Re: [gentoo-user] Re: Digest of gentoo-user@lists.gentoo.org issue 1554 (82878-82927)
On Fri, 2008-08-08 at 02:10 -0500, toefraz wrote:
unsubscribe
NO!
--
Iain Buchanan iaindb at netspace dot net dot au
Interestingly enough, since subroutine declarations can come anywhere,
you wouldn't have to put BEGIN {} at the beginning, nor END {} at the
end. Interesting, no? I wonder if Henry would like it. :-) --lwall
[gentoo-user] Add drivers to custom live CD
Hi, I created a live CD by following the steps in this tutorial: http://gentoo-wiki.com/HOWTO_create_a_custom_distro_(in_less_than_three_hours) Everything works perfectly, except for one thing: I don't know how to add drivers. If I just make install them with DESTDIR=/real, the drivers don't appear when I chroot to the real directory and while making I get these mismatching errors: WARNING: vmlinux - Section mismatch: reference to .init.text: from .text between 'rest_init' (at offset 0xc01010ec) and 'try_name' WARNING: vmlinux - Section mismatch: reference to .init.text: from .text between 'iret_exc' (at offset 0xc0313e76) and '_etext' WARNING: vmlinux - Section mismatch: reference to .init.data:initkmem_list3 from .text between 'set_up_list3s' (at offset 0xc014e226) and 's_start' WARNING: vmlinux - Section mismatch: reference to .init.data:logo_linux_clut224 from .text between 'fb_find_logo' (at offset 0xc021621a) and 'lcd_class_release' WARNING: vmlinux - Section mismatch: reference to .init.data:logo_linux_mono from .text between 'fb_find_logo' (at offset 0xc021621f) and 'lcd_class_release' WARNING: vmlinux - Section mismatch: reference to .init.text: from .text between 'iret_exc' (at offset 0xc03144fe) and '_etext' Does anyone know how to properly add drivers to the live CD? Thanks in advance
[gentoo-user] python-updater keeps catching same packages
voodoo adam # python-updater * Starting Python Updater from 2.4 to 2.5 : * Adding to list: =x11-libs/vte-0.16.13 * Adding to list: =app-office/dia-0.96.1-r1 * Adding to list: =dev-libs/boost-1.34.1-r2 And if i run python-updater again after letting it rebuild these packages it picks them up again. emerge --depclean -pv says; dev-lang/python selected: 2.4.4-r13 protected: none omitted: 2.5.2-r6 And revdep-rebuild is clean. So - is it safe to remove python 2.4.4?
[gentoo-user] RE: python-updater keeps catching same packages
Sorry, should have googled first; http://bugs.gentoo.org/show_bug.cgi?id=232467
Re: [gentoo-user] Adding a gentoo workstation to Active Directory network
On 9 Aug 2008, at 14:52, Yoav Luft wrote: In an ideal world users should use their domain username password to log on when they sit down at the Linux box. And they should be mounting the directories they need off the file server by (double-clicking on a drive icon on their KDE desktop if necessary and) using their same unique credentials (*not yours!*). If you want to fully implement this then it's not a two minute job; you shouldn't need much from the Windows IT admins except the name of the domain and perhaps the resolvable name of the domain master server - you should be able to test using your own domain \user:pass That is, actually, what I'm trying to achieve, but what is crucial to the usability of the linux box is that each user (a would be developer) would have access to his own files and the departments files on the server without any knowledge of the working of Linux, Samba, or others. It would be especially nice if logon names would be taken from the server, and those relief users to manually add and configure more users. I can think on an awkward solution, making a script that sets up a new user and assumes the user name is the same as the one in the domain. But I am sure there is a cleaner, better solution, only that I haven't found it yet. So, I will sum up shortly what I want, starting from most important: 1. Users will have access to the departments files without root access with their own privileges rather then mine (achieved through given sudo to mount, and putting it all in a script). 2. Users will have access to their own personal files (achieved through the same script. Not sure if it is run automatically when a user logs on) 3. Any user on the domain will be able to log on to the machine, and have access to his files, will automatically authenticate himself to network services, etc. http://www.google.com/search?q=authenticating+linux+users+against +windows+domain Sorry to say read teh g0ggles, newb, but I'd need to read a number of these pages myself before I could say you want to do it this way not that or before I was even aware of the advantages disadvantages of the different approaches. Mostly you shouldn't need much from the Windows admins. If you were to install XP Pro on a new PC and bring it into the office, all you'd need to do is right-click on my computer and change from My Workgroup to My Domain (or BobsElectricals or whatever) - the next time the machine starts you'll need to log on using your username:password on the domain. Likewise all you *should* need to add the Linux box to the domain is its name, and perhaps the hostname / ip address of the master domain server. The approach for mounting shares isn't obvious to me right now, but hopefully will become clear to you during the days that you spend setting the authentication up. On a Linux Samba box there is a special share called homes and mounting that seems to automatically use the ~ of the user authenticating; on Windows you can refer to % user%, although you probably can't combine these two methods directly. I don't use Linux on the desktop, but KDE or Gnome or whatever probably has a facility to run scripts upon logon; write a Bash script calling var=`whoami` ; mount \\domainserver\$var and put it in /etc/skel (or the KDE / Gnome equivalent). Stroller.
Re: [gentoo-user] Adding a gentoo workstation to Active Directory network
Hello, I recently set up samba to allow authentification against Active Directory for file sharing on a CentOS 4.5. Even if their installer is supposed to do it correctly, it didn't work the way I wanted, so I had to understand how to set it up manually. The main problem I found with documentations is that there's no one-shot documentation that allows you to join a domain if you meet so many obscure error messages like I had. I have more knowledge on Gentoo than centOs (so redhat), but what I say here has only been tested on centOS. Unfortunately for you, I'm on hollydays and won't go back to office until second part of October, so I can only tell you what I remember : You need : - a Kerberos client - a ntp daemon to set your clock according to your domain controller (more than 5 minutes offset will lead kerberos not to deliver tickets) - samba with winbind support - manually record your machine in the DNS used by AD Set up samba with ads security (refer to the official samba howto) Be sure your smb.conf has winbind configuration directives Files I remember I updated (CentOS architecture) : - /etc/samba/smb.conf - /etc/sysconfig/network (for the hostname of your machine to be the FQDN e.g. tux.mywindows.domain.corp and `hostname --fqdn` must immediately answer) = /etc/conf.d/hostname on gentoo - /etc/nsswitch.conf to add winbind for a few things (passwd,group,shadow if I remember, with less priority than file; otherwise it will be long to log in as a local user) - /etc/krb5.conf /etc/krb.conf[backward compatibility, may not be needed on gentoo; try without that's one file less to manage] (documentations give the few lines required) You'll also have to modify PAM config files for local access matching against AD, but I didn't tried it. Before you frag your brain out with samba and winbind, you must succeed a `kinit mywindowsuser` and see your ticket with `klist`. And be sure you can resolve local names with a nslookup. Some recommend you set the name and ip of your Domain Controller (DC) in /etc/hosts to avoid DNS failure. To join a domain, use the net join ads command, as explained in the docs : it must work. If it don't, don't look forward: solve this problem as it means you cannot access your DC. There's no need to configure LDAP if you use an AD architecture. And unless your DC is configured otherwise, it should offer you all required services (kerberos, ntp, dns). Don't hesitate to set up the log level of samba to 4 or the example value of the man page to get what's wrong. Don't look for complex configuration : a few simple lines does the job for matching AD. If you can identify against AD for file shares, then you just ( :D ) have to set up pam for the main login. I'd say there are 3 or 4 winbind directives (uid/gid range, auto append defautl domain, etc) in and 5 important samba directives smb.conf. I hope this fragment can help you a little bit, Jil.
Re: [gentoo-user] Advice about setting up split home directory
Alan E. Davis wrote: Norberto and Josh: Thank you for the suggestion. It's on the back burner. I have the space to experiment with it now. I have balked for the time being on basis of, partly, my need to be able to swap drives in and out, and have it clear in mind which partitions belong to what. Also my main drive is a 1 RPM faster drive, and I'd like to keep the partitions or directories that are mainly for storage separated. I really do notice a difference in the performance of the drive. this is somewhat of a conundrum: how to keep the current projects focused on the faster drive. Interestingly (to me) while I carefully planned for swap on the faster drive, since I moved to 2GB of RAM, I think I've only touched swap two or three times, and then only passingly! I definitely wouldn't want to put / into LVM. If I do LVM it will be the easy way, the most clearcut way. As one that's used LVM and other similar software in both Windows and the BSDs, be sure you understand the risk involved. While the idea of one big drive sounds appealing (which is why I used it), lose one drive and you lose everything in the LV unless you are mirroring, using parity, or some combination of both. I have been bit by this time and time again and have finally decided that LVM is not worth the hassle for me any longer, especially since a 1 TB drive can be found easily for less than $200 (US). Anyway, I'm not knocking those who use LVM. Just understand the risk. :) Cheers, Drew -- Be a Great Magician! Visit The Alchemist's Warehouse http://www.alchemistswarehouse.com
Re: [gentoo-user] Re: Kernel Linux 2.6.25-r7 with gentoo patches fail to boot
Thanks, it works! On Sat, Aug 9, 2008 at 5:20 PM, Nikos Chantziaras [EMAIL PROTECTED] wrote: Yoav Luft wrote: mount /boot copied the kernel, reconfigurated grub, and rebooted. You can also emerge debianutils (I think it's installed by default, but check first) which will enable you to do a make install instead of copying the files by hand to /boot. It will name the files correctly (vmlinuz symlink to vmlinuz-version along with System.map and config).
Re: [gentoo-user] Changes in the way USB devices are mounted
On Tue, 22 Jul 2008 19:03:34 +0100 Mick [EMAIL PROTECTED] wrote: I suspect that something changed recently on the way that USB devices are mounted. I noticed that mounting a USB flash drive has stopped working as follows: [snip] $ pmount /dev/sda Error: device /dev/sda is not removable Same problem here after upgrade to 2.6.25-r7. It seems that there is a bug in sysfsutils: http://bugs.gentoo.org/show_bug.cgi?id=220347. Robert -- Robert Cernansky E-mail: [EMAIL PROTECTED] Jabber: [EMAIL PROTECTED]
[gentoo-user] [extremly, wildly, obscenely OT] Is there a Linux system without GNU userlands?
Hi, I ask it here because I really don't know where to ask it. Is there a Linux system somewhere with a *non-GNU* userland? I wonder in particular if: - there are Linux systems using the BSD userlands - there are Linux systems using completely non-standard userlands... let's say, non-Unix tools on top of a Linux kernel. Only thing I can think about is (maybe) embedded systems or things using busybox, but in the latter case just imitating gnu or bsd userlands. Not that I have a real purpose for such a bizarre beast, I'm just curious. m.
[gentoo-user] mplayer and fontconfig
I recently started using tovid again and it seems to be having a lot of problems with the programs it uses. The last time I used it was in November and all worked well. First there was a problem with libGL.so.1, it didn't seem to be on my system. ldd `which mplayer` said it wasn't found. After making a link (which I'm not sure is the correct fix) /usr/lib/libGL.so.1 now links to the /usr/X11R6/lib/libGL.so file. This got me a little further with tovid. Second tovid now complains when it runs mplayer. The error is: The fontconfig option can't be used in a config file. Error parsing option fontconfig=1 at line 132 Warning unknown option subfont-osd-scale at line 133 Warning unknown option subfont-text-scale at line 134 The /etc/mplayer/mplayer.conf file has fontconfig=1 in it. If I leave this value tovid's mplayer command fails and if I use mplayer for something else it doesn't work. If I comment out fontconfig=1 the tovid mplayer command hangs, but then I can use mplayer to play videos. Any help is greatly appreciated. Thanks, dave
Re: [gentoo-user] [extremly, wildly, obscenely OT] Is there a Linux system without GNU userlands?
b.n. schrieb: Hi, I ask it here because I really don't know where to ask it. Is there a Linux system somewhere with a *non-GNU* userland? I wonder in particular if: - there are Linux systems using the BSD userlands - there are Linux systems using completely non-standard userlands... let's say, non-Unix tools on top of a Linux kernel. Only thing I can think about is (maybe) embedded systems or things using busybox, but in the latter case just imitating gnu or bsd userlands. Not that I have a real purpose for such a bizarre beast, I'm just curious. m. Damn Small Linux and Damn Small Linux Not (50MB and ~100MB total size, respectively) use busybox. signature.asc Description: OpenPGP digital signature
[gentoo-user] Re: [extremly, wildly, obscenely OT] Is there a Linux system without GNU userlands?
On 2008-08-10, b.n. [EMAIL PROTECTED] wrote: Hi, I ask it here because I really don't know where to ask it. Is there a Linux system somewhere with a *non-GNU* userland? What do you mean by non-Gnu? SW for which the FSF doesn't own the copyrights? SW which is distributed under something other than one of the Gnu Public Licenses? I wonder in particular if: - there are Linux systems using the BSD userlands Are there even any BSD systems with 100% non-GNU userlands? The last time I did a basic FreeBSD install, it included Gnu user-land stuff (e.g. gcc). - there are Linux systems using completely non-standard userlands... let's say, non-Unix tools on top of a Linux kernel. I'm sure there are embedded Linux systems which contain little or no FSF-owned code. But I don't know if you'd consider those systems to even _have_ a userland. Only thing I can think about is (maybe) embedded systems or things using busybox, but in the latter case just imitating gnu or bsd userlands. -- Grant
[gentoo-user] Re: [extremly, wildly, obscenely OT] Is there a Linux system without GNU userlands?
Grant Edwards wrote: On 2008-08-10, b.n. [EMAIL PROTECTED] wrote: Hi, I ask it here because I really don't know where to ask it. Is there a Linux system somewhere with a *non-GNU* userland? What do you mean by non-Gnu? SW for which the FSF doesn't own the copyrights? SW which is distributed under something other than one of the Gnu Public Licenses? Software that does not belong to the GNU project. The GNU project is a fixed and well defined set of software. Are there even any BSD systems with 100% non-GNU userlands? Probably not, but almost. The last time I did a basic FreeBSD install, it included Gnu user-land stuff (e.g. gcc). Anything else besides GCC? I installed FreeBSD just for the kicks of it, and I'm not sure there's anything GNU in there other than GCC, not sure.
Re: [gentoo-user] [extremly, wildly, obscenely OT] Is there a Linux system without GNU userlands?
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 b.n. wrote: Hi, I ask it here because I really don't know where to ask it. Is there a Linux system somewhere with a *non-GNU* userland? I wonder in particular if: - there are Linux systems using the BSD userlands - there are Linux systems using completely non-standard userlands... let's say, non-Unix tools on top of a Linux kernel. Only thing I can think about is (maybe) embedded systems or things using busybox, but in the latter case just imitating gnu or bsd userlands. Not that I have a real purpose for such a bizarre beast, I'm just curious. m. You might possibly be missing one of the most basic (in organization) differences between any BSD and any Linux is that BSD's are all built and packaged with a set of userland programs. This doesn't include many user applications, just the kind of things that you think of as being part of any base (like shells, or utilities like the various filesystem tools, grep, find, like that) Linux, OTOH, is only a kernel. Any time you go after a distribution that has more than the kernel (and ONLY the kernel) its because the group putting together that distribution has decided to attach those parts, but the Linux developers are concerned with the kernel alone. So, when you talk about, say, FreeBSD, you're talking about kernel + userland base. This isn't truie with Linux, so all linuxes are just a little bit different in their choice of userland tools. Some Linux distros cater more to developers, some to businesspeople, some to newbies, some to professionals. FreeBSD is FreeBSD. There are good reasons why both are as they are, neither is (without your own opinion making it so) better. It is usually true that Linuxes all have better coverage of device drivers. It is also usually true BSD's are usually more evenly planned. But, there are differences. What you ought to do is to read as many different OSes as yo have time for, because it sure makes a great hobby. -BEGIN PGP SIGNATURE- Version: GnuPG v2.0.9 (FreeBSD) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkifWQgACgkQz62J6PPcoOlHmgCfRZFD/GhB0Isz/ZJ2MOt/nU5i BYAAnR4ahD7qLaX1RmAMpT56egSIbbah =lTA4 -END PGP SIGNATURE-
[gentoo-user] Re: [extremly, wildly, obscenely OT] Is there a Linux system without GNU userlands?
On 2008-08-10, Nikos Chantziaras [EMAIL PROTECTED] wrote: Is there a Linux system somewhere with a *non-GNU* userland? What do you mean by non-Gnu? SW for which the FSF doesn't own the copyrights? SW which is distributed under something other than one of the Gnu Public Licenses? Software that does not belong to the GNU project. The GNU project is a fixed and well defined set of software. It would then help if you could point us to that definition? Answering a question about your terminology with my term is well defined isn't very helpful. The last time I did a basic FreeBSD install, it included Gnu user-land stuff (e.g. gcc). Anything else besides GCC? I can't think of anything else off the top of my head, but it's been a couple years since I've had a FreeBSD system. -- Grant Edwards grante Yow! I'm sitting on my at SPEED QUEEN ... To me, visi.comit's ENJOYABLE ... I'm WARM ... I'm VIBRATORY ...
Re: [gentoo-user] [extremly, wildly, obscenely OT] Is there a Linux system without GNU userlands?
Chuck Robey ha scritto: You might possibly be missing one of the most basic (in organization) differences between any BSD and any Linux is that BSD's are all built and packaged with a set of userland programs. This doesn't include many user applications, just the kind of things that you think of as being part of any base (like shells, or utilities like the various filesystem tools, grep, find, like that) Linux, OTOH, is only a kernel. Any time you go after a distribution that has more than the kernel (and ONLY the kernel) its because the group putting together that distribution has decided to attach those parts, but the Linux developers are concerned with the kernel alone. Ehm, thanks for the lesson, but I am actually well aware of that. I installed and used a lot of Linux distros and, to a lesser extent, BSD and other exotic systems (Hurd anyone?). Instead, maybe you might possibly be missing the fact that kernel-BSD systems with GNU userlands have been attempted (Debian GNU/kFreeBSD being one - dunno about the Gentoo/FreeBSD port -is it still alive, by the way?). I wondered if there is the contrary, as a startpoint. So, when you talk about, say, FreeBSD, you're talking about kernel + userland base. This isn't truie with Linux, so all linuxes are just a little bit different in their choice of userland tools. That's why I asked if there is some Linux that is not a little bit but *wildly* different, as to be almost unrecognizable as the Linux we're all familiar with (that usually is done by a bash/zsh/ksh shell + other gnu coreutils etc.) For a (theoretical) example, imagine a system that boots in the Windows Powershell on top of the Linux kernel. m.
[gentoo-user] Re: [extremly, wildly, obscenely OT] Is there a Linux system without GNU userlands?
Grant Edwards wrote: On 2008-08-10, Nikos Chantziaras [EMAIL PROTECTED] wrote: Is there a Linux system somewhere with a *non-GNU* userland? What do you mean by non-Gnu? SW for which the FSF doesn't own the copyrights? SW which is distributed under something other than one of the Gnu Public Licenses? Software that does not belong to the GNU project. The GNU project is a fixed and well defined set of software. It would then help if you could point us to that definition? Answering a question about your terminology with my term is well defined isn't very helpful. Sorry. Here is the definitive list of all software provided by the GNU project: http://directory.fsf.org/GNU Given how much software GNU provides, I imagine it's pretty much impossible to find a Linux or BSD these days that is 100% GNUless.
Re: [gentoo-user] [extremly, wildly, obscenely OT] Is there a Linux system without GNU userlands?
On Sonntag, 10. August 2008, b.n. wrote: Hi, I ask it here because I really don't know where to ask it. Is there a Linux system somewhere with a *non-GNU* userland? linux + uclibc + busybox? yes. And maybe you even get X or KDE run on it - google and tell us your results ;) http://www.linuxfromscratch.org/hints/downloads/files/uclibc.txt
[gentoo-user] Re: [extremly, wildly, obscenely OT] Is there a Linux system without GNU userlands?
On 2008-08-10, Nikos Chantziaras [EMAIL PROTECTED] wrote: Sorry. Here is the definitive list of all software provided by the GNU project: http://directory.fsf.org/GNU Given how much software GNU provides, I imagine it's pretty much impossible to find a Linux or BSD these days that is 100% GNUless. It's going to be pretty hard to find a Linux-based system (that's not an embedded/appliance setup) which doesn't at least use gcc, binutils, and autogen/automake. -- Grant Edwards grante Yow! Jesuit priests are at DATING CAREER DIPLOMATS!! visi.com
Re: [gentoo-user] Re: [extremly, wildly, obscenely OT] Is there a Linux system without GNU userlands?
Grant Edwards wrote: On 2008-08-10, Nikos Chantziaras [EMAIL PROTECTED] wrote: The last time I did a basic FreeBSD install, it included Gnu user-land stuff (e.g. gcc). Anything else besides GCC? I can't think of anything else off the top of my head, but it's been a couple years since I've had a FreeBSD system. This is pretty easy to find out: they keep all the GNU project stuff separated out in their CVS tree and build environment (for both legal and technical efficiency, I suspect). I know there is an ongoing effort to replace any GPL'd userland tools with equivalent BSD licensed ones, but, of course, if the best tool for the job is under the GPL they're going to keep using it. Of the stuff that's still being included, the most items notable GNU projects are: binutils/gcc/gdb rcs texinfo groff grep They also use several standard, non-GNU GPL'd tools, like cvs, man, and patch.
Re: [gentoo-user] [extremly, wildly, obscenely OT] Is there a Linux system without GNU userlands?
On Mon, 2008-08-11 at 00:54 +0200, b.n. wrote: That's why I asked if there is some Linux that is not a little bit but *wildly* different, as to be almost unrecognizable as the Linux we're all familiar with (that usually is done by a bash/zsh/ksh shell + other gnu coreutils etc.) For a (theoretical) example, imagine a system that boots in the Windows Powershell on top of the Linux kernel. Now that you mention it, I think there is a PLC manufacturer that has a real time windows CE environment. It works because they use a Linux kernel with the low latency and real-time options, which only hands over run time to windows when it decides it can. I was reading about it a while back... If you want I could dig up some more? cya, -- Iain Buchanan iaindb at netspace dot net dot au You can do more with a kind word and a gun than with just a kind word. - Al Capone
Re: [gentoo-user] [extremly, wildly, obscenely OT] Is there a Linux system without GNU userlands?
Volker Armin Hemmann ha scritto: On Sonntag, 10. August 2008, b.n. wrote: Hi, I ask it here because I really don't know where to ask it. Is there a Linux system somewhere with a *non-GNU* userland? linux + uclibc + busybox? yes. And maybe you even get X or KDE run on it - google and tell us your results ;) http://www.linuxfromscratch.org/hints/downloads/files/uclibc.txt Wow! To bring back the thread on a Gentoo topic, I found neat howtos on the wiki: http://gentoo-wiki.com/TinyGentoo http://gentoo-wiki.com/Embedded_Gentoo I guess I'll try when I'll have some really spare time... Thanks for the cool link. The next step, I guess, is things that differ conceptually from the familiar Linux we're accustomed to. That is, if you follow newslogs like OSNews, you'll see a lot of hobbysts and engineers like to create new kernels. There is less interest in doing conceptually novel userlands (novel shells etc.) or it is just my impression? Maybe a more boring task? Sorry for the naive question, I am *by no mean* a system programmer (All I know is some decent Python, and I am just *now* learning the basics of C++ ,that's it), so I'm sure I am plain wrong or there are rock solid reasons behind it... m.
Re: [gentoo-user] [extremly, wildly, obscenely OT] Is there a Linux system without GNU userlands?
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 b.n. wrote: Chuck Robey ha scritto: You might possibly be missing one of the most basic (in organization) differences between any BSD and any Linux is that BSD's are all built and packaged with a set of userland programs. This doesn't include many user applications, just the kind of things that you think of as being part of any base (like shells, or utilities like the various filesystem tools, grep, find, like that) Linux, OTOH, is only a kernel. Any time you go after a distribution that has more than the kernel (and ONLY the kernel) its because the group putting together that distribution has decided to attach those parts, but the Linux developers are concerned with the kernel alone. Ehm, thanks for the lesson, but I am actually well aware of that. I installed and used a lot of Linux distros and, to a lesser extent, BSD and other exotic systems (Hurd anyone?). Instead, maybe you might possibly be missing the fact that kernel-BSD systems with GNU userlands have been attempted (Debian GNU/kFreeBSD being one - dunno about the Gentoo/FreeBSD port -is it still alive, by the way?). I wondered if there is the contrary, as a startpoint. So, when you talk about, say, FreeBSD, you're talking about kernel + userland base. This isn't truie with Linux, so all linuxes are just a little bit different in their choice of userland tools. That's why I asked if there is some Linux that is not a little bit but *wildly* different, as to be almost unrecognizable as the Linux we're all familiar with (that usually is done by a bash/zsh/ksh shell + other gnu coreutils etc.) For a (theoretical) example, imagine a system that boots in the Windows Powershell on top of the Linux kernel. m. Sorry. Not to be insulting, but it really sounded like a newbie question, which is why I reacted that way. On your own rereading, doesn't it sound a bit that way to you, a bit? I apologize, then. -BEGIN PGP SIGNATURE- Version: GnuPG v2.0.9 (FreeBSD) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkifhewACgkQz62J6PPcoOnGyQCfVJeYfaVDjZGChV/U92F3B6ve pqoAni0TBcjaapnxKEmgK20+FcOS/X55 =g/B1 -END PGP SIGNATURE-
[gentoo-user] cannot compile net-fs/mount-cifs using gcc 4.3.1
Hi. I tried to compile net-fs/mount-cifs because when I used smbfs I
got a very prominent message from the kernel saying that smbfs would
be removed in 2.6.27 and I should use cifs instead. However when I
tried to compile it, I got the following error:
Compiling source in
/var/tmp/portage/net-fs/mount-cifs-3.0.28/work/mount-cifs-3.0.28
...
mount.cifs.c: In function 'main':
mount.cifs.c:1095: error: 'PATH_MAX' undeclared (first use in this
function)
mount.cifs.c:1095: error: (Each undeclared identifier is reported only
once
mount.cifs.c:1095: error: for each function it appears in.)
*
* ERROR: net-fs/mount-cifs-3.0.28 failed.
* Call stack:
* ebuild.sh, line 49: Called src_compile
* environment, line 2041: Called die
* The specific snippet of code:
* $(tc-getCC) ${CFLAGS} mount.cifs.c -o mount.cifs || die make
* mount.cifs failed;
* The die message:
* make mount.cifs failed
*
* If you need support, post the topmost build error, and the call
stack if relevant.
* A complete build log is located at
'/var/log/portage/net-fs:mount-cifs-3.0.28:20080810-213024.log'.
* The ebuild environment file is located at
'/var/tmp/portage/net-fs/mount-cifs-3.0.28/temp/environment'.
*
ccs:/usr/src# script -c 'CFLAGS=-O2 -mtune=nocona -pipe
-D_GNU_SOURCE emerge net-fs/mount-cifs' temp.txt
Script started, file is temp.txt
These are the packages that would be merged, in order:
Calculating dependencies ... done!
[ebuild N] net-fs/mount-cifs-3.0.28 0 kB
Total: 1 package (1 new), Size of downloads: 0 kB
Would you like to merge these packages? [Yes/No] y
Verifying ebuild manifests
Building (1 of 1) net-fs/mount-cifs-3.0.28 for /
* mount-cifs-3.0.28.tar.bz2 RMD160 SHA1 SHA256 size ;-) ...
[ ok ]
* checking ebuild checksums ;-) ...
[ ok ]
* checking auxfile checksums ;-) ...
[ ok ]
* checking miscfile checksums ;-) ...
[ ok ]
cfg-update-1.8.2-r1: Checksum index is up-to-date ...
Unpacking source...
Unpacking mount-cifs-3.0.28.tar.bz2 to
/var/tmp/portage/net-fs/mount-cifs-3.0.28/work
* Applying user-unmount-ioctl-fix-bug4784.diff ...
[ ok ]
Source unpacked.
Compiling source in
/var/tmp/portage/net-fs/mount-cifs-3.0.28/work/mount-cifs-3.0.28 ...
mount.cifs.c: In function 'main':
mount.cifs.c:1095: error: 'PATH_MAX' undeclared (first use in this
function)
mount.cifs.c:1095: error: (Each undeclared identifier is reported only
once
mount.cifs.c:1095: error: for each function it appears in.)
*
* ERROR: net-fs/mount-cifs-3.0.28 failed.
* Call stack:
* ebuild.sh, line 49: Called src_compile
* environment, line 2041: Called die
* The specific snippet of code:
* $(tc-getCC) ${CFLAGS} mount.cifs.c -o mount.cifs || die make
* mount.cifs failed;
* The die message:
* make mount.cifs failed
*
* If you need support, post the topmost build error, and the call
stack if relevant.
* A complete build log is located at
'/var/log/portage/net-fs:mount-cifs-3.0.28:20080811-002815.log'.
* The ebuild environment file is located at
'/var/tmp/portage/net-fs/mount-cifs-3.0.28/temp/environment'.
*
My emerge --info is below -- any assistance would be appreciated.
I did try adding -DGNU_SOURCE to CFLAGS which made a few other
packages work, but no joy for this one.
Portage 2.2_rc3 (default/linux/x86/2008.0/desktop, gcc-4.3.1,
glibc-2.8_p20080602-r0, 2.6.25-gentoo-r6 i686)
=
System uname: [EMAIL PROTECTED]
Timestamp of tree: Sun, 10 Aug 2008 20:36:01 +
app-shells/bash: 3.2_p39
dev-java/java-config: 1.3.7, 2.1.6-r1
dev-lang/python: 2.4.4-r13, 2.5.2-r5
dev-python/pycrypto: 2.0.1-r6
sys-apps/baselayout: 1.12.11.1
sys-apps/sandbox:1.2.18.1-r3
sys-devel/autoconf: 2.13, 2.62-r1
sys-devel/automake: 1.4_p6, 1.5, 1.6.3, 1.7.9-r1, 1.8.5-r3, 1.9.6-r2, 1.10.1-r1
sys-devel/binutils: 2.18-r3
sys-devel/gcc-config: 1.4.0-r4
sys-devel/libtool: 1.5.26
virtual/os-headers: 2.6.25-r4
ACCEPT_KEYWORDS=x86
ALSA_CARDS=ali5451 als4000 atiixp atiixp-modem bt87x ca0106 cmipci emu10k1
emu10k1x ens1370 ens1371 es1938 es1968 fm801 hda-intel intel8x0 intel8x0m
maestro3 trident usb-audio via82xx via82xx-modem ymfpci
ALSA_PCM_PLUGINS=adpcm alaw asym copy dmix dshare dsnoop empty extplug file
hooks iec958 ioplug ladspa lfloat linear meter mmap_emul mulaw multi null plug
rate route share shm softvol
ANT_HOME=/usr/share/ant-core
APACHE2_MODULES=actions alias auth_basic auth_digest authn_anon authn_dbd
authn_dbm authn_default authn_file authz_dbm authz_default authz_groupfile
authz_host authz_owner authz_user autoindex cache dav dav_fs dav_lock dbd
deflate dir disk_cache env expires ext_filter file_cache filter headers ident
imagemap include info log_config logio mem_cache mime mime_magic negotiation
proxy proxy_ajp proxy_balancer proxy_connect proxy_http rewrite setenvif so
speling status unique_id userdir usertrack vhost_alias
APACHE2_MPMS=prefork
Re: [gentoo-user] [extremly, wildly, obscenely OT] Is there a Linux system without GNU userlands?
On Montag, 11. August 2008, b.n. wrote: Volker Armin Hemmann ha scritto: On Sonntag, 10. August 2008, b.n. wrote: Hi, I ask it here because I really don't know where to ask it. Is there a Linux system somewhere with a *non-GNU* userland? linux + uclibc + busybox? yes. And maybe you even get X or KDE run on it - google and tell us your results ;) http://www.linuxfromscratch.org/hints/downloads/files/uclibc.txt Wow! To bring back the thread on a Gentoo topic, I found neat howtos on the wiki: http://gentoo-wiki.com/TinyGentoo http://gentoo-wiki.com/Embedded_Gentoo I guess I'll try when I'll have some really spare time... Thanks for the cool link. The next step, I guess, is things that differ conceptually from the familiar Linux we're accustomed to. That is, if you follow newslogs like OSNews, you'll see a lot of hobbysts and engineers like to create new kernels. There is less interest in doing conceptually novel userlands (novel shells etc.) or it is just my impression? Maybe a more boring task? there are many shells. sh, bash, bsh. korn, csh, zsh, dash, tcsh, why make a new one, if you can do incredible stuff with zsh? A shell is not so easy to create. A new kernel is not so hard to do. The problem are the drivers - and all the quirks. It is one thing to write a little task scheduler for your little pet project, but if it crashs constantly it becomes a bitch to fight through all the errata. But at the beginning a simple kernel is much easier to do than stuff that runs on it (simple is the important work. A non-simple kernel is very hard). Another thing are libcs. A libc is a bitch. Luckily there is a whole bunch to choose from. glibc, bsd's libc, uclibc, dietlibc, ... so why re-invent the wheel? Or look at X. X is horrible. A convoluted mess of grown cruft and standards to hold the pile together. But where is the replacement? Fiasco/Berlin? failed. Y-window? failed. Because X works good enough. And if you aren't writing toolkits or apps using xlib directly, you don't need to care about most of the stuff. So hobbyist concentrate on the easy stuff - and a userland is not easy. Userland is not boring - it is very hard. And the best userland doesn't help you if no 3rd party software runs on it.
[gentoo-user] Re: cannot compile net-fs/mount-cifs using gcc 4.3.1
John covici wrote: Hi. I tried to compile net-fs/mount-cifs because when I used smbfs I got a very prominent message from the kernel saying that smbfs would be removed in 2.6.27 and I should use cifs instead. However when I tried to compile it, I got the following error: less /usr/portage/net-fs/mount-cifs/ChangeLog There it says: 11 Jun 2008; Tiziano Müller [EMAIL PROTECTED] +mount-cifs-3.0.30.ebuild: Version bump (fixes bug #225509, patch included) http://bugs.gentoo.org says about bug #225509: net-fs/mount-cifs-3.0.28 /w glibc-2.8 update: 'PATH_MAX' undeclared That means you have to use mount-cifs-3.0.30. Why I wrote all that down? To point out that looking at the ChangeLog of an ebuild can give an answer without even asking ;)
[gentoo-user] Re: cannot compile net-fs/mount-cifs using gcc 4.3.1
on Monday 08/11/2008 Nikos Chantziaras([EMAIL PROTECTED]) wrote John covici wrote: Hi. I tried to compile net-fs/mount-cifs because when I used smbfs I got a very prominent message from the kernel saying that smbfs would be removed in 2.6.27 and I should use cifs instead. However when I tried to compile it, I got the following error: less /usr/portage/net-fs/mount-cifs/ChangeLog There it says: 11 Jun 2008; Tiziano Müller [EMAIL PROTECTED] +mount-cifs-3.0.30.ebuild: Version bump (fixes bug #225509, patch included) http://bugs.gentoo.org says about bug #225509: net-fs/mount-cifs-3.0.28 /w glibc-2.8 update: 'PATH_MAX' undeclared That means you have to use mount-cifs-3.0.30. Why I wrote all that down? To point out that looking at the ChangeLog of an ebuild can give an answer without even asking ;) Now that is strange -- I did a google search and did not see that bug or the changelog or anything pointing to such -- this is why I wrote to the list. Thanks much. -- Your life is like a penny. You're going to lose it. The question is: How do you spend it? John Covici [EMAIL PROTECTED]
[gentoo-user] Shell problem
Hi all, I was working in a shell with tar and I changed something where now when I type * it interprets it as --exlucde i.e $ * bash: --exlucde: command not found or $ ls * ls: unrecognized option `--exlucde' Try `ls --help' for more information. how can I fix this? thanks, Ivan
Re: [gentoo-user] Shell problem
looks like an alias, maybe you have accidentaly edited .bashrc or .bash_profile hope this helps Francisco On Mon, Aug 11, 2008 at 9:57 AM, Ivan Alden [EMAIL PROTECTED] wrote: Hi all, I was working in a shell with tar and I changed something where now when I type * it interprets it as --exlucde i.e $ * bash: --exlucde: command not found or $ ls * ls: unrecognized option `--exlucde' Try `ls --help' for more information. how can I fix this? thanks, Ivan -- If you have an apple and I have an apple and we exchange apples then you and I will still each have one apple. But if you have an idea and I have one idea and we exchange these ideas, then each of us will have two ideas. - George Bernard Shaw
Re: [gentoo-user] Shell problem
2008/8/11 Francisco Ares [EMAIL PROTECTED]: looks like an alias, maybe you have accidentaly edited .bashrc or .bash_profile Just input alias in shell to check if the alias about * exists. hope this helps Francisco On Mon, Aug 11, 2008 at 9:57 AM, Ivan Alden [EMAIL PROTECTED] wrote: Hi all, I was working in a shell with tar and I changed something where now when I type * it interprets it as --exlucde i.e $ * bash: --exlucde: command not found or $ ls * ls: unrecognized option `--exlucde' Try `ls --help' for more information. how can I fix this? thanks, Ivan -- If you have an apple and I have an apple and we exchange apples then you and I will still each have one apple. But if you have an idea and I have one idea and we exchange these ideas, then each of us will have two ideas. - George Bernard Shaw -- BR, Zhou Rui
Re: [gentoo-user] Shell problem
On Montag, 11. August 2008, Ivan Alden wrote: Hi all, I was working in a shell with tar and I changed something where now when I type * it interprets it as --exlucde i.e $ * bash: --exlucde: command not found or $ ls * ls: unrecognized option `--exlucde' Try `ls --help' for more information. how can I fix this? thanks, Ivan unalias?
Re: [gentoo-user] Shell problem
On Mon, Aug 11, 2008 at 12:57:26PM +, Ivan Alden wrote: Hi all, I was working in a shell with tar and I changed something where now when I type * it interprets it as --exlucde i.e $ * bash: --exlucde: command not found or $ ls * ls: unrecognized option `--exlucde' Try `ls --help' for more information. Does this happen in every directory, or do you have a file named --exlucde that you created by mistake in the dir where this happens? That name would tend to sort first ahead of most other names. I can recreate it like this: $ touch ./--exlucde $ ls * ls: unrecognized option '--exlucde' Try `ls --help' for more information. $ and I can fix it like this: $ rm ./--exlucde $ -- ... _._. ._ ._. . _._. ._. ___ .__ ._. . .__. ._ .. ._. Felix Finch: scarecrow repairman rocket surgeon / [EMAIL PROTECTED] GPG = E987 4493 C860 246C 3B1E 6477 7838 76E9 182E 8151 ITAR license #4933 I've found a solution to Fermat's Last Theorem but I see I've run out of room o
