Re: [gentoo-user] crossdev/x86dev screwdup ?!
On Mon, Oct 01, 2012 at 12:50:40PM +0200, meino.cra...@gmx.de wrote: hi, Something broke my crossdev installation... I installed crossdev and did a crossdev --target armv7a-unknown-linux-gnueabi which produces a useable cross development toolchain. A few days ago there was an update related to thsi crossdev toolchain (dont remember exactly), which also installs fine. But now, calling armv7a-unknown-linux-gnueabi-gcc results in gcc-config: error: could not run/locate 'armv7a-unknown-linux-gnueabi-gcc' [1]25787 exit 1 armv7a-unknown-linux-gnueabi-gcc . Calling armv7a-unknown-linux-gnueabi-gcc-4.7.2 gives the exspected error message no input files. The output of gcc-config -l is [1] armv7a-unknown-linux-gnueabi-4.7.2 [2] x86_64-pc-linux-gnu-4.4.7 [3] x86_64-pc-linux-gnu-4.5.4 * seems that the armv7a compiler isn't 'active' in the gcc-config output (no asterix after the [1] option), did you try to 'reactivate' it with gcc-config, ie # gcc-config 1 ? My output here: julka ~ # gcc-config -l [1] armv4tl-softfloat-linux-gnueabi-4.4.6 * [2] armv4tl-softfloat-linux-gnueabi-4.5.3 [3] x86_64-pc-linux-gnu-4.4.4 [4] x86_64-pc-linux-gnu-4.5.3 * (Though it is possible the asterix won't show up if there is only one option, but reactivating it should do anything bad...) . Before i screw up my native toolchain I better ask for help, how I can reactivate my crossdev toolchain, so that it is available via armv7a-unknown-linux-gnueabi-gcc . How can I acchieve this? Thank you very much in advance for any help! Best regards, mcc yoyo
Re: [gentoo-user] crossdev/x86dev screwdup ?!
On Mon, Oct 01, 2012 at 02:13:49PM +0200, YoYo Siska wrote: On Mon, Oct 01, 2012 at 12:50:40PM +0200, meino.cra...@gmx.de wrote: The output of gcc-config -l is [1] armv7a-unknown-linux-gnueabi-4.7.2 [2] x86_64-pc-linux-gnu-4.4.7 [3] x86_64-pc-linux-gnu-4.5.4 * seems that the armv7a compiler isn't 'active' in the gcc-config output (no asterix after the [1] option), did you try to 'reactivate' it with gcc-config, ie # gcc-config 1 ? My output here: julka ~ # gcc-config -l [1] armv4tl-softfloat-linux-gnueabi-4.4.6 * [2] armv4tl-softfloat-linux-gnueabi-4.5.3 [3] x86_64-pc-linux-gnu-4.4.4 [4] x86_64-pc-linux-gnu-4.5.3 * (Though it is possible the asterix won't show up if there is only one option, but reactivating it should do anything bad...) ... shouldn't do anything bad... ;) yoyo
Re: [gentoo-user] crosscompiling...the point of view
On Mon, Aug 06, 2012 at 06:37:07PM +0200, meino.cra...@gmx.de wrote: Paul Hartman paul.hartman+gen...@gmail.com [12-08-06 17:36]: On Mon, Aug 6, 2012 at 8:01 AM, meino.cra...@gmx.de wrote: Hi, I am asking, because I found not only one description of somehow complicated setups to compile a distribution (namely gentoo) for a platform A (Beaglebone TI OMAP4) on that platform with distcc to speedup things or with emulated chroot environments based on qemu... I thought it would be the easiest to compile the whole stuff on a host system B with a crosscompilation toolchain...but may be I have overlooked something important... Seems the best way to me, too. So - is there any logical reason, which prevents the process of the compilation of a complete distribution/rootfs/boot-mechanism for a platform A on a hostsystem of the platform B if the cross compilation toolchain is already installed on B and no emulated environment is wanted? So you want to install the packages into a virtual filesystem image on the compiler machine to create a whole disk image for the target, basically? Hmmm. Maybe something like Scratchbox can help with this. Hi Paul, ...yes, exactly. But theproblem remains...is there a logical reason, which renders this attempt useless ? Crosscompilation is the preffered way, but there can be a lot of problems with it . The cross-toolchain isn't a problem, gennto has a nice tool called crossdev that will generate it for you and with a bit of env. variables and make.conf tweaks, you can easily do cross-compile emerges into a specific sysroot (gentoo has nice documentation at [1]) The problem is usually with build systems that don't take croos-compilation into account: things like getting paths wrong (including the path to the root of your croscompiled stuff or using paths in your host system, (older libtool used to generate .la files with paths in your host system)), using information from your life system instead of the target system, not correctly handling cases where, as part of the installation, tools are compiled that are then directly executed. Most build systems have ways to correctly express these things, but authors often get them wrong etc... Then there are cases where some post-installation steps are required to be run in the target system (remeber those Updating desktop mime database or Updating shared mime info database messages after most desktop apps ?) that usually require executing the installed programs, which means that either a host version has to be compiled (which can correctly work with stuff in the targe root directory) or they have to be somehow started on the first boot of the target system. Because of this, doing it in an emulated enviroment or on the actual device makes most of these problems go away. So if somebody doesn't have time to sort out these cross compilation problems but doesn't mind if the installation runs few times longer (but without problems), its better for them... I last did a cross compiled install of gentoo about two years ago, mostly it was without any problems, though there were some problematic packages (IIRC python used to be a problematic one, not sure how it is now). Recently I have done a few emulated root installs and they went absolutely without any problems (at least as far the 'compile packages and get a rootfs) step is concerned... getting things to run correctly on an embedded device without any display etc has its own challenges ;) yoyo [1] http://www.gentoo.org/proj/en/base/embedded/cross-development.xml
Re: [gentoo-user] distcc to compile Gentoo on the laptop
On Thu, Jun 14, 2012 at 01:49:16AM +0530, Yohan Pereira wrote: On Wed, Jun 13, 2012 at 08:43:20AM +0200, Daniel Wagener wrote: On Wed, 13 Jun 2012 01:04:19 -0500 What about a different approach: gentoo in a VM on the desktop Would that not be much easier? Of course some processor power is used for the VM itself, but it should still significantly decrease compilation time on the laptop. Plus you can easily equip other machines with that VM and use their power too. What about a chroot? that would be much lighter compared to a VM. Alternativly you can use the chroot to build binnary pkgs that can be installed on the laptop. There are 3 solutions I can think of: 1. Get a cross compiler for i686 working under fedora (have no idea about that, except manually setuping and compiling gcc) and setup the Fedora's distcc to use that correctly (this should go more or less according to the docs on the gentoo wiki regarding cmake and i686 vs amd64) 2. Create a gentoo chroot on the Fedora OS. Much lightweight that a virtual machine, you can setup the cross toolchain and distcc according to the wiki, you just have to play a bit with how to start the distcc inside the chroot (a plain /etc/init.d/distcc start inside the chroot won't work). Note that you can run a i686 chroot in an amd64 system, so you actually don't need to set up any fancy crosscompiler inside the chroot. Just run the chroot as 'linux32 chroot /mnt/chrooot ...' to make sure uname and similar get correct info... 3. Mount the laptops root filesystem through nfs on the fast computer (use no_root_squash on the laptop export to have correct root access to files) bind-mount something local (disk or tmpfs if you have enough mem) over /var/tmp/portage, chroot into it (don't forget to mount /proc, maybe /sys and maybe bind-mount /dev, though that should not be needed and don't forget 'linux32 chroot') and run emerges there... you will actually be running everything on the fast computer, only the access to the laptops disk will be through the network. With a fast network it should be a lot faster then working ont the slower notebook (note that if you bind-mount /var/tmp/portage inside the chroot, most of the compilation will be working with a local disk...) Compared to the distcc method, even the configure phases will be much faster... I do this often with my Pentium M 1.6Ghz thinkpad laptop and my quad core amd64 desktop... yoyo
Re: [gentoo-user] Fetch List
On Tue, Jun 05, 2012 at 01:37:11PM +0200, Silvio Siefke wrote: Hello, is there a option that i can build a list of Requirements to download of a ebuild? I want build inkscape, but i has only mobile connection at moment. Can i build a list which i can then later load in a cybercafe? In the Handbook i found only a list for world. Thank u. use -p -f (pretend and fetchonly) emerge --pretend --fetchonly package... it will output a line per each file/url you need to fetch (and put in the distfiles dir), one line can contain multiple alternate urls (if there are alternative mirrors etc...), you need to fetch only one of them IIRC, it has a small downside, that it also lists the files you allready have downloaded in distfiles (ie those for which emerge -pv would show 0kB as size of downloads...) so you might check the list against what you allready have... yoyo
Re: [gentoo-user] Portage telling me what it's doing
On Sat, Jun 02, 2012 at 02:44:31AM -0400, Walter Dnes wrote:
On Sat, Jun 02, 2012 at 02:08:39PM +0800, Andrew Lowe wrote
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
Are you using konsole ? As walter said, emerge sets the title of the
shell/terminal window to include the current package, however konsole
does not (by default) show the titles set by programs, but is configured
to only show the working dir and command name...
Just go to Settings - Edit current profile - Tabs (tab ;) and
you will see two things: Tab title format and Remote tab title format
(the defautl for the first is '%d: %n' == 'current dir: current
command', which is exactly what you wrote), just change them to '%w' and
you will get what every other decent terminal app shows ;) (you can
click the Insert dropdown to see other options)
yoyo
I use xterm under ICEWM (a simple WM). The title bar at the top of the
xterm lists how far in the list you are, and the current ebuild...
emerge:(1 of 2) www-client/midori-0.4.3 Compile
see attached top few lines of a screen shot. Note that even if you
minimize the xterm, you can still see the info by doing either of...
* holding down {ALT-TAB} to bring up the programs menu
* hovering the mouse pointer over the location on the program bar list
of running programs.
Both of these simply duplicate what shows up on the title bar.
--
Walter Dnes waltd...@waltdnes.org
Re: [gentoo-user] ppp-gentoo woes cont'd
On Tue, Mar 20, 2012 at 09:09:06PM -0600, Maxim Wexler wrote: Since route and other things *are* getting set, I have the same strong suspicion Bill and YoYo have... DNS is likely not getting set properly in /etc/resolv.conf I always assumed that DHCP was writing this file automatically, so I never checked, but this time I made sure to check and viola! there they were. with ppp connections you are not using a dhcp client, pppd gets the nameserver ip addressess as part of the connection negotiation (if peerdns is set) and the aforemetioned script in /etc/ppp/ip-up.d/40-dns.sh writes those to /etc/resolv.conf saved a bunch of likely files across the partition from ubuntu ... did that include dropping them into place on the Gentoo side naturlich MW yoyo
Re: [gentoo-user] ppp-gentoo woes cont'd
On Mon, Mar 19, 2012 at 10:33:27AM -0600, Maxim Wexler wrote: Just got back from gentoo land. Arrrgh, gmail won't let me attach files, just sits there spinning. So I'll have to make do with pastebin. http://paste.ubuntu.com/890854/ hmm, pppd seems to bring the connection up ok and also to get the DNS servers, however it might not set them correctly... what's in /etc/resolv.conf after you connect ? can you ping directly the other side of the ppp connection (ie 161.184.0.199 according to your logs): ping 161.184.0.199 can you ping anything on internet through ip addresss: ping 8.8.8.8 (8.8.8.8 are the google public dns servers) 30-wins.sh http://paste.ubuntu.com/890854/ 40-dns.sh http://paste.ubuntu.com/890857/ 50-initd.sh http://paste.ubuntu.com/890857/ 90-ntpd.sh http://paste.ubuntu.com/890857/ The first two don't apply. 50-initd.sh, I don't quite grok. you gave the same link for the last 3 ;) but 50-initd.sh takes care of the case, when pppd is run through gentoo scrips... in that case, when you run /etc/init.d/net.ppp0 start, the net.ppp0 service is marked as 'inactive' istead of 'started' because pppd didn't connect yet... when it connects, this script marks the net.ppp0 service as started... btw, i newer used pon, i either used gentoo initscripts, or directly started pppd ;), or sometimes used kppp from kde3 but if your problem is with dns, maybe the pon script / program doesn't start pppd with the usepeerdns option... (you might consider also adding the debug options to pppd options..., but you have to find out how to do that with pon) yoyo
Re: [gentoo-user] How are Fn-F# ACPI events mapped?
On Sun, Mar 11, 2012 at 03:23:28PM -0700, Mark Knecht wrote: Hi, I'm trying to figure out how my Asus laptop maps function key events. This is being driven by an emerge message telling me that the acpi4asus package is being obsoleted and removed in 30 days and replaced by an in-kernel driver. I've removed the package and rebuilt my kernels to use this driver, and for vanilla-sources-3.2.7 the results are similar as with the acpi4asus package. don't know anything about the assus packages/drivers, but the general direction in all such drivers is to move these things where they belong: to the input subsystem, so my guess is the new driver doesn't produce acpi events, but insted create a input device and produce key press/release events on that device... (note that sleep / hibernate actions are usually still reported also throgh acpi, because some programs expect them to come from there, that would explaing your Fn-F1/sleep working) to test this, just run as root on the linux console (not in a termnal in X): showkey -s and then pres the keys (ie Fn-F4,) to stop showkeys, just don't press anything for 10 seconds... if you see numbers appearing after each keypress / release, then the key directly generates keyboard ivents and its possible you will see them directly in X, for that jus run (now under X) xev and (the window that appears must have focus / be active) press the keys again, xev will print the X keycodes/keysyms to its output... if you see reasonable names there, then you should be able to map those keys in the programs you are using (ie global hotkeys in kde, etc...) note however that qt/kde doesn't recognise some of the more exotic keysyms... (in my case XF86TouchpadToggle produced by Fn-F8 on thinkpads ;) if you can see the key in showkey -s but not in xev, the problem might be in kernel keyboard map (though i'm not sure if the x's evdev driver uses that) or in the evdev driver not mapping that key basically the kernel driver reports scan codes (what showkey -s shows), kernel translates that to keycodes (showkey -k to see them) and then X's input driver (ie evdev) translates those to the X keycodes X server again trasnlates them to keysym-s yoyo However, for vanilla-sources-3.2.9 the only key that is doing anything seems to be Fn-F1 which says 'button/sleep' (using acpi_listen) but actually just turns on the screen saver as best I can tell. Note that even with 3.2.7 most keys don't actually work, but at least they all produce acpi_listen events. The only ones that do work in 3.2.7 and earlier are: Fn-F1 - screen saver Fn-F5 - turns off screen but doesn't seem to generate an ACPI event in acpi_listen (may be hardware mapped) Fn-F11 - turn volume down Fn-F12 - turn volume up I haven't tested the external monitor one. The ones I really want to figure out are Fn-F3 F4 as they turn the keyboard lighting up and down. With 3.2.7 I had lighting, but with 3.2.9 I have no keyboard lighting at all so it will have hard to use in the dark. Before I call this a 3.2.9 regression I figured I should determine if I'm supposed to configure this stuff by hand, or maybe load some new machine specific package that sets up the mappings. Thanks in advance, Mark vanilla-sources-3.2.9 slinky events # acpi_listen button/sleep SLPB 0080 0003 button/sleep SLPB 0080 0004 vanilla-sources-3.2.7 slinky ~ # acpi_listen button/sleep SLPB 0080 0001 hotkey ATKD 005d hotkey ATKD 007e hotkey ATKD 00c5 hotkey ATKD 00c4 hotkey ATKD 002e hotkey ATKD 001a hotkey ATKD 0034 hotkey ATKD 0033 hotkey ATKD 0034 0001 hotkey ATKD 0033 0001 hotkey ATKD 0061 hotkey ATKD 006b hotkey ATKD 0032 hotkey ATKD 0032 0001 hotkey ATKD 0032 0002 hotkey ATKD 0031 hotkey ATKD 0031 0001 hotkey ATKD 0031 0002 hotkey ATKD 0030 hotkey ATKD 0030 0001 hotkey ATKD 0030 0002 hotkey ATKD 0030 0003
Re: [gentoo-user] Best file system for portage tree?
On Mon, Mar 12, 2012 at 10:23:49AM +0100, Alex Schuster wrote: José Romildo Malaquias writes: On Sat, Mar 10, 2012 at 07:36:07PM +0100, YoYo Siska wrote: mke2fs -f -b1024 -i2048 /usr/img_portage The -f option from mke2fs is to specify a fragment size and expects an argument. Do you -F (which forces mke2fs to create a filesystem, even if the specified device is not a partittion on a block special device)? I'm pretty sure that's what he meant, without the -F you need to confirm that you really want to create the FS. I forgot to mention this when I also quoted this line in my reply. yes, that's exactly what I had in mind was writing that from memory and forgot the the case... thanks for the correction yoyo
Re: [gentoo-user] Best file system for portage tree?
On Sat, Mar 10, 2012 at 03:35:05PM +, Neil Bothwick wrote: On Sat, 10 Mar 2012 14:30:15 +0100, Alex Schuster wrote: Any tips on this? Does it make sense to use a special file system just for the portage tree? What would be best? Would it help to re-create this file system from time to time in case it gets slower with every sync? I use an ext2 filesystem for portage, it's still the fastest out there. Journals are unnecessary because its such a small filesystem, and if it does get damaged I can just reformat and sync again. I use an ext2 partition in a 500MB file image on most of my computers. Its important to check the inode count on such small filesytem, as mke2fs' default inode ration for such size is 4096, which is too low for portage: dd bs=$((500*1024*1024)) count=1 if=/dev/zero of=/usr/img_portage mke2fs -f -b1024 -i2048 /usr/img_portage fstab: /usr/img_portage/usr/portage/ ext2loop,noatime 0 0 (this is from desktop, on servers I usually only mount it manually when emerging) # df -h Filesystem Size Used Avail Use% Mounted on /dev/loop0 469M 306M 139M 69% /usr/portage # df -i FilesystemInodes IUsed IFree IUse% Mounted on /dev/loop0256032 152044103988 60% /usr/portage yoyo
Re: [gentoo-user] photo viewer other than gthumb?
On Thu, Mar 08, 2012 at 10:48:30PM -0500, Daddy wrote: On March 8, 2012 at 10:34 PM Michael Mol mike...@gmail.com wrote: On Thu, Mar 8, 2012 at 10:14 PM, Daddy da...@happypenguincomputers.com wrote: On March 8, 2012 at 9:20 PM Grant emailgr...@gmail.com wrote: Can anyone recommend a photo browser/viewer other than gthumb which is in portage or an overlay? - Grant media-gfx/gqview gqview became geeqie, FWIW. I don't recall the full story, but IIRC, gqview stagnated, and geeqie is a fork. -- :wq Thanks ... I'm an old CLI dinosaur and display has been working for me fine for so many years. Something recently changed where it's not going from one file to the next in a directory. Also, there was something else I used to generate thumbnails, but I forgot. Just emerge geeqie ... btw ... gqview was just recommended to me 2 days ago in #gentoo ;) For quick commandline viewing I use xv, it is simple, just shows the image, can show multiple images (space, backspace to go through them, q to quit,) and a few other resize / rotate options, if you rightclick on the image, you get a 'menu' window with all the options and a list of the images... yoyo
Re: [gentoo-user] How can i use GRUB to boot my windows?
On Fri, Mar 09, 2012 at 04:12:40PM +0800, 赵佳晖 wrote: Hello , everyone , just now , i have installed gentoo from ubuntu 11.10 , And when i install the grub with the command : grub-install --no-floppy /dev/sda . And when i reboot , i aware that i have override my MBR. And Now how can i edit my grub in the gentoo to boot my windows 7 on /dev/sda1 ? For my windows boot manager has been override , should i use a windows live CD , and go to DOS to execute the command : fdisk /mbr , after that ,add the ubuntu and gentoo to the windows boot manager ? Can i boot the windows 7 from GRUB dircetly? PS: my grub version in gentoo is : GNU GRUB 0.97 Windows has (should have in a standard install ;) its own 'boot loader' in its partition, it usually only installs a simple 'select the active partition and boot from that' loader to the MBR, so in 99% it is enough to just add title Windows rootnoverify (hd0,0) chainloader +1 makeactive to /boot/grub/grub.conf, change (hd0,0) to whatever your windows partition is ( hd0,0 is /dev/sda1, in hdX,Y X is the disk number (sda - 0, sdb -1, ..), Y is the partition number, counted from 0, ie sda4 - hd0,3) yoyo
Re: [gentoo-user] How can i use GRUB to boot my windows?
On Fri, Mar 09, 2012 at 09:15:20PM +0800, 赵佳晖 wrote: i changed hd(0,0) to hd(0,1) , and it also comes Error 11, and when i run the command : fdisk -l .The /dev/sda1 has the boot * flag .. 2012/3/9 Mick michaelkintz...@gmail.com On Friday 09 Mar 2012 12:40:45 赵佳晖 wrote: OK, /etc/fstab: /dev/sda8 / ext3 noatime 0 1 /dev/cdrom /mnt/cdrom autonoauto,ro 0,0 grub.conf: default 0 timeout 30 splashimage=(hd0,7)/boot/grub/splash.xpm.gz title Gentoo root(hd0,7) kernel /boot/kernel-3.2.1-gentoo-r2 root=/dev/sda8 title Windows 7 rootnoverify hd(0,0) makeactive chainloader +1 it should be (hd0,0) and not hd(0,0) that would explain the Error 11, which means the string describing the disc is wrong / cannot be parsed... yoyo
Re: [gentoo-user] Clone live system as a simple backup?
On Wed, Mar 07, 2012 at 02:47:08PM -0500, Joshua Murphy wrote:
On Wed, Mar 7, 2012 at 8:55 AM, gand...@d-danks.co.uk wrote:
Hi,
I'm interested in the idea of cloning a live, complicated hardware
system onto a single external hard drive as a simple backup. I would
like this external drive to be completely bootable. What's the best
way to approach doing this? I was considering just doing a Gentoo
install from scratch but figured maybe there's a way to clone enough
of the live system to get me there less painfully?
The system I'm playing with has five 500MB hard drives with most
partitions in linked together in various forms of RAID. (1, 5 6)
That said, the total storage that this system presents KDE and the
users is about 600GB.
I have an external 1TB eSATA drive which is therefore large enough
to hold everything on this system, albeit without the reliability of
RAID which is fine for this purpose.
The system looks more or less like:
/dev/sda1 - /boot (50MB)
/dev/sdb1 - /boot copy
/dev/sdc1 - /boot copy
c2stable ~ # df
Filesystem 1K-blocks Used Available Use% Mounted on
rootfs 51612920 31862844 17128276 66% /
/dev/root 51612920 31862844 17128276 66% /
rc-svcdir 1024 92 932 9% /lib64/rc/init.d
udev 10240 476 9764 5% /dev
shm 6151284 0 6151284 0% /dev/shm
/dev/md7 389183252 350247628 19166232 95% /VirtualMachines
tmpfs 8388608 0 8388608 0% /var/tmp/portage
/dev/sda1 54416 29516 22091 58% /boot
c2stable ~ # cat /proc/mdstat
Personalities : [linear] [raid0] [raid1] [raid10] [raid6] [raid5]
[raid4]
md6 : active raid5 sdb6[1] sdc6[2] sda6[0]
494833664 blocks super 1.1 level 5, 64k chunk, algorithm 2 [3/3]
[UUU]
md7 : active raid6 sdb7[1] sdc7[2] sda7[0] sdd2[3] sde2[4]
395387904 blocks super 1.2 level 6, 16k chunk, algorithm 2 [5/5]
[U]
md3 : active raid6 sdb3[1] sdc3[2] sda3[0] sdd3[3] sde3[4]
157305168 blocks super 1.2 level 6, 16k chunk, algorithm 2 [5/5]
[U]
md126 : active raid1 sdc5[2] sda5[0] sdb5[1]
52436032 blocks [3/3] [UUU]
unused devices: none
c2stable ~ #
/dev/md3 is a second Gentoo installation that doesn't need to be
backed up at this time. md6 is an internal RAID used to back up md7
daily. It doesn't need to be backed up, but if the machine totally
failed killing all the drives that wouldn't survive so currently I
back up md126 to md6 daily, and then back up md6 weekly to an external
eSATA drive.
What I'd like to do is clone
1) /boot (sda1) including grub and everything required to make it bootable
2) back up the system portions of dev/md126 (/ )
3) Add some swap space on the external drive
4) back up /dev/md7 which is all of my VMs
5) back up /home to a separate partition on the external drive
6) back up some special things like /var/lib/portage/world and
/usr/portage/packages
My thought is that this drive is basically bootable, but over time
gets out-of-sync with the system. However should the system fail I've
got a bootable external drive with all the binary packages required to
get it running again quickly. However I can always boot the drive, do
an emerge -ek @world, and basically be back to where I am as of the
last backup.
The external drive will look something like:
/dev/sdg1 - /boot
/dev/sdg2 - swap
/dev/sdg3 - / (not including /home, /usr/portage/distfiles, etc)
/dev/sdg5 - /usr/portage/packages
/dev/sdg6 - /dev/md7
etc
I will of course have to modify grub.conf and /etc/fstab to work
from this drive but that's no big deal.
What are folks best ideas about how to approach doing something like
this?
Thanks,
Mark
Hi,
Why don't you something like bind mount the folders you want to copy
and rsync them to the eSATA disk, after creating a similar partition
layout on it. Remember to exclude system files like /proc/*, /dev/*
and /sys/* as well as the ones you want to exclude yourself from the
rsync. When you want to sync the clone again just do the same again
and rsync the changes.
Regards,
Derek
As an added note on this, rsync's --one-file-system (-x) flag is handy
for avoiding grabbing unneeded things, but will typically leave you
without the base few device nodes needed to boot the backup, those can
either be grabbed from a stage3, or created with (courtesy of Linux
From Scratch's section 6.2.1. Creating Initial Device Nodes):
mknod -m 600 ${backup}/dev/console c 5 1
mknod -m 666 ${backup}/dev/null c 1 3
The best way to copy a filesystem without any sub-mounts is to
mount-bind it to some directory and copy it from there:
mkdir /tmp/root
mount --bind / /tmp/root
rsync -a /tmp/root/ /mnt/backup/
Note that copying a rw filesystem (especially /) on a
Re: [gentoo-user] What is the best audio system?
On Tue, Feb 28, 2012 at 03:30:24AM +0100, Frank Steinmetzger wrote:
dmix *may* be able to handle multiple audio streams (in practice, in
my personal experience, it always requires more work than PA); but it
will never be able to do the other stuff PA handles.
This seems like a dumb question (for I was a strict PA denier until recently
and have been using alsa-only since always), but does PA handle OSS
applications better than alsa/dmix? Whenever I want to use sidplay, which only
speaks OSS, I need to stop all other audio programs (e.g. press Stop in the
Clementine player if it's only paused), or else /dev/dsp was busy.
PA doesn't care about oss (/dev/dsp). It opens the soundcard through
normal alsa interface (which means /dev/dsp becomes busy). You can
either kill pulseaudio, or tell pulseaudio to suspend the correspondig
sink (not sure what exactly happens if an audio stream through PA is active
etc..).
Regarading oss (/dev/dsp) and plain alsa, it is the same, if something
opens the soundcard through alsa, /dev/dsp becomes busy... (even when
using dmix in alsa, because /dev/dsp is handled by a kernel modules,
dmix is userspace).
There is however a way to amke oss work with dmix through aoss (a small
program that preloads a binary, that 'hijacks' calls to open /dev/dsp
and 'reroutes' that to alsa, works most of the time, but can have
problems if the program does some weird things...)
In that case aoss opens the alsa device pcm.dsp (or dsp0, i'm not sure
right now), which you can easily point to dmix...
from my /etc/asond.conf:
pcm.dsp {
type plug
slave.pcm duplex
}
pcm.dsp0 {
type plug
slave.pcm duplex
}
pcm.!default {
type plug
slave.pcm duplex
}
pcm.duplex {
type asym
playback.pcm dmix:0
capture.pcm dsnoop:0
}
Then you can run (even multiple) 'aoss mpg123 file.mp3 ...'
You can also make this to work alongside pulseaudio, if you configure
pulseaudio to use the dmix device instead of directyly using the hw
device.
Nnote that this might cause problems, you have to disable pulseaudio's
autodetect and configure all soundcards manually, and using dmix
introduces some additional overhead and probles, also such setup is most
probably not supported by pulseaudio etc... However it also enables you
tu run plain alsa apps alongside PA (officialy you should just configure the
!default device to use the PA alsa plugin is simpler and it should work
better, though I had some problems with some apps) and more importantly
to run multi PAs simultaneusly (ie for multiple users...)
yoyo
Re: [gentoo-user] Freeing up disk space problem!!
On Tue, Feb 28, 2012 at 11:37:44AM +, trevor donahue wrote: Hi everyone, I'm experiencing a major problem right now. I've been using gentoo for several months now and I simply lllooove it! So here's the thing. When I use gentoo for a long time, even without updating the current pack of installed software (emerge -uD world), I am left without disk space... In situations like this I start deleting /var/tmp/*, /tmp/*, /usr/portage/distfiles/*, maybe do even a revdep-rebuild to fix something, but even then I'm left with no more then 100 mb, which obviously is not enough ... So this time I googled a bit and I deleted all the /usr/share/doc/ and this left me with 2.5 gb of space (wow). My usual suspect for disk space in /usr/share/doc is kdelibs, with the doc use flag turned on it installs the whole kde api documentation, which takes a lot of space ... so I either set -doc for kde-base/kdelibs, or just set -doc globally and just enable it for things i now I might need... (note that that won't remove all of the /usr/share/doc dirs / files, but removes most of the large ones...) yoyo So the questions are ... in cases like this, what should be done? what is storing this much space? logs?
Re: [gentoo-user] Konsole question
On Mon, Jan 30, 2012 at 07:51:45AM -0800, Mark Knecht wrote: On Mon, Jan 30, 2012 at 7:08 AM, Philip Webb purs...@ca.inter.net wrote: When an URL appears in a console, it is usually possible to R-click, then choose 'open link' a browser(-tab) opens for that link. With Xfce's Terminal, it opens in a running instance of Firefox. With KDE's Konsole, it opens Konqueror, which I don't usually have running. However, since around KDE 4.7.1 continuing in 4.8.0 , the version of Konqueror which opens is the file browser, which I never otherwise use which is useless for this purpose. Clearly, what sb opened is Konqueror in browser mode. Has anyone else encountered this problem ? I've tried to find where the browser to open is configured, but can't. Have I missed a setting somewhere or is this a KDE bug I should report ? -- ,, SUPPORT ___//___, Philip Webb ELECTRIC /] [] [] [] [] []| Cities Centre, University of Toronto TRANSIT `-O--O---' purslowatchassdotutorontodotca I'm running 99.99% stable here so takes that for what it is. Here, with Firefox open I run in konsole eix ardour and then right click on the ardour.org link and choose Open Link. I get the Ardour web site in the running version of Firefox. With Firefox not running KDE/konsole still opens Firefox. Konsole should open the default browser set in KDE. You can check that in systemsettings - Workspace Appearance and Behavior - Default Applications - Web Browser IIRC some browsers allow you to 'set them as default' somewhere in their menu (usually as an annoying popup at startup, until you disable it ;) which should work with kde.. (at least i think it worked the last time i tested it with chromium... but i mostly set it up manualy) yoyo
Re: [gentoo-user] Cross Compiling in Gentoo
On Tue, Jan 17, 2012 at 01:55:25AM -0500, Chris Walters wrote: My question is, does anyone know of any good resources (mailing lists, sites, etc.) on cross compiling on a GNU/Linux platform for a W32/W64 platform? The searches I've run have directed me to sites that talk about using MSYS and Mingw on a W32 platform (I don't have all year to build a single package). I am looking to build GraphicsMagick, and some helpful tools for W64 (though I'd accept W32, if that's the only way). You should also ask on the gentoo-embedded mailling list. Mingw might not be reallly 'embedded' but that list would still be the most relevant place to ask gentoo-related crosscompile questions... Gentoo also has a crossdev tool that should also support mingw, though I only ever used it for i686 and arm crosscompiles. yoyo
Re: [gentoo-user] Init Scripts Not Starting
On Wed, Jan 04, 2012 at 07:10:07PM -0800, Dan Cowsill wrote: On Wed, Jan 4, 2012 at 8:53 AM, Neil Bothwick n...@digimed.co.uk wrote: On Wed, 4 Jan 2012 06:53:18 -0800, Dan Cowsill wrote: Running rc or rc default returns immediately. I am sure I am starting into the default runlevel because ntp-client runs on default and it starts no problem. Have you enabled logging in rc.conf? What does /var/log/rc.log show? I have indeed. The logs show that everything but xdm and virtualbox-guest-additions start normally. The logs do not show any attempt to start either of those. It seems this system is full of lies. Did you check the file permissions on those scripts? I have. Permissions in the /etc/init.d folder are consistent across the board. If anything else, you can try running 'strace -e trace=file' on rc-status and rc-update show and might notice something there that would explain while those two tools come up with different sets of scripts... I've attached the strace results from rc-status, rc-update show and rc itself. If you guys can make any sense of them, I'd be grateful. Seems that rc-status sees the xdm initscript... but why doesn't it show it then... no idea... My advice would be to submit a bug, the inconsistent output from rc-status vs rc-update is suspicious enough... You might also try to re-emerge the packages that installs /etc/init.d/xdm to see if that doesn't help... (should be x11-base/xorg-server) yoyo
Re: [gentoo-user] Init Scripts Not Starting
On Tue, Jan 03, 2012 at 08:31:08PM -0800, Dan Cowsill wrote: Hey list, A little while after I compiled Gnome and got things running, I lost the ability to add scripts to the default runlevel. I can run rc-update add xdm default, for example, and the xdm symlink will appear in /etc/runlevels/default, and that symlink will indeed point to /etc/init.d/xdm, but xdm will not start. Further to that, there's no evidence to indicate that RC is even trying to start it. No errors, no logs, no nothing. Same goes for virtualbox-guest-additions and sysklogd. I tried logging rc and got absolutely nowhere. There's nothing overt in dmesg either. The really fun part is these scripts function perfectly if I run them after boot. Since there's no evidence of this problem in any logs or during the startup process, I assume there is no problem and I am doing it wrong. Is it possible that you are booting into a different runlevel that default ? (there's a softlevel=... kernel cmdline parameter) What happens if (after boot) you just run rc (should start all service in the current runlevel, that are not started yet) or rc default (should swithc to 'default' runlevel and start all services) ? Could you post the output of rc-status -a ? And maybe also grep rc /etc/inittab ? For xdm there is one additional thing to check: xdm can be disabled through a 'nox' kernel cmdline option or an /etc/.noxdm file... But in that case the initscript itself should start and just print a message, that it is not starting the DM. yoyo
Re: [gentoo-user] Init Scripts Not Starting
On Wed, Jan 04, 2012 at 06:53:18AM -0800, Dan Cowsill wrote: On Wed, Jan 4, 2012 at 3:17 AM, YoYo Siska y...@gl.ksp.sk wrote: On Tue, Jan 03, 2012 at 08:31:08PM -0800, Dan Cowsill wrote: Hey list, A little while after I compiled Gnome and got things running, I lost the ability to add scripts to the default runlevel. I can run rc-update add xdm default, for example, and the xdm symlink will appear in /etc/runlevels/default, and that symlink will indeed point to /etc/init.d/xdm, but xdm will not start. Further to that, there's no evidence to indicate that RC is even trying to start it. No errors, no logs, no nothing. Same goes for virtualbox-guest-additions and sysklogd. I tried logging rc and got absolutely nowhere. There's nothing overt in dmesg either. The really fun part is these scripts function perfectly if I run them after boot. Since there's no evidence of this problem in any logs or during the startup process, I assume there is no problem and I am doing it wrong. Is it possible that you are booting into a different runlevel that default ? (there's a softlevel=... kernel cmdline parameter) What happens if (after boot) you just run rc (should start all service in the current runlevel, that are not started yet) or rc default (should swithc to 'default' runlevel and start all services) ? Could you post the output of rc-status -a ? And maybe also grep rc /etc/inittab ? For xdm there is one additional thing to check: xdm can be disabled through a 'nox' kernel cmdline option or an /etc/.noxdm file... But in that case the initscript itself should start and just print a message, that it is not starting the DM. yoyo Hey, thanks for the reply. Running rc or rc default returns immediately. I am sure I am starting into the default runlevel because ntp-client runs on default and it starts no problem. Hmm, the weird thing is that 'rc-status -a' and 'rc-update show' show different things (no xdm and virtualbox-... in the first one) but I have no idea what might be causing that... Few random thoughts: Did you check the file permissions on those scripts? Any pending etc-update stuff (esp. in /etc/init.d) ? (not that it should affect this in any way...) If anything else, you can try running 'strace -e trace=file' on rc-status and rc-update show and might notice something there that would explain while those two tools come up with different sets of scripts... You can also look around the dirs/files in /lib/rc/init.d/ to se a more complete state of the rc system... (/lib/rc/init.d/softlevel should show the current runlevel) yoyo Output from grep rc /etc/inittab: si::sysinit:/sbin/rc sysinit rc::bootwait:/sbin/rc boot l0:0:wait:/sbin/rc shutdown l1:1:wait:/sbin/rc single l2:2:wait:/sbin/rc nonetwork l3:3:wait:/sbin/rc default l4:4:wait:/sbin/rc default l5:5:wait:/sbin/rc default l6:6:wait:/sbin/rc reboot su0:S:wait:/sbin/rc single Output from rc-status -a: Runlevel: default net.eth1 [ started ] dbus [ started ] net.eth0 [ started ] netmount [ started ] ntp-client[ started ] sshd [ started ] udev-postmount[ started ] local [ started ] Runlevel: sysinit dmesg [ started ] udev [ started ] devfs [ started ] Runlevel: boot hwclock [ started ] modules [ started ] fsck [ started ] root [ started ] mtab [ started ] localmount[ started ] sysctl[ started ] bootmisc [ started ] urandom [ started ] net.lo[ started ] termencoding [ started ] swap
Re: [gentoo-user] KDE won't start up....
On Wed, Jan 04, 2012 at 06:36:59PM +0100, Volker Armin Hemmann wrote: Am Donnerstag, 5. Januar 2012, 01:04:53 schrieb Andrew Lowe: Hi all, I had a running KDE 4 setup and this afternoon did an: emerge -NuD world There were no errors reported, the kernel source had been updated, so I compiled the new kernel, and copied it into place, recompiled my nvidia driver and also evdev drivers and then rebooted the machine. Now, the machine boots up, I get all the usual booting messages, starting ntp, mounting drives, getting IP addresses, exporting nfs and so on, the screen goes black, the hour glass of the KDE log in screen briefly appears then the screen is blanked and I'm back at a text login. I've logged into the machine from the text login and recompiled the kernel, copied it into place, recompiled nvidia and evdev and still the problem persists. I've looked at the xorg kdm logs and there are a few errors there that google searches seem to say are OK. The thing that is confusing is that when at the text prompt, I can start up bog standard X via startx and the mouse and keyboard work but I can't get KDE to start. Does anyone have any ideas as to what could be causing this problem? Any thoughts on where, besides the two obvious logs, that I can try and track down what's going wrong here, or steps I can take to debug the KDE startup? Any thoughts greatly appreciated, Andrew .xsession-errors Xorg.0.log please. If both are huge, upload them somewhere. Also make sure that the permissions of /tmp and /var/tmp are ok. Had it in the past that some update did some very scary things to both places. Also /var/log/kdm.log would be of use (if you are (well, want to ;) using KDM as the login manager)... because it seems that the problem might be with KDM greeter or something similar. I had it segfaulting once, the symptoms were very similar, KDM did bring up X and start kdmgreet which crashed, X went down immediately, because they had no clients, KDM noticed that X was running only for a very short time and interpreted that (correctly) as a sign that something went wrong and didn't restart the X and insted just stopped... that would also explain why running startx works ... btw, does startx /usr/bin/startkde start a KDE session? yoyo
Re: [gentoo-user] Display name and Wacom tablet
On Sun, Nov 13, 2011 at 04:07:42PM -0500, Daniel D Jones wrote: I have an Nvidia GeForce GTX 560 Ti running nvidia-drivers 275.09.07. It supports dual monitors via Twinview. I have a Wacom Inspire3 6 x 8 Tablet. The tablet is working but it covers the entire display across both monitors and I'd like to restrict it to one monitor. This is supposed to be done via xsetwacom set Wacom Intuos3 6x8 pad MapToOutput VGA1 VGA1 is supposed to be the name of the display you want to restrict it to, and that name is supposed to be available via xrandr. xrandr gives me the following output: ddjones@kushiel ~ $ xrandr xrandr: Failed to get size of gamma for output default Screen 0: minimum 2048 x 768, current 3360 x 1050, maximum 3360 x 1050 default connected 3360x1050+0+0 0mm x 0mm 3360x1050 50.0* 2048x768 51.0 ddjones@kushiel ~ $ xrandr --verbose xrandr: Failed to get size of gamma for output default Screen 0: minimum 2048 x 768, current 3360 x 1050, maximum 3360 x 1050 default connected 3360x1050+0+0 (0x166) normal (normal) 0mm x 0mm Identifier: 0x165 Timestamp: 13703 Subpixel: unknown Clones: CRTC: 0 CRTCs: 0 Transform: 1.00 0.00 0.00 0.00 1.00 0.00 0.00 0.00 1.00 filter: 3360x1050 (0x166) 176.4MHz *current h: width 3360 start0 end0 total 3360 skew0 clock 52.5KHz v: height 1050 start0 end0 total 1050 clock 50.0Hz 2048x768 (0x167) 80.2MHz h: width 2048 start0 end0 total 2048 skew0 clock 39.2KHz v: height 768 start0 end0 total 768 clock 51.0Hz I've tried guessing at the display name, trying VGA, DVI and LVDS with various numbers appended but xsetwacom simply complains that the display does not exist. I've also tried setting the Coordinate Transformation Matrix as described here: http://sourceforge.net/apps/mediawiki/linuxwacom/index.php?title=Dual_and_Multi- Monitor_Set_Up#Dual_Monitors I can set the matrix via the xinput command and xinput list-props for the device confirms that the matrix is set to the new value but it does not alter the behaviour of the tablet - it still spans both displays. I set and confirmed the matrix for the pad, the eraser and the cursor. Any advice or suggestions on how to either either identify the display names (or fix whatever issue causes xrandr not to display the info) or to otherwise restrict the tablet to one monitor would be greatly appreciated. Nvidia driver doesn't use the xrandr protocol when dealing with multiple monitors... they have their own extension and they say that it is better and that xrandr is bad, and they report through xrandr only one output (monitor) that covers all nvidia outputs. Thhat's the 'default' output in your xrandr output, it's widht is the sum of widhts of both monitors etc.. They report the physicall layout of monitors through the xinerama extension, but I guess that wacom uses xrandr (xinerama only numbers the display, it doesn't have names, reports only a subset of randr information and is generally older...) I guess your only chance is to find parameters in the wacom display that allow you to restrict the tablet to a certain area by setting the coordinates... (and that would ofcourse work only for a specific resolution...) yoyo
Re: [gentoo-user] The LIGHTEST web server (just for serving files)?
On Sat, Nov 12, 2011 at 07:40:08PM +0700, Pandu Poluan wrote: On Nov 12, 2011 7:00 PM, Mick michaelkintz...@gmail.com wrote: I've been using boa just for this purpose for years: * www-servers/boa Available versions: ~ 0.94.14_rc21 ~x86 ~sparc ~mips ~ppc ~amd64 [doc] Homepage:http://www.boa.org/ Description: A very small and very fast http daemon. It can be easily locked down for internet facing roles. I've also used thttpd (you can throttle its bandwidth if that's important in your network), but it's probably more than required for this purpose: * www-servers/thttpd Available versions: 2.25b-r7 amd64 ~hppa ~mips ppc sparc x86 ~x86-fbsd [static] ~ 2.25b-r8 ~amd64 ~hppa ~mips ~ppc ~sparc ~x86 ~x86-fbsd [static] Homepage:http://www.acme.com/software/thttpd/ Description: Small and fast multiplexing webserver. Thanks for all the input! During my drive home, something hit my brain: why not have the 'master' server share the distfiles dir via NFS? So, the question now becomes: what's the drawback/benefit of NFS-sharing vs HTTP-sharing? The scenario is back-end LAN at the office, thus, a trusted network by definition. NFS doesn't like when it looses connection to the server. The only problems I had ever with NFS were because I forgot to unmout it before a server restart or when I took a computer (laptop) off to another network... Otherwise it works well, esp. when mounted ro on the clients, however for distfiles it might make sense to allow the clients download and save tarballs that are not there yet ;), though I never used it with many computer emerging/downloading same same stuff, so can't say if locking etc works correctly... And with NFS the clients won't duplicate the files in their own distfiles directories ;) yoyo
Re: [gentoo-user] Unable to login to gentoo
On Tue, Nov 01, 2011 at 04:54:06PM -0500, Dale wrote: Michael Schreckenbauer wrote: Am Dienstag, 1. November 2011, 23:00:57 schrieb Alan McKinnon: On Tue, 01 Nov 2011 14:49:30 -0500 Dalerdalek1...@gmail.com wrote: Vishnupradeep wrote: On Tue, Nov 1, 2011 at 2:48 PM, Mickmichaelkintz...@gmail.com mailto:michaelkintz...@gmail.com wrote: emerge -1aDv app-portage/gentoolkit revdep-rebuild -v --ask i can't use emerge command as it is shown as command not found. can u tell the correct path of emerge. /usr/bin/emerge Little bit of learning here: root@smoker / # which emerge /usr/bin/emerge root@smoker / # Now you know where it is and how I found out where it is. Trick is remembering a command you rarely use. That won't work. The $PATH is broken so the shell can't find ls and emerge. Well, it won't find which either :-) Solution: You run which which and tell the OP which directory contains which so he can run /path/towhich emerge to find out where emerge is. Or just use type -a, because that's a builtin. ~ $ type -a emerge emerge is /usr/bin/emerge Best, Michael That is better. Now to remember that command. ;-) Dale Except that both won't work in OP's case, because both look for the executable in directories in PATH which is the OP's problem ;) yoyo
Re: [gentoo-user] Consistency checking
On Tue, Nov 01, 2011 at 04:45:25AM +0100, meino.cra...@gmx.de wrote: Hi, I know of three commands to check the consistency of a Gentoo system: eix-sync emerge --color=n -p -v --newuse --update --deep world emerge -p -v --depclean revdep-rebuild --ignore -p -v of course, one has to remove the -p and -v flags after checking the putput of the commands. What else can be checked and should be checked from time to time or after each update? python-updater perl-cleaner to rebuild needed perl/python packages after an update, maybe module-rebuild after a kernel upgrade... personally I'd realy like if all the revdep-rebuild / python-updater / perl-cleaner / ... were combined into a single tool or at least named consistently :) (perl-cleaner is the one I can never remember ;) yoyo
Re: [gentoo-user] Unable to login to gentoo
On Tue, Nov 01, 2011 at 02:19:59PM +0530, Vishnupradeep wrote: On Tue, Nov 1, 2011 at 1:56 PM, Mick michaelkintz...@gmail.com wrote: revdep-rebuild -v -- --ask localhost / # revdep-rebuild -v --ask bash: revdep-rebuild: command not found localhost / # ls bash: ls: command not found localhost / # maybe only the environment is somehow broken try echo $PATH or env and try using full paths: /bin/ls (or /usr/bin/ls) and /usr/bin/revdep-rebuild yoyo
Re: [gentoo-user] Unable to login to gentoo
On Wed, Nov 02, 2011 at 01:00:12AM +0530, Vishnupradeep wrote: yes full path works. On Tue, Nov 1, 2011 at 6:51 PM, YoYo Siska y...@gl.ksp.sk wrote: and try using full paths: /bin/ls (or /usr/bin/ls) What about the env or echo $PATH output? Might be that some files used to set up the enviroment are broken. If PATH and maybe some other things are not set up correctly, that would explain why login into KDE or anything graphical doesn't work... You can try env-update source /etc/profile if that doesn't help, you will have to check the files that are read by bash, mostyle /etc/profile (which should source /etc/profile.env among others) Check also files in users/roots homedir (~/.bash_profile / ~/.bashrc), but if the problem happes both with root and regular user, I don't think the prob. is in those files. yoyo
Re: [gentoo-user] Re: How to configure eth1:1 ?
On Mon, Oct 17, 2011 at 10:19:32PM +, Grant Edwards wrote: On 2011-10-17, Florian Philipp li...@binarywings.net wrote: Ugh, sorry. Just ignore that. I didn't see the second line in config_eth1. The odd quoting confused me. Sorry about that. I was trying various quoting schemes I'd found in examples. My current configuration works: modules_eth0=( !plug ) config_eth0=( 192.168.8.4/16 ) routes_eth0=( default via 192.168.0.254 ) modules_eth1=( !plug ) config_eth1=( 10.0.0.1/8 192.168.250.1/24 ) $ /sbin/ip address show 1: lo: LOOPBACK,UP,LOWER_UP mtu 16436 qdisc noqueue state UNKNOWN link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00 inet 127.0.0.1/8 scope host lo 2: eth0: BROADCAST,MULTICAST,UP,LOWER_UP mtu 1500 qdisc pfifo_fast state UP qlen 1000 link/ether 00:1b:21:b1:d1:e9 brd ff:ff:ff:ff:ff:ff inet 192.168.8.4/16 brd 192.168.255.255 scope global eth0 3: eth1: BROADCAST,MULTICAST,UP,LOWER_UP mtu 1500 qdisc pfifo_fast state UP qlen 1000 link/ether 00:16:17:84:a7:b3 brd ff:ff:ff:ff:ff:ff inet 10.0.0.1/8 brd 10.255.255.255 scope global eth1 inet 192.168.250.1/24 brd 192.168.250.255 scope global eth1 4: eth2: BROADCAST,MULTICAST,UP,LOWER_UP mtu 1500 qdisc pfifo_fast state UNKNOWN qlen 1000 link/ether 00:18:e7:08:20:33 brd ff:ff:ff:ff:ff:ff And the lack of eth1:1 is presumably explained if the system is using the iproute2 module instead of the ifconfig module. My current theory is that iproute2 is getting used because I have openvpn installed with the iproute2 use flag. [I'm not actually using openvpn, but it's still istalled from a couple years ago when I was using it.] From /usr/share/doc/openrc-0.7.0/net.example: ## # INTERFACE HANDLERS # # We provide two interface handlers presently: ifconfig and iproute2. # You need one of these to do any kind of network configuration. # For ifconfig support, emerge sys-apps/net-tools # For iproute2 support, emerge sys-apps/iproute2 # If you don't specify an interface then we prefer iproute2 if it's installed # To prefer ifconfig over iproute2 #modules=ifconfig But for some reason modules=ifconfig doesn't seem to work for me (eth0:1 doesn't get created) but modules=!ifconfig works ;) Also it seems that modules_ethX shouldn't be an array, modules_eth0=!plug !iproute2 uses ifconfig, modules_eth0=( !plug !iproute2 ) uses iproute2 ;) As to why you have iproute2 installed... I always install it, so I can't say which packages might be pulling it in, but you can use equery to find out: tableta ~ # equery depends iproute2 * These packages depend on iproute2: net-misc/openvpn-2.2.0-r1 (iproute2 ? sys-apps/iproute2[-minimal]) yoyo
Re: [gentoo-user] Listing partition labels
On Tue, Aug 02, 2011 at 10:12:33AM +0800, Andrew Lowe wrote: Greetings all, I'm probably in the situation where I can't see the wood for the trees so a bit of help would be appreciated. I've decided to go the LABEL route in fstab and have set the labels on my partitions a few days ago. I now want to update fstab but can't remember the names. I can't find a command that will list the partitions and the names I've given them. I'm sure fdisk does not list them when I do just fdisk at the command prompt, but then again as I said above, I think I'm in the wood/forest mode at the moment. Any idea on the command? Any thoughts greatly appreciated, Andrew Others have replied with possible options, here is another one, maybe the easiest one that shows all the labels/uuids at once ;) ls -l /dev/disk/by-label/ and ls -l /dev/disk/by-uuid/ yoyo
Re: [gentoo-user] Nvidia kernel 3.0
On Tue, Aug 02, 2011 at 11:35:01AM -0400, Philip Webb wrote: Has anyone else run into a problem trying to compile Nvidia with kernel 3.0 ? AFAIK I have the correct symlink to the kernel source root:601 src pwd /usr/src root:602 src ls -l lrwxrwxrwx 1 root root 19 Aug 2 11:04 linux - linux-3.0.0-gentoo/ drwxr-xr-x 24 root root 1648 Dec 14 2010 linux-2.6.33-gentoo-r1 drwxr-xr-x 24 root root 1640 Jan 21 2011 linux-2.6.37-gentoo drwxr-xr-x 24 root root 1640 Apr 11 03:06 linux-2.6.38.2 drwxr-xr-x 24 root root 1640 Apr 10 08:51 linux-2.6.38-gentoo drwxr-xr-x 24 root root 1640 Aug 2 11:16 linux-3.0.0-gentoo yet when I try to emerge 'nvidia-drivers-270.41.19', it tells me *** Unable to determine the target kernel version. *** I originally had the default kernel name '3.0', which brought the same error msg, so changed it to '3.0.0', but both names cause the same result. 275.21 from ~ work ok with 3.0... 3.0 seems to be still ~, so i guess that for unstable kernel you might need unstable drivers... yoyo -- ,, SUPPORT ___//___, Philip Webb ELECTRIC /] [] [] [] [] []| Cities Centre, University of Toronto TRANSIT`-O--O---' purslowatchassdotutorontodotca
Re: [gentoo-user] Re: Openoffice being replaced?
On Thu, Jul 28, 2011 at 03:25:48PM -0500, Paul Hartman wrote: On my machine, it took this: root@fireball / # genlop -t libreoffice * app-office/libreoffice Thu Jul 28 12:33:11 2011 app-office/libreoffice-3.3.1 merge time: 1 hour, 9 minutes and 33 seconds. root@fireball / # I did stop it with a ctrl Z for about 5 minutes. I was deleting stuff to give it some more room. This is OOo: Tue Jul 5 05:15:01 2011 app-office/openoffice-3.2.1-r1 merge time: 50 minutes and 27 seconds. That's about a average. Some were binary installs. I can't recall why I did that now but it was since it only took a minute or so. That's the report from this rig. AMD 4 cores running at 3.2Ghz with 16Gbs of ram. No tmpfs this time. That wouldn't be fair since I had to stop it for a few minutes. Dale Here is mine: Wed May 25 10:07:18 2011 app-office/libreoffice-3.3.2 merge time: 34 minutes and 10 seconds. On my Intel Core i7 920 with 12GB of RAM using tmpfs. :) Do those merge times include download time? I wonder... BTW, I was emerging libreoofice yesterday, and it took ~3.5 hours, although OOo usually took ~1 hour. It seemed as if it wasn't doing anything paralell, so I blamed my MAKEOPTS=-j -l4 (to which I changed some time ago, but all the OOo compiles were before that). I just tried today again with MAKEOPTS=-j4 and it took only ~1 hour ;). Guess dmake doesn't know the --load-average stuff ;) can't give a nice genlop -t output, as the first time i was building it first with FEATURES=buildpkgonly and then installing with --usepkg (just after i removed OOo), as portage didn't wan't to resolve the blocker, and i didn't want to be without it while it compiled ... ;) # genlop -t libreoffice * app-office/libreoffice Fri Jul 29 08:57:10 2011 app-office/libreoffice-3.3.3 merge time: 1 minute and 59 seconds. Fri Jul 29 13:55:42 2011 app-office/libreoffice-3.3.3 merge time: 1 hour, 7 minutes and 25 seconds. the first one is for the --usepkg binary merge ;), but I checked the timestamps in emerge.log and the compile was a small bit over 3.5h yoyo
Re: [gentoo-user] Running out of space on /var partition
On Mon, Jul 25, 2011 at 11:02:34AM +0100, Mick wrote: After some deliberation I've started emerging libreoffice. It gave the usual office suite warnings at the beginning that there isn't enough space in /var (I have 5.8G and it was asking for more than 7G+). Half way through the emerge I noticed that I have only 74M left and is going down fast! O_O OpenOffice was able to emerge in the past using this partition size without a problem. I've flushed logs and what not to free some space, but I'm thinking of extending the partition somehow. I don't run LVM on this machine so that's not a solution for this circumstance. Is there anything I can do with mount --rbind and could I do this in the middle of an emerge? I have another partition with loads of space in it, but it has a different fs on it (reiser4 instead of /var's ext4). You can mount --bind a dir from a large partition to /var/tmp/portage before an emerge, but you can't do that during a running emerge. However with a reasonable buildsystem, you could in theory stop the emerge, copy over the files in /var/tmp/portage to new location (preserving timestamps) and then try resuming the emerge with FEATURES=keepwork noclean, though i don't know if that works well with openoffice... Also you don't have to mount anything to /var/tmp/portage, you can just change the dir by setting PORTAGE_TMPDIR to a directory on a partition with enough space (I normally have my DISTDIR, PKGDIR and PORTAGE_TMP set on a different partition...) yoyo
Re: [gentoo-user] Running out of space on /var partition
On Mon, Jul 25, 2011 at 11:40:55AM +0100, Mick wrote: On Monday 25 Jul 2011 11:24:33 YoYo Siska wrote: On Mon, Jul 25, 2011 at 11:02:34AM +0100, Mick wrote: After some deliberation I've started emerging libreoffice. It gave the usual office suite warnings at the beginning that there isn't enough space in /var (I have 5.8G and it was asking for more than 7G+). Half way through the emerge I noticed that I have only 74M left and is going down fast! O_O OpenOffice was able to emerge in the past using this partition size without a problem. I've flushed logs and what not to free some space, but I'm thinking of extending the partition somehow. I don't run LVM on this machine so that's not a solution for this circumstance. Is there anything I can do with mount --rbind and could I do this in the middle of an emerge? I have another partition with loads of space in it, but it has a different fs on it (reiser4 instead of /var's ext4). You can mount --bind a dir from a large partition to /var/tmp/portage before an emerge, but you can't do that during a running emerge. However with a reasonable buildsystem, you could in theory stop the emerge, copy over the files in /var/tmp/portage to new location (preserving timestamps) and then try resuming the emerge with FEATURES=keepwork noclean, though i don't know if that works well with openoffice... Also you don't have to mount anything to /var/tmp/portage, you can just change the dir by setting PORTAGE_TMPDIR to a directory on a partition with enough space (I normally have my DISTDIR, PKGDIR and PORTAGE_TMP set on a different partition...) Oh yes! Of course, PORTAGE_TMPDIR what was I thinking?!! Thank you. I never understood properly how the mount --bind/rbind works. I understand that the original partition content becomes visible on a second partition, but I'm not at all sure what happens when the space on the first partition runs out? Not on a second partition but under another mount point i.e. another path... When you try to access a file by its filename, the kernel takes the absolute file name, compares it to all the moutend mount points, takes the best match and tries to find (create) the file in that filesystem/partition... Lets say you do something like: mount /dev/sda1 / (well... you don't actually do this... ;) mount /dev/sda2 /mnt/sda2 mount --bind /mnt/sda2/bigtmp /tmp Then path /home/yoyo/something is accessing file home/yoyo/something on partition sda1. The path /mnt/sda2/somedir/somefile is accessing file somedir/somefile on partion sda2. The path /tmp/somedir/somefile is accessing file bigtmp/somdir/somefile on partition sda2. The files (and free space) under /mnt/sda2 and /tmp are actually on partition sda2, everything else is on sda1... So if sda1 runs out of space (by writing to other places than /mnt/sda2 and /tmp), it doesn't in any any way affect /mnt/sda2 and /tmp which have their free space from sda2. Conversely if you fill up sda2 by writing to /tmp, your system partition still has free space ;) yoyo
Re: [gentoo-user] Re: Kernel panics and more info
On Fri, Jul 22, 2011 at 12:22:32PM -0500, Dale wrote: ... using either. I clicked on the link to download and the window popped up to ask me whether to open it or save it. I selected to save it as I have done countless times before. As soon as I clicked that, the window popped up asking where to save it to then kernel panic. This was in Seamonkey. ... So, when Seamonkey or Firefox try to download something, besides the web pages itself, I get a kernel panic. Is this weird or what? BTW, as any other browser (well, new enough..), firefox starts downloading as soon as you click on a link (ie, it dowloads it while you are choosing where to save it, so that by the time you choose the dir/filename, smaller files are allready downloaded ;). You can easilly see this if you have some kind of network traffic monitoring widget/applet/app... I guess it starts to download it to a temp file, than moves it to the file you choose (never looked into it)... so the problem would be most likely in that operation.. yoyo
Re: [gentoo-user] nvidia-settings over ssh sees my local GPU?
On Thu, Jun 23, 2011 at 11:31:29AM -0700, Mark Knecht wrote: On Thu, Jun 23, 2011 at 10:39 AM, YoYo Siska y...@gl.ksp.sk wrote: On Thu, Jun 23, 2011 at 05:21:07PM +0100, Neil Bothwick wrote: On Thu, 23 Jun 2011 08:54:01 -0700, Mark Knecht wrote: My question is about running nvidia-settings. I'm finding that if I shell into his machine using ssh -X -Y -C IP-address and run nvidia-settings I get it displayed here, as it should be. The problem is it is seeing my GTX 465 and not his 8400GS. Looking at the man page, it appears you need to use the -ctrl-display parameter or the $DISPLAY env var. The man page mentions that nvidia-settings queries the X server, which is running locally. It looks like this setting may force it to use another. as neil wrote, it is nvidia-settings -c :0 nvidia-settings connects to the remote xserver to communicate with the graphics card (through a special nvidia xtenstion to the x protocol), so you need to be able to access the remote xserver, if you are logged in as the user running the xserver, you should be ok yoy Yeah, I've been tripping over doing this right since Neil pointed me toward the -c command. I think the problem is that I don't have the permissions set correctly to allow this to work right. The owner of the remote machine is logged in and possibly using X. I'm not sure about that but I'm not 'running the X server' in any meaningful way. I'm just remotely trying to access his GPU with nvidia-settings but display the GUI here. The problem seems to be getting the right number of his server or else permissions. This page is one of the better ones I've found about running nvidia-settings remotely, specifically section 6 which gives this example: http://www.makelinux.com/man/1/A/alt-nvidia-173-settings (issued from bartok.nvidia.com) xhost +stravinsky.nvidia.com (issued from schoenberg.nvidia.com) xhost +stravinsky.nvidia.com nvidia-settings --display=bartok.nvidia.com:0 --ctrl-display=schoenberg.nvidia.com:0 which allows all X clients run on stravinsky.nvidia.com to connect and display on bartok.nvidia.com's X server and configure schoenberg.nvidia.com's X server. this seems pretty old... defaults on most distros these days are that X server does not listen on tcp/ip (ip-address:0) only on a local unix sockets (:0), see below for more It seems this program allows you to run it from machine1, display it on machine2 which controlling machine3? So, locally I ran mark@c2stable ~ $ xhost access control enabled, only authorized clients can connect mark@c2stable ~ $ xhost +DWP-Linux DWP-Linux being added to access control list mark@c2stable ~ $ xhost access control enabled, only authorized clients can connect INET:DWP-Linux mark@c2stable ~ $ which I think allows the remote machine access here in my server. I then log in which creates the .Xauthority file: mark@c2stable ~ $ ssh -XYC DWP-Linux Password: Last login: Thu Jun 23 14:11:33 EDT 2011 from c-67-161-57-1.hsd1.ca.comcast.net on pts/3 /usr/bin/xauth: file /home/mark/.Xauthority does not exist mark@DWP-Linux ~ $ ls -al .Xauthority -rw--- 1 mark mark 55 Jun 23 14:21 .Xauthority mark@DWP-Linux ~ $ cat .Xauthority DWP-Linux11MIT-MAGIC-COOKIE-1��:��T'6�@R��mark@DWP-Linux ~ $ mark@DWP-Linux ~ $ On that machine I see this $DISPLAY: mark@DWP-Linux ~ $ echo $DISPLAY localhost:11.0 mark@DWP-Linux ~ $ so I run mark@DWP-Linux ~ $ nvidia-settings -c :11 which sees my GPU, not his, presumably because I said to control my system with -c :11. However if I try something like mark@DWP-Linux ~ $ nvidia-settings -c :0 I get a bunch of stuff ending with ERROR: Unable to assign attribute XVideoSyncToDisplay specified on line 62 of configuration file '/home/mark/.nvidia-settings-rc' (no Display connection). No protocol specified ERROR: Cannot open display ':0'. mark@DWP-Linux ~ $ I'm a bit lost at this point. (OBVIOUSLY!) :-) To connect to an Xserver, you need to know its address, which is something like machine_or_ip:NUMNER for tcp/ip connection or just :NUMBER for (local) unix socket connection. When you do a ssh -X , ssh creates a tunnel from the remote computer to your local Xserver (:0 at your side), and creates a new address, usually localhost:10 (or 11, 12, which on is free) if i'm behing a laptop named 'tabletka' and there is also a desktop named 'yoyo' (its the same as my username... on both), you could do: yoyo@tabletka ~ $ DISPLAY=:0 xterm which runs xterm here (tabletka), displays it here yoyo@tabletka ~ $ DISPLAY=yoyo:0 xterm runs xterm here, displays it on yoyo however as i said, most distributions don't allow Xservers to listen on tcp/ip and only allow it to listen on local sockets... you can however use ssh to tunnel X traffic, so you could do: yoyo@tabletka ~ $ echo $DISPLAY :0.0 yoyo@tabletka ~ $ ssh -X yoyo yoyo@yoyo:~$ echo
Re: [gentoo-user] nvidia-settings over ssh sees my local GPU?
On Thu, Jun 23, 2011 at 05:21:07PM +0100, Neil Bothwick wrote: On Thu, 23 Jun 2011 08:54:01 -0700, Mark Knecht wrote: My question is about running nvidia-settings. I'm finding that if I shell into his machine using ssh -X -Y -C IP-address and run nvidia-settings I get it displayed here, as it should be. The problem is it is seeing my GTX 465 and not his 8400GS. Looking at the man page, it appears you need to use the -ctrl-display parameter or the $DISPLAY env var. The man page mentions that nvidia-settings queries the X server, which is running locally. It looks like this setting may force it to use another. as neil wrote, it is nvidia-settings -c :0 nvidia-settings connects to the remote xserver to communicate with the graphics card (through a special nvidia xtenstion to the x protocol), so you need to be able to access the remote xserver, if you are logged in as the user running the xserver, you should be ok yoy
Re: [gentoo-user] /etc/locale vs /etc/env.d/02locale?
On Thu, Jun 16, 2011 at 09:23:16AM -0700, Mark Knecht wrote: On Thu, Jun 16, 2011 at 9:00 AM, Paul Hartman paul.hartman+gen...@gmail.com wrote: On Thu, Jun 16, 2011 at 10:45 AM, Mark Knecht markkne...@gmail.com wrote: Is there a simple explanation concerning the difference between the two locales I have seen on Gentoo machines? 1) /etc/locale, as specified in the installation documents 2) /etc/env.d/02locale as has been discussed on the list recently I'm not near a Gentoo machine right now, but off the top of my head IIRC: /etc/locale.gen contains a list of locales to be compiled when glibc is emerged. These will be available to be used. /etc/env.d/02locale specifies which of those locales you actually want to use for the system-wide default (the LC variables) Thanks for the response Paul. Does that mean that the /etc/locale.gen is used only by glibc and not really by the system? If so, what is glibc doing with these beyond letting me system run programs? If 02locale specifies what the system is using, then should it be 02locale that's in the install documents vs off in an optional Gentoo Localization guide? Note that the /etc/locale.gen stuff is marked optional in the guide so presumably it isn't actually needed. All I've determined about it is that it reduces the amount of time emerge spends buildingglibc/gcc. locale.gen is in the install docs, because it allows you to choose which locales should be built, ie after emerging libc, which locales you can choose from... if you don't modify it, you get a lot of usual locales built... /etc/env.d/02locale is used to actually choose which one of the built ones will be used as the default locale for (almost) everything that runs... I gues it might deserve a mention in the install guide... though it actullly isn't any special file... the actuall locale is set by setting an enviroment variable (LANG or the specific LC_...), you could set it in your .bashrc / .bash_profile only for your user, or anywhere where it would apply to most programs, ie /etc/profile ... Gentoo has the mechanism, that anything that gets put into /etc/env.d is then (through env-update, which you have certainly run from time to time ;) merged together to /etc/profile.env, which is in turned sourced by /etc/profile (and posibly other things) so that it is just logical to put it there... but the actual name of the file doesn't really matter ;) yoyo
Re: [gentoo-user] Override DHCP-provided DNS
On Wed, Jun 15, 2011 at 04:07:43PM +0200, Florian Philipp wrote: Hello list! for some wireless access points, I want to get an IP via DHCP but not use the provided DNS-server (I use an openvpn setup with its own DNS server, domain name, etc.). In /usr/share/doc/openrc-0.8.2-r1/net.example it reads: # Setting name/domain server causes /etc/resolv.conf to be overwritten # Note that if DHCP is used, and you want this to take precedence then # please put -R in your dhcpcd options But dhcpcd does not seem to have a -R option. It does have a --static option, though. While this is good enough for simply setting the DNS server, it does not seem to allow specifying domain names or search-domains (at least it is not shown in the man-page). Please tell me what the proper way is and whether the mention of -R is a documentation bug. Thanks in advance, Florian Philipp from the man page, this seems to do what you want (never tried, i use dhclient and its /etc/dhcp/dhclient.conf): -C, --nohook script Don't run this hook script. Matches full name, or prefixed with 2 numbers optionally ending with .sh. So to stop dhcpcd from touching your DNS or MTU settings you would do:- dhcpcd -C resolv.conf -C mtu eth0 yoyo
Re: Odp: [gentoo-user] Re: polish fonts xorg.conf
On Wed, Jun 15, 2011 at 04:46:54PM +0200, fajfu...@wp.pl wrote: Dnia 14-06-2011 o godz. 21:51 walt napisał(a): On 06/14/2011 09:02 AM, fajfu...@wp.pl wrote: Hello When I execute: setxkbmap pl I can type polish fonts in xterm and other X programs. But when I generate xorg.conf file with Xorg -configure and add the following to it I cannot type the polish fonts (I copied it to /etc/x11/xorg.conf) Section InputDevice Identifier Keyboard0 Driver kbd Option XkbModel pc105 Option XkbLayoutpl EndSection Xorg.0.log: [ 29007.715] (==) Using config file: /etc/X11/xorg.conf [ 29008.100] (II) XINPUT: Adding extended input device Power Button (type: KEYBOARD) [ 29008.100] (**) Option xkb_rules evdev [ 29008.100] (**) Option xkb_model evdev [ 29008.100] (**) Option xkb_layout us The only problem I can see at the moment is that the log file says that your keyboard is using the 'evdev' driver but your xorg.conf specifies the 'kbd' driver. Try changing the Driver to evdev instead of 'kbd'. I have reconfigured xorg.conf as follows: Section InputDevice Identifier Keyboard0 Driver evdev Option XkbModel pc105 Option XkbLayoutpl EndSection or Section InputDevice Identifier Keyboard0 Driver evdev Option XkbModel evdev Option XkbLayoutpl EndSection Unfortunatelly it didn't help. I attach the complete Xorg.0.log. Do you have another suggestions. Thank you for help [ 24703.710] (**) Keyboard0: always reports core events [ 24703.710] (EE) Keyboard0: No device specified. [ 24703.710] (II) UnloadModule: evdev [ 24703.710] (EE) PreInit returned NULL for Keyboard0 you defined a (new) keyboard in the config, which doesn't actually point to a device (the old kbd driver didn't need a device, but it didn't work for other reasons...) so that X basically ignored that section and your real keyboard device did get added automatically later albeit without your settings... the most correct way with a newer xorg is to create a file in /etc/xorg.conf.d where you put an entry, which would *match* your keyboard device (that gets automatically created) and add the options to it, basically something like: Section InputClass Identifier pl keyboard layout MatchIsKeyboard on Option XkbModel evdev Option XkbLayoutpl EndSection can't find gentoo specific doc / page for this stuff, but man xorg.conf (search for InputClass) and google for xorg.conf.d and/or InputClass should give you something usefull fex: http://fedoraproject.org/wiki/Input_device_configuration yoyo
Re: [gentoo-user] RE: Kernel Modules
On Sat, Jun 11, 2011 at 08:35:52AM +0700, Pandu Poluan wrote: -original message- Subject: Re: [gentoo-user] Re: Kernel Modules From: Dale rdalek1...@gmail.com Date: 2011-06-11 03:05 I notice a really long list of things when I do this: eselect bashcomp list Is there a way to just enable them all? Is there some that should NOT be enabled, maybe for good reason? Personally, I do some cherry-picking and enable a bashcomp when I found out I need it. I have 2 concerns (which may or may not be true): 1. It will make bash (or the whole system) slower well, only when you are hitting tab ... ;) I know it can be annoying to have to wait a long time when you accidentally hit tab on a complex command..., but when you know how to do the explicit filename only completion... 2. For some commands I *might* want the standard completion meta-/ (or ESC then /) for the complete-filename, there are also others for some other things (variable, username...) man bash /Completing yoyo
Re: [gentoo-user] Kernel Modules
On Thu, Jun 09, 2011 at 11:52:42AM +0100, Ignas Anikevicius wrote: Hello list, I was wondering if it is possible to have a tool with which it would be possible to have external modules installed for _all_ kernel versions in my computer. Now I am using 2.6.38 kernel, but would like to try 2.6.39 and the thing is that I would like to have tp_smapi and phc-intel modules in both kernels. Is it possible to have it without any serious hacking? I have only 3 ideas how I could achieve that: * Making a custom ebuild, which would build the modules, but install itself as a different package depending on the kernel version (eg tp_smapi-2.6.39-gentoo)? * Making a custom ebuild, which would build the modules for all kernel versions in one go... (is this possible?) * patching the gentoo-sources each time. kernel modules are CONFIG_PROTECTED, so they are not automatically removed when you uninstall / remerge the package (you have to remove them manually), so you just have to remerge the package after you change the /usr/src/linux symlink there is also the module-rebuild utility, that automatically remerges packages that installed a kernel module i.e. ln -sfn linux-VERSION1 /usr/src/linux module-rebuild -X rebuild ln -sfn linux-VERSION2 /usr/src/linux module-rebuild -X rebuild ... you might have to do module-rebuild populate before the first time... yoyo
Re: [gentoo-user] Cannot mount USB stick using Dolphin
On Sun, Jun 05, 2011 at 11:03:54PM +0100, Mick wrote: Both consolekit and polkit are running. What could be the problem? $ ps axf | grep polkit 8961 pts/1SN+0:00 \_ grep --color=auto polkit 5678 ?Sl 0:00 /usr/libexec/polkitd $ ps axf | grep console 5594 ?Ssl0:00 /usr/sbin/console-kit-daemon 9088 pts/1SN+0:00 \_ grep --color=auto console I'm starting e17 WM running startx on this box, unlike another similarly configured machine where mounting devices works fine but I start that with /etc/init.d/xdm (kdm). that console-kit-daemon is running doesn't mean your session is 'registered' with it, you can check with ck-list-sessions yoyo@tabletka ~ $ ck-list-sessions Session18: unix-user = '1000' realname = '(null)' seat = 'Seat18' session-type = '' active = FALSE x11-display = '' x11-display-device = '' display-device = '/dev/ssh' remote-host-name = ... is-local = FALSE on-since = '2011-06-06T07:48:11.669544Z' login-session-id = '' Session1: unix-user = '1000' realname = '(null)' seat = 'Seat1' session-type = '' active = TRUE x11-display = ':0' x11-display-device = '/dev/tty7' display-device = '' remote-host-name = '' is-local = TRUE on-since = '2011-06-01T17:09:37.282998Z' login-session-id = '' the defaults for disk mounting should be that only active (local?) sessions can do it... here it seems that a normaln console (text) / ssh seesions are registered in ck through pam, but X session's have to do that manulally, which i guess is done either in DM's xsession script or in the startup script of the chosen DE ... if your session is not thare, then just add ck-launch-session to the script you use to start the X session (~/.xinitrc in your case i guess, or the script you specify to startx) yoyo
Re: [gentoo-user] xpdf won't run - missing libpoppler.so.5
On Thu, Apr 28, 2011 at 09:45:55AM +, Alan Mackenzie wrote: Hi, Gentoo! I try to run xpdf from within Gnome. It doesn't run. On the virtual terminal, the error message is: xpdf: error while loading shared libraries: libpoppler.so.5: cannot open shared object file: No such file or directory . However there is a library /usr/lib/libpoppler.so.7. ^ revdep-rebuild ? Seems like poppler has been updated, but xpdf wasn't recompiled yet to use the new lib. I look in /var/db/pkg/app-text/xpdf-3.02-r4/DEPENDS, and see: =app-text/poppler-0.12.3-r3[xpdf-headers] ^^ ^ I currently have installed poppler-0.14.5-r1, which satisfies DEPENDS. Presumably, though, it incorporate so.7, which is useless for xpdf. This situation, requiring an older version of a dynamic library, can't be unheard of. What's the canonical way of getting .so.5? As a matter of interest, how do you say, in DEPENDS, that you need a library _between_ two version numbers rather than just = than one? Am I missing anything else important? you can say ( =something something) but my guess is, that xpdf doesn't _need_ the old version of lib, it just needs to be recompiled with the new version... yoyo
Re: [gentoo-user] win key takes me from X to VT
On Mon, Apr 18, 2011 at 12:27:03AM +0200, Jesús J. Guerrero Botella wrote: El día 18 de abril de 2011 00:01, Jesús J. Guerrero Botella jesus.guerrero.bote...@gmail.com escribió: Try to reset all shortcuts with: setxkbmap -option It doesn't change anything. The problem starts in kdm, before loging in, so it's nothing specific to a given user account. Oh, I forgot, it is nothing specific to kdm either. What I meant above is that it happens since I enter X. Or rather, since this is the default behavior in the console, we could more correctly say that it *continues* happening when I enter X, where it should not happen. I tested the lxde login manager and it has the same problem. It seems like X didn't switch the keyboard to raw mode or something like this... The win key on linux console swithes to a previou vt (don't know if it is intentional, or just a side effect of the kernel not correctly handling it) Sometimes, when an app freezes the whole X (usually when it grabs the keyboard and freezes) I have to use the magic sysrq keys to unraw the keyboard, which means I can that use alt-fX to swtich to text VTs, kill the app and return to X... however from that moment on until I restart X, the keyboard is not in raw mode and alt-fX and also the winkey switch consoles (like you describe) yoyo
Re: [gentoo-user] Problem with X fonts when restarting from hibernate-to-disk
On Wed, Mar 16, 2011 at 07:26:58PM -0400, Walter Dnes wrote: I'm not sure where to look. I've recently started having font problems when restarting my PC from hibernate-to-disk. Here are the symptoms... * I hibernate my home desktop machine when not using it * when I restart from hibernate, the following problems *SOMETIMES* occur * existing GUI windows are OK * launching new programs or dialogues (YES/NO, Apply/Cancel, etc) seems to have zero-size fonts * shutting down X and restarting it solves the problem, but if I have to go through that, it's almost as cumbersome as sudo /usr/sbin/poff and booting fresh I'm attaching a partial screen-capture of an existing gnumeric spreadsheet and a new one launched after waking up from hibernation. The tracker.gnumeric spreadsheet that was opened before hibernation is OK. The Book1.gnumeric spreadsheet was opened after waking up from hibernate. Note the missing File Edit View etc text menus. The location and formula areas are very thin, compared to the same areas in tracker.gnumeric. This happens to Opera, AbiWord, etc. I did this partial screencapture with Gimp. I left it open before hibernating. The radiobuttons and icons were OK in the screencapture dialogue, but the fonts were screwed up on the dialogue. I managed to do the screencapture from memory of which button was which. Might be that something happens to X's DPI settings during the suspend/resume... You can try to comaper the outputs of xdpyinfo | egrep dimensions|resolution and maybe xrandr befor and after the suspend. Also all the applications you metion are GTK V(well, I'm not sure about opera...), does the same happen to qt, qk, plain xlib applications also? yoyo
Re: [gentoo-user] Anyone get Acer laptop internal microphone working in Gentoo?
On Sun, Feb 20, 2011 at 08:33:22PM -0500, Walter Dnes wrote: On Sun, Feb 20, 2011 at 12:09:27PM +, Mick wrote Indeed, the Gentoo Alsa Guide still says pretty much the same thing: Please note that for ease of use, all examples show ALSA built as modules. It is advisable to follow the same as it then allows the use of alsaconf which is a boon when you want to configure your card. I've added some options for my alsa modules in /etc/modprobe.d/alsa-base.conf, e.g.: options snd-hda-intel enable_msi=1 This is exasperating. After re-building alsa sound support as kernel modules, rather than built into the kernel, I see some improvement. When I blow into the internal mic I hear it from the laptop speakers. With mic-boost turned up, I can hear myself echoing when I talk into the internal mic or into an external mic. Turn up the boost high enough, and the external mic generates a mean high-ptched feedback squeal, unless I also plug in headphones. So the hardware is working now, *BUT I STILL CAN'T RECORD THE BLEEPING THING*. When I try ffmpeg -f oss -i /dev/dsp audio.wav it thinks it's recording, but the output file is only hiss. Ditto for the command ffmpeg -f alsa -i plughw:0,0 audio.wav. Seems that the mic works, but is not selected for capture... (routing it to speakers recording from it are two different things ;) You basically need to tell alsa which sources to record from, which may be different from which are just unmuted in the mixer (and thus playing from the speakers). I've seen a few different sets of alsa controls on different sound cards that control the record sources, these are the two I remember: Run alsamixer, pres F4 (capture) and either look for - a mic control (along with CD, Line, Aux,... controls), then make sure it has a red CAPTURE text (spacebar toggles CAPTURE) - if you dont't have separate Mic, Line, ... controls, try looking for a Input source control, you should be able to use up/down arrows to change it to mic Also in both cases make sure a Capture control has CAPTURE on too.. You might have more Mic sources two choose from, or you might also have a Mic Select control, with which you can switch between different mics, so you might have to experiment to find which is the correct one... yoyo
Re: [gentoo-user] irritating cron habit : solved
On Mon, Feb 21, 2011 at 12:39:43AM -0500, Philip Webb wrote: 110220 Philip Webb wrote: 110220 Florian Philipp wrote: Just change your cron job to look like 'test -e /var/run/dhcpcd.pid fetchmail' That's by far the simplest it still fetches the mail, so we'll see if it also avoids the occasional internal spam msgs. Indeed it does: I tried delaying starting the I/net connection there is no 'dead.letter' file. Thanks again. PS the file is /var/run/dhcpcd-eth0.pid , not as above. btw, if I need to check if the network is up in a script, I usually do ping -q -c1 -w4 some.remote.host /dev/null 21 command-to-run-if-remote-host-reachable It the advantage that it checks directly connection to the host you wish to connect to, so it also won't run the command if your network is up, but the remote host is inaccessible... If your server doesn't respond to pings, just use some other server (eg google's public dns 8.8.8.8) The -w4 paramater controls how long to wait for reply if the network is up, but the reply is not comming (ie a network problem, if your network connect is down, ping will return immediately with a network unreachable or unknown host error) yoyo
Re: [gentoo-user] k3b: Drive not found ...
On Sun, Jan 30, 2011 at 09:11:35AM +0100, meino.cra...@gmx.de wrote: Hi Jon, thanks for your reply. Unfortunately it does not change anything here. The problem remains... Best regard, mcc Try checking / restarting hal and you could also try to re-run etc-update, to make sure, that /etc/dbus-1/system.d/hal.conf is updated (iirc there were some changes in that file some time ago...) I don't know if dbus-monitor --system or a dbus log would tell you which call wasn't allowed, maybe increasing dbus's debug level would include that in the log... yoyo Jon Cox hypers...@gmail.com [11-01-30 09:08]: Hello, I was recently having similar issues with a computer I picked up from my university's surplus that I put a fresh Gentoo install on. The issues went away after I restarted dbus, a la: $ /etc/init.d/dbus restart It may not fix your issues, but it did make mine vanish. Best of luck! -Jon On Sunday 30 January 2011 00:01:25 meino.cra...@gmx.de wrote: Hi, My setup is ASUS Crosshair IV formula. Attached to the box is a USB-IDE converted. Attached to this converter there is my old DVD burner. This setup has worked for reading and writing DVDs/CDs in the past. But there must be a update or something which kills that There is: solfire:/rootl /dev/sr0 brw-rw 1 root cdrom 11, 0 2011-01-30 07:51 /dev/sr0 solfire:/root Therefore the burner is detected by the kernel an appropiate permissions are given to it. I am member of the group cdrom. When k3b is started from the commandline I see this output: K3bQProcess::QProcess(0x0) QStringList Solid::Backends::Hal::HalManager::findDeviceByDeviceInterface(const Solid::DeviceInterface::Type) error: org.freedesktop.DBus.Error.AccessDenied K3bQProcess::QProcess(0x0) k3b(22035)/kdeui (kdelibs): Attempt to use QAction view_projects with KXMLGUIFactory! k3b(22035)/kdeui (kdelibs): Attempt to use QAction view_dir_tree with KXMLGUIFactory! k3b(22035)/kdeui (kdelibs): Attempt to use QAction view_contents with KXMLGUIFactory! k3b(22035)/kdeui (kdelibs): Attempt to use QAction location_bar with KXMLGUIFactory! solfire:/home/mccramerQStringList Solid::Backends::Hal::HalManager::findDeviceByDeviceInterface(const Solid::DeviceInterface::Type) error: org.freedesktop.DBus.Error.AccessDenied QStringList Solid::Backends::Hal::HalManager::findDeviceByDeviceInterface(const Solid::DeviceInterface::Type) error: org.freedesktop.DBus.Error.AccessDenied QStringList Solid::Backends::Hal::HalManager::findDeviceByDeviceInterface(const Solid::DeviceInterface::Type) error: org.freedesktop.DBus.Error.AccessDenied QStringList Solid::Backends::Hal::HalManager::findDeviceByDeviceInterface(const Solid::DeviceInterface::Type) error: org.freedesktop.DBus.Error.AccessDenied QStringList Solid::Backends::Hal::HalManager::findDeviceByDeviceInterface(const Solid::DeviceInterface::Type) error: org.freedesktop.DBus.Error.AccessDenied QStringList Solid::Backends::Hal::HalManager::findDeviceByDeviceInterface(const Solid::DeviceInterface::Type) error: org.freedesktop.DBus.Error.AccessDenied QStringList Solid::Backends::Hal::HalManager::findDeviceByDeviceInterface(const Solid::DeviceInterface::Type) error: org.freedesktop.DBus.Error.AccessDenied QStringList Solid::Backends::Hal::HalManager::findDeviceByDeviceInterface(const Solid::DeviceInterface::Type) error: org.freedesktop.DBus.Error.AccessDenied k3b(22035)/kdecore (services) KMimeTypeFactory::parseMagic: Now parsing /usr/share/mime/magic k3b(22035)/kdecore (services) KMimeTypeFactory::parseMagic: Now parsing /home/mccramer/.local/share/mime/magic Unfortunately I have no clue, what it wants to tell me. Only this Permission denied alerts me...but where to patch/modify what? A hint for a fix is very appreciated ;) Thanks a lot in advance! Best regards, mcc
Re: [gentoo-user] Re: Simultaneously emerging multiple packages with same dependencies
On Thu, Jan 27, 2011 at 03:33:21PM +0200, Nikos Chantziaras wrote: On 01/27/2011 03:11 PM, Dale wrote: [...] I am using the -j option for the first time now. I'm updating KDE. It seems to work fine. It doesn't scroll all the stuff like with a regular emerges but this new rig is so fast, I can't read it anyway. I did have a package to fail and it spit out the error for me to read. You don't need that if you have MAKEOPTS set in your make.conf, which is preferred. The -j option of emerge emerges multiple packages, while with MAKEOPTS set to -j4 or whatever does a parallel build in the same package (meaning compiling multiple source files at the same time). It's preferred because with emerge -jN the last package will only use one CPU, while with -jN in MAKEOPTS even the last package will use N CPUs. Furthermore, emerge can't always build N packages at the same time because one can depend on the other, so it will have to wait until the dependency is built. On the other hand, unpacking, configure and install stages are not parallel and emerge can do those in parallel for different packages... The best would be somewhere in the middle ;) There are also the load-average options to -j, i.e.: MAKEOPTS=-j -l5 emerge -j --load-average=5 which makes make spawn parallel processes while load average is below 5 and the same for emerge spawning parallel ebuilds (when make isn't parallel enough) yoyo
Re: [gentoo-user] Re: Simultaneously emerging multiple packages with same dependencies
On Thu, Jan 27, 2011 at 03:12:49PM +0100, J. Roeleveld wrote: On Thursday 27 January 2011 15:05:25 YoYo Siska wrote: On Thu, Jan 27, 2011 at 03:33:21PM +0200, Nikos Chantziaras wrote: On 01/27/2011 03:11 PM, Dale wrote: [...] I am using the -j option for the first time now. I'm updating KDE. It seems to work fine. It doesn't scroll all the stuff like with a regular emerges but this new rig is so fast, I can't read it anyway. I did have a package to fail and it spit out the error for me to read. You don't need that if you have MAKEOPTS set in your make.conf, which is preferred. The -j option of emerge emerges multiple packages, while with MAKEOPTS set to -j4 or whatever does a parallel build in the same package (meaning compiling multiple source files at the same time). It's preferred because with emerge -jN the last package will only use one CPU, while with -jN in MAKEOPTS even the last package will use N CPUs. Furthermore, emerge can't always build N packages at the same time because one can depend on the other, so it will have to wait until the dependency is built. On the other hand, unpacking, configure and install stages are not parallel and emerge can do those in parallel for different packages... The best would be somewhere in the middle ;) There are also the load-average options to -j, i.e.: MAKEOPTS=-j -l5 emerge -j --load-average=5 which makes make spawn parallel processes while load average is below 5 and the same for emerge spawning parallel ebuilds (when make isn't parallel enough) yoyo Hmmm... didn't know about that one yet. Does that mean that by doing it like that, the emerge-process (and compile- processes) will try to keep the load average at 5 and if that is lower, it will keep adding more processes? Yes. It might not be perfect, but mostly it works pretty well. Once make started 10 or so process, which ate all my ram, because I forgot to reenable swap, when I was playing with something before that :) yoyo
Re: [gentoo-user] Re: Paste into vim keeping indention or original?
On Thu, Jan 27, 2011 at 02:28:47PM -0500, Mike Gilbert wrote: On Thu, Jan 27, 2011 at 2:22 PM, Mark Knecht markkne...@gmail.com wrote: I solved it by creating a .vimrc file and putting set pastetoggle=F2 Running :set paste will do the job as well if you don't want to assign a hot key for it. BTW, if - vim has access to X (you run it on your local machine or from ssh -X or something similar) - is compiled with X support (check with vim --version | grep +X11) - and you :set mouse=a then you can paste by middle clicking in vim (not shift-middle click), which should paste the text as is... The difference is that with shift-middle click, or with vim that cannot talk to X, the terminal sends the selected text to vim as normal input (as if you would type it) and thus its get indented/formated/etc.. If you have mouse=a set and vim can talk to X, when you middle click it will ask X for the selection and insert it as is without any formatting yoyo
Re: [gentoo-user] Re: Simultaneously emerging multiple packages with same dependencies
On Thu, Jan 27, 2011 at 08:18:34PM +0100, J. Roeleveld wrote: On Thursday 27 January 2011 19:56:23 Allan Gottlieb wrote: On Thu, Jan 27 2011, Neil Bothwick wrote: On Thu, 27 Jan 2011 17:09:27 +0200, Nikos Chantziaras wrote: So on a 20 package world update, only 19 are faster while the 20th runs at the same speed? Where's the loss there? Even if the last were slower, it would be worth it. Given the amount of time unpack/configure/install of most packages needs (very short), my observation is that it would not be worth it. Even if that were true, how much time would you have to save to justify adding -j 2 to EMERGE_DEFAULTS in make.conf? But it's not true, large packages spend a lot of time on these phases of the install. OK I'm convinced since I know that those phases do take noticeable time. I have a 4 processor i7 model 620 (2 cores, doubled for hyperthreading) and have set MAKEOPTS=-j5. If I add -jobs=2 to EMERGE_DEFAULT_OPTS, should I lower MAKEOPTS to 3 (to 4)? thanks, allan You could, as if you leave it at -j5, you can end up with 2 * 5 = 10 processed, eg: similar as if running with MAKEOPTS=-j10 I think the option that YoYo came with is a good compromise: # MAKEOPTS=-j -l5 emerge -j --load-average=5 Next time I am doing a big upgrade, I'm going to test that to see how it behaves. I was just building the whole system for my notebook in a chroot on my desktop machine ( I use FEATURES=buildpkg to build binary packages in the chroot on a fast desktop machine and then upgrade the notebook with the binary packages) and I used exactly that (-j -l5 for bot make and emerge). Can't say if it really is better or not ;) but most of the time all four cores were busy, though sometimes I saw even 6 or 7 gcc-s simultaneously in top ;) emerge was running 3 to 4 jobs most of the time, sometimes dropping to 1 and once I saw it emerging about 10 parallel packages ;) (mostly small things, which I guess were doing a lot of unpacking/configuring/installing but almost nothing of compiling ;) btw, just now I got this error from dev-lang/v8: SCons error: option -j: invalid integer value: '-l4' seems scons honors MAKEOPTS, but doesn't understand the loadaverage version (-j -l4) yoyo
Re: [gentoo-user] Something like nullfs (FreeBSD) available in linux
On Wed, Jan 26, 2011 at 01:37:38PM +0100, Matthias Fechner wrote: Dear list, I want to build up a NFS share for several host. The base system is always equal, only some files are different (files in /etc and /var, maybe more). The structure should look like: Base system is on /basesystem Then I want a directory for every host: /nfsroot/host1 /nfsroot/host2 .. /nfsroot/hostn Client 1 will boot via lan and mount the /nfsroot/host1 via NFS. On the server the mount could look like: mount -t nullfs /basesystem /nfsroot/host1 mount -t nullfs /basesystem /nfsroot/host2 If the host1 changes now a file the changed file is not stored in /basesystem but in /nfsroot/host1. (so you can think about a filesystem in two layers, the ground layer is /basesystem and readonly, changes will go to /nfsroot/hostn) If i upgrade the /basesystem all files should be available for all hosts expect the files the host changed by itself (they are stored in the nfsroot/hostn directory). In FreeBSD there is a file system available called nullfs which could be used for this. Is there a similar filesystem for Linux available, too? Unionfs or aufs should be able to do that. yoyo
Re: [gentoo-user] Emerging package via NFS ?
On Wed, Dec 15, 2010 at 10:15:12AM +0100, Thomas Drueke wrote:
Hi,
is it possible to emerge packages to a $ROOT directory mounted via NFS ?
The setup is
- machine A is equipped with a Quad core CPU
- machine B is equipped with an N330 Atom-CPU
- machine A is doing the system update on a local chroot-environment
for machine B and generates binary packages. These packages are
installed on machine B using the binary package feature of portage.
I expected that the above setup would give an performance improvement
over letting machine B do the portage update itself. However a trial run
did not show significant improvement that justifies the effort. Machine
B still needs a reasonable amount of time to fetch unpack and install
the packages.
An alternative way might be to mount machine B's / directory via NFS
and change make.conf's $ROOT variable to that mount point.
Does that sound as a reasonable approach ?
I had a very old machine, that was really slow. Compiles could be
offloaded by distcc, but even the ./configure-s and portage stuff
(checking, upacking, ...) was reaaly slow...
So I just used to export / through nfs, mounted it on a fast amd64 and
basically did (other is the slow machine)
mount other:/ /mnt/other
mount -t proc proc /mnt/other/proc
mount --bind /dev /mnt/other/dev
mkdir /tmp/other
mount --bind /tmp/other /mnt/other/var/tmp/portage
mkdir /home/gentoo-other
mount --bind /home/gentoo-other /mnt/other/home/gentoo
linux32 chroot /mnt/other /bin/bash
emerge.
For the last mkdir/mount, I have DISTDIR=/home/gentoo/distfiles and
PKGDIR=/home/gentoo/packages in make.conf, you can do that with the
standart /usr/portage/{distfiles,packages}
This way most of the compile is done localy on the fast machine.
yoyo
Regards,
Thomas
Re: [gentoo-user] New monitor problems. sighs
On Wed, Dec 15, 2010 at 11:20:33AM +0100, J. Roeleveld wrote: On Wednesday 15 December 2010 11:10:03 Helmut Jarausch wrote: On 12/15/10 10:56:21, Adam Carter wrote: I confirmed with lsmod and even rmmod and modprobed it back in a few times. Still no joy. I really think it is either the card or the xorg.conf file. Thing is, when Mandriva failed to work, that makes me wonder about the card. It was using the generic nv driver and it still wouldn't work. IMO its a waste of time to worry about any of this other stuff - the EE's in the log file are ERRORs, directly pointing at serious issues with missing modules. So either the modules dont exist, or xorg cant find them. There's no evidence of a problem with the nvidia module or the hardware. I don't think that's true. In my Xorg.0.log I do have (EE) Failed to load module dri (module does not exist, 0) (EE) Failed to load module dri2 (module does not exist, 0) and still, X11 is running just fine. I think these modules are superfluous with a recent Xorg version. Helmut. I agree, just checked on all the machines I am running Gentoo on with Nvidia cards and they all have those 2 EE-lines. Both of these have 2 monitors hooked up and that is working correctly. IIRC they are used by the other (open) drivers to communicate with the dri modules in kernel... nvidia X driver communicates with the nvidia kernel driver on its own and doesn't use those two X modules eselect opengl creates those two as symlinks whnen you select xorg-x11, and when you select nvidia, they are just not created on intel machine: tabletka ~ # ls -l /usr/lib/xorg/modules/extensions/libdri.so lrwxrwxrwx 1 root root 45 Dec 13 16:02 /usr/lib/xorg/modules/extensions/libdri.so - ../../../opengl/xorg-x11/extensions/libdri.so tabletka ~ # qfile /usr/lib/opengl/xorg-x11/extensions/libdri.so x11-base/xorg-server (/usr/lib/opengl/xorg-x11/extensions/libdri.so) that said, on all my nvidia setups (with eselect opengl nvidia) I get those errors and everything works ok, on all others (well, intel only ;) with eselect opengl xorg-x11 the modules are there... yoyo
Re: [gentoo-user] New monitor problems. sighs
On Wed, Dec 15, 2010 at 05:39:27AM -0600, Dale wrote: ... I'm attaching both the new just tried xorg.conf and the xorg.log file. No grep or anything this time. Let me know if you see anything fishy or that needs changing. Thanks. Dale ... [ 2082.083] (II) Loading extension NV-GLX [ 2082.101] (EE) NVIDIA(0): Failed to obtain a shared memory identifier: Function not [ 2082.101] (EE) NVIDIA(0): implemented ... [ 2082.120] (EE) Failed to initialize GLX extension (Compatible NVIDIA X driver not found) [ 2082.138] Fatal server error: [ 2082.138] Failed to initialize the OpenGL server this looks like the problem, but I don't know whats wrong (it certainly should not be a mismatch between nvidia X driver and nvidia kernel module, that produces a different error ;) google seems to find a lot of stuff for the Failed to obtain, but I didn't look into it.. yoyo
Re: [gentoo-user] KDE-4 multi-monitor + fullscreen applications
On Wed, Nov 17, 2010 at 11:41:25PM +0100, Florian Philipp wrote: Am 17.11.2010 23:26, schrieb Alan McKinnon: Apparently, though unproven, at 00:08 on Thursday 18 November 2010, Florian Philipp did opine thusly: Hi list! Today, KDE nearly killed a presentation I held and now I want to understand what's going on: Following setup: One laptop, two outputs (internal display + projector). Now I configure KDE to expand the desktop on both (instead of simple cloning). So far, so good. For anyone to help at all, we'll need to know your hardware and video drivers, plus versions in use of X.org and it's drivers, plus relevant config stuff. Everything else is highly configurable and subject to the whim of driver writers and the user. And there's always nVidia's stance to be taken into account as well Ah, right, forgot about that. Intel GMA HD graphics (i915 driver), x11-base/xorg-server-1.8.2 (USE=udev -hal) and x11-base/xorg-drivers-1.8 No xorg.conf. Tried it with composite effects off and on. KDE is on version 4.4.5 and some packages 4.4.7 (current stable). First question: How does KDE choose on which output the standard desktop ends up and which gets the second set of desktop background + plasma widgets? It seems like the one with the higher resolution is standard and on a draw, it is the right-most. Is that correct? Can it be configured? Now that I have both desktops, I open Acroread or Okular and start the fullscreen/presentation mode. What happens is that the presentation is deterministically opened on one of the displays. What I don't understand is how it chooses which one it uses? xrandr 1.3 has a new option to say which output should be 'primary' you can try something like xrandr --output LVDS1 --mode 1024x768 --pos 0x0 --primary --output VGA1 --mode 1024x768 --righ-of LVDS1 However IIRC kde used to ignore which display was primary (reported as xinerama screen 0) and somehow decided on its own order... Here okular works correclty (well, at least Current screen and Screen XX used to work, don't remember for Default screen and can't test right now...), but right now I'm using fluxbox as window manager and not kwin, but I it would be weird if it actually made things break.. ;) yoyo It doesn't depend on the placement of the window (although other applications like Flash in Firefox, MPlayer, Kaffeine and Gwenview do). It doesn't always open on the secondary or standard desktop (as specified above). It rather seems like it always opens on the one with the higher resolution and if both are equal, it opens on the left-most. So, what happened when I tried to hold my presentation? The projector had a low resolution (1024x768) and therefore neither Acroread nor Okular showed on fullscreen on the projector. None of my previous tests showed that problem since I used two displays with equal resolution. Great fun! In the end, I cloned the output and thereby gave Okular no other choice. (Lucky me that I didn't any additional notes or anything on the other display ...) What can I do to influence this behavior? Edit: I just noticed that both applications have settings for this. However, they are ignored and the setting in Acroread is even reset to Current display each time I close the settings dialog! What is going on here? Thanks in advance! Florian Philipp
Re: [gentoo-user] which NIC is which?
On Mon, Nov 15, 2010 at 12:35:35PM +0100, Stefan G. Weichinger wrote: Am 15.11.2010 10:39, schrieb Steffen Loos: Maybe a little bit late but: As a summary-tool all the info is gattered and shown by lshw. yep, thanks. Although it should be possible to just ask the kernel somehow, shouldn't it? I usually do (especially from a livecd, when I want to know which drivers to enable in the kernel for a new device ;) y...@desktop ~ $ ls -l /sys/class/net/eth?/device/driver lrwxrwxrwx 1 root root 0 2010-11-15 14:38 /sys/class/net/eth1/device/driver - ../../../../bus/pci/drivers/tulip lrwxrwxrwx 1 root root 0 2010-11-15 14:38 /sys/class/net/eth2/device/driver - ../../../../bus/pci/drivers/r8169 yoyo
Re: [gentoo-user] Setting up two monitors
On Fri, Nov 05, 2010 at 09:38:07PM +, Mick wrote: On Friday 05 November 2010 11:11:04 YoYo Siska wrote: On Thu, Nov 04, 2010 at 11:08:23PM +, Mick wrote: On Thursday 04 November 2010 21:36:46 Florian Philipp wrote: Am 04.11.2010 21:17, schrieb Mick: [...] Then I ran xrandr again as Florian suggested and this is what it shows: $ xrandr --output DVI-0 --auto --this gives 1920x1080 $ xrandr --output DVI-0 --right-of-VGA-0 --verbose xrandr: screen cannot be larger than 1920x1920 (desired size 3200x1080) As a result it does not place the DVI on the right of the VGA driven monitor. Can you please explain this error to me - why does it complain? Hmm, do you still have an xorg.conf file or changed settings in /etc/X11/xorg.conf.d? If you have, can you post it please? I think it is related to the 'SubSection Device Virtual xdim ydim' setting but I'm not sure. In any case, if I were you, I'd try running without any xorg.conf and see whether auto-configuration can handle it. Oh, and if you are still on x11-base/xorg-server-1.7.*, please try x11-base/xorg-server-1.8.2 with USE=udev -hal Thanks again Florian, I do not have an xorg.conf. I am running x11-base/xorg-server-1.7.7-r1. I have been waiting on 1.8.2 to go stable. Googling around I suspect I know what the error is: $ xrandr -q Screen 0: minimum 320 x 200, current 1280 x 1024, maximum 1920 x 1920 is telling me that my ATI X600 can only do a max of 1920 x 1920. Above that I will need to set up a virtual screen (and it won't be able to do dri). Without an xorg.conf file it is failing because it is not given a virtual screen to expand its physical capability beyond 1920x1920. Any idea if I can set up a virtual screen using the .fdi files? Intel drivers (for my thinkpad notebook) had a similar problem. If you didn't use an xorg.conf, they would set up the max screen size to the maximum possible resolution on one of the monitors... I haven't found a way to change that without an xorg.conf... (didn't have much motivation as I just always used an xorg.conf, event with hal... and I'm on ~arch, so its not much of an issue now...) yoyo PS right now, the current intel driver I have seems to have a hard maximum of 2048x2048 on my card, though I remember going above that in the past... ;(( (I was wondering how come MSWindows works fine - not sure if it uses virtual screens ...) Are you saying that the maximum mode of the video card is determined by the driver? Two different ati cards here, both show 1920x1920 as the maximum. The card I am having this problem with has 256M memory. The other has 1G memory (in MSWindows) while Gentoo only shows: Memory at d000 (32-bit, prefetchable) [size=256M] I/O ports at 2000 [size=256] Memory at cfef (32-bit, non-prefetchable) [size=64K] [virtual] Expansion ROM at cfe0 [disabled] [size=128K] If the maximum mode available changes with the driver version, does this mean that one day I need to set up a virtual screen size and next day the driver is updated and virtual screen is no longer required? From what I know (but I may be completely wrong ;) its this way: the maximum size xrandr reports is what X thinks is the maximum possible framebuffer size... Its reported by the graphics card driver, which (I think) should be the maximum resolution the graphics card supports. This depends on the card, the amount of memory it has (which gets a bit complicated with cards with shared memory, that can dynamically allocate how much they actually need) etc... AFAIK this value is constant for X (it can't change without restart), and X will never allow you to have a large 'virtual screen' (i.e. the space in which all outputs have to fit) But I've seen drivers that don't report the maximum they support, but the maximum resolution of the actually connected display: The driver should report to X, what display devices are connected to the card and which resolutions they support -- the things you see in xrandr output. It seems that that some drivers report the maximum of these resolutions (ati and older intel, though newer intel drivers seem to report 2048x2048 or 4096x4096, I can't say for newer ati, as I don't have an ati card...) I guess that this is mostly a 'historical' issue from the times when Xserver/drivers did not support 'dynamic' monitor configuration (ie adding/removing monitors) without restarting the Xserver... You can override this value with the Virtual option (It used to be in the screen section of xorg.conf, now the correct place seems to be in the Device section). IIRC, the driver will still change it to the maximum it supports, if you made it bigger, but not to the maximum resolution of the connected displays ;) Also, some
Re: [gentoo-user] Setting up two monitors
On Thu, Nov 04, 2010 at 11:08:23PM +, Mick wrote: On Thursday 04 November 2010 21:36:46 Florian Philipp wrote: Am 04.11.2010 21:17, schrieb Mick: [...] Then I ran xrandr again as Florian suggested and this is what it shows: $ xrandr --output DVI-0 --auto --this gives 1920x1080 $ xrandr --output DVI-0 --right-of-VGA-0 --verbose xrandr: screen cannot be larger than 1920x1920 (desired size 3200x1080) As a result it does not place the DVI on the right of the VGA driven monitor. Can you please explain this error to me - why does it complain? Hmm, do you still have an xorg.conf file or changed settings in /etc/X11/xorg.conf.d? If you have, can you post it please? I think it is related to the 'SubSection Device Virtual xdim ydim' setting but I'm not sure. In any case, if I were you, I'd try running without any xorg.conf and see whether auto-configuration can handle it. Oh, and if you are still on x11-base/xorg-server-1.7.*, please try x11-base/xorg-server-1.8.2 with USE=udev -hal Thanks again Florian, I do not have an xorg.conf. I am running x11-base/xorg-server-1.7.7-r1. I have been waiting on 1.8.2 to go stable. Googling around I suspect I know what the error is: $ xrandr -q Screen 0: minimum 320 x 200, current 1280 x 1024, maximum 1920 x 1920 is telling me that my ATI X600 can only do a max of 1920 x 1920. Above that I will need to set up a virtual screen (and it won't be able to do dri). Without an xorg.conf file it is failing because it is not given a virtual screen to expand its physical capability beyond 1920x1920. Any idea if I can set up a virtual screen using the .fdi files? Intel drivers (for my thinkpad notebook) had a similar problem. If you didn't use an xorg.conf, they would set up the max screen size to the maximum possible resolution on one of the monitors... I haven't found a way to change that without an xorg.conf... (didn't have much motivation as I just always used an xorg.conf, event with hal... and I'm on ~arch, so its not much of an issue now...) yoyo PS right now, the current intel driver I have seems to have a hard maximum of 2048x2048 on my card, though I remember going above that in the past... ;(( yoyo
Re: [gentoo-user] Setting up two monitors
On Thu, Nov 04, 2010 at 08:43:25AM +0100, Florian Philipp wrote: Am 04.11.2010 08:38, schrieb Mick: PS. Another thing I noticed with the WinXP setup is that the application windows seem to be screen aware. On the left monitor they will maximise only to cover fully the left hand screen not the right hand. The same happens when maximising an application window on the right. I don't remember seeing this in Linux - applications I think maximised across both screens. Again, I don't know what desktop environment you are using but that works flawlessly on KDE. Just to make it a bit more clear: xrandr is used to setup the resolution and position of the monitors (you can make them clone each other, overlap, be alongside / above / below the other...) How the windows / panels behave depends on your windows manager/desktop environment (or on the panels themselves). X server provides them with enough information about the layout of the monitors, and they have to use it. So it depends on which DE or window manager you use... In kde3, there was a configuration option for kwin, whether windows should be maximized across all screens or on single screen... I can't find it in kde4 settings right now, but I have only single head card here and I guess it would be under Multiple Monitors option in settings, which just says You don't appear to have this configuration for me ;) Plasma in kde4 manages things per monitor, so panels should be only on one monitor (and you can't get them across multiple monitors, you have to have a separate panel on each)... Recent versions of fluxbox allow you to have the toolbar on a certain monitor (head) or across all heads... Don't know how it is when maximizing windows (some time ago I used to patch it to make it an option, didn't play with it lately...) I can't say anything for gnome or other DEs/WMs... yoyo
Re: [gentoo-user] su in konsole takes much longer to complete in KDE 4.5.1
On Mon, Sep 20, 2010 at 11:53:34PM +0200, Thomas Drueke wrote: Thanks for hints, but no luck so far. Yohan, using xterm instead of konsole results in the same delay. To rule some other things out, you could also try: unset DISPLAY su - DISPLAY is one of the differences between a text konsole and anything under X... Might be that some bashrc/profile script tries to do something with X if it sees DISPLAY, but isn't able to connect to X under root... (maybe some xauth stuff..) yoyo Alan, hosts contains the hostname (FQDN) for eth0 and also alocalhost entry. Plus wireshark didn't show any network traffic during the delay (for both eth0 and lo). Is there any of the new services from KDE 4 which requires some configuration concerning DNS or similar network services ? Regards, Thomas Am 20.09.2010 23:11, schrieb Alan McKinnon: Apparently, though unproven, at 20:08 on Monday 20 September 2010, Thomas Drueke did opine thusly: Hi, I installed KDE 4.5.1 over the weekend following the remove-all-old-kde-packages-first approach on the gentoo webpage. So far everything seems to be fine except one thing. When I type su - in konsole it takes 20-30 seconds to complete. Doing the same on a text console the command completes immediately. I don't have NIS or LDAP enabled. strace su - came back with an authentication failure immediately so no much info from there. Also top didn't show any suspicious process consuming the time. I found a thread from may which might be related to my observation (KDE takes ages to show password screen after suspend). The solution there was to upgrade to KDE 4.4.4 which does not fit here. Google didn't show much on this topic as well. Any ideas what might cause the delay or how to get more close to the root cause ? 20-30 second delays due to DNS timeouts have hit me so many times it's always the first thing I check, even when it seems irrelevant. Does your machine have a local hostname, and do you have an entry for it in either DNS or /etc/hosts?
Re: [gentoo-user] Booting Gentoo from USB stick
On Wed, Sep 15, 2010 at 08:34:33AM +1000, Jake Moe wrote: On 15/09/10 04:28, YoYo Siska wrote: On Fri, Sep 10, 2010 at 07:29:01AM -0400, David Relson wrote: On Fri, 10 Sep 2010 11:05:12 +0200 J. Roeleveld wrote: On Friday 10 September 2010 10:43:30 Jake Moe wrote: On 10/09/2010 5:27 PM, Maciej Grela wrote: 2010/9/10 Jake Moejakesaddr...@gmail.com: Hello all, I've been thinking about creating a Gentoo USB stick for install and rescue purposes (and, of course, just to see if I could). I've mostly followed the Gentoo handbook (I used a single 4GB partition for the whole system, and no swap). I've used genkernel for the kernel (so I can have a multi-system capable kernel). I've gotten GRUB installed and working. My problem comes in after what I believe is the init process: * Checking root filesystem ... fsck.ext2: No such file or directory while trying to open /dev/sda1 /dev/sda1: The superblock could not be read or does not describe a correct ext2 filesystem. If the device is valid and it really contains an ext2 filesystem (and not swap or ufs or something else), then the superblock is corrupt, and you might try running e2fsck with an alternate superblock: e2fsck -b 8193device * Filesystem couldn't be fixed :( [ !! ] Give root password for maintenance (or type Control-D to continue): If I give the root password, I can find no /dev/sda1. However, mount shows /dev/sda1 on /, and there *is* a /sys/block/sda folders, with a sda1 folder in that as well. It's almost like it had /dev/sda1, but then lost it somehow. Does anyone have any idea what's going on here? Any help would be appreciated. Have you seen http://www.sysresccd.org/Main_Page ? It's based on Gentoo, you could check what they did to boot from a usb stick. Br, Maciej Grela Excellent, thanks for that, I hadn't found it in my previous searches. I'll have a look there. Jake Moe Had a similar issue a while ago when I was playing around with this myself. Take a look at the linux boot parameters. The 'theoretical' part is: You need to let the kernel initialize the USB-stick before trying to access it. (This can take some time) There is a delay-option, just can't remember the proper name off-hand. -- Joost I've got USB booting working in a syslinux environment. A delay of 12 seconds is working for me. The syslinux.cfg stanza I use is: LABEL usb KERNEL linux APPEND rootdelay=12 root=/dev/sda2 The usual way for linux on removable usb sticks / disks is to use LABEL or UUID to identify the disks and not the device names, because they will be different in different computers ;) The downside is that you need an initrd to mount the root partition... I think that the usual initrd generated by genkernel works... If you created the rootfs with: mkfs.ext2 -j -LUSBGentoo /dev/sdXY then you can change the kernel parameter to root=LABEL=USBGentoo and your fstab to: LABEL=USBGentoo / ext3 ... You can also use the uuid of the filesystem, find it out with dumpe2fs -h /dev/sdb2 | grep UUID and then use UUID=XXX instead of LABEL=XXX I never really played around with grub and USB booting, so I use syslinux. I create a small FAT partition with syslinux, kernel and initrd image (it gets also pretty handy when you sometimes need to copy something from a windows machine ;) and a second regular ext3 partition for the rootfs. Basically you would do: - partition the stick, mark the FAT partition as bootable/active - format the partitions: - mkfs.vfat -nUSBData /dev/sdX1 - mkfs.ext2 -j -LUSBGentoo /dev/sdX2 - install syslinux (on the FAT partition): - syslinux /dev/sdX1 - mount /dev/sdX2, install gentoo in the usual way - compile the kernel and initrd, make sure required USB stuff is in the kernel (theoretically it could be as modules in initrd... but in-kernel is safer :) if you are in a hurry, or don't know how to create them, get them from a gentoo livecd ;) don't forget to also copy the modules (/lib/modules-XXX/...) from the livecd to the rootfs. - put the kernel and initrd on the FAT partition (I name them vmlinuz.img and initrd.img) - edit syslinux.cfg (on the FAT partition), see http://syslinux.zytor.com/wiki/index.php/SYSLINUX#How_do_I_Configure_SYSLINUX.3F a very simple one from my USB disk: DEFAULT linux LABEL linux SAY Now booting USBGentoo KERNEL vmlinuz.img APPEND root=LABEL=USBGentoo initrd=initrd.img you might also add rootdelay=10 to the options if the usb stick/disk isn't detected quick enough umount, reboot, set the computer to boot from usb, enjoy... ;) Xorg without a config seems to work pretty well on most computers these days, IIRC the alsa modules for soundcards are also autoloaded, so you don't need any fancy hw detection to have a desktop running from USB stick ;) yoyo BTW there is also a manual way
Re: [gentoo-user] Booting Gentoo from USB stick
On Fri, Sep 10, 2010 at 07:29:01AM -0400, David Relson wrote: On Fri, 10 Sep 2010 11:05:12 +0200 J. Roeleveld wrote: On Friday 10 September 2010 10:43:30 Jake Moe wrote: On 10/09/2010 5:27 PM, Maciej Grela wrote: 2010/9/10 Jake Moejakesaddr...@gmail.com: Hello all, I've been thinking about creating a Gentoo USB stick for install and rescue purposes (and, of course, just to see if I could). I've mostly followed the Gentoo handbook (I used a single 4GB partition for the whole system, and no swap). I've used genkernel for the kernel (so I can have a multi-system capable kernel). I've gotten GRUB installed and working. My problem comes in after what I believe is the init process: Gentoo Linux; http://www.gentoo.org Copyright 1999-2009 Gentoo Foundation; Distributed under the GPLv2 Press I to enter interactive boot mode * Mounting proc at /proc ... [ ok ] * Mounting sysfs at /sys ... [ ok ] * Mounting /dev ... [ ok ] * Starting udevd ... [ ok ] * Populating /dev with existing devices through uevents ... [ ok ] * Waiting for uevents to be processed ... [ ok ] * Mounting devpts at /dev/pts ... [ ok ] * Checking root filesystem ... fsck.ext2: No such file or directory while trying to open /dev/sda1 /dev/sda1: The superblock could not be read or does not describe a correct ext2 filesystem. If the device is valid and it really contains an ext2 filesystem (and not swap or ufs or something else), then the superblock is corrupt, and you might try running e2fsck with an alternate superblock: e2fsck -b 8193device * Filesystem couldn't be fixed :( [ !! ] Give root password for maintenance (or type Control-D to continue): If I give the root password, I can find no /dev/sda1. However, mount shows /dev/sda1 on /, and there *is* a /sys/block/sda folders, with a sda1 folder in that as well. It's almost like it had /dev/sda1, but then lost it somehow. Does anyone have any idea what's going on here? Any help would be appreciated. Have you seen http://www.sysresccd.org/Main_Page ? It's based on Gentoo, you could check what they did to boot from a usb stick. Br, Maciej Grela Excellent, thanks for that, I hadn't found it in my previous searches. I'll have a look there. Jake Moe Had a similar issue a while ago when I was playing around with this myself. Take a look at the linux boot parameters. The 'theoretical' part is: You need to let the kernel initialize the USB-stick before trying to access it. (This can take some time) There is a delay-option, just can't remember the proper name off-hand. -- Joost I've got USB booting working in a syslinux environment. A delay of 12 seconds is working for me. The syslinux.cfg stanza I use is: LABEL usb KERNEL linux APPEND rootdelay=12 root=/dev/sda2 The usual way for linux on removable usb sticks / disks is to use LABEL or UUID to identify the disks and not the device names, because they will be different in different computers ;) The downside is that you need an initrd to mount the root partition... I think that the usual initrd generated by genkernel works... If you created the rootfs with: mkfs.ext2 -j -LUSBGentoo /dev/sdXY then you can change the kernel parameter to root=LABEL=USBGentoo and your fstab to: LABEL=USBGentoo / ext3 ... You can also use the uuid of the filesystem, find it out with dumpe2fs -h /dev/sdb2 | grep UUID and then use UUID=XXX instead of LABEL=XXX I never really played around with grub and USB booting, so I use syslinux. I create a small FAT partition with syslinux, kernel and initrd image (it gets also pretty handy when you sometimes need to copy something from a windows machine ;) and a second regular ext3 partition for the rootfs. Basically you would do: - partition the stick, mark the FAT partition as bootable/active - format the partitions: - mkfs.vfat -nUSBData /dev/sdX1 - mkfs.ext2 -j -LUSBGentoo /dev/sdX2 - install syslinux (on the FAT partition): - syslinux /dev/sdX1 - mount /dev/sdX2, install gentoo in the usual way - compile the kernel and initrd, make sure required USB stuff is in the kernel (theoretically it could be as modules in initrd... but in-kernel is safer :) if you are in a hurry, or don't know how to create them, get them from a gentoo livecd ;) don't forget to also copy the modules (/lib/modules-XXX/...) from the livecd to the rootfs. - put the kernel and initrd on the FAT partition (I name them vmlinuz.img and initrd.img) - edit syslinux.cfg (on the FAT partition), see
Re: [gentoo-user] revdep-rebuild + minimal output
On Sat, Jul 10, 2010 at 05:11:49AM -0400, Willie Wong wrote: On Sat, Jul 10, 2010 at 06:52:35AM +0200, Enrico Weigelt wrote: is there an option to revdep-rebuild to only do output if it has something to rebuild ? This should be run via cron to notify me via email. Usually is there an option... questions are well-answered by man program name. And in this case the closest thing from the man page is the -q option, which doesn't do exactly what you want. Cheers, revdep-rebuild --help ... -p, --pretendDo a trial run without actually emerging anything (also passed to emerge command) it would basically show you what would be emerged... -q and -P might be usefull to get rid of unwated output... yoyo
Re: [gentoo-user] One hard drive much slower for some reason.
On Tue, Jun 01, 2010 at 04:26:55AM -0500, Dale wrote: Paul Hartman wrote: On Tue, Jun 1, 2010 at 12:26 AM, Dalerdalek1...@gmail.com wrote: I am in the process of moving my OS from drive to drive and thought I would test to see which drive is the fastest. I got some strange results when I tested them One drive is MUCH slower than the others on the buffered disk reads but I can't see any reason why that would be so. Check dmesg to see if the drives show any differences... If they are SATA drives check to see if there is a jumper which forces it into compatibility mode, slow mode, something like that... If they are SATA and you use an intel chipset motherboard check to be sure that the SATA header of that drive is set to AHCI and not IDE mode... Those are just ideas, things I have encountered. :) This is from dmesg: smoker-new ~ # dmesg | grep hda hda: Maxtor 6E040L0, ATA DISK drive hda: host max PIO5 wanted PIO255(auto-tune) selected PIO4 hda: UDMA/133 mode selected hda: max request size: 128KiB hda: 78156288 sectors (40016 MB) w/2048KiB Cache, CHS=65535/16/63 hda: cache flushes supported hda: hda1 hda2 hda5 hda6 hda7 hda8 hda9 hda10 smoker-new ~ # dmesg | grep hdb hdb: WDC WD800BB-00DKA0, ATA DISK drive hdb: host max PIO5 wanted PIO255(auto-tune) selected PIO4 hdb: UDMA/100 mode selected hdb: max request size: 512KiB hdb: 156301488 sectors (80026 MB) w/2048KiB Cache, CHS=16383/255/63 hdb: cache flushes supported hdb: hdb1 hdb2 hdb5 smoker-new ~ # dmesg | grep hdc hdc: Maxtor 6Y080P0, ATA DISK drive hdc: host max PIO5 wanted PIO255(auto-tune) selected PIO4 hdc: UDMA/133 mode selected hdc: max request size: 128KiB hdc: 160086528 sectors (81964 MB) w/7936KiB Cache, CHS=65535/16/63 hdc: cache flushes supported hdc: hdc1 hdc2 hdc5 hdc6 hdc7 REISERFS (device hdc6): found reiserfs format 3.6 with standard journal REISERFS (device hdc6): using ordered data mode REISERFS (device hdc6): journal params: device hdc6, size 8192, journal first block 18, max trans len 1024, max batch 900, max commit age 30, max trans age 30 REISERFS (device hdc6): checking transaction log (hdc6) REISERFS (device hdc6): Using r5 hash to sort names REISERFS (device hdc7): found reiserfs format 3.6 with standard journal REISERFS (device hdc7): using ordered data mode REISERFS (device hdc7): journal params: device hdc7, size 8192, journal first block 18, max trans len 1024, max batch 900, max commit age 30, max trans age 30 REISERFS (device hdc7): checking transaction log (hdc7) REISERFS (device hdc7): Using r5 hash to sort names Adding 976712k swap on /dev/hdc5. Priority:-1 extents:1 across:976712k Adding 976712k swap on /dev/hdc5. Priority:-1 extents:1 across:976712k Adding 976712k swap on /dev/hdc5. Priority:-1 extents:1 across:976712k Adding 976712k swap on /dev/hdc5. Priority:-1 extents:1 across:976712k smoker-new ~ # It appears that hdb is using UDMA/100 but it has always done that. It's a older drive. This drive used to get somewhere in the 40Mb/sec range tho. I want to say it used to be about 47Mb/sec or so. All three of those drives are ATA. I have a SATA drive but I didn't list it since it is working good and fast all things considered. It's hooked to a PCI card. Thoughts? just a quess, did you use 80 or 40 wire ata cable for that disk? yoyo
Re: [gentoo-user] In which order services are started?
On Wed, Apr 21, 2010 at 05:05:46AM -0500, Dale wrote: Alex Schuster wrote: Jarry writes: Is there any way to find out in which order services are started during boot-up (except for looking at boot-up screen and making notes)? I think the output of 'rc-status' shows the services in the right order. Wonko It may be a coincidence but mine are alphabetical. Also, mine only shows the ones in the current runlevel, default at the moment. It does not list the ones in the boot runlevel. rc-status --all yoyo
Re: [gentoo-user] Re: I want my Ctrl+Alt+Backspace back
On Sun, Apr 18, 2010 at 07:28:00PM -0500, Harry Putnam wrote: Harry Putnam rea...@newsguy.com writes: Nikos Chantziaras rea...@arcor.de writes: Read more details here: http://www.gentoo.org/proj/en/desktop/x/x11/xorg-server-1.6-upgrade-guide.xml HAL is deprecated and will not be supported in X anymore, so it's not the new way ;) Well, its just not the NEWEST way. But what is the newest (post hal) way? And will the xorg.conf technique work anyway? I should have mentioned that after posting the OP, I discovered I've had that stanza in xorg.conf for mnths... I forgot I had taken it from a post by Florien P., but I do not get the use of Ctrl+alt+bkspk to quit X. I guess it works for you though eh, Mick? Works here (either in xorg.conf or in fdi files if let X use hal). If you set up a keyboard layout in kde / gnome, it may be possible that it resets the xkb options... though you seem to have the layout set in xorg.conf, so that it seem improbable that you use the gnome/kde layout configuration... yoyo ---- ---=--- - From xorg.conf: Section InputDevice Identifier Keyboard1 Driver kbd # [HP 100709_111603 From post on gentoo.user ## From: Florian Philipp li...@f_philipp.fastmail.net ## Subject: Re: Fed up with Xorg + hal mess ## Date: Tue, 06 Oct 2009 19:22:46 +0200 ## Message-ID: 4acb7ce6.10...@f_philipp.fastmail.net ## Restablishes Ctrl-Alt-Bkspc to quit X Option XkbOptions terminate:ctrl_alt_bksp # ] Option AutoRepeat 500 30 Option XkbRules xorg Option XkbModel pc104 Option XkbLayout us EndSection
Re: [gentoo-user] No Mousewheel in X and ctrl-alt-backspace not working
On Sun, Apr 18, 2010 at 07:53:43AM +, Konstantinos Agouros wrote: Hi, for whatever reason my mousewheel is no longer working after upgrading zu the latest X-Org server (and corresponding libraries). Here's the excerpt from the xorg.conf that is relevant: Section InputDevice Identifier Mouse1 Driver mouse Option ProtocolIMPS/2 Option Device /dev/input/mouse0 Option ZAxisMapping 4 5 6 7 I had to rebuild the mouse-drive for ABI changes but that didn't fix the problem. Any input is welcome. Also ctrl-alt-backspace to kill X does not work anymore. DonZap is commented out. Clues in that regard would be welcome as well. this changed some while back, try adding Option XkbOptions terminate:ctrl_alt_bksp to your keyboard InputDevice yoyo
Re: [gentoo-user] Recompile system but omit package?
On Sun, Apr 18, 2010 at 11:57:48AM -0400, Tanstaafl wrote: On 2010-04-18 11:45 AM, Johannes Kimmel wrote: well... you could use --keep-going and kill something when gcc compiles. not very nice, but will work without breaking anything. Dang - I already started the emerge... You can still break the emerge (for example with ctrl-c) when it starts to emerge gcc, the continue the emerge process with emerge --resume --skipfirst that's what I usually do with openoffice and similar apps when I do a quick update and fail to notice the large app in there.. ;) yoyo
Re: [gentoo-user] Recompile system but omit package?
On Sat, Apr 17, 2010 at 04:59:07PM -0400, Tanstaafl wrote: Is there a way to emerge, say, system, but omit one package in it? For example, I've already recompiled gcc 4.3.4 with itself... is there a way to now do something like: emerge system -gcc (where '-gcc' serves to tell portage to compile everything *but* gcc)? Its not a big deal, I'm just curious... -- Charles You can do something like: emerge -pe world | sed -e /^.ebuild/ ! d; s/.*] /=/; s/ .*//; list ... edit list and remove anything you don't want to reinstall ... emerge -av1 `cat list` the -1 (or --oneshot), means that the packages won't be added to the world file (they would normally because you are listing them all on the commandline) yoyo
Re: [gentoo-user] Recompile system but omit package?
On Sun, Apr 18, 2010 at 12:52:26PM -0400, Tanstaafl wrote: On 2010-04-18 12:29 PM, YoYo siska wrote: On Sun, Apr 18, 2010 at 11:57:48AM -0400, Tanstaafl wrote: On 2010-04-18 11:45 AM, Johannes Kimmel wrote: well... you could use --keep-going and kill something when gcc compiles. not very nice, but will work without breaking anything. Dang - I already started the emerge... You can still break the emerge (for example with ctrl-c) when it starts to emerge gcc, the continue the emerge process with emerge --resume --skipfirst To clarify - I can do this with the currently running emerge (that did not specify --keep-going)? So, when it gets to gcc (its on package # 181 of 355 now, hasn't hit either of the gcc's or glibc yet), hit ctrl-c, then: emerge --resume --skipfirst ? Do I need to add the -ev world in there? Or does emerge just know where to pick up all by itself? yes, it knows what the last emerge was, so you just say --resume but if you do another emerge in between, it will forget the previous interrupted one --resume just resumes the last interrupted (or failed) emerge , starting with the package that was interrupted, so that you can fix the problem if it was a compilation failure, and then continue... no need to give any special args to the first emerge. --skipfirst makes it skip the first package - the one that was interrupted handy when the emerge fails on a packages that isn't a depency of something other, you can just skip it then, very much like a manual --keep-going ;) This is good info to have. Also - is it ok to do this during the actual compile? Or do I need to catch it before the actual compiling starts? you can break it whenever you want.. --resume than starts the package again from beginning so you just waste the time/work it allready did... which does not really matter if you are going to do --skipfirst ;) yoyo
Re: [gentoo-user] Recompile system but omit package?
On Sun, Apr 18, 2010 at 03:54:38PM -0400, Tanstaafl wrote: On 2010-04-18 3:49 PM, Tanstaafl wrote: On 2010-04-18 1:09 PM, YoYo siska wrote: On Sun, Apr 18, 2010 at 12:52:26PM -0400, Tanstaafl wrote: On 2010-04-18 12:29 PM, YoYo siska wrote: On Sun, Apr 18, 2010 at 11:57:48AM -0400, Tanstaafl wrote: On 2010-04-18 11:45 AM, Johannes Kimmel wrote: well... you could use --keep-going and kill something when gcc compiles. not very nice, but will work without breaking anything. Dang - I already started the emerge... You can still break the emerge (for example with ctrl-c) when it starts to emerge gcc, the continue the emerge process with emerge --resume --skipfirst Worked a treat, many thanks! Saved me at least an hour of compile time... :) Hmmm... one last question... Will etc-update still prompt for all necessary changes for config files for *all* of the installs done, considering I did ctrl-c 3 times (glibc, and both gcc's)? yes, if new config files got installed, etc-update will show them (I think it uses 'find' to find them ;) yoyo
Re: [gentoo-user] KDE4 doesn't heed hal keyboard settings
On Thu, Mar 25, 2010 at 01:49:04PM +0100, Frank Steinmetzger wrote: Hello I'm not sure how to specify the topic in more detail. Over the last to hours I tried to get KDE4 to behave like it did before I rebuilt my system. My goal is to have a German layout with dead keys. In the process I've even gotten the system to react to the menu key again. But apart from that, for some reason, KDE4 won't behave like I want it to. I say KDE4 because KDE3 apparently works. I know because login with KDM4 doesn't work at the moment, so I'm using KDM3. There I can input accented letters of all sorts (in the username input field). But not in KDM4 and not in my KDE4 environment. I also tried it with and without KDE's own keyboard layout settings, where I set up a de layout of default variant (thus with dead keys). But to no avail. Here's the content of my hal config file: ?xml version=1.0 encoding=UTF-8? deviceinfo version=0.2 device match key=info.capabilities contains=input.keyboard merge key=input.x11_options.XkbRules type=stringevdev/merge merge key=input.x11_options.XkbModel type=stringevdev/merge merge key=input.x11_options.XkbLayout type=stringde/merge merge key=input.xkb.options type=strlistmenu:Multi_key,terminate:ctrl_alt_bksp/merge /match /device /deviceinfo It worked before[TM], and I have no idea where else to look. Any ideas please? I think taht KDE4 do not touch your initial keyboard settings (xorg.conf or hal,...) unless you change something in its keyboard config (systemsettings -regionallanguage-keyboard layout ) maybe that you have wrong keyboard set explicitly there? I have disable keyboard layouts under the layout tab selected, but I can't say if it really works correctly (keeps the X configuration defaults) right now, because I can't restart X right now, and I change between 'setxkbmap -option grp:shifts_toggle sk,us qwerty,' and a simple 'setxkbmap us' with some scripts, because the dual layout confuses some apps like synergy, x2x, rdesktop, even some vnc clients... what happens if you you type setxkbmap de in an terminal (konsole, xterm,...) after kde starts up? yoyo
Re: [gentoo-user] KDE4 doesn't heed hal keyboard settings
On Thu, Mar 25, 2010 at 02:30:49PM +0100, YoYo siska wrote: On Thu, Mar 25, 2010 at 01:49:04PM +0100, Frank Steinmetzger wrote: Hello I'm not sure how to specify the topic in more detail. Over the last to hours I tried to get KDE4 to behave like it did before I rebuilt my system. My goal is to have a German layout with dead keys. In the process I've even gotten the system to react to the menu key again. But apart from that, for some reason, KDE4 won't behave like I want it to. I say KDE4 because KDE3 apparently works. I know because login with KDM4 doesn't work at the moment, so I'm using KDM3. There I can input accented letters of all sorts (in the username input field). But not in KDM4 and not in my KDE4 environment. I also tried it with and without KDE's own keyboard layout settings, where I set up a de layout of default variant (thus with dead keys). But to no avail. Here's the content of my hal config file: ?xml version=1.0 encoding=UTF-8? deviceinfo version=0.2 device match key=info.capabilities contains=input.keyboard merge key=input.x11_options.XkbRules type=stringevdev/merge merge key=input.x11_options.XkbModel type=stringevdev/merge merge key=input.x11_options.XkbLayout type=stringde/merge merge key=input.xkb.options type=strlistmenu:Multi_key,terminate:ctrl_alt_bksp/merge /match /device /deviceinfo It worked before[TM], and I have no idea where else to look. Any ideas please? I think taht KDE4 do not touch your initial keyboard settings (xorg.conf or hal,...) unless you change something in its keyboard config (systemsettings -regionallanguage-keyboard layout ) maybe that you have wrong keyboard set explicitly there? I have disable keyboard layouts under the layout tab selected, but I can't say if it really works correctly (keeps the X configuration defaults) right now, because I can't restart X right now, and I change between 'setxkbmap -option grp:shifts_toggle sk,us qwerty,' and a simple 'setxkbmap us' with some scripts, because the dual layout confuses some apps like synergy, x2x, rdesktop, even some vnc clients... what happens if you you type setxkbmap de in an terminal (konsole, xterm,...) after kde starts up? and of course you can check your current settings with setxkbmap -print yoyo
Re: [gentoo-user] Playing Apple Trailers
On Mon, Mar 22, 2010 at 07:30:49PM +, Mick wrote: On Monday 22 March 2010 19:18:48 Christian Schulze wrote: Hi, I installed www-plugins/gecko-mediaplayer version 0.9.9.2 (masked with keywords ~alpha ~amd64 ~hppa ~ppc ~ppc64 ~x86) recently and this works great with the Apple Trailers. It pulls in gnome-mplayer, Thanks Christian, I am not sure I want to install a gnome front end to mplayer. However, it seems that the gecko-mediaplayer plugin only has one option. Anyone knows if there's an alternative? Qt4 front end like smplayer would be preferable to me. don't know about alternatives, but gnome-mplayer with -gnome use flag (and the same holds for gecko-mplayer) is just a simple gtk app, just like firefox itself, nothig gnome specific about it and theoretically it can look better in firefox than qt apps, because they both use the same engine/theme ;) with -gnome it also doesn't bring in any gnome dependencies... its just that it is named gnome-... (well, because when you enable it, it can becamu a full gnome app..., though I really don't know what gnome features it uses ;) yoyo
Re: [gentoo-user] Can't see /dev/hda1,2,3 but I know they exist...
On Fri, Feb 19, 2010 at 12:49:47AM -0500, James Homuth wrote: I performed a bit of an update on my laptop a day or two ago, and after reboot, I lost the ability to do anything with /dev/hda*. I currently have 0 swap space, and according to stat, ls etc, they don't exist. But, booting to an install CD I burned for diagnostic purposes, it sees them just fine. Also, and this is the strange part. It boots no problem, so the OS is able to mount at least /dev/hda3, even though from the command line I'm not seeing it. I'm probably missing something completely dead obvious (it's after midnight here and all), and Google's turning up nothing, so if someone could kindly slap me in the face with it, that'd be appreciated. Thanks either way for whatever help comes my way. Hi, I just had to restart my computer (power issues :( ) in the middle of an update (well, it was more like 'just before the end';) and after restart I have the same problem as you, no /dev/sd[ab]* files... My first guess was that I rebooted without updating the config files, so I ran etc-update (there were some udev config files as well as init script) and rebooted, but that didn't help. It is certainly not a problem with drivers not being in kernel, as the kernel sees the disks and partitions (see below), so I just run tail -n +3 /proc/partitions | while read maj min size name ; do mknod /dev/$name b $maj $min ; done /etc/init.d/localmount pause; /etc/init.d/localmount start to get everything mounted again... That means it will have to be an udev (or even openrc) problem. The last update of udev did in fact say this: * Checking for suitable kernel configuration options... * CONFIG_SYSFS_DEPRECATED:should not be set. But it is. * CONFIG_SYSFS_DEPRECATED_V2: should not be set. But it is. * CONFIG_IDE: should not be set. But it is. * Please check to make sure these options are set correctly. * Failure to do so may cause unexpected problems. * * udev-151 does not support Linux kernel before version 2.6.25! * For a reliable udev, use at least kernel 2.6.27 * Your kernel version (2.6.28-gentoo-r2) is new enough to run udev-151 reliably. I didn't want to mess with the kernel right now, but I gues that's the first thing to try... I'll report when I rebuild reboot... yoyo === Kernel can see the partitions just fine: julka dev # cat /proc/partitions major minor #blocks name 70 512000 loop0 80 199148544 sda 81 18940603 sda1 82 32218357 sda2 832152710 sda3 84 1 sda4 85 145830006 sda5 8 16 312571224 sdb 8 17 312568641 sdb1 julka dev # ls /sys/block/ hda/ loop1/ loop3/ loop5/ loop7/ ram1/ ram11/ ram13/ ram15/ ram3/ ram5/ ram7/ ram9/ sdb/ loop0/ loop2/ loop4/ loop6/ ram0/ ram10/ ram12/ ram14/ ram2/ ram4/ ram6/ ram8/ sda/ julka dev # ls /sys/block/sd* /sys/block/sda: bdi dev ext_range power range rosda2 sda4 sizestat uevent capability device holdersqueue removable sda1 sda3 sda5 slaves subsystem /sys/block/sdb: bdi capability dev device ext_range holders power queue range removable ro sdb1 size slaves stat subsystem uevent -- _ | YoYo () Siska === http://www.ksp.sk/
Re: [gentoo-user] Can't see /dev/hda1,2,3 but I know they exist...
On Mon, Feb 22, 2010 at 04:13:40PM +0100, YoYo siska wrote: On Fri, Feb 19, 2010 at 12:49:47AM -0500, James Homuth wrote: I performed a bit of an update on my laptop a day or two ago, and after reboot, I lost the ability to do anything with /dev/hda*. I currently have 0 swap space, and according to stat, ls etc, they don't exist. But, booting to an install CD I burned for diagnostic purposes, it sees them just fine. Also, and this is the strange part. It boots no problem, so the OS is able to mount at least /dev/hda3, even though from the command line I'm not seeing it. I'm probably missing something completely dead obvious (it's after midnight here and all), and Google's turning up nothing, so if someone could kindly slap me in the face with it, that'd be appreciated. Thanks either way for whatever help comes my way. Hi, I just had to restart my computer (power issues :( ) in the middle of an update (well, it was more like 'just before the end';) and after restart I have the same problem as you, no /dev/sd[ab]* files... My first guess was that I rebooted without updating the config files, so I ran etc-update (there were some udev config files as well as init script) and rebooted, but that didn't help. It is certainly not a problem with drivers not being in kernel, as the kernel sees the disks and partitions (see below), so I just run tail -n +3 /proc/partitions | while read maj min size name ; do mknod /dev/$name b $maj $min ; done /etc/init.d/localmount pause; /etc/init.d/localmount start to get everything mounted again... That means it will have to be an udev (or even openrc) problem. The last update of udev did in fact say this: * Checking for suitable kernel configuration options... * CONFIG_SYSFS_DEPRECATED:should not be set. But it is. * CONFIG_SYSFS_DEPRECATED_V2: should not be set. But it is. * CONFIG_IDE: should not be set. But it is. * Please check to make sure these options are set correctly. * Failure to do so may cause unexpected problems. * * udev-151 does not support Linux kernel before version 2.6.25! * For a reliable udev, use at least kernel 2.6.27 * Your kernel version (2.6.28-gentoo-r2) is new enough to run udev-151 reliably. I didn't want to mess with the kernel right now, but I gues that's the first thing to try... I'll report when I rebuild reboot... yop, that was it though you wrote about /dev/hda*, which means you should be a bit more carefull if you used the IDE drivers (under ATA/ATAPI/ support, thats the CONFIG_IDE option) and disabled the CONFIG_IDE options, you have to enable it under Serial ATA (prod) and Parallel ATA (experimental) drivers (CONFIG_ATA) and also your device might get renamed to sd* instead of hd* (I don't know, I have only a cdrom, that becomes sr0 ;) But I think that the real problem was with those SYSFS_DEPRECATED options, so you might be able to get things working with just disabling those and leaving IDE as it was... btw, I found this bug afterwards: http://bugs.gentoo.org/show_bug.cgi?id=302173 yoyo
Re: [gentoo-user] Re: kde4 panelbar recovery
On Mon, Feb 22, 2010 at 07:19:41PM +, James wrote: Nikos Chantziaras realnc at arcor.de writes: With the mouse. Must be something wrong. These panels that fire up are disfunctional. Cant move add or delete too them Lock/Unlock widgets in the context menu? Btw you can fire up add applets on the desktop and drag them to a panel.. doesn't really make a difference ;) You might want to delete your ~/.kde4 folder instead to get back at the defaults. Keep the stuff you want though (like settings for other programs, like Amarok, Kopete, etc.) It seems really stupid there is not way to recover kicker and such without deleting the entire folder. More kde4 snafus I found lots of evidence where folks had done the exact same thing, with no simple recovery... Very disappointed in KDE4.again This recovery in kde3 was simple. Panel and desktop settings are in .kde/share/config/plasma* files removing just them (ideally when logged out of kde) should bring the default desktop/panels back.. theoretically plasma-desktop-appletsrc should be enough to delete... Don't see how different from kde3 this is.. if you messed up your kicker configuration you had to either delete it, or rebuild the panel (create it, place it at the correct position, add correct applets...) (well, the only difference is that desktop and panel were separate programs...) yoyo
Re: [gentoo-user] Xmonad, beagle-search, and mime types
On Sun, Feb 21, 2010 at 05:33:35PM +0100, Damian wrote: On Sun, Feb 21, 2010 at 4:05 PM, Willie Wong ww...@math.princeton.edu wrote: On Sun, Feb 21, 2010 at 03:45:14PM +0100, Damian wrote: So I tried to see how xdg-open works, but the man page didn't give me any useful information. The related command, xdg-mime, doesn't work as I expected. I asked a similar question a week or so back. I searched in my mails but I couldn't find it. Sorry, I probably didn't enter a relevant search string. But xdg-open (and therefore beagle-search) still refuses to open jpeg images with geeqie. Any ideas? xdg-open is just a shell script. If you are interested, take a look at less `which xdg-open` and you will be enlightened as to why it is a complete piece of crap unless you are using KDE, GNOME, or XFCE. (Hint, notice how nowhere in the script does it read whatever you modified with xdg-mime.) A possible way to work around it (depends on your application, which, in your case, is beagle, which I am not familiar with) is to go into the offending application that is calling xdg-open and see if you can configure MIME types in there yourself. The application that made me look this up, Jabref, does allow that configuration. Your mileage can of course vary. Thanks Willie for your answer. Sadly beagle-search doesn't offer any option. The developers must use only gnome. I guess I will have so find another beagle front end, or choose a different desktop search engine. xdg-open tries to determine the desktop enviroment you are running and uses its way to open files... If you are in a kde session, it will just run kfmclient XXX (kde way to open files in their application), for gnome it will use gnome-open or something... So even if you are not logged in a full gnome or kde session, but have some of its packages installed, you can trick xdg-open to use that enviroment for kde just export KDE_FULL_SESSION=true, for gnome GNOME_DESKTOP_SESSION_ID=something so you can run beagle-search with GNOME_DESKTOP_SESSION_ID=something beagle-search you have to set which applications to use in corresponding config (for kde, just run konqueror/dolphin and right click a file, don't know for gnome) for kde, you need the package kde-base/kfmclient, which should depend juast on kdelibs, for gnome-open you need gnome-base/libgnome, which shouldn't have much dependences that you don't have allready if you have some gtk app.. btw, gnome-open/kfcmclient will open files in any available program that correctly registers its mime-types, not only in kde/gnome apps..., so you really need only kfmclient/libnome to use it... one last remark, if set KDE_FULL_SESSION/GNOME_DESKTOP_SESSION_ID for xdg-open and it starts a gnome/kde app, that app might think that there are some desktop specific things running, that are not, though I haven't seen any real problems with it... You might however just edit the detectDE function in xdg-open to always behave like gnome/kde without settings the variables... yoyo PS the idea of xdg-open using a browser, when it cannont detect which DE are you running, is that a browser usually knows how to open which files... and you should have your preffered browser set in $BROWSER...
Re: [gentoo-user] changing nvidia settings dynamically
On Fri, Jan 22, 2010 at 08:59:43AM +0930, Iain Buchanan wrote: On Sat, 2010-01-16 at 12:52 +, Neil Bothwick wrote: On Fri, 15 Jan 2010 16:55:33 +0930, Iain Buchanan wrote: Whenever I dock / undock I have to run nvidia-settings to change the resolution from the virtual 3840x1200 to 1920x1200 or vice versa. Also since two screens is the Default I have to do this when I log in with only the laptop. I am looking for a way to use the command line nvidia-settings (as much as I've studied the help I can't find out how to do it - all attributes seem read-only to the command line nvidia-settings) Can you do this with the --config and --load-settings options? OK, I finally got around to trying this, but no it doesn't work. When trying to switch from laptop-only to laptop and LCD I just get: ERROR: Invalid display device DFP-2 specified on line 40 of configuration file '.nvidia-settings-rc-twin' (the currently enabled display devices are DFP-0 on orpheus:0.0). ERROR: Invalid display device DFP-2 specified on line 42 of configuration file '.nvidia-settings-rc-twin' (the currently enabled display devices are DFP-0 on orpheus:0.0). And this is when using the config file I saved with two screens working. I have to run nvidia-settings by hand and select detect displays before DFP-2 appears. ideas? thanks, from time to time I tried to use the commandline options of nvidia-settings to change between two sets of outputs (main+tv and main+second monitor) but it seems, that nvidia-settings just can't enable/disable outputs from commandline ... it wasn't very important for me, so I didn't ask on any more nvidia specific forums... but your mail made me look at the sorces of nvidia-settings, and there's a samples directory with some examples, and it seems that nv-control-dpy can do that... you should be able to enable the correct outputs with --set-associated-dpys and then load your configs as you tried... (I didn't try that, just checked in the gui that the output changed to off insteadd of disabled) maybe it is even enough to run the --probe-dpys according to what you say about selecting detect displays in the gui... yoyo btw, to get the program, just run (change the version/dirs...): ebuild /usr/portage/media-video/nvidia-settings/nvidia-settings-190.53.ebuild unpack cd /var/tmp/portage/media-video/nvidia-settings-190.53/work/nvidia-settings-1.0/samples/ make y...@julka samples $ ./nv-control-dpy --probe-dpys Using NV-CONTROL extension 1.18 on :0.0 Connected Display Devices: CRT-1 (0x0002): Philips 200WS DFP-0 (0x0001): Philips 200WS Display Device Probed Information: number of GPUs: 1 display devices on GPU-0 (GeForce 8600 GT): CRT-1 (0x0002): Philips 200WS DFP-0 (0x0001): Philips 200WS y...@julka samples $ ./nv-control-dpy --get-associated-dpys Using NV-CONTROL extension 1.18 on :0.0 Connected Display Devices: CRT-1 (0x0002): Philips 200WS DFP-0 (0x0001): Philips 200WS associated display device mask: 0x0001 y...@julka samples $ ./nv-control-dpy --set-associated-dpys 0x10002 Using NV-CONTROL extension 1.18 on :0.0 Connected Display Devices: CRT-1 (0x0002): Philips 200WS DFP-0 (0x0001): Philips 200WS set the associated display device mask to 0x00010002 y...@julka samples $ ./nv-control-dpy --get-associated-dpys Using NV-CONTROL extension 1.18 on :0.0 Connected Display Devices: CRT-1 (0x0002): Philips 200WS DFP-0 (0x0001): Philips 200WS associated display device mask: 0x00010002 y...@julka samples $ ./nv-control-dpy --set-associated-dpys 0x1 Using NV-CONTROL extension 1.18 on :0.0 Connected Display Devices: CRT-1 (0x0002): Philips 200WS DFP-0 (0x0001): Philips 200WS set the associated display device mask to 0x0001
Re: [gentoo-user] How to determine if a NIC is playing gigabit?
On Mon, Jan 18, 2010 at 11:50:55AM +, Stroller wrote: Hi there, Yesterday I reseated the network cable between my server cupboard and my desk, and it now lights up on the switch by my desk as gigabit. But a file-transfer today is slower than I might have hoped. I'm not ruling out the cable, because it's pretty beat up (but the switch *is* lighting up as 1000), but how do I determine, please, that the Linux server at the other end is recognising the NIC and negotiating as gigabit speeds? mii-tool (net-tools) or ethtool should be able to tell you that The hard-drives on the server are using an older PCI SATA card, and the NIC is also PCI. But I would have expected it to be a bit faster than 100Mbps. Any estimates over what kind of speed I should be seeing for large file-transfers over Samba? Wildly ball-park is fine - I wouldn't expect a 10x speed increase, but maybe 2x or 3x - 4x would be great! don't know about samba, but with scp or nfs I can get about 20MByte/s which is the speed of my disk (and for scp almost what my cpu can manage ;) scp-ing /dev/zero gets me something short of 30MBye/s but that is because my CPU cannot manage more ;) You can see an estimate of your raw speed between the two machines by running nc -l -p | pv /dev/null on one computer and pv /dev/zero | nc OTHER_COMPUTER on the other. I don't have a 1gbit switch here right now, so can't give you an estimate (with two notebooks connected directly by cable I just got 100MByte/s, which is near enough to the theoretical maximum ;) (pv is like cat, but displays a progressbar with act. speed, sys-apps/pv) you can also try netperf for more precise benchmarks I'll be testing between my Macs (both on the desktop switch, ruling out both the Linux box and the suspicious cable) later today, I'd just like some ideas of where I should be starting from. Right now I'm seeing 10 gigs of .mp4 files (1gb - 2gb per video file) taking about an hour - that's about what I'd expect from old 100Mbps networking, not this shiny new stuff. hmm, that seems a bit low even for 100mbit, I have usually no problem getting cca 10 MByte/s with 100mbit switches (without other traffic), though I use either nfs or scp the only time I remember using samba was with a winxp server, which didn't go above 1MB/s, but I suspect that the problem was either on the win side or some misunderstanding between win and linux ;) I'm not seeing any difference commenting uncommenting aio read size = 1, aio write size = 1 (separate lines) from /etc/samba/smb.conf and then running `/etc/init.d/samba reload`, but maybe I shouldn't expect that to make any difference on an existing transfer. I just don't want to interfere with this right now - I just want to copy as much as possible on to my laptop before I go out, and I'll take a look at this performance issue when I get home. Thanks in advance for any suggestions or pointers, Stroller. yoyo
Re: [gentoo-user] Re: How can I move system to new disk?
On Sun, Jan 17, 2010 at 08:48:21AM +0200, Nikos Chantziaras wrote: On 01/17/2010 12:40 AM, YoYo siska wrote: On Sat, Jan 16, 2010 at 03:21:32PM +0200, Nikos Chantziaras wrote: On 01/15/2010 07:33 PM, Jarry wrote: [...] I'll just copy the instructions I have someone else here: You can clone the existing Gentoo installation into the new partition and boot from it. You can do this while the system is actually running. The new partition can be anything you want (different size, different file system). This usually means: rsync your existing / to your target / (except /dev, /sys and /proc and of course mount points that belong to a different filesystem, /boot or /home for example if you're using dedicated partitions for those). If you mounted your target / as /root/newpart, this is done with: rsync -ax / /root/newpart If this copied directories it shouldn't have (like /sys or /proc), simply delete them again. [...] If you are doing it this way (on a running system with mounted dev/proc/sys...), you can just bind-mount your current / to another directory. That copy will not contain any sub-mounts rsync -ax / /target shouldn't copy any sub-mounts either, because of the -x option. See man rsync. I mentioned it just in case ;) yes, but it will miss any files hidden under those mounts, though normally that menas only /dev/, the others are empty... and i like it more, because it makes a more exact copy ;) yoyo
Re: [gentoo-user] Re: How can I move system to new disk?
On Sat, Jan 16, 2010 at 03:21:32PM +0200, Nikos Chantziaras wrote: On 01/15/2010 07:33 PM, Jarry wrote: Hi, I'm facing this problem: I want to exchange hard-drive in my computer for other, bigger one. I do not want to add new hard-drive somewhere on mount-point permanently, I just want to copy everything from the old drive to the new one and then get rid of the old one. And of course, I'd like to use my computer as before. What is the best (maybe I should ask for safest) way to acomplish this? First I thought about cp -a. But I'm not sure which directories I should skip (/proc, maybe some other like /dev?). And I do not know how cp handles links (if I first copy link and later target, where is the link pointing? to the original file or its copy?). Maybe dump/restore is better solution? Or something else? I'll just copy the instructions I have someone else here: You can clone the existing Gentoo installation into the new partition and boot from it. You can do this while the system is actually running. The new partition can be anything you want (different size, different file system). This usually means: rsync your existing / to your target / (except /dev, /sys and /proc and of course mount points that belong to a different filesystem, /boot or /home for example if you're using dedicated partitions for those). If you mounted your target / as /root/newpart, this is done with: rsync -ax / /root/newpart If this copied directories it shouldn't have (like /sys or /proc), simply delete them again. Then: mkdir /root/newpart/dev mkdir /root/newpart/proc mkdir /root/newpart/sys mknod /root/newpart/dev/console c 5 1 mknod /root/newpart/null c 1 3 touch /root/newpart/dev/.keep touch /root/newpart/proc/.keep touch /root/newpart/sys/.keep If you are doing it this way (on a running system with mounted dev/proc/sys...), you can just bind-mount your current / to another directory. That copy will not contain any sub-mounts (as if you accessed it from a livecd), so you could just do (mount your /mnt/new_root) mkdir /mnt/current_root mount -o bind / /mnt/current_root rsync -aHAX /mnt/current_root/ /mnt/new_root/ i always remout / readonly first, for that you usually have to go to single user, or stop most of the services and programs... yoyo Now chroot into it to set up the boot loader (I assume you use Grub): mount -t proc none /root/newpart/proc mount -o bind /dev /root/newpart/dev chroot /root/newpart /bin/bash Now edit /etc/fstab to use the new partition and edit /boot/grub/grub.conf and reinstall grub: grub root (hd0,0) -- sustitute with what you really have/want setup quit You're ready. Leave the chroot and unmount: exit umount /root/newpart/dev umount /root/newpart/proc If you've set up grub correctly while in the chroot, you can now reboot and the system should come up using the new partition. If you used a different filesystem for the new partition (for example going from ext3 to ext4), make sure your kernel supports the new filesystem.
Re: [gentoo-user] What standalone GTK2 Engines are available?
On Sun, Jan 03, 2010 at 01:51:55PM +, Ognjen Bezanov wrote: Hello, I used to use xfce-mcs-manager as a lightweight standalone gtk2 engine for my enlightenment desktop (for using GTK2 themes to render). I've just reinstalled Gentoo on my machine, and now xfce-mcs-manager is no more (seems it's been merged deeper in xfce now, so no longer standalone). An alternative that has been proposed to me is gnome-settings-daemon, but I'd rather install something lighter that doesn't need all the gnome libraries installed. Is there any such thing available? Thanks! lxde-base/lxappearance x11-themes/gtk-chtheme x11-themes/gtk-theme-switch the last one allows one to specify the theme on commandline... but you don't have to run them all the time like gnome-settings-daemon... just run them once, select your theme and they will set up your ~/.gtkrc-2.0 so that all gtk apps will use that theme from now on... (unless you start up gnome-settings-daemon which will enforce the theme selected in gnome settings... don't know why they did it that way...) yoyo
Re: [gentoo-user] net.eth0 started by udev-postmount
On Tue, Apr 14, 2009 at 05:27:53PM +0200, Arnau Bria wrote: Hi all, I removed my net.eth0 service from default/boot level: amparo ~ # rc-update show|grep eth0 amparo ~ # but now, udev-postmount tries to start it. I don't want it cause I don't plug any wire to my laptop, and udev hang my start for a minute... how may I disable udev from starting net.eth0? I remember in the past I had something that checked is the wire was pluged, and if so, it started my network... anyone could help me to remember that program? TIA, Arnau look at the rc_hotplug option in /etc/rc.conf something like rc_hotplug=!net.eth0 should work for you... the other option is, if you emerge ifplugd, the net.eth0 service will use ifplugd to see if there is a cable connected... it will start but will not set up the interface until you plug in a cable... yoyo
Re: [gentoo-user] Double firefox references in taskbar
On Sat, Apr 11, 2009 at 02:13:06PM +0100, Etaoin Shrdlu wrote: Since some time (I think since firefox 3), when I launch firefox using the button in kde's menu, I get two Mozilla firefox references in the task bar, one of which with the turning hourglass. Then, the main window of firefox appears. The bouncing firefox icon continues to be near the pointer arrow. After say 10 or 15 seconds, the bouncing icon and the taskbar reference with the turning hourglass disappear, and I'm left with only one (as should be in the first place imho). The aplication associated with the kde menu button is /usr/bin/firefox %U (I tried removing the %U, no difference). Mozilla-firefox does have a startup-notification use flag, do you have it turned on? If I run /usr/bin/firefox from the command line, everything is fine (ie, a single reference in the task bar). Well, kde doesn't know that you are starting an app, so there's no startup-notification magic involved ;) Anyone else has seen this? Kde version 3.5.9. Not a big problem, but I'm just curious. yoyo
Re: [gentoo-user] [OT] KDE4 session saving
On Mon, Apr 06, 2009 at 06:31:54PM +0400, Andrew Gaydenko wrote: On Monday 06 April 2009 18:11:13 YoYo siska wrote: On Mon, Apr 06, 2009 at 01:31:57PM +0400, Andrew Gaydenko wrote: I have found it is possible to select a restoring of a manually saved KDE4 session. But have not found how to save :-) Help! It just needs a relogin to show the option in the logout dialog, last time I tried it, it worked this way: - switch to restore manually saved session in systemsettings/wherever, no save this session appears in the logout dialog... - just log out of kde and log in again, now there should be the save this session option in the logout dialog... yoyo Have tried (config dialog settings is attached), but after relogin still has 'Logout' and 'Cancel' buttons in log out dialog. Now that I looked at it, it really isn't in the logout dialog, but it is in the Leave tab in the start menu: http://people.ksp.sk/~yoyo/screenshots/2009/save_session.png This is a bit old SVN version (i think just before 4.2) yoyo -- _ | YoYo () Siska === http://www.ksp.sk/
Re: [gentoo-user] [OT] KDE4 session saving
On Mon, Apr 06, 2009 at 01:31:57PM +0400, Andrew Gaydenko wrote: I have found it is possible to select a restoring of a manually saved KDE4 session. But have not found how to save :-) Help! It just needs a relogin to show the option in the logout dialog, last time I tried it, it worked this way: - switch to restore manually saved session in systemsettings/wherever, no save this session appears in the logout dialog... - just log out of kde and log in again, now there should be the save this session option in the logout dialog... yoyo -- _ | YoYo () Siska === http://www.ksp.sk/
Re: [gentoo-user] Xinerama vs TwinView for dual monitor setup
On Mon, Oct 13, 2008 at 09:10:34AM +0200, Alan McKinnon wrote: On Monday 13 October 2008 04:02:19 Iain Buchanan wrote: [snip] I've configured it with TwinView as in: Option TwinView True Yes. Some output : $ sudo grep -i -e xinerama -e twinview /var/log/Xorg.0.log (**) Option Xinerama 1 (**) Xinerama: enabled (**) NVIDIA(0): Option TwinView 1 (**) NVIDIA(0): Option TwinViewXineramaInfoOrder DFP-0 (**) NVIDIA(0): TwinView enabled (II) Initializing built-in extension XINERAMA $ sudo grep -i -e xinerama -e twinview /etc/X11/xorg.conf Option Xinerama 1 Option TwinView 1 Option TwinViewXineramaInfoOrder DFP-0 The viewports are aligned along the top edge you mean move the mouse up and it appears on the next screen? Don't you want them aligned left / right of each other? My description wasn't clear. I mean the screens are physically and logically laid out like so: +--+ | | | |1 | 2 | | |---+ +--+ 1 is the notebook screen 2 is the external lcd below 2 is dead space. The mouse works correctly. and the panel/kicker/plasma/whatever on every desktop environment insists on trying to stretch across both monitors, into dead space on the right hand one. Sounds like you haven't compiled stuff with the xinerama USE flag. I put it in make.conf, and then did a emerge --newuse. OK, I did that. The packages that got rebuilt are: $ equery hasuse xinerama [ Searching for USE flag xinerama in all categories among: ] * installed packages [I--] [ ~] x11-apps/xdpyinfo-1.0.3 (0) [I--] [ ~] x11-libs/qt-3.3.8b (3) [I--] [ ~] x11-libs/gtk+-2.14.3-r2 (2) [I--] [ ~] x11-libs/qt-gui-4.4.2 (4) [I--] [ ] x11-misc/engage- (0) [I--] [ ~] kde-base/ksplash-4.1.2 (4.1) [I--] [ ~] kde-base/plasma-workspace-4.1.2 (4.1) [I--] [ ~] kde-base/ksplashml-3.5.10 (3.5) [I--] [ ~] kde-base/systemsettings-4.1.2 (4.1) [I--] [ ~] kde-base/kwin-4.1.2 (4.1) [I--] [ ~] kde-base/libplasma-4.1.2 (4.1) [I--] [ ~] kde-misc/knetworkmanager-0.2.2_p20080528 (0) [I--] [ ~] kde-misc/filelight-1.0-r1 (0) [I--] [ ] media-libs/libsdl-1.2.13 (0) [I--] [ ] media-libs/xine-lib-1.1.15-r1 (1) [I--] [ ] net-libs/xulrunner-1.8.1.17 (1.8) [I--] [ ~] media-sound/kid3-1.0 (0) [I--] [ ~] media-sound/amarok-1.4.10-r1 (0) [I--] [ ~] media-video/mplayer-1.0_rc2_p27725-r1 (0) [I--] [ ] media-video/xine-ui-0.99.5-r1 (0) [I--] [ ~] media-video/gxine-0.5.903 (0) [I--] [ ~] app-cdr/k3b-1.0.5-r3 (0) tabletka ~ # equery hasuse xinerama | wc -l 285 most of them are apps from kde-base/* (3.5.9), seems that it changed between 3.5.9 and 3.5.10, plus iwndow managers like fluxbox, openbox... Seems like the only things that would affect kde-3 apps is qt-3.3.8b. Plus x11-libs/libXinerama and x11-proto/xineramaproto (both latest unstable) are installed. [snip] I'd appreciate some pros and cons feedback from the list before I embark on a huge emerge -e world to include Xinerama support. Why would you do -e world? How about `emerge -uN world` The N being --newuse. or `emerge -vauDN world`. I was running /bin/think --exaggerate --frustrated --logic-level -3 when I typed that :-) check out my blog for how I did it: http://nthrbldyblg.blogspot.com/2008/08/nvidia-xinerama-on-dell-m6300.html Nice blog :-) I'll fiddle some more with these tips later in the day, but first a conceptual question: I read that huge collection of docs from nvidia-drivers, and concluded that Xinerama and TwinView are fundamentally different and incompatible. i.e. Xinerama starts with two classic X screens and joins them in software to make one big display - an abstraction layer if you will. TwinView rips out the guts of X, dispenses with the notion of separate screens for a TwinView display and gives you one giant screen with no API for an app to see how this big screen is composed. So, you either use Xinerama or TwinView, but not both. Obviously, this understanding of mine is flawed. Which bit did I get wrong? Xinerama consists basically of two parts, the protocol to communicate the position/sizes of screen between the Xserver and the applications (which you usually get by enabling the xinerama use flag) and an xserver part (module?) that you can use to set up the screens. What you said is correct for the Xserver setup part... You use either xinerama setup to put together completely different displays (might be different cards, such as one nvidia, one ati, ...) or twinview in case of a dualhead nvidia setup. But both this setups use the xinerama protocol to let the apps/wm know the placement of the monitors. -- alan dot mckinnon at gmail dot com -- _ | YoYo () Siska === http
Re: [gentoo-user] Xinerama vs TwinView for dual monitor setup
On Sat, Oct 11, 2008 at 11:34:10PM +0200, Alan McKinnon wrote: Hi, My notebook has this graphics hardware. [EMAIL PROTECTED] ~ $ sudo lspci | grep VGA 01:00.0 VGA compatible controller: nVidia Corporation GeForce 8600M GT (rev a1) [EMAIL PROTECTED] ~ $ sudo xdpyinfo | grep -A4 'screen #0' screen #0: print screen:no dimensions:1920x1200 pixels (332x210 millimeters) resolution:147x145 dots per inch depths (7):24, 1, 4, 8, 15, 16, 32 I also have a second LCD monitor at work, a 1280x1024 that is physically slightly larger than the notebook screen, with a corresponding lower dpi. I've configured it with TwinView to have the second monitor on the right, and how I usually use it is to put a user's support mail on that where I can read it and fix their issues using the tools on the main monitor. So it's a very unsophisticated setup, I have no need for massive 3D accel for eg games, or even for placing windows across two monitors. Windows are always on one screen or the other (because of the huge dpi difference). There are two smallish issues: The viewports are aligned along the top edge and the panel/kicker/plasma/whatever on every desktop environment insists on trying to stretch across both monitors, into dead space on the right hand one. I'm getting use to right-click on panel, configure, set width to 57% at work, 100% at home. If I align the viewports on the bottom edges, windows managers tend to want to position new windows with their title bars in the dead space at the top. You probably haven't emerged the applications with Xinerama support. This is especially true for kde 3. Twinview uses the xinerama protocol (well,its an extension of the X protocol... ;) to inform applications about the layout of monitors. kdm and entrance want to stretch over both monitors. I definitely do not want this. Murphy dictates that all useful DM menus will end up in the dead space regardless of the theme I use g My research into nvidia's docs leads me to believe that TwinView is designed to make the presence of two physical monitors invisible and present one giant X screen, with a funky API for dead spaces (which may or may not work). I'm thinking Xinerama is the better option, despite the fact that it's old, clunky, hopeless at dealing with XRandR and can't be changed on the fly. I'm happy to set up two ServerLayouts to deal with this. As I said, twinview uses the xinerama protocol to inform apps about the monitors, so there wouldn't be any difference in the way applications behave. You would only loose the advantages of twinview (you can look at it as an enhanced, nvidia only, in-driver version of xinerama) Even xrandr 1.2 provides xinerama style info for the applications, so you certainly want your application to be compiled with xinerama support, independently of the way you set up the X server. BTW in my experince kde compiled without xinerama supp. handles multiple (independent) screens O, but not xinerama (well, that could be expected), and with xinerama support it handles xinerama ok, but fails with independent screen ;) I'd appreciate some pros and cons feedback from the list before I embark on a huge emerge -e world to include Xinerama support. -- alan dot mckinnon at gmail dot com yoyo -- _ | YoYo () Siska === http://www.ksp.sk/
Re: [gentoo-user] libqt-mt.so.3:
On Wed, Sep 17, 2008 at 08:48:48PM -0400, sean wrote: Iain Buchanan wrote: sean wrote: sounds like you need to run revdep-rebuild Gentoo is kind-of different to other distros, so unless the references you found were all about gentoo, they're probably leading you up the wrong path. I think you've upgraded qt, hence whatever program you're trying to run hasn't been compiled with the new lib versions. HTH, Thanks to Dales reply, I have found libqt. See below. The system is a fresh install, but it is also amd64 bit, as the result shows below. equery b libqt-mt.so.3 [ Searching for file(s) libqt-mt.so.3 in *... ] x11-libs/qt-3.3.8-r4 (/usr/qt/3/lib64/libqt-mt.so.3 - libqt-mt.so.3.3 Here is a ls of /usr/qt/3/lib lrwxrwxrwx 1 root root 15 Sep 16 10:00 libqt-mt.so.3 - libqt-mt.so.3.3 lrwxrwxrwx 1 root root 17 Sep 16 10:00 libqt-mt.so.3.3 - libqt-mt.so.3.3.8 -rwxr-xr-x 1 root root 8628296 Sep 16 10:00 libqt-mt.so.3.3.8 So it is linked. The application is called Firstclass, www.firstclass.com , an email client my employment uses, so I thought I would try to get it running here. Though I would not use it for my own purposes. I figure the 64bit OS is giving things a headache. Just to make sure, is the app you're trying to run 32bit or 64bit? If its 32bit, you would need some of the emul-linux-x86 packages, perhaps app-emulation/emul-linux-x86-qtlibs to get the 32bit libqt-mt.. yoyo
Re: [gentoo-user] Re: ctrl+alt+fx doesn't work [SOLVED]
On Sat, Sep 06, 2008 at 12:43:04AM +0200, pat wrote: Problem is this line in the keyboard section: Option XkbLayout us,cz Simply, enabling another language disable switching to console (I've check it for another languages too). Ugh =8-() If someone is able to explain why that happen and how to solve it for two languages I'll be glad :-) But right now I'm setting up the dualhead display, so I need to switch to console ... after that I'll turn on the language :-D Thanks to Dale and others for help Pat I remember that some (older) versions of evdev drivers had problems with VT switching (and layouts in general ;) but it seems to work for me for some time now... btw, if you are really desperate you can always map ctrl-alt-fX to commands like sudo chvt 1 :) yoyo -- _ | YoYo () Siska http://www.ksp.sk/
Re: [gentoo-user] Clone a running gentoo machine onto another machine
Benyamin Dvoskin wrote: It is a running gentoo system in this case But it doesnt make a difference to me. I want to know generally. anyway I will try what everyone wrote here and we'll see how it goes. Thanks again. Btw You can also do a mount --bind / /mnt/something and then you will see the original root in /mnt/something without any of the other filesystems. This is sometimes better if you want an exact copy, because fex /dev usually has some basic nodes which get covered by the udev's tmpfs, and althought you normally don't need them... ;) I cloned a few running systems this way (copied it to an usb disk, setup lilo and took the disk to another machine ;) but it was always with mount -o remount,ro / and the systems were minimal (system+few packages, almost nothing running, so it was possible to remount it ro) yoyo -- gentoo-user@lists.gentoo.org mailing list
Re: [gentoo-user] Routing problem ?
Holla wrote: Hi, I think I have a routing problem with network shown below (hope my ascii art survives) From PC2, I cannot ping 192.168.1.1 and no internet. Also cannot ping ISP's DNS servers. But there is full connectivity between PC1 and PC2. At PC2, # traceroute 192.168.1.1 traceroute to 192.168.1.1 (192.168.1.1), 30 hops max, 40 byte packets 1 * * * 2 * * * I reached upto this point by following up the gentoo howtos, but now stuck. Any pointers ? as someone other said, you should setup NAT, there should be enough information on the wiki, but basically iptabales -t nat -A POSTROUTING -o eth0 -s 192.168.2.0/24 -j MASQUERADE on PC1 should do it, but there might be better ways ;) (note that you need some iptables stuff in the kernel) one other thing, if nat doesn't work, some wireless aps (i'm thinking about the 192.168.2.1) need to have correctly set up default gateway etc... they sometimes try to be to smart and I had sometimes problems when the router was connected as a wireless client to them... btw, why don't you use the wireless on the ROUTER1 (doesn't seem you want to do any firewalling on the PC1)? It might make things much simpler... you could setup the other ap to connect to it in client mode and all your network could then be on the 192.168.1.0/24 and I would gues that your provider NATs the whole subnet... yoyo 192.168.1.1 +-+ ++ | |---| Router1 |=ASDL conn | | ++ | | | | | | | |192.168.1.23 +---+ 192.168.2.43 | |--| PC1 |))). +-+ +---+ . . Passive Hub . 192.168.2.1. ++ . | Router2|--))).. ++ | | +--+ | PC2 | +--+ 192.168.2.24 -- Router1 (UTSStarCom ISP supplied) : - router IP 192.168.1.1 - wireless enabled but not used -- PC1: (gentoo) - eth0 (192.168.1.23) and wireless (192.168.2.43) - no iptables configuration - routing table entries Kernel IP routing table Destination Gateway Genmask Flags Metric RefUse Iface 192.168.2.0 * 255.255.255.0 U 0 00 ra0 192.168.1.0 * 255.255.255.0 U 0 00 eth0 loopback* 255.0.0.0 U 0 00 lo default 192.168.1.1 0.0.0.0 UG0 00 eth0 # echo 1/proc/sys/net/ipv4/ip_forward # Kernel Networking options # CONFIG_UNIX=y CONFIG_XFRM=y CONFIG_INET=y CONFIG_IP_ADVANCED_ROUTER=y CONFIG_ASK_IP_FIB_HASH=y CONFIG_IP_FIB_HASH=y CONFIG_IP_ROUTE_VERBOSE=y CONFIG_INET_XFRM_MODE_TRANSPORT=y CONFIG_INET_XFRM_MODE_TUNNEL=y CONFIG_INET_XFRM_MODE_TRANSPORT=y CONFIG_INET_XFRM_MODE_TUNNEL=y CONFIG_TCP_CONG_BIC=y -- Router2 (WRT54GL) - router IP 192.168.2.1 - wireless enabled and used -- PC2 (gentoo) - static IP address 192.168.2.24 - routing table entries Kernel IP routing table Destination Gateway Genmask Flags Metric RefUse Iface 192.168.2.43* 255.255.255.255 UH0 00 eth0 192.168.2.0 * 255.255.255.0 U 0 00 eth0 192.168.1.0 192.168.2.43255.255.255.0 UG0 00 eth0 loopback* 255.0.0.0 U 0 00 lo default 192.168.2.430.0.0.0 UG0 00 eth0 -- gentoo-user@lists.gentoo.org mailing list
Re: [gentoo-user] Re: Is GWN dead?
Eddie Mihalow Jr wrote: Has anyone on this list ever used a PXE boot image to install? well, not exactly a PXE boot image... i had to install gentoo on thinkpad X41T (no cd) some time ago, and I wasn't able to boot from usb directly ( don't really remember why ;) so i dumped the minimal cd on my usb stick, copied the kernel and initrd to other machine and booted it through PXE, the kernel/initrd found the usb stick (and thought it to be the livecd ;) and everything worked fine... yoyo -- gentoo-user@lists.gentoo.org mailing list
Re: [gentoo-user] x looses ctrl, alt and shift keys
Sascha Hlusiak wrote: I have this too when using vmware and it seems that somehow just the keymap screws up. I run kcontrol, activate the keyboard layout switcher (with 2 languages in it) and once it's activated I can deactivate it again (or switch the language back and forth). Then it works again. What about a simple setxkbmap us? (or other layout you are using) Does that help? (You might need to emerge setxkbmap... ;) yoyo -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] zcat /usr/share/doc/conky-1.4.0-r1/conkyrc.sample.bz2 ~/.conkyrc didn't work.
Chen Xianwen wrote: Hello all, I'm a noob. I tried zcat /usr/share/doc/conky-1.4.0-r1/conkyrc.sample.bz2 ~/.conkyrc but it didn't work. Please help me. bzcat instead of zcat? yoyo -- [EMAIL PROTECTED] mailing list
