, the cooler does not become hot, and in the BIOS there are not high temperatures begin reported. But one thing was strange: I updated Calligra from 2.4 to 2.5 (I think), and it took ages, at least 8 hours. I thought there may b something strange with the build process of this new version, forcing MAKEOPTS=-j1 and such, but still this is very long. But when working with it, I did not notice anything strange like sluggish reactions, and videos played fine. But I did not use it as much as I normally do, and maybe even when overheated and throttled down it would have been fast enough for me to not notice this. I watch the syslog normally, but maybe I just did not look closely that day, I was busy doing other stuff. CPUs don't just die, do they? Even when overheating, I think these days throttle down, so no permanent harm should be done? So maybe it's the board? It looks okay, no bent or leaking capacitors. This is really annoying. Of course most of my passwords are in my KDE wallet I cannot access. There's also Wiki, CVS and Git repositories, not needed every day, but still important. And the timinig is very bad, I just started my new job the day the problem happened, and I do not have much time for this now. Before, I was working at home, so I would have had all day to diagnose and try things. It's an AMD FX-4100 Quad-Core CPU, and an ASRock 880GMH/U3S3 board. Wonko
Re: [gentoo-user] My PC died. What should I try?
meino.cra...@gmx.de writes: ...shot in the dark: Remove as much as possible of the cards, addons, connections etc from the PC ... make in as much bare bone as possible. Done already. Check All coolers (the little ones also) for dust. Remove all dust even if it is not completly covered with it. They are clean. Dont forget the internals of the power supply. Detach all cables. Remove the power supply. Go outside ;) and blow the dust inside away. I did not remove it yet... but if it's a temperature problem, it should not happen right after 30 seconds, when Grub already fails. The voltages reported in the BIOS are okay, but I don't know it this information is accurate and reliable. Put the power supply back into the PC again an attach the cables. If I only could find a spare one... I have it, but I don't know where. Remove all RAM, carefully clean the contacts, insert as less RAM as possible. Did that, using only 4 of 16 GB, and I switched the modules. Remove even the HD if it is possible to get into the BIOS without any HD attached. I also did that, only the CD-ROM is attached. Remove the BIOS battery, wait at least a day and insert it again. That's worth a try. My old PC had a jumper which I could short circuit to instantly drain it, not sure if this was normal. Start the PC and go directly into the BIOS. Check the date/time. If it shows the current date/time, the battery wasn't removed long enough. Check the battery voltage. Reinsert the battery. If your board has a BIOS reset: Reset the BIOS. Then: In the BIOS enter a page which does something (reports continously temperatures for example). If this is possible, let the PC run for a while that BIOS page and see, whether it hangs again or not. Okay, I will do this. If all went fine, add ONE component and try it again. Add the HD at last to sort out hardware from software bugs... Nah, I cannot even boot from my USB stick any more. I don't have a boot partition on my hard drive, so it is not involved there. May be one of the components and not the CPU or motherboard causes the problem and you will be able to identify it by this procedure... I hope it's the power supply, this would mean the least effort. I'd simply buy a new one, and I would not have to think about what board or which CPU I would like to get. HTH! GOOD LUCK! Thanks! I can need it. Wonko
Re: [gentoo-user] My PC died. What should I try?q
Randolph Maaßen writes: Aaa aAaa aaa a Am 17.08.2012 10:31 schrieb meino.cra...@gmx.de mailto:meino.cra...@gmx.de: Hi Alex, ...shot in the dark: Remove as much as possible of the cards,aadwqqqaaa www wpa www a weißes www aa Array www www www a aaa aa aadwqqqaaa aaa w aadwqqqaaa www aa aaa a aaa www Awaa aaa aa quattro Aquarellw aaa aa aa Webauftritt aaa a aaa aA aaa aAaAaAaq aaawa addons, connections etcwo from the PC ... make ian as much bare bone aaa stwww wwwaaa www qaaa wwwas a. www www waslittle ones also) for dust. Removeaa wwwaa all dust even if it is not completly covered with ait. Woow! What is going on here? Wonko
Re: [gentoo-user] My PC died. What should I try?q
Randolph Maaßen writes: 2012/8/17 Alex Schuster wo...@wonkology.org mailto:wo...@wonkology.org Woow! What is going on here? Damn!! Sorry for this bad post, somehow my phone unlocked in my pocket. I'm happy for every reply, and this was a very special one :) -- Mit freundlichen Grüßen / Best regards Randolph Maaßen The signature seems to be separated correctly by -- instead of --, yet my Thunderbird does not recognize it as such. Maybe it has a problem with quoted-printable format? Wonko
Re: [gentoo-user] My PC died. What should I try?
v...@ukr.net writes: If the system behaves in such an unpredictable way (freezing at a random point), I usually check the following things: - RAM; - bloated capacitors on the Motherboard; - bloated or dried capacitors in the power supply unit; If your PC is only half a year old, it is unlikely that the capacitors dried. But they could easily bloat, especially if they were of bad quality or situated near some hot surface like heat sinks. Testing the power supply needs not only visual analysis. It would be good to attach the oscilloscope to the output and see the voltage level. It should not have large peaks (voltage jumps). But this is usually true for the old units with dried capacitors, as I said. The power supply is older, I re-used it from the PC I had before this one. I hope it causes the trouble, and will try another one this evening. Thanks for this information, this strengthens my confidence that I do not have to buy a new board or CPU. Now I am driving home with a bag of three PSUs I had lent to a friend (and already forgotten). If I were you, I'd tried to temporarily replace the memory with a 100% working module, and if it does not help - replace the power supply unit (if you do not have the necessary equipment to test it thoroughly). I wish I had :) The RAM is okay, I think, I cannot imagine different memory modules to suddenly go bad all at once. And memtest86 found one error only after an hour, while the crashes happen after a few minutes already. And one more simple test: turn on the PC, enter the BIOS setup utility and keep it running in this state. If it runs ok for some time (like a couple of hours), I'd say the problem is in RAM. It once crashed after ten minutes. That was not reproducable, but I did not try that often. Wonko
Re: [gentoo-user] My PC died. What should I try?
Volker Armin Hemmann writes: sounds like a power problem. Either psu is gone bad (get a new one) Well, I got three old ones instead :) or your mainboard's power circuitry gone bad (if replacement of psu does not help, get a new one). It did not help :( Too bad, I probably need a new mainboard. And I cannot get one before monday evening, I have to go to a wedding tomorrow (not mine) and I doubt I will have time to find a hardware store there. But first thing first: disconnect your hdds! No reason to risk them. I did that soon. I already had trouble with one two weeks ago, it had bad blocks on the home partition. The replacement drive also had bad blocks, I had to get yet another one. It's a good thing to have recent backups :) And there, it just crashed while in the BIOS setup. Wonko
Re: [gentoo-user] My PC died. What should I try?
Paul Hartman writes: If you are using a video card (instead of built-in/on-board video) I would try a different video card, if you have an old or spare one. I have had lots of video cards die from overheating and power spikes. Sorry, I did not mention that I do not have a video card, it's onboard video. I do not need great video power, and I wanted to have a quiet PC that also saves power. I only had one motherboard ever die, a computer I gave to my father died after a few months... it was ASRock brand but I'm sure that is a coincidence. :) It had blown/cracked capacitors all over the motherboard. It did not die completely at once. It would kind of work, but started to crash randomly and became worse and worse until finally it wouldn't boot at all. I replaced the MB, but kept the same CPU, RAM everything else, and it has been working ever since. That was after we bought a new power supply that didn't make any difference. I'd also say this is unusual. I had a board die, but that was my own error :) Well, all I can do now is to get a new board and see if things will be okay then. Thanks for all your responses! I know this is not really related to Gentoo, but that's what I love this list for, people are very helpful and competent here. Wonko
Re: [gentoo-user] Cannot see Grub menu
Frank Steinmetzger writes: So after the recent thread here about 32bit/64bit and some arguments from a friend, I made the switch from 32 bit to 64 bit (with a clean install from scratch of course). There’s one big problem I’m having: I cannot see the Grub (legacy) boot menu. It still functions alright, but I don’t see it. Weird, I have no idea. Just want to say that I am using legacy Grub on ~amd64 just fine. Not grub-static, and the static USE flag is not set. Never had a problem with that. Do you use a splashimage in your grub.conf? Maybe without you will get a working text mode Grub. Not that this should matter, but anyway. Wonko
Re: [gentoo-user] Want to seriously test a NEW hard drive
Dale writes: I have seen where people use dd to do this sort of thing to. I read somewhere that if you do a dd and put in all 1's, then all 0's then back again that it is very hard to get any data back off the drive. I think if you do it like over a dozen times, it is deemed impossible to get anything back. I think that is the Government standard of it's gone. There's no need for multiple passes of dd with different values. http://www.h-online.com/security/news/item/Secure-deletion-a-single-overwrite-will-do-it-739699.html Wonko
Re: [gentoo-user] Anyone compiled libreoffice-3.6.0.4 yet?
walt writes: This has been slow and painful so far. First, the build stops repeatedly because of zero-length library files. I can restart the ebuild manually and each iteration builds one more (real) library. I've been doing this iterating for hours and I think I may have gotten past that part, but the build is far from done... Second, the ebuild is using only one CPU out of four, so this is taking much longer than before. Ugh. Anyone else seeing these problems? No, I just built it today, in the usual time of about 2.5 hours. Wonko
Re: [gentoo-user] Want to seriously test a NEW hard drive
Dale writes: I finally got me a 3Tb drive on the way. Should be here Wednesday. I have seen some reviews where it would not work right. I think some of it may be BIOS related since some BIOS's don't like drives that large. Anyway, I want to test this thing real good to really make sure it is up to the task before putting my data on it. It's going to be so much data, there is really no way to do back-ups at this point. Come on, 2 to 3Tbs on 4Gb DVDs. Really? lol Maybe a external drive later on but for now, well. I have heard of bonnie and friends. I also think dd could do some testing too. Is there any other way to give this a good work and see if it holds up? Oh, helpful hints with Bonnie would be great too. I have never used it before. Maybe someone has some test that is really brutal. smartctl -t long /dev/sdb will make the drive start a selftest. This will take a while, and even more if the drive is being used otherwise, as this test should not impact its performance. Use smartctl -l selftest to view the results. As long as there is no number in the 'LBA_of_first_error' column, it should be okay. That is a reading test only, badblocks -sw /dev/sdb will make it perform a write-mode test. It uses four different patterns, I would be okay with only one test, so I'd either stop it when it is done writing and comparing the first pattern, or supply a test pattern with option -t. Wonko
Re: [gentoo-user] Udev rules for identical hard drives
Canek Peláez Valdés writes: On Wed, Aug 1, 2012 at 7:42 PM, Alex Schuster wo...@wonkology.org wrote: Canek Peláez Valdés writes: [ snip ] Oh, and I forgot; doesn't the links in /dev/disk/by-id, /dev/disk/by-label, /dev/disk/by-uuid do what you want to? Those seem to list partitions only, not whole drives. A label for a drive would be nice to have. I'm pretty sure whole drives are there also: $ ll /dev/disk/by-id ... ata-SAMSUNG_HD160JJ_S08HJ10YC13279 - ../../sda ... That's a whole drive right there. Wow, now I feel really stupid :) You are so right, they are there, and I don't why I overlooked them... too many entries there maybe, I have 140. But still. Stuuupid! Thanks, Canek! Wonko
Re: [gentoo-user] Udev rules for identical hard drives
Alex Schuster writes: Canek Peláez Valdés writes: $ ll /dev/disk/by-id ... ata-SAMSUNG_HD160JJ_S08HJ10YC13279 - ../../sda ... That's a whole drive right there. Wow, now I feel really stupid :) You are so right, they are there, and I don't why I overlooked them... too many entries there maybe, I have 140. But still. Stuuupid! I looked again in the terminal at what I did this night, and at least feel a little less stupid now. I had searched for my /dev/sdd drive, and this one just has no label. Only its partitions do, they appear twice, as ata-SAMSUNG_SP1614N_0735J1FW815459-part[15678] and wwn-0x50f0-part[15678]. This drive is an older PATA drive, maybe that's the difference? Wonko
Re: [gentoo-user] Udev rules for identical hard drives
Mark Knecht writes:
Check out the very nice 'lsdrv' script by Phil Turmel. Run it, save a
copy of the output for bad times.
https://github.com/pturmel/lsdrv
That doesn't work here, and I do not understand why. In line 305 it tries
and fails to create /dev/block, which is already existing.
if not os.path.exists('/dev/block'):
os.mkdir('/dev/block', 0755)
Uh, is this a python bug? It works fine with python 2.7, but not with
3.2. But os.path.exists() is quite a basic function, if that wouldn't
work, I'd expect all things to break, including emerge.
Nice script. Much similar to lshw I think, but it shows more stuff, like
LVM names and UUIDS. Thanks!
Wonko
Re: [gentoo-user] Udev rules for identical hard drives
Walter Dnes writes:
You can get the ATTRS{serial} (i.e. serial number). See the printer
example at http://www.reactivated.net/writing_udev_rules.html and adapt
to your hard drive. Serial numbers should be unique, even amongst
otherwise identical drives...
==
I power on my printer, and it is assigned device node /dev/lp0. Not
satisfied with such a bland name, I decide to use udevinfo to aid me in
writing a rule which will provide an alternative name:
# udevinfo -a -p $(udevinfo -q path -n /dev/lp0)
looking at device '/class/usb/lp0':
KERNEL==lp0
SUBSYSTEM==usb
DRIVER==
ATTR{dev}==180:0
looking at parent device
'/devices/pci:00/:00:1d.0/usb1/1-1':
SUBSYSTEMS==usb
ATTRS{manufacturer}==EPSON
ATTRS{product}==USB Printer
ATTRS{serial}==L72010011070626380
My rule becomes:
SUBSYSTEM==usb, ATTRS{serial}==L72010011070626380,
SYMLINK+=epson_680
That's exactly what I would like to have! I have a working solution, but
using UDEV would seem more adequate.
But: I cannot find a serial number for my hard drives in the output. And
shouldn't there be a file named 'serial' in /sys? I have some, but not
for my block devices, only for USB and in /sys/{bus,pci}/drivers/.
BTW, sys-fs/udev-187 does not have the 'udevinfo' command, it seems to be
'udevadm info' now.
Wonko
Re: [gentoo-user] Udev rules for identical hard drives
Dale writes:
Alex Schuster wrote:
Mark Knecht writes:
Check out the very nice 'lsdrv' script by Phil Turmel. Run it, save a
copy of the output for bad times.
https://github.com/pturmel/lsdrv
That doesn't work here, and I do not understand why. In line 305 it
tries and fails to create /dev/block, which is already existing.
if not os.path.exists('/dev/block'):
os.mkdir('/dev/block', 0755)
Uh, is this a python bug? It works fine with python 2.7, but not with
3.2. But os.path.exists() is quite a basic function, if that wouldn't
work, I'd expect all things to break, including emerge.
[...]
I'm amd64 and it works here.
root@fireball / # equery l python
* Searching for python ...
[IP-] [ ] dev-lang/python-2.7.3-r2:2.7
[IP-] [ ] dev-lang/python-3.2.3:3.2
Um, but did you use eselect to make 3.2 the current version?
Wonko
Re: [gentoo-user] Udev rules for identical hard drives
Mark Knecht writes:
On Thu, Aug 2, 2012 at 8:02 AM, Alex Schuster wo...@wonkology.org
wrote:
Mark Knecht writes:
Check out the very nice 'lsdrv' script by Phil Turmel. Run it, save a
copy of the output for bad times.
https://github.com/pturmel/lsdrv
That doesn't work here, and I do not understand why. In line 305 it
tries and fails to create /dev/block, which is already existing.
if not os.path.exists('/dev/block'):
os.mkdir('/dev/block', 0755)
Uh, is this a python bug? It works fine with python 2.7, but not with
3.2. But os.path.exists() is quite a basic function, if that wouldn't
work, I'd expect all things to break, including emerge.
Nice script. Much similar to lshw I think, but it shows more stuff,
like LVM names and UUIDS. Thanks!
Dunno about the python-3.2 thing. Are you set to use 3.2 by default?
(How aggressive of you!) ;-) I'm set to use 2.7 as default which I
think is the overall recommendation of dummies like me:
Portage should work well with 3.2 now, but I wouldn't wonder much if
something would break. I don't mind much about this, when it happens I
file a bug report, and use 2.7 again. But the problem with
os.path.exists() seems weird to me.
c2stable ~ # eselect python list
Available Python interpreters:
[1] python2.7 *
[2] python3.2
c2stable ~ #
The script has been around awhile and updated now and again. Possibly
it's just not tested with python-3.2?
I guess so. Hmm, does anybody want to provide an ebuild on
bugs.gentoo.org for it? It would be nice to have it in portage.
Wonko
[gentoo-user] Udev rules for identical hard drives
Hi there!
I do not understand the numbering of my hard drives. There may be some
inherent logic, but whenever I make some changes, like replacing drives,
or changing BIOS settings, the order changes. Maybe it's even more random.
So I made some udev rules like this, and my drives are called /dev/hd1,
hd2 and hd3:
SUBSYSTEMS==scsi, KERNEL==sd?, ATTRS{model}==SAMSUNG HD154UI,
SYMLINK=hd1
This works fine, and this way I can address them in scripts, smartd and
hdparm config files and such. But now I have two identical drives. I had
this before with the drive above, but while being identical models, the
two drives differed a little in size, so I just had to add ATTR{size}.
This does not help with my current drives, and I find nothing
in /sys/block/sd?/device/ that differs. Could there be another way to
distinguish the drives, like looking at the partition scheme or something?
Wonko
Re: [gentoo-user] Udev rules for identical hard drives
Canek Peláez Valdés writes:
On Wed, Aug 1, 2012 at 6:59 PM, Canek Peláez Valdés can...@gmail.com
wrote:
On Wed, Aug 1, 2012 at 6:34 PM, Alex Schuster wo...@wonkology.org
wrote:
[...]
Could there be another way to distinguish the drives, like looking
at the partition scheme or something?
If you want to distinguish partitions, I would recommend using labels
(in fstab too); those never change unless you specifically change
them. Then, no matter how you put them in your machine, they will get
mounted correctly, and then you don't need to fuzz with udev rules.
Also, as a superficial bonus, they get mounted using the label and it
looks nice in your file browser.
I'm aware of that, and I would use this, if I weren't using LVM and
encryption on top of that. So I do not deal with raw partitions at all,
but with partitions like /dev/mapper/root or /dev/weird/portage.
Oh, this gives me an idea of what to use as workaround: If what I would
like to have is not possible, I will add a little start script
in /etc/local.d/ which calls pvscan to check which volume groups belong
to which drives, and creates the symlinks.
The drives themselves I see no reason to recognize them, why do you
need to do that?
Well, I don't really *need* this. But it's convenient.
- I have a monitoring plasmoid on my desktop that shows whether a drive
is active or on standby, and also gives the temperature of my always
running system drive. If there were a mixup, calling hddtemp on a
sleeping drive would wake it up.
- I have different idle time settings in /etc/conf.d/hdparm, and I spin
down two drives immediately after I have booted.
- Same goes for a little script I use for suspend-to-ram. It makes use of
the rtcwake command to make the PC wake up in the morning (before I get
up), and along other stuff spins down drives.
- And I have different settings in /etc/smartd.conf.
Oh, and I forgot; doesn't the links in /dev/disk/by-id,
/dev/disk/by-label, /dev/disk/by-uuid do what you want to?
Those seem to list partitions only, not whole drives. A label for a drive
would be nice to have.
Uh, and here's the little start script I just wrote. No idea why I call
my drives hd1 to hd4 instead of using the name of the only volume group
they have, but I'll keep it like that for now.
str=$( pvscan )
hd()
{
hd=$( echo $str | grep $1 | head -n 1 | awk '{print $2}' )
echo ${hd//[0-9]/}
}
ln -s $( hd weird ) /dev/hd1
ln -s $( hd weird2 ) /dev/hd2
ln -s $( hd weird3 ) /dev/hd3
ln -s $( hd pata1 ) /dev/hd4
Wonko
Re: [gentoo-user] Python TK
Silvio Siefke writes: on my Netbook i use Sabayon, because all compile from source need much time and it was not so really run. Can i ask here a question, because i has problems with emerge. Sure, and there doesn't even seem to be a Sabayon mailing list anyway. I has Install the Game PySolFC, a python solitaire Game. It want not run, because it miss imagingtk. So i try to rebuilt Python and Imaging with the normal Gentoo Way. Mistake in Game http://nopaste.info/c04fddda9d.html Recompile Python and Imaging http://nopaste.info/090baf194d.html I follow the advice what says emerge, but emerge do nothing. Did you put tk in your USE flags for python? Like, having dev-lang/python tk in /etc/portage/package.use, if you want to have it for all versions of python. If you only want that for 2.7, use this line: dev-lang/python:2.7 tk gentoo-mobile siefke # emerge --newuse --update =dev-lang/python-2.7.2-r3 Calculating dependencies... done! Auto-cleaning packages... Should work, when the tk USE flag is set now, but wasn't set when python 2.7 was compiled. You can use emerge -pv dev-lang/python:2.7 to see which USE flags are set. Wonko
Re: [gentoo-user] Python TK
Neil Bothwick writes: On Thu, 26 Jul 2012 12:50:09 +0200, Silvio Siefke wrote: gentoo-mobile siefke # emerge --newuse --update =dev-lang/python-2.7.2-r3 Calculating dependencies... done! Auto-cleaning packages... No outdated packages were found on your system. gentoo-mobile siefke # That's correct, because you used --update, so the package is only built if a newer version is available. Drop --update and it will build the package. It would build it, but with the same USE flags, so this would make no change. If Silvia had changed the USE flag for python, --newuse would make emerge rebuild it, with or without --update. BTW, also add --oneshot / -1 to the emerge options, so the packages you build manually do not end up in your world file. Unless you explicitly want that, but when rebuilding existing things, they either are already in world, or they are dependencies that do not need to be in world. Wonko
Re: [gentoo-user] How can I rewrite all empty sectors with zeros?
Jarry writes: I want to backup my whole hard-drive (8 partitions) with: # dd if=/dev/sda | gzip /path/image.gz In order to achieve good compression level I'd like to wipe out all empty space with zeros. How can I do that? You can create files containing only zeros on all partitions until they are full. Like this: for i in 5 6 7 8 9 10 11 12 do mount /dev/sda$i /mnt dd if=/dev/zero of=/mnt/zero rm /mnt/zero umount /mnt done Wonko
Re: [gentoo-user] Need a clue about merging logical volumes/groups with lvm2
walt writes: I know there are a few lvm2 experts lurking here :) I have a 500gig disk that is split roughly in half between two volume groups, each containing four physical volumes, and each vg is formatted into an ext4 filesystem of roughly 250GB. What I plan to do is merge the two volume groups into one, containing one big ext4 filesystem, which will contain all of the files currently on the disk. Can this be done without copying one of the existing ext4 filesystems to a separate drive first, and then copying it back after extending the remaining vg/filesystem? (One filesystem has 24GB free and the other has 25GB free.) I'm expecting a no but I'd like to be wrong :) I think you are right. But if you had more free space, it might be possible. So your physical volumes are about 63 G each. If you free that much space on one filesystem, reduce the file system, then reduce the LV, you can use pvmove to move stuff from one PV you want to empty to the others. When done, you can remove the now empty PV from the VG with vgreduce. Then use vgextend to add the PV to the other VG. Extend the LV of that VG and enlarge the file system, copy stuff from the other FS over until you can free another PV. And so on. But copying all stuff of one VG to another location would be much easier. And less error-prone. Wonko
Re: [gentoo-user] Re: The system's font display problem
Mark Knecht writes: On Sun, Jul 15, 2012 at 8:09 AM, Nikos Chantziaras rea...@gmail.com wrote: Do not set anything other than LANG and LC_COLLATE. Then only set vars that differ from LANG. Your /etc/env.d/02locale should look like this: LANG=en_US.UTF-8 LC_COLLATE=C LC_CTYPE=zh_CN.UTF-8 [...] Just double checking here. Is the file /etc/locale.gen now totally depreciated or is it still required? The install guide still has it in chapter 8 where the file /etc/locale.gen ends up looking pretty much identical to the 02locale file. Or maybe they serve different purposes somehow? /etc/locale.gen defines which locales are supported on your system. /etc/env.d/02locale defines which of these locale you are actually using by setting LANG and LC_* environment variables. Files in /etc/env.d/ end up in /etc/profile.env (by running the env-update command), which is evaluated from /etc/profile and as such by every shell. If you want different settings for your user, override that stuff in your ~/.bash_profile. Wonko
Re: [gentoo-user] Everything disappeared from world list
Doug Hunley writes: On Wed, Jul 11, 2012 at 12:17 PM, Alex Schuster wo...@wonkology.org wrote: How would you do that? I'm currently using ~amd64 and can't yet use sets for some reason. Then you probably need portage 2.2 for this. Which will never ever become stable it seems, but I'm using it just fine for three years now. Yes, 2.2 is needed. And same here. Using it forever and no issues Well, I _had_ some issues with that when the preserved-libs feature was new, a few times after an emerge @preserved-rebuild nothing changed, and emerge @preserved-libs would emerge the same stuff over and over again. I had to delete /var/lib/portage/preserved_libs_registry manually. That was in 2009, all was fine since then. But I fear that one day one of the many many portage updates might introduce a nasty bug, and portage will be broken. The chance may be small, but then there already were over 150 portage updates in my case. Is emerge @preserved-rebuild (with FEATURES=preserve-libs) also a 2.2 feature? I like this most, I no longer need to use revdep-rebuild. I always considered having to use it a bug, fixing broken things after breaking them, instead of preventing breakage. Indeed, also a 2.2 thing and the biggest reason I use it honestly Me too. revdep-rebuild reports stuff that is broken, and I do not want to have anything like that on my systems. What if I need such a broken application before revdep-rebuild has fixed it? For a large package this might take hours. And what if I run into a build problem? Using the preserved-libs feature I feel much safer. Wonko
Re: [gentoo-user] Everything disappeared from world list
Claudio Roberto França Pereira writes: On Fri, Jul 6, 2012 at 11:13 AM, Doug Hunley doug.hun...@gmail.com wrote: This is why I keep an empty world file and use /etc/portage/sets/ exclusively. I'm backing up /etc/portage anyway (package.use and friends), so it just makes sense to have 'world' in there ;) How would you do that? I'm currently using ~amd64 and can't yet use sets for some reason. Then you probably need portage 2.2 for this. Which will never ever become stable it seems, but I'm using it just fine for three years now. You have to put this into /etc/portage/package.accept_keywords: sys-apps/portage- ~* Or just enable a specific version, and put that in your local overlay, if you don't trust the updates which happen every few days. Is emerge @preserved-rebuild (with FEATURES=preserve-libs) also a 2.2 feature? I like this most, I no longer need to use revdep-rebuild. I always considered having to use it a bug, fixing broken things after breaking them, instead of preventing breakage. Wonko
Re: [gentoo-user] profiling:/var/tmp/portage/xfce-base ... Cannot create directory
v...@ukr.net writes: On Mon, 9 Jul 2012 08:51:32 +0800 microcai micro...@fedoraproject.org wrote: You forgot to disable pgo when you compile some xfce-session . What is 'pgo'? Thanks, I just used this question to finally look up the man page of the euses command, which I thought would give the description of use flags. It turns out that it's simply 'euses pgo', which is much less to type than my usual 'grep :pgo /var/portage/tree/profiles/use.*'. Anyway, it returns that only firefox and torbrowser make use of this optimization for gcc-4.5. About the free inodes question, use df -i for that. Wonko
Re: [gentoo-user] nxserver - Connection with remote peer broken
Joseph writes: After upgrade I enabled KMS in the kernel for my for my Radeon card and now I can not connect client to nxserver I'm getting an error message: Connection with the remote server was shut down. Please check the state with your remote connection. My remote ssh connection is working OK. Here is the log from remote nxserver: [...] nxagentXkbGetRules: WARNING! Failed to stat file [/usr/X11R6/lib/X11/xkb/rules/xorg]: Unknown error -1. /usr/lib64/NX/bin/nxagent: symbol lookup error: /usr/lib64/libXtst.so.6: undefined symbol: _XGetRequest [...] nxagent needs the _XGetRequest symbol, and looks for it in libXtst.so.6. It's not directly defined there, but in libX11.so, ldd -r /usr/lib64/libXtst.so.6 will show that this library is also being searched for. There the symbol is found on my system, but I assume nm -D /usr/lib64/libX11.so | grep XGetRequest will list nothing for you. I am using net-misc/nx-3.5.0-r3 just fine. And I also have xorg-server-1.12.2 installed (with use flags ipv6 nptl udev xorg). There is a discussion about older versions of libX11.so not having this symbol, but this should not matter here: http://www.linuxquestions.org/questions/slackware-14/radeon-problem-4175414183/ So, I don't know what is going on. Wonko
Re: [gentoo-user] better alternative to NX
Joseph writes:
Are there better alternative to NX?
I don't know. The commercial original version from nomachine.org
(net-misc/nxserver-freeedition) was said to be somewhat faster than FreeNX
(net-misc/nxserver-freenx), not sure if this is still true, but as
FreeNX is dead, it's probably right. For up to two users, original NX is
free, you only need to pay for more users.
Then there's net-misc/neatx from Google, I don't know about this. The
successor of FreeNX is X2Go (net-misc/x2go{server,client}), but I did not
have any success with that yet. I didn't try much, though.
Wonko
Re: [gentoo-user] nxserver - Connection with remote peer broken
Joseph writes: On 07/09/12 19:07, Alex Schuster wrote: Joseph writes: Here is the log from remote nxserver: [...] nxagentXkbGetRules: WARNING! Failed to stat file [/usr/X11R6/lib/X11/xkb/rules/xorg]: Unknown error -1. /usr/lib64/NX/bin/nxagent: symbol lookup error: /usr/lib64/libXtst.so.6: undefined symbol: _XGetRequest [...] nxagent needs the _XGetRequest symbol, and looks for it in libXtst.so.6. It's not directly defined there, but in libX11.so, ldd -r /usr/lib64/libXtst.so.6 will show that this library is also being searched for. There the symbol is found on my system, but I assume nm -D /usr/lib64/libX11.so | grep XGetRequest will list nothing for you. I am using net-misc/nx-3.5.0-r3 just fine. And I also have xorg-server-1.12.2 installed (with use flags ipv6 nptl udev xorg). [...] When run nm -D /usr/lib64/libX11.so | grep XGetRequest I get: 00047600 T _XGetRequest so it seems to me it finds it there, so I'm not sure why I closes the connection. It just started after I upgraded xorg and enabled KMS in the kernel. That's strange. I assume ldd -r /usr/lib64/libXtst.so.6 will output libX11.so.6, and libX11.so and libX11.so.6 in /usr/lib64/ are both links to libX11.so.6.3? There is an error: nxagentXkbGetRules: WARNING! Failed to stat file [/usr/X11R6/lib/X11/xkb/rules/xorg]: Unknown error -1. but I have no clue what to do with it. I see this here, too. I think the file has moved to /usr/share/X11/xkb/rules/xorg, if you want you can try to symlink it to /usr/X11R6/lib/X11/xkb/rules/xorg. I'd also rebuild nx, maybe this is needed after the X.org upgrade. Wonko
Re: [gentoo-user] better alternative to NX
Joseph writes: Do you have a good link how to setup net-misc/neatx on Gentoo? No. I installed Beatx once, but that was on Fedora I think. The system got another distro soon after, so Neatx was abandoned. There is no development for it any more, so the next remote desktop service was FreeNX again. I'm not having much luck with nxserver-freenx, how I have key authentication problem. nxsetup --setup-nomachine-key maybe? This manual step is necessary after installation. Wonko
Re: [gentoo-user] freenx - no /var/lib/nxserver/home/ drectory
Joseph writes: I'm setting up again freenx and following the instructions from: http://www.gentoo-wiki.info/HOWTO_FreeNX_Server but after running: emerge -av nxserver-freenx nxsetup --install --setup-nomachine-key --clean --purge the installation did not create /var/lib/nxserver/home/.ssh/client.id_dsa.key file At which point that file should be created? I thought that was done automatically for me by nxserver --setuip-nomachine-key, but you can try the nxkeygen command, which should create the file. Wonko
Re: [gentoo-user] Gentoo Wiki test page?
Walter Dnes writes: I've got USB devices automounting on mdev and I'd like to set up a page on wiki.gentoo.org, describing the steps, and link to it from https://wiki.gentoo.org/wiki/Mdev Is there a way to play around on a test page and import it to the final destination? Another option might be to simply work on the new page, and not publicize the URL until it's ready. You can use the sandbox as a playground: https://wiki.gentoo.org/wiki/Gentoo_Wiki:Sandbox Wonko
Re: [gentoo-user] Distorted Mirroed Overlapping screen with ATI Rage card
Christopher Lemire writes: I tried disabling r128 and building the raedon driver into the kernel. However, I am getting the message raedon module not found. It's not a module. It's built into the kernel. Xorg.0.log. I mistakenly typed raedom, but then fixed it to raedon, so if you see that in log, I corrected it and tried starting X again. Do you have VIDEO_CARDS=radeon in make.conf? That should pull in x11-drivers/xf86-video-ati, I think that's what has the missing radeon X module, which is different from the kernel module. I think! http://vpaste.net/K3N1v Whoa, now THAT is a weird Xorg.0.log file! Um, it's the wrong file :) Wonko
Re: [gentoo-user] Failed gnome3 upgrade
William Kenworthy writes: On Thu, 2012-07-05 at 14:24 +0200, Philipp Riegger wrote: On 05.07.2012 14:00, William Kenworthy wrote: Any hints where to look next as the gdm logs arent very informative. Hmm, do you have a user polkitd with invalid home directory in /etc/passwd? Not polkitd but a polkituser - what should I have? polkituser:x:118:1021:added by portage for polkit:/dev/null:/sbin/nologin I have a similar entry, but with UID 991. But I wonder why you don't have a polkitd user. Maybe this is your problem? The entry should be like this, at least for sys-auth/polkit-0.106-r7: polkitd:x:129:991:added by portage for polkit:/var/lib/polkit-1:/sbin/nologin For me, it's still set to /var/empty, which also seems to work. When the directory was /dev/null, one of the symptoms was Gnome3 refusing to start. But maybe my setup is different because I'm on ~amd64, and the polkitd user has been introduced by a version of polkit later than yours. Wonko
Re: [gentoo-user] Is there a Gentoo Expert in NYC?
David Kuhl writes: I need a Gentoo Expert to take a look at this. Are there any in NYC around West 72nd? I've got to get this laptop working. After following the recomendations on building the latest kernel I don't have a system anymore. Everything on my LVM2 partitions are gone or at least not working. Use a live-cd (I recommend systemrescuecd), and see what the lvscan oer lvdisplay command gives. Does it still find the LVMs? Then your data should be okay. You may need to enter 'vgchange -a y' to make them active. Then try to mount the partititions. pvscan / pvdisplay and vgscan / vgdisplay also give some information. I added genkernel ~amd64 to the portage/package.keywords as suggested to get the latest genkernel to build . . . which it did. Now the kernel (3.3.8) which was suppose to fix the xorg-server problem destroyed the system, I can't boot to the old kernel either it's the same thing. This is getting worse. How can I fix this? Is there anyone near by? Thanks. What exactly happens? Do you have a separate /usr? I assume you only see the root partitition, and most services do not start because of missing partitions? Do you get a rescue shell only? Does lvscan work there? Maybe vgchange -a y and exit will continue the boot process? Give us some information on what exactly happens, and we'll try to help you. I can imagine how you feel, I've been there, too. If you cannot open your partitions even from a live-cd, maybe someone will allow you to log in to his system via ssh -L :1234:localhost:22 user@host, and the helping person could then log into your system with ssh -p 1234 localhost and have a look. I'd do this, but here it's time to get some sleep now, so I can't help at the moment. Wonko
Re: [gentoo-user] Re: Gentoo Laptop Looks to be Trashed
Philip Webb writes: 120704 Nikos Chantziaras wrote: Am 03.07.2012 20:00, schrieb David Kuhl: What's the best way to get this back without loosing the system? ... install Ubuntu (or one of its spin-offs). Ik ! Why does he need KSM ? -- Google found an article which advises : if you need to run multiple virtual machines on a host where memory is a constraint, then KSM is your solution. Why would he need to do that on a laptop ? KMS != KSM :) Wonko
Re: [gentoo-user] Suspend, hibernate and mount stopped working as a regular user.
Yohan Pereira writes: This happened after a recent world upgrade. I am currently using kde 4.8.4. I can however suspend, hibernate using the pm-utils as root. Google has lead me to believe this has something to do with consolekit. Or maybe sys-auth/polkit? There were issues lately with a nonexisting home directory, the elog message tells us to fix this with: usermod -d /var/lib/polkit-1 polkitd Wonko
Re: [gentoo-user] Is it safe to change the start of the first partition?
Helmut Jarausch writes: modern fdisk puts the first partition at block 63 while older version have put it at block 1. No, it's older versions that use 63, while the new fdisk uses 2048. This way the new 4K sectors of huge drives are aligned well. It does not need to be 2048, as long as it's dividable by 8. But fdisk apparently does not allow smaller values. Now, I'm going to upgrade an older system. Is it safe to repartition it by letting the first partition start 63 and shortening the first partition by 62 blocks? (I will loose the first partition which doesn't matter in my case.) I thought this would work, but I'm not sure now. But why do you want to change the partition start anyway? Wonko
Re: [gentoo-user] Not receiving emails
Hi David! I only recieved one email since signing up on this list yesterday. I expected to see more traffic. There's nothing going to spam, I'm not sure it I should repost or not. The forum doesn't seem to have it either. Your mails arrive just fine, I see six altogether. You can see them here, for example: http://old.nabble.com/gentoo-user-f12640.html And your initial question about the initramfs has been answered already: https://bugs.gentoo.org/show_bug.cgi?id=424579 Posted mailed, Wonko
Re: [gentoo-user] Gentoo Laptop Looks to be Trashed
David Kuhl writes: I'm so stuck with this Gentoo laptop. It started with a standard update which was the first in three months. Then when the X didn't run due to xorg-server getting upgraded, the 3.3.8 gen kernel was suppose to be built with KSM. That failed due to mkfs_ext2.h. The a beta of genkernel was used and which built the kernel and initramfs, but the rest of the machine looks like it's gone. All the LVM2 partitions are broke: /home /var /opt /usr. What's the best way to get this back without loosing the system? Thanks Did you use genkernel with --lvm, or is it activated in genkernel.conf? Does lvscan still list the logical volumes? Assuming you get to a prompt and can use your system without the missing partitions, if not you need to use a live CD. Maybe they are just inactive, and vgchange -a y will make them active and you can continue booting? Does your old kernel (the one without KMS and so without X, but with a working text console) find the LVMs? Wonko
Re: [gentoo-user] Grub2 and is the upgrade a tooth puller.
Dale writes: Neil Bothwick wrote: Seriously though, why not use make install? That way you know the right files get copied and given the expected names. Because I name my kernel and config the same thing. I also don't like the way it does that link thingy it does. It seems to expect to keep only two kernels around and I'm real bad to have more than that, sometimes way more than that. Plus, if I do it myself, I know what I am doing. If I use make install, I don't know if something was changed in how it does it. It's just me being me. lol No, me too. In my history of using Linux, I very often had trouble with new kernels. When I had an NVidia graphics card, that often caused trouble. Nowadays it's ISDN sometimes. The fact that I build a new kernel does not necessarily mean that I want to boot it yet. And I want to keep old kernels around, several, not only the last one. I do not reboot often, so sometimes multiple kernel versions have been installed since the last reboot. I would not want my current kernel to have vanished, just in case I will need it again when the new ones do not work. With kernel = 3.4.3 I had two weird panics in the last two weeks, I am still using it, but maybe I will need 3.3.5 again, which would be the sixth-newest one. And I think that maybe hibernation and ISDN used to work longer ago, maybe I will give the last 2.6 kernel a try again. So I use genkernel to build and install new kernels, and modify grub.conf manually to add this kernel to the menu. The .config is also being copied to the boot partition, using a similar name as the kernel and the initramfs. I'll continue to use the old Grub, as it's working fine for me. I understand it very well, probably because there is not much to understand. Ususally it only takes root (hd0,0) and setup (hd0) commands to install, and the config file is very easy to edit. I had some painful experiences with Grub2 on Ubuntu, and did not understand for a while what to do. There's too much automagic involved, scripts creating the actual grub.cfg file. Config files in /etc/grub.d and /etc/default/grub. There's grub-install, grub-setup, update-grub, and what else. The Grub menu is shown only if there are multiple operating systems installed, it took me quite a while to figure out how to make it appear at all. Gentoo is a distro for experts they say, but for me it seems to be actually easier than other distros like Ubuntu which are supposed to be easy. Yes, they are, but only when you do standard things. If your setup is somewhat special, it's actually harder to figure out what is necessary to do, at least that's my experience. Wonko
Re: [gentoo-user] Hanging mount
Alex Schuster writes: Helmut Jarausch writes: On 06/27/2012 04:44:34 AM, meino.cra...@gmx.de wrote: i accidentally tried to mount the extended partition /dev/sda4 from this disk: [...] which results in a hanging mount process, which cannot be killed. I cannot tell you if kernel hanging in this case is normal. But, if there is any problem during mount, the kernel seems to hang. As far as I remember, it only has a very long time out, but it will 'kill' that mount request some time. Let's see how long 'very long' is, it's hanging for over one hour now. I will wait some more hours, but that would be one really long timeout. All I can say now is that the timeout, if there is one, must be larger than 24 hours. Isn't this weird? Wonko
Re: [gentoo-user] Hanging mount
Helmut Jarausch writes: On 06/27/2012 04:44:34 AM, meino.cra...@gmx.de wrote: Hi, i accidentally tried to mount the extended partition /dev/sda4 from this disk: [...] which results in a hanging mount process, which cannot be killed. I was urged to use the sysreq-key to reboot and get rid of that process. This happens with kernel 3.2.21 and 3.4.4. Is this the expected bahviour? At least it happens here, too. And the mount process uses 100% of one of my cores, this was not expected. I cannot tell you if kernel hanging in this case is normal. But, if there is any problem during mount, the kernel seems to hang. As far as I remember, it only has a very long time out, but it will 'kill' that mount request some time. Let's see how long 'very long' is, it's hanging for over one hour now. I will wait some more hours, but that would be one really long timeout. Wonko
Re: [gentoo-user] Re: Nvidia 295.59 driver on kernel 3.3.8
Nikos Chantziaras writes: On 24/06/12 13:49, Samuraiii wrote: Hello, yesterday I run emerge and run into problem with Nvidia 295.59 driver on kernel 3.3.8. It won't compile (also broadcom-sta driver won't rebuild) [...] Linux-3.2.12-gentoo-x86_64-Intel-R-_Core-TM-2_Duo_CPU_T8300_@_2.40GHz-with-gentoo-2.1 You're running 3.2.12. Your active kernel source is 3.3.8. This cannot work. Why not? Either use 3.2.12 as your active source, or build a 3.3.8 kernel. He has built 3.3.8 already, he is just not using it yet. Wonko
Re: [gentoo-user] Nvidia 295.59 driver on kernel 3.3.8
Samuraiii writes: yesterday I run emerge and run into problem with Nvidia 295.59 driver on kernel 3.3.8. [...] Unpacking source... Unpacking NVIDIA-Linux-x86_64-295.59.run to /tmp/portage/x11-drivers/nvidia-drivers-295.59/work Source unpacked in /tmp/portage/x11-drivers/nvidia-drivers-295.59/work Preparing source in /tmp/portage/x11-drivers/nvidia-drivers-295.59/work ... * Converting /kernel/Makefile.kbuild to use M= instead of SUBDIRS= ... [ ok ] Source prepared. Configuring source in /tmp/portage/x11-drivers/nvidia-drivers-295.59/work ... Source configured. Compiling source in /tmp/portage/x11-drivers/nvidia-drivers-295.59/work ... * Preparing nvidia module make -j7 HOSTCC=x86_64-pc-linux-gnu-gcc CROSS_COMPILE=x86_64-pc-linux-gnu- 'LDFLAGS=-m elf_x86_64' ARCH=x86_64 IGNORE_CC_MISMATCH=yes V=1 SYSSRC=/usr/src/linux SYSOUT=/lib/modules/3.3.8-gentoo/build CC=x86_64-pc-linux-gnu-gcc clean module Everything is very similar up to here, but I do not get any of your later output. I built against gentoo-sources-3.4.4 on ~amd64. No idea what's the problem with your setup. Wonko
Re: [gentoo-user] dev-python/PyQt4-4.9.2 requires SIP v4.13.3 or later
Dale writes: I'm doing my KDE4 upgrades and ran into this: [...] Error: This version of PyQt requires SIP v4.13.3 or later [...] I notice tho that portage seems to have failed to notice this was needed. Should I file a bug report or is this just me? File a bug. There is a DEPEND line in the ebuild, but that states it needs =dev-python/sip-4.13.1, not 4.13.3. Wonko
Re: [gentoo-user] OT: Linus ranting about Gnome3
Pandu Poluan writes: Just in case anyone missed it: https://plus.google.com/102150693225130002912/posts/UkoAaLDpF4I That URL cannot be found. This seems to work for me: https://plus.google.com/102150693225130002912/posts/StQB1ftUp8D Wonko
Re: [gentoo-user] OT: Linus ranting about Gnome3
Alex Schuster writes: Pandu Poluan writes: Just in case anyone missed it: https://plus.google.com/102150693225130002912/posts/UkoAaLDpF4I That URL cannot be found. This seems to work for me: https://plus.google.com/102150693225130002912/posts/StQB1ftUp8D Argh, but it's the wrong post. How did this end up imy clipboard? This one is correct (your URL, with a lowercase 'I' at the end): https://plus.google.com/102150693225130002912/posts/UkoAaLDpF4i Wonko
Re: [gentoo-user] dracut + UUID : a problem solved
Mick writes: PS. [OT] what key am I supposed to press to be able to see some more verbose output on the console while *ubuntu is booting? Press the shift key during startup, so the Grub menu will appear. Then E to edit, and remove the 'quiet' kernel parameter. The Grub menu still does not appear? Then you may have to comment the GRUB_HIDDEN_TIMEOUT* values in /etc/default/grub, and run update-grub. It's so simple, isn't it! Wonko
Re: [gentoo-user] Adding a use flag: hwdb
Alan McKinnon writes: On Sun, 10 Jun 2012 19:57:42 -0700 Chris Stankevitz chrisstankev...@gmail.com wrote: Today emerge is asking me to add =sys-fs/udev-171-r6 hwdb to package.use to appease udisk. Just as before, this looks fishy to me and I would like to get your opinion about how to properly satisfy this seemingly system-required-use-flag. My gut instinct is that this USE flag requirement should be handled by the Gentoo team in the profile or in some other place that I never look at. It's not a question of system vs personal preference It's a question of treating USE flags as global or local in scope. The USE variable in make.conf applies globally, and the package.use file applies to individual packages. Portage itself couldn't care how you view the use of your flags, so you should organize them how you see fit. See here: $ euses -sf hwdb sys-fs/udev:hwdb - read vendor/device string database and add it to udev database The flag applies to only one package. Some USE flags do not have a sane default so there's no choice the devs can make on your behalf. Especially convenience features like this one - some folk want it, others do not. So the devs delegate the choice to you to apply in any way you see fit. I think Chris' question is more about why he has to manually activate this USE flag, as it seems to be necessary anyway, in his case. Looking at my own setup, I have built udev with hwdb, but I do not know why. There is no hwdb in make.conf or package.use, emerge --info does not show it, but emerge --info sys-fs/udev does. I'm on version 182-r3. What do you guys think? Should I append udev hwdb to package.use right after my long list of personal preference customizations? Yes. Or use the --autounmask-write option for emerge (may need the new portage), this will add it automatically, with a comment. Wonko
Re: [gentoo-user] Building a binary package without installing
Neil Bothwick writes: On Wed, 06 Jun 2012 13:25:38 +0300, Nikos Chantziaras wrote: I wish to build a binary package of glibc-2.14.1-r3. I don't want to actually install it on my system. Just build a tbz2 for it to use on another system. Is there a way to do that? emerge -B atom Nope, this does not work either. Wonko
Re: [gentoo-user] Portage telling me what it's doing
Andrew Lowe writes: I've just kicked off an emerge -NuD world and will now head out for a while. My emerge has to do, amongst others, gcc, libreoffice, Firefox Thunderbird. Now when I get back I'll want to know where the emerge is up to so, in my ignorance of portage/emerge in great depth and with only compiler output spewing up the screen, I'll fire up another terminal, and now don't laugh, I'll do emerge --pretend -NuD world. That will tell me what's currently being compiled as it will be the top thingy on the list. There has to be a better way Using the --jobs / -j option to emerge will give a nice output, omitting all the compiler output. It can also speed up emerging, because it will build packages in parallel. I really really like this feature. Is there a way so that the terminal that the emerge is happening in can display additional info? At the moment, I get: /home/agl: emerge can I get, say: /home/agl: emerge www-client/firefox by setting some config variable? Yes, but I do not know how. Failing that is there a log file that lists just what's been emerged, not a whole lot of checking this, checking that, compiling this file, linking that library, whoops, error here... sort of thing. tail -f /var/log/emerge.log, or better emerge app-portage/genlop, then use genlop -l | tail. Wonko
Re: [gentoo-user] Can't emerge any gcc
Ezequiel Garcia writes: I found something strange. What should I have in /usr/i686-pc-linux-gnu? Currently: $ ls bin binutils-bin gcc-bin lib Same here on ~amd64, except for an additional i686-pc-linux-gnu/lib/ directory, containing broken symlinks only, which belongs to no package. I may be missing sys-include dir? any of you have it? No, nowhere. Neither sys-include nor sysinclude. Perhaps I messed up stuff when playing with crossdev, and friends ;) I did that, too, maybe those broken symlinks are results from that? Now I'm building i686 stuff in a chroot. Wonko
Re: [gentoo-user] How can I control size of /run (tmpfs)?
Dale writes: Jarry wrote: On 26-May-12 22:01, Dale wrote: Jarry wrote: after updating baselayout from 2.0.3 to 2.1-r1 /run is mounted as tmpfs. But I can not find any mount-option for controlling how much memory is (or could be) used for it. Filesystem 1K-blocksUsed Available Use% Mounted on tmpfs8223848 224 8223624 1% /run I know it does not use 8GB right now, yet I'd like to reduce it to some lower value, not half of my physical memory. How can I do it? Can I simply add line in fstab like: none /run tmpfs size=128m 0 0 ??? Just try it :) I don't know if this would work, probably yes. But you can change it later with mount -o remount,size=128m /run Holy smoke ! Mine is doing the same thing. tmpfs 7.9G 260K 7.9G 1% /run But I also have this: tmpfs 7.9G 0 7.9G 0% /var/tmp/portage Now have a look at /dev/shm... So, between those two, I could run out of ram since I have 16Gbs. But only if you copy stuff to /run yourself, otherwise this will never happen. There is now TWO people that needs a answer to this question. Why does it need that much anyway? It looks to me like a few hundred Mbs, like Jarry posted, would be plenty. Jeepers creepers. lol It doesn't need it, it's just the maximum sitze, which it will never reach. I suppose default size for tmpfs is half of physical memory, if it is not configured somewhere else. BTW, is there any way to turn this great feature off? What is it good for? I do not see any advantage in having /run on tmpfs... In case of power failure or lockup, the contents are lost, and will not cause confusion on the next reboot when /run is still populated by stuff. Just an idea, I do not know if it would really matter. But it does no harm, so why not juest keep it like it is. I had no idea it was doing this either until your post. I got the same questions as you do. Why is it there? Why so much is allocated to it? Where can we change the settings for this questionable feature? I'm hoping someone will come along and answer both our questions. I'm really hoping for a place we can change the settings. I don't mind it being there so much if it is useful. I would like to know its purpose tho. I don't know the details, but I'd think it does not matter. There will nothing be put into /run that uses a lot of memory, so it will never actually use its default size of half of your RAM. Wonko
Re: [gentoo-user] Epiphany freezes
Jakub Daniel writes: every time i open facebook and enter anything into the search field and click any of the results epiphany freezes, then i can only force quit. I know this is not the best description but could anyone, please, suggest how to try to debug this? Does it also happen with another user who never used Epiphany before? This way you can test whether it is affected by some of your settings. Wonko
Re: [gentoo-user] udevd boot messages
Tanstaafl writes: *Especially* for servers, there really, REALLY needs to be a way to see this kind of warning BEFORE updating... ie, the warning should be printed to the screen during an 'emerge -pvuDN world' or something, so I know that a reboot will be required for this update. /pet-peeve Indeed! I think eselect news read should show this, at least. Wonko
Re: [gentoo-user] [OT] CPU temperature monitoring?
walt writes: Now it's hotter than Hades here and I'm very much aware of the fan noise, but I can't tell if the fan is beginning to fail (the noise sounds a bit harsh to me) or something is merely controlling the speed appropriately. The machines BIOS has no settings whatever concerning the fan or temperature warnings, etc. so I have no idea how to find out the CPU temp. Strange. The k10temp kernel module loads automatically at boot with no errors, so I just hope something (somewhere) is taking care of this stuff automatically. But I'm only hoping, not knowing. Any ideas how to find out for sure? emerge sys-apps/lm_sensors, run 'sensors-detect', and hopefully the 'sensors' command will show you the fan speeds and temperatures then. And maybe the output makes sense. Here it does that only partially, this is what it looks like: k10temp-pci-00c3 Adapter: PCI adapter temp1: +9.9°C (high = +70.0°C) (crit = +70.0°C, hyst = +67.0°C) fam15h_power-pci-00c4 Adapter: PCI adapter power1: 86.04 W (crit = 95.04 W) nct6775-isa-0290 Adapter: ISA adapter Vcore:+0.93 V (min = +0.00 V, max = +1.74 V) in1: +1.66 V (min = +0.00 V, max = +0.00 V) ALARM AVCC: +3.30 V (min = +0.00 V, max = +0.00 V) ALARM +3.3V:+3.30 V (min = +0.00 V, max = +0.00 V) ALARM in4: +0.06 V (min = +0.00 V, max = +0.00 V) ALARM in5: +1.86 V (min = +0.00 V, max = +0.00 V) ALARM in6: +0.06 V (min = +0.00 V, max = +0.00 V) ALARM 3VSB: +3.42 V (min = +0.00 V, max = +0.00 V) ALARM Vbat: +3.50 V (min = +0.00 V, max = +0.00 V) ALARM fan1:1155 RPM (min =0 RPM, div = 8) ALARM fan2:1054 RPM (min =0 RPM, div = 128) ALARM fan3: 540 RPM (min =0 RPM, div = 64) ALARM fan4: 0 RPM (div = 128) SYSTIN: +37.0°C (high = +0.0°C, hyst = +0.0°C) ALARM sensor = thermistor CPUTIN: +37.0°C (high = +80.0°C, hyst = +75.0°C) sensor = thermistor AUXTIN: +127.5°C (high = +80.0°C, hyst = +75.0°C) ALARM sensor = thermistor cpu0_vid:+0.000 V intrusion0: ALARM Wonko
