Re: Lynx and patches
Scott wrote: > On Fri, Jan 29, 2010 at 09:07:33PM -0500, rblythe wrote: > >> Whenever I use Lynx to download a patch, I do the following: >> >> 1) select the patch >> 2) right arrow -> (this will show the patch in text format) >> 3) press p (for print) >> 4) select save to a local file >> 5) at the bottom of the screen it shows the name of the file (just press >> enter) >> > > Why not select the patch and press 'd' to download it? That's what I > do all the time, and have never had a problem. > > Scott Swanson > You're right, that works too. Six years ago (when I was completely new GNU/Linux) I was having a heck of a time downloading patches with lynx (and I don't remember why). Printing them to a file worked back then so I just stuck with it. Now I will (re)add downloading patches to my personal knowledge base. Six years and still (re)learning. Thanks for the reminder. rblythe -- http://linuxfromscratch.org/mailman/listinfo/blfs-support FAQ: http://www.linuxfromscratch.org/blfs/faq.html Unsubscribe: See the above information page
Re: Lynx and patches
> > On 29 January 2010 22:32, Dr. Edgar Alwers wrote: > >> Hallo, >> >> building BLFS svn-20100128 I can download all packages through lynx. However, >> nearly every patch downloaded in this way is corrupted, the patch cannot be >> applied. Nothing similar happens, if I download the patches e.g. through >> Firefox on another box. Is such a problem known ? >> >> Thanks in advance >> Edgar >> -- Whenever I use Lynx to download a patch, I do the following: 1) select the patch 2) right arrow -> (this will show the patch in text format) 3) press p (for print) 4) select save to a local file 5) at the bottom of the screen it shows the name of the file (just press enter) The patch file is then saved properly on my system in my home directory. Years ago, I had this same problem and that was the method someone gave me to save patch files properly with lynx. I don't remember what I was doing wrong back then, but I was also getting 'corrupted' files. I hope this helps rblythe -- http://linuxfromscratch.org/mailman/listinfo/blfs-support FAQ: http://www.linuxfromscratch.org/blfs/faq.html Unsubscribe: See the above information page
Re: hal and users mounting sticks
lux-integ wrote:
> read/write access for usb-sticks can be made by editing /etc/auto.misc
> as:-
>
> ustick -fstype=vfat,rw,gid=8,umask=002 :/dev/sdXY
>
> the 'kludge' is to create a few (2 or 3) for different values of X and Y
> so that if/when the usb device is plugged out then in again (and gets
> another usb-bus-ID), it can still be accessed.
Perhaps it does not have to be a 'kludge'
Note: I am using udev-113, and autofs-4.1.4 but I believe this will
apply to later/current version of each package. My kernel is 2.6.23.9
I have been using autofs with udev for a few years now
To get persistent naming try this:
1) Plug in you usbstick (if it is already plugged in, unplug it first
and then plug it in again)
2) type: dmesg | tail -n25
At the end of this output you should see what your kernel sees the
device as
On my system, the kernel sees:
usb-storage: waiting for device to settle before scanning
scsi 5:0:0:0: Direct-Access SanDisk Cruzer Micro 0.1 PQ: 0 ANSI: 2
sd 5:0:0:0: [sdg] 1000944 512-byte hardware sectors (512 MB)
sd 5:0:0:0: [sdg] Write Protect is off
sd 5:0:0:0: [sdg] Mode Sense: 03 00 00 00
sd 5:0:0:0: [sdg] Assuming drive cache: write through
sd 5:0:0:0: [sdg] 1000944 512-byte hardware sectors (512 MB)
sd 5:0:0:0: [sdg] Write Protect is off
sd 5:0:0:0: [sdg] Mode Sense: 03 00 00 00
sd 5:0:0:0: [sdg] Assuming drive cache: write through
sdg: sdg1
sd 5:0:0:0: [sdg] Attached SCSI removable disk
sd 5:0:0:0: Attached scsi generic sg6 type 0
usb-storage: device scan complete
It is the 'sdg' part that I am interested in (similar for your system)
Leave the device plugged in
3) type: udevinfo -a -p /sys/block/sdg | tee usbstick_udev.output
# Use your information in place of sdg
# you can do this as a regular user *not* root
# The 'tee' command is so you can log the output for reading later
# you can use 'less usbstick_udev.output' to read the output
# my output is *not* included to keep this post short(er)
4) Now we want to make a persistent rule for you usbstick so that it is
named
the same thing every time the device is created:
I created a file: /etc/udev/rules.d/10-card_reader.rules (as root user)
and placed the following line in it:
SUBSYSTEMS=="scsi", ATTRS{model}=="Cruzer Micro|JUMPDRIVE|Flash Reader",
NAME{all_partitions}="JumpDrive"
I can't remember why I made the rule early (10-*) but I think it is
important
The above is all one line. see
http://www.reactivated.net/writing_udev_rules.html
for details on how to make good udev rules. Make sure you read it and
follow exactly what it says. The website explains how to decipher the
above output and how rules are made from that output
Once your udev rule is created:
5) Modify you autofs.misc (or what ever you have named yours) to
something like this:
JumpDrive -fstype=vfat,rw,gid=disk,umask=002 :/dev/JumpDrive1
see man 5 autofs for details on syntax and options. Also man mount for
more information
Now for extra measure, I created a symlink in /media like this:
ln -s /var/autofs/JumpDrive /media/JumpDrive which looks like this with
ls -l /media:
JumpDrive -> /var/autofs/JumpDrive
I did this so I could go to a consistent location to see the contents of
my usbstick without trying to remember where the actual autofs drive is
for the device. This is only for personal preference and not necessary.
Once all of this is done you should have:
1) A persistent name /dev/ustick (or whatever your preference is)
2) A way to automount the device when it is plugged in (autofs
functionality)
3) Possibly an easy way to find the device from the command line or GUI
file manager
4) A device that will auto-unmount once it is not in use or unplugged
I hope this helps you get your device working and handled like you want.
rblythe
--
http://linuxfromscratch.org/mailman/listinfo/blfs-support
FAQ: http://www.linuxfromscratch.org/blfs/faq.html
Unsubscribe: See the above information page
Re: AutoFS not working (will not create mount points for card reader & usb) [SOLVED]
rblythe wrote:
> I am having some trouble with autofs that isn't making much sense to me.
>
> Current Situation:
> Kernel = 2.6.23.9
> LFS Version = SVN-20071223
> BLFS Version = SVN-20071217 (not sure how this is earlier than LFS)
> AutoFS Version = 4.1.4 (this was the version at the time of my build)
> UDEV Version = 113 (this was the version at the time of my build)
> I am *not* running KDE, GNOME, XFCE, FluxBox, etc. (I use ude as my
> window manager)
>
>
> UDEV rules (specific to my situation) (devices are properly created in /dev)
> 10-cardreader.rules:
> SUBSYSTEMS=="scsi", ATTRS{model}=="IC1210CF",
> NAME{all_partitions}="CompactFlash"
> SUBSYSTEMS=="scsi", ATTRS{model}=="IC1210MS",
> NAME{all_partitions}="MemoryStick"
> SUBSYSTEMS=="scsi", ATTRS{model}=="IC1210MMC/SD",
> NAME{all_partitions}="MMC_SD_Card"
> SUBSYSTEMS=="scsi", ATTRS{model}=="IC1210SM",
> NAME{all_partitions}="SmartMedia"
> SUBSYSTEMS=="scsi", ATTRS{model}=="Cruzer Micro|JUMPDRIVE",
> NAME{all_partitions}="JumpDrive"
>
> /etc/auto.master:
> # Begin /etc/auto.master
> /media /etc/auto.misc
> # End /etc/auto.master
>
> /etc/auto.misc:
> # $Id: auto.misc,v 1.2 2003/09/29 08:22:35 raven Exp $
> #
> # This is an automounter map and it has the following format
> # key [ -mount-options-separated-by-comma ] location
> # Details may be found in the autofs(5) manpage
>
> # cd-fstype=iso9660,ro,nosuid,nodev :/dev/cdrom
>
> # The following reference devices created in
> # /etc/udev/rules.d/10-cardreader.rules
> # Review 'man mount' for proper fstype options# This must match UDEV
> rules
> CompactFlash--fstype=vfat,rw,gid=disk,umask=002
> :/dev/CompactFlash1 --ghost
> MMC_SD_Card --fstype=vfat,rw,gid=disk,umask=002
> :/dev/MMC_SD_Card1 --ghost
> SmartMedia --fstype=vfat,rw,gid=disk,umask=002
> :/dev/SmartMedia1 --ghost
> MemoryStick --fstype=vfat,rw,gid=disk,umask=002
> :/dev/MemoryStick1 --ghost
> JumpDrive --fstype=vfat,rw,gid=disk,umask=002
> :/dev/JumpDrive1 --ghost
>
> # All devices have '1' at the end on purpose. I could not manually mount to
> # /dev/CompactFlash, but I can manually mount to /dev/CompactFlash1
>
> When checking my kernel for proper configuration, I get the following
> output:
> grep AUTOFS /boot/config-2.6.23.9
> CONFIG_AUTOFS_FS=y
> CONFIG_AUTOFS4_FS=y
>
> grep NFS /boot/config-2.6.23.9
> CONFIG_NFS_FS=y
> CONFIG_NFS_V3=y
> # CONFIG_NFS_V3_ACL is not set
> # CONFIG_NFS_V4 is not set
> # CONFIG_NFS_DIRECTIO is not set
> CONFIG_NFSD=y
> CONFIG_NFSD_V3=y
> # CONFIG_NFSD_V3_ACL is not set
> # CONFIG_NFSD_V4 is not set
> CONFIG_NFSD_TCP=y
> CONFIG_ROOT_NFS=y
> CONFIG_NFS_COMMON=y
>
> grep SMB /boot/config-2.6.23.9
> # CONFIG_SMB_FS is not set
>
> UDEV does create the devices *before* autofs starts
>
> What could possibly be wrong with autofs? When I plug in my usb pen
> drive, autofs does not create the mount point /media/JumpDrive, but I
> can manually mount the drive (as root) with the following command:
> mount -t vfat -o uid=disk /dev/JumpDrive1 /media/JumpDrive
> so I believe udev is working properly.
>
> Any help or direction you can provide would be greatly appreciated.
>
> rblythe
>
>
>
It turns out this was a simple syntax error in my /etc/auto.misc file.
I had to change all "--fstype=*" to "-fstype=*" and completely remove
the "--ghost" syntax from each line in the file. Lastly, I changed my
/etc/auto.master file to read:
/var/atuofs /etc/auto.master --ghost
After completing those two steps, I created symlinks in /media like this:
ln -sf /var/autofs/CompactFlash1 /media/CompactFlash
Once done, I rebooted the system and now everything works perfectly.
rblythe
--
http://linuxfromscratch.org/mailman/listinfo/blfs-support
FAQ: http://www.linuxfromscratch.org/blfs/faq.html
Unsubscribe: See the above information page
AutoFS not working (will not create mount points for card reader & usb)
I am having some trouble with autofs that isn't making much sense to me.
Current Situation:
Kernel = 2.6.23.9
LFS Version = SVN-20071223
BLFS Version = SVN-20071217 (not sure how this is earlier than LFS)
AutoFS Version = 4.1.4 (this was the version at the time of my build)
UDEV Version = 113 (this was the version at the time of my build)
I am *not* running KDE, GNOME, XFCE, FluxBox, etc. (I use ude as my
window manager)
UDEV rules (specific to my situation) (devices are properly created in /dev)
10-cardreader.rules:
SUBSYSTEMS=="scsi", ATTRS{model}=="IC1210CF",
NAME{all_partitions}="CompactFlash"
SUBSYSTEMS=="scsi", ATTRS{model}=="IC1210MS",
NAME{all_partitions}="MemoryStick"
SUBSYSTEMS=="scsi", ATTRS{model}=="IC1210MMC/SD",
NAME{all_partitions}="MMC_SD_Card"
SUBSYSTEMS=="scsi", ATTRS{model}=="IC1210SM",
NAME{all_partitions}="SmartMedia"
SUBSYSTEMS=="scsi", ATTRS{model}=="Cruzer Micro|JUMPDRIVE",
NAME{all_partitions}="JumpDrive"
/etc/auto.master:
# Begin /etc/auto.master
/media /etc/auto.misc
# End /etc/auto.master
/etc/auto.misc:
# $Id: auto.misc,v 1.2 2003/09/29 08:22:35 raven Exp $
#
# This is an automounter map and it has the following format
# key [ -mount-options-separated-by-comma ] location
# Details may be found in the autofs(5) manpage
# cd-fstype=iso9660,ro,nosuid,nodev :/dev/cdrom
# The following reference devices created in
# /etc/udev/rules.d/10-cardreader.rules
# Review 'man mount' for proper fstype options# This must match UDEV
rules
CompactFlash--fstype=vfat,rw,gid=disk,umask=002
:/dev/CompactFlash1 --ghost
MMC_SD_Card --fstype=vfat,rw,gid=disk,umask=002
:/dev/MMC_SD_Card1 --ghost
SmartMedia --fstype=vfat,rw,gid=disk,umask=002
:/dev/SmartMedia1 --ghost
MemoryStick --fstype=vfat,rw,gid=disk,umask=002
:/dev/MemoryStick1 --ghost
JumpDrive --fstype=vfat,rw,gid=disk,umask=002
:/dev/JumpDrive1 --ghost
# All devices have '1' at the end on purpose. I could not manually mount to
# /dev/CompactFlash, but I can manually mount to /dev/CompactFlash1
When checking my kernel for proper configuration, I get the following
output:
grep AUTOFS /boot/config-2.6.23.9
CONFIG_AUTOFS_FS=y
CONFIG_AUTOFS4_FS=y
grep NFS /boot/config-2.6.23.9
CONFIG_NFS_FS=y
CONFIG_NFS_V3=y
# CONFIG_NFS_V3_ACL is not set
# CONFIG_NFS_V4 is not set
# CONFIG_NFS_DIRECTIO is not set
CONFIG_NFSD=y
CONFIG_NFSD_V3=y
# CONFIG_NFSD_V3_ACL is not set
# CONFIG_NFSD_V4 is not set
CONFIG_NFSD_TCP=y
CONFIG_ROOT_NFS=y
CONFIG_NFS_COMMON=y
grep SMB /boot/config-2.6.23.9
# CONFIG_SMB_FS is not set
UDEV does create the devices *before* autofs starts
What could possibly be wrong with autofs? When I plug in my usb pen
drive, autofs does not create the mount point /media/JumpDrive, but I
can manually mount the drive (as root) with the following command:
mount -t vfat -o uid=disk /dev/JumpDrive1 /media/JumpDrive
so I believe udev is working properly.
Any help or direction you can provide would be greatly appreciated.
rblythe
--
http://linuxfromscratch.org/mailman/listinfo/blfs-support
FAQ: http://www.linuxfromscratch.org/blfs/faq.html
Unsubscribe: See the above information page
Re: Trouble building OpenOffice-2.3.1 [SUCCESS]
DJ Lucas wrote: ( snipped by rblythe) > Is odd. Only thing I can see is that ln uses -s and not -sf, but the > link is removed two lines above! :-/ I suspect that the previou build > failure was in this directory as well. In OOo build, is the easiest to > remove the sub-project delivery directory and try again (this > unxlngi6.pro dir goes for all OOo subprojects, though I'm not sure about > the target dir on other arches). > > 'rm -rf stlport/unxlngi6.pro' and then try dmake again. If it errors > again, don't try to fix it without recording the error so that we'll be > able to provide better assistance. Maybe 'dmake 2>&1 | tee -a > ../OOo-build-log' would be useful (-a appends). > > -- DJ Lucas > > The post that contained my complete log was help because it was over the message limit even as a tar.gz file. It turns out that the failure had something to do with the partition I was building it on. Here is the entry from /etc/fstab concerning the partition: /dev/hda2 /mnt/hda2 vfat auto,rw,uid=1000,gid=100,umask=022 0 0 The build kept giving me permission errors in building STLport4.5 I also tried this scheme for the partition: /dev/hda2 /mnt/hda2 vfat auto,rw,uid=1000,gid=100,umask=000 0 0 And I received the same errors. I guess it goes to the fact that I still don't have a good enough grasp on partitions, permissions, and umask, so I still have a lot more to learn (never stop learning) What I did to get this to build successfully: I booted into another working BLFS installation and chroot'ed in the new one. I performed the strip command (under "Stripping Again" in LFS) being careful to only strip-debug so I wouldn't mess up my whole install. Rebooted into the new BLFS system Followed the instruction exactly (deviating only in configure options) and waited 5+ hours later, I had a completely new OpenOffice-2.3.1 I am now going to research the permissions, partitions, and umask information so that I can figure out what wen wrong. Thanks again -- http://linuxfromscratch.org/mailman/listinfo/blfs-support FAQ: http://www.linuxfromscratch.org/blfs/faq.html Unsubscribe: See the above information page
Trouble building OpenOffice-2.3.1
7;unxlngi6.pro/misc/build/so_built_so_stlport' ---* tg_merge.mk *--- ERROR: Error 65280 occurred while making /mnt/hda2/OpenOfficeBuild/OOG680_m9/stlport dmake: Error code 1, while making 'build_instsetoo_native' ---* *--- I mistakenly wrote over my first log which showed the detailed information on solenv and boost but I had no errors there. I have searched the internet and previous posts to see if I could find an answer but no such luck. I don't understand why I would get an "Operation not permitted" error on something that is part of a build procedure. Any and all help on solving this would be greatly appreciated. Thank you, rblythe -- http://linuxfromscratch.org/mailman/listinfo/blfs-support FAQ: http://www.linuxfromscratch.org/blfs/faq.html Unsubscribe: See the above information page
Re: Question about compiling gcc (f77 in particular)
Randy McMurchy wrote: > rblythe wrote: > >> I am going to be installing f77 based on the information from the blfs >> wiki on compiling gcc-3.4.6 with c and f77 compilers only. My questions >> are: >> >> 1) Is it possible to only install f77 (--enable-languages=f77) ? >> > > If I remember correctly, you must install the 'C' language > as well as any other. Which means you cannot install just > the fortran language. > > > >> When I built BLFS-SVN a year ago, I think there was something that >> needed f77/g77 and not f95/g95 (I currently have f77 and not f95 on my >> host, but I can't remember why I went through the trouble before). >> >> 2) Can f77 be installed using --prefix=/usr instead of >> --prefix=/opt/gcc-3.4.6 ? >> > > Sure, but you stand a chance or ruining your current GCC > 'C' language installation. Be careful. > > Thanks, I believe I understand now. I will stick with the current method. rblythe -- http://linuxfromscratch.org/mailman/listinfo/blfs-support FAQ: http://www.linuxfromscratch.org/blfs/faq.html Unsubscribe: See the above information page
Question about compiling gcc (f77 in particular)
I am going to be installing f77 based on the information from the blfs wiki on compiling gcc-3.4.6 with c and f77 compilers only. My questions are: 1) Is it possible to only install f77 (--enable-languages=f77) ? When I built BLFS-SVN a year ago, I think there was something that needed f77/g77 and not f95/g95 (I currently have f77 and not f95 on my host, but I can't remember why I went through the trouble before). 2) Can f77 be installed using --prefix=/usr instead of --prefix=/opt/gcc-3.4.6 ? Any help you can provide in clearing this up would be greatly appreciated. Thank you, rblythe -- http://linuxfromscratch.org/mailman/listinfo/blfs-support FAQ: http://www.linuxfromscratch.org/blfs/faq.html Unsubscribe: See the above information page
Re: How to read BLFS in LFS
Dan Nicholson wrote: > On 10/9/07, Walter Barnes <[EMAIL PROTECTED]> wrote: > >>> From: Simon Geard <[EMAIL PROTECTED]> >>> >>> Installing one or two tools like Links isn't exactly a major change to >>> the system - it's a trivial build that'll take a minute or two of your >>> time to perform. Would you really prefer to spend hours writing scripts >>> to do xml transformations, just to avoid installing a simple package? >>> >> That's a very good point. Except that I want to start creating pacman >> packages for everything new I add to the system and as I haven't done this >> before it would require some experimentation. >> > > First thing to note is that what's in LFS isn't supposed to be a > feature complete system. It tries to create a general development > system for you to build on. So, if links needs to be part of your base > install, you shouldn't feel like you're doing something wrong. My > "base install" is considerably larger than what's in LFS since my > barrier for "minimum for usability" is higher. Nothing wrong with > that. > > If you're experimenting with a package manager, it may be best to just > install the needed tools now and just overwrite them once you get your > actual packages in place. Not ideal, but you can always smooth the > edges for the next run. > > As an aside, what's the difference between less and links? I.e., why > is it OK to have less be an unpackaged part of LFS, but links must be > a pacman package? I guess that sort of goes back to the first thing I > said: the borders of LFS are blurry at best. > > -- > Dan > Take a look at this: http://linux.maruhn.com/sec/unhtml.html It is called unhtml. It turns html pages into text. I used it when I built the BLFS system I am using now and I also included it in this hint: http://www.linuxfromscratch.org/hints/downloads/files/copy-paste-blfs-book-commands.txt (shameless plug, I know ;-)) It might get you started. rblythe -- http://linuxfromscratch.org/mailman/listinfo/blfs-support FAQ: http://www.linuxfromscratch.org/blfs/faq.html Unsubscribe: See the above information page
Re: directfb
Barry Loo wrote:
> After doing some research, it seems like directfb would be the easiest
> way to put gimp on on an lfs machine. It also seems that installing
> it isn't easy enough for _me_. Is there any interest among the blfs
> developers to write/update the directfb build instructions? Directfb
> will be very popular among your other users, too, if the step-by-step
> instructions are written. (Note: there's no need to send me that link
> {www.directfb.org/wiki/index.php/Projects:GTK_on_DirectFB} to the wiki
> that describes how to build it; I have already failed pretty good at
> that tutorial).
>
> Loo
> FILM ROCKS!!
>
It would help if you would let us know:
1) What LFS/BLFS version you are attempting to build against?
2) What errors have you received during the build of directFB?
3) Did you have any problems installing the dependencies required for
directFB?
4) Is Frame Buffer support in your kernel?
I am not a BLFS developer, but I have found that giving everyone on the
list more information may just get you the answers you need to build
successfully.
rblythe
--
http://linuxfromscratch.org/mailman/listinfo/blfs-support
FAQ: http://www.linuxfromscratch.org/blfs/faq.html
Unsubscribe: See the above information page
Re: read blfs on a lfs system
Andreas wrote: > Hi to everyone > I have set up a lfs this week and want to go on with parts of the blfs > book. But I did not found any possibility to read blfs on my pure lfs > system. Is there any possibility or perhaps a text only version of blfs? > I have searched the blfs site but I did only found a html and a pdf version. > thanks for your answers. > Andreas > Andreas, You could try this hint: http://www.linuxfromscratch.org/hints/downloads/files/copy-paste-blfs-book-commands.txt located in the hints section of the LFS website. I used to build the system I am using right now. I also wrote the hint. It may help you. Good Luck! rblythe -- http://linuxfromscratch.org/mailman/listinfo/blfs-support FAQ: http://www.linuxfromscratch.org/blfs/faq.html Unsubscribe: See the above information page
Re: domainname - proper for NSS-3.11.3 and for /etc/issue
juras256 wrote: The Book says that most tests of NSS-3.11.3 will fail without defined proper domainname in DUMSUF. I think I have defined proper domainname, but I'm in doubt what does it mean "proper"... The only place I defined something like domainname in my system is /etc/hosts file. 192.168.0.101 kaptah.homelinux.net kaptah and I defined export DOMSUF=homelinux.net Still I have only 670 tests passed and there should be 800. Something must have gone wrong... :-( May be the problem lies somewhere else... At the very beginning of setting up my system I noticed that the login banner defined in /etc/issue shows domainname=(none) when I use \o in it. Does it have anything to do with DOMSUF and how to define domainname for /etc/issue? These are probably two completely different stories... Can someone explain any of them? - Panorama Internetu - prognoza pogody, poczta e-mail z największym załącznikiem, SMS, wyszukiwarki: Gooru, Anonser, serwisy: randki, ogłoszenia, wakacje, program TV, Kina, muzyka, DVD, newsy, inne. http://www.panoramainternetu.pl/ (http://www.epf.pl/) I believe the trouble may be with: 192.168.0.101 kaptah.homelinux.net kaptah ^ Try changing it to 127.0.0.1 kaptah.homelinux.net kaptah 127.0.0.1 is the loop back (home address) of your computer. I had this same problem went I tried to build NSS a few months back. Thanks to many here who helped me through it. See this thread: http://archives.linuxfromscratch.org/mail-archives/blfs-support/2006-June/060076.html for more details of what I went through. Also, when rebuilding NSS, start from a clean build tree. Remove the current NSS directory, untar the package, and start from the beginning command again. Starting clean when rebuilding NSS also was a factor in my success. HTH rblythe -- http://linuxfromscratch.org/mailman/listinfo/blfs-support FAQ: http://www.linuxfromscratch.org/blfs/faq.html Unsubscribe: See the above information page
Re: OpenOffice-2.0.3 configure errors
Luca wrote:
Hi... again !
Well the configuration of OpenOffice took place but in the making it
ends up with:
./configure: line 24672:
/usr/lib/pkgconfig:/usr/X11R7/lib/pkgconfig:/usr/local/lib/pkgconfig:/opt/gnome-1.4/lib/pkgconfig:/opt/gnome-2.14.2/lib/pkgconfig:/opt/gnome/lib/pkgconfig:/opt/kde-3.5.4/lib/pkgconfig:/opt/mono/lib/pkgconfig:/opt/qt-3.3.6/lib/pkgconfig:/opt/qt/lib/pkgconfig:
No such file or directory
Did you add the location of pkgconfig to PKG_CONFIG_PATH as per the BLFS
instructions for this? It appears that it is looking at your
PKG_CONFIG_PATH variable, but it cannot find it. It is not shown as a
dependency in the instructions, but I do believe you need it.
*** Your version of pkg-config is too old. You need version 0.9.0 or newer.
*** See http://www.freedesktop.org/software/pkgconfig
./configure: line 24768:
/usr/lib/pkgconfig:/usr/X11R7/lib/pkgconfig:/usr/local/lib/pkgconfig:/opt/gnome-1.4/lib/pkgconfig:/opt/gnome-2.14.2/lib/pkgconfig:/opt/gnome/lib/pkgconfig:/opt/kde-3.5.4/lib/pkgconfig:/opt/mono/lib/pkgconfig:/opt/qt-3.3.6/lib/pkgconfig:/opt/qt/lib/pkgconfig:
No such file or directory
*** Your version of pkg-config is too old. You need version 0.9.0 or newer.
*** See http://www.freedesktop.org/software/pkgconfig
checking libxml2 no ... checking for libxml2 libraries >= 2.4.2...
checking libxml2 ./libxml2-config ... yes ('')
checking for libxslt libraries >= 1.0.20... no
checking for openssl libraries >= 0.9.6... no
checking for gnutls libraries >= 0.8.1... no
./configure: line 25769:
/usr/lib/pkgconfig:/usr/X11R7/lib/pkgconfig:/usr/local/lib/pkgconfig:/opt/gnome-1.4/lib/pkgconfig:/opt/gnome-2.14.2/lib/pkgconfig:/opt/gnome/lib/pkgconfig:/opt/kde-3.5.4/lib/pkgconfig:/opt/mono/lib/pkgconfig:/opt/qt-3.3.6/lib/pkgconfig:/opt/qt/lib/pkgconfig:
No such file or directory
*** Your version of pkg-config is too old. You need version 0.9.0 or newer.
*** See http://www.freedesktop.org/software/pkgconfig
./configure: line 25866:
/usr/lib/pkgconfig:/usr/X11R7/lib/pkgconfig:/usr/local/lib/pkgconfig:/opt/gnome-1.4/lib/pkgconfig:/opt/gnome-2.14.2/lib/pkgconfig:/opt/gnome/lib/pkgconfig:/opt/kde-3.5.4/lib/pkgconfig:/opt/mono/lib/pkgconfig:/opt/qt-3.3.6/lib/pkgconfig:/opt/qt/lib/pkgconfig:
No such file or directory
*** Your version of pkg-config is too old. You need version 0.9.0 or newer.
*** See http://www.freedesktop.org/software/pkgconfig
checking for nspr libraries >= 4.0... no
checking for nss libraries >= 3.2... no
Do you have NSS installed per the BLFS instructions? It is looking for
it here.
checking for crypto library... configure: error: At least one crypto
library should exist for xmlsec1
dmake: Error code 1, while making
'./unxlngi6.pro/misc/build/so_configured_so_xmlsec1'
'---* tg_merge.mk *---'
ERROR: Error 65280 occurred while making /sources/OOC680_m7/libxmlsec
dmake: Error code 1, while making 'build_instsetoo_native'
'---* *---'
What's happening in here?
Who can help me with this?
Luca
Also, why all of the deviation from what is in the BLFS instructions
concerning configuration? Did you try the basic instructions and they
just not work for you? If they didn't work, where did they fail?
Are you passing the special configuration flags according to the BLFS
instructions for OpenOffice? From the book:
"If you want to optimize the build, edit the appropriate makefile in
solenv/inc/ and add the desired optimization flags to the CFLAGSOPT
variable."
My suggestion is that you check to make sure your dependencies are in
place first. Second, determine if all the optimizations are really
necessary. Third, try again following the instructions on optimizations
and configuring.
Good luck!
rblythe
--
http://linuxfromscratch.org/mailman/listinfo/blfs-support
FAQ: http://www.linuxfromscratch.org/blfs/faq.html
Unsubscribe: See the above information page
Re: Problem to configue my dhcpcd.
nadav vinik wrote: I uncomment these line and set to: DHCP_START="" DHCP_STOP="-k" -- but I get the same error and warning - Unable to process /etc/sysconfig/network-devices/ifconfig.eth0/dhcpcd. Either the SERVICE variable was not set, or the specified service cannot be executed. [ FAIL ] Interface eth0 doesn't exist. On 8/9/06, Dan Nicholson <[EMAIL PROTECTED]> wrote: On 8/9/06, nadav vinik <[EMAIL PROTECTED]> wrote: > > the dhcpd file is: > > ONBOOT="yes" > SERVICE="dhcpcd" > #DHCP_START="" > #DHCP_STOP="-k " > > # Set PRINTIP="yes" to have the script print > # the DHCP assigned IP address > PRINTIP="no" > > # Set PRINTALL="yes" to print the DHCP assigned values for > # IP, SM, DG, and 1st NS. This requires PRINTIP="yes". > PRINTALL="no" > > > I didn't found in the man dhcpcd what to set the DHCP_START and DHCP_STOP. Try `man dhcpcd' and see what options are interesting to you. Generally, though, you can just have DHCP_START="" DHCP_STOP="-k" But you can't comment out those variables. -- Dan -- http://linuxfromscratch.org/mailman/listinfo/blfs-support FAQ: http://www.linuxfromscratch.org/blfs/faq.html Unsubscribe: See the above information page You might have to try (in your dhcpcd file) - make a back-up first: ONBOOT="yes" SERVICE="dhcpcd" DHCP_START="eth0 -t 20" # See man dhcpcd for definition of -t and the reason for eth0 spec DHCP_STOP="-k " # Set PRINTIP="yes" to have the script print # the DHCP assigned IP address PRINTIP="no" # Set PRINTALL="yes" to print the DHCP assigned values for # IP, SM, DG, and 1st NS. This requires PRINTIP="yes" PRINTALL="no" And for your ipv4 file: (make a back-up first) place comment (#) symbol in front of everything except SERVICE. change SERVICE variable to SERVICE="dhcpcd" I know this appears to be redundant, but even though I use a wireless connection (my file has ra0 instead of eth0), it connects without problem every time. Also, what does your /etc/resolv.conf look like? If dhcpcd is working correctly, it should have created a back-up named /etc/resolv.conf-eth0.sv or something similar (also in the man page). Last thing: Was you eth0 working before you tried dhcpcd? Did you name it something different when you set it up in LFS (using the udev rules as outlined in the LFS instructions)? If you named it something else, then technically eth0 no longer exists and the new name of your interface does. You would then need to change your information accordingly when setting up dhcpcd in BLFS. rblythe -- http://linuxfromscratch.org/mailman/listinfo/blfs-support FAQ: http://www.linuxfromscratch.org/blfs/faq.html Unsubscribe: See the above information page
Re: What happened to my /usr/bin/run-mozilla.sh ?
Ken Moffat wrote: Firefox does it by running the 'firefox' script (resolving the symlinks). You seem to have something missing from your find command, for me it prints out every file on the system followed by an error message for the second parameter. I'd use find / -xdev -name 'run*.sh*' 2>/dev/null [ -name is what you seem to be missing ] I'd expect it to be in /usr/lib/firefox (or /usr/lib/firefox-1.X.Y depending on how you installed firefox). Ken Thanks. Firefox is in /usr/lib/firefox-1.5.0.3/run-mozilla.sh Thunderbird is in /usr/ib/thunderbird-1.5.0.2/run-mozilla.sh This is a real relief. -- http://linuxfromscratch.org/mailman/listinfo/blfs-support FAQ: http://www.linuxfromscratch.org/blfs/faq.html Unsubscribe: See the above information page
What happened to my /usr/bin/run-mozilla.sh ?
I just installed Nvu (N-view) the HTML editor from http:/www.nvu.com/index.php. After it was installed and I restarted my system, I could not access Firefox or Thunderbird. It appears that Nvu installed run-mozilla.sh OVER the existing file that handled starting Firefox & Thunderbird. I immediately (as root) renamed this newly installed run-mozilla.sh to run-mozilla.sh.old. I then changed the nvu script to reference this new file and it seems to be working. Once I renamed the file, I was able to call firefox and thunderbird from both my gui menu and the command line. I also looked into the firefox script (/usr/bin/firefox - initially I did not know this was a script) and I found these two comments: # Use run-mozilla.sh in the current dir if it exists # If not, then start resolving symlinks until we find run-mozilla.sh My problem is, when I used the find command (find / run-mozilla.sh -- searching everything) the system could not find it. How does Firefox find it (and Thunderbird for that matter)? This is only an issue in case in the future I can't access the packages. As I said, right now everything "seems" to be working correctly. I have completely restarted (not cold boot) my system twice since I did this and I am e-mailing you from Thunderbird right now. I am just curious. BTW, I checked on the Nvu website and they mentioned that I needed a newer version of Mozilla (>=1.7.8) in order for run-mozilla.sh to function without modification, so I am baffled (and thankful) that my system "just works". Any insight into this would be greatly appreciated. Thanks, rblythe -- http://linuxfromscratch.org/mailman/listinfo/blfs-support FAQ: http://www.linuxfromscratch.org/blfs/faq.html Unsubscribe: See the above information page
Re: Starting hplip before CUPS Server? (An update)
DJ Lucas wrote:
rblythe wrote:
I have managed to install hplip and it's dependencies and I set the
printer up with CUPS and printed a test page.
Kernel = 2.6.12.5
Build = BLFS-SVN (20060411) when I started
CUPS is installed and functioning (used BLFS bootscript)
HPLIP-0.9.10 with dependencies listed at hplip.sourceforge.net installed
I read somewhere (can't remember the link) that hplip must be started
before cups.
I have looked at the /etc/inittab that was created during LFS and found
this line:
id:3:initdefault: so if I understand this correctly, my system boots
into rin level 3 (multi-user mode with networking)
Looking at /etc/rc.d/rc3.d (command ls -l) I see (among other items)
S25cups -> ../init.d/cups
I managed to find the script supplied with the hplip-0.9.10 sources
(hplip.sh) and I copied (not moved in case of error on my part):
cp hplip.sh /etc/rc.d/init.d/hplip (done as root) and made it
executable: chmod +x hplip
Now here is where I am stuck:
Do I only need to created this one link in rc3.d: ln -s ../init.d/hplip
S23hplip (I am using S23 because it is before S25)
- or - do I need to create simlinks in the other runlevels as well?
I don't plan on installing the desktop environment until then end but
when I do that and change the /etc/inittab to "5" instead of "3" am I
right in thinking that I will have to create ln -s ../init.d/hplip
S23hplip in rc5.d as well?
My reason for asking is that I have viewed each one of the links in the
various rc?.d directories and I noticed many of them appear more than
once and I just want to make sure of doing this correctly. Also, if I
have and "S" for start to I also need a corresponding "K" for kill (stop)?
Thanks,
rblythe
Ideally, you should have S23hplip links in rc{3,4,5}.d/ to start in
multiuser, multiuser-test, and graphical user modes reps. You should
also have K35hplip links in rc{0,1,6}.d/ to stop on shutdown, single
user mode, and reboot repsectively. Runlevel 2 is questionable. You
should have one or the other but it depends on weather the service is
dependent on the network being up. RL2 is no network. I don't know
much about hplip. My question is, do you need it? Personally I need
only the IJS portion of it, so that is all that is built and installed.
-- DJ Lucas
I finally managed to get this to work using the hplip script that comes
with the package. My system was missing /var/lock/subsys, so I simply
created the directory using (as root): mkdir /var/lock/subsys. Now
thinking about it, I could have used install -d /var/lock/subsys as
well. Anyway, it is the functions start() and stop() inside the hplip
script that are looking for this directory to exist otherwise it returns
a value of 1 and causes the print_error_msg() function inside of
$rc_base/init.d/functions to be triggered.
I have not started on trying to use the $rc_base/init.d/functions way of
doing things in starting hplip yet so I don't get the cool colors that
are generated [OK], [WARN], [FAIL] etc. when this script is invoked, but
instead I get the message from inside the hplip script. That is another
project for another time.
Sorry for not truncating this post, but my hope is that someone else who
used BLFS and has an HP Printer (Multifunction type) that utilized HPLIP
will be able to use this.
HTH
rblythe
--
http://linuxfromscratch.org/mailman/listinfo/blfs-support
FAQ: http://www.linuxfromscratch.org/blfs/faq.html
Unsubscribe: See the above information page
Re: I need some help with Thunderbird please [SOLVED!!]
Randy McMurchy wrote: rblythe wrote these words on 07/02/06 20:49 CST: Randy McMurchy wrote: Dedicate some time to *learning*, this is how you will solve your problems. Cool. I will do it. That attitude alone is enough. You will succeed. You will find the issue that is holding you back. If you sense progress but still have not achieved your goals, then ask the group why (whatever you did that you thought would work and didn't) and we will help to the best of our ability. It is a two-way street. If we sense you are doing what it takes to help yourself, we will bend over backwards to help. I am not the only one with this attitude. There are many here that can offer years of experience and wisdom to help folks like you. Just show an effort that you've tried to research on your own. First, I would like to day Thank You to Randy and Dan for your help in getting me to understand this for myself. I am e-mailing you from my BLFS that is sharing the same e-mail folder with Mepis, Suse, and Windows XP. In doing research, I found this site: http://home.planet.nl/~elst0093/motub/multboot.html After reading that thoroughly, this is what I did to accomplish my goal (hope this helps someone else): 1) Booted into the distro that I created the partition (/dev/hda2) and folder (/mnt/hda2/Thunderbird_Mail) with. 2) typed this command: ls -l /mnt/hda2 This provided me with the output of user id (uid) and group id (gid) of all the directories on the partition 3) wrote down the numbers (uid=1000, gid=100). Sorry for the redundancy, but this is important 4) Rebooted the machine into the BLFS system 5) checked my (rblythe) uid in /etc/passwd and checked my gid in /etc/group. Because the uid and gid in BLFS was the same (1000,100 respectively) I could easily do step number 6. (note: I had to research where I could find this on my new system) 6) edited /etc/fstab (I used vi, but any text editor will do the job). Must be logged in as root for this step. The new line for this entry in /etc/fstab is as follows: /dev/hda2 /mnt/hda2 vfat auto,rw,uid=1000,gid=100,umask=000 0 0 It is vfat because it was originally created to use with Windows. If you are not expecting to access your e-mail from a Windows system, then make sure you use the file type (ext3 ,ext2, reiserfs, etc.) that you used to create the partition originally. auto is for mounting automatically at boot-up. rw for read/write. Setting uid and gid is so I (rblythe) own the partition while booted into BLFS (this is critical) 7) Reboot into BLFS 8) Open Thunderbird. Edit account settings (Server Settings and Local Folders) to point to the mail on /mnt/hda2/Thunderbird_Mail. Save and exit Thunderbird completely) 9) Open Thunderbird again. Now I can see all of my e-mails, sub-folders, filters, etc. It is more simple for me to do than to type out these instructions. I am not going into great detail with this post because the reference web-site is very detailed. There are also details on doing this if your uid and gid on BLFS (or any distro) is different than what is on the distro that created the partitions/files/etc. in the first place. As I mentioned before, thanks for letting me find this out on my own. Now I will never forget it. There is so much I still have to learn. rblythe -- http://linuxfromscratch.org/mailman/listinfo/blfs-support FAQ: http://www.linuxfromscratch.org/blfs/faq.html Unsubscribe: See the above information page
Re: I need some help with Thunderbird please
Randy McMurchy wrote: Dan Nicholson wrote these words on 07/02/06 20:18 CST: man mount, man fstab Read, learn, absorb, google, exhaust every channel of knowledge. You would learn so much if you did this stuff on your own. You have the research skills, Randy, I sense that. Simply find some stuff on the internet that is relevant to your issue (not Thunderbird specific, but overall Unix disk access permissions issues) and learn everything there is to know about it. Then, move to the next issue that you're not 100% confident in. Study it, learn, become an expert. And so forth and so on. Simply writing in to a support list hoping for a shot in the dark that will fix your problem is no way to go about *learning*. I mean this in the most positive way. You have it in you. Dedicate some time to *learning*, this is how you will solve your problems. Cool. I will do it. I have changed /etc/fstab 3 times now ( an rebooted each time) and I still get a Permission denied message. I have also added myself to the disk group with no luck. I will keep searching, reading, and learning. I know there is an answer to this and it is probably staring me in the face. I just can't see it yet. My main confusion about permissions is that when I am in my host distro, I can mkdir on mnt/hda2 as rblythe with no problem. Well, back to the drawing board. I will let you know when I figure this out. Thanks, rblythe -- http://linuxfromscratch.org/mailman/listinfo/blfs-support FAQ: http://www.linuxfromscratch.org/blfs/faq.html Unsubscribe: See the above information page
Re: I need some help with Thunderbird please
Dan Nicholson wrote:
On 7/2/06, rblythe <[EMAIL PROTECTED]> wrote:
When I try this, I get a message when I restart Thunderbird that says an
instance of Thunderbird is already running and that I should close that
one and try again. I then completely reboot the system and try again
but I get the same message.
1. Try running thunderbird from a terminal so you can see the output
from the shell script. If that doesn't help, edit
/usr/bin/thunderbird and add "set -ex" somewhere near the top.
2. Try running "thunderbird -ProfileManager". This might help the the
profile get rebuilt.
3. Are there any running instances of thunderbird?
I tried changing profiles.ini as a normal user and as root and I get the
same results.
I just tried it on my machine by moving my profile to /tmp and it worked.
I still have access to all of my e-mails (old and new) and I think I am
just going to start over with a NEW shared e-mail storage area and try
again. If I figure this out, I will definitely post my results.
I wouldn't do that just yet. Maybe the issue has nothing to do with
the profile. I've seen this error on Mozilla products on *nix a
million times. You can usually fix them.
--
Dan
On a whim, I tried to create a directory on my separate partition as a
normal user and I received a Permission denied error.
The entry in /etc/fstab for this partition:
/dev/hda2 /mnt/hda2 vfat,ext3,ext2 defaults 0 0
I am wondering if this all boils down to a permission issue. Also
looking at /etc/group I can see that I (rblythe) am not part of the disk
group, but accoding to /etc/udev/rules.d/25-lfs.rules, the hd[a - z]
items belong to the "disk" group. I can make directories on /dev/hda12
(where {B}LFS is) without problem. I am not sure what needs to be
defined in order for me to mkdir on /mnt/hda2.
Also, I am running Thunderbird from a terminal. I do not have a window
manager or desktop environment installed yet. There is no output in the
terminal when I am running Thunderbird. Also, when I run thunderbird
-profilemanager, I get a menu showing me the only created user (default)
and options to create, rename, or delete the user. If I click on start
Thunderbird, it does the obvious and starts Thunderbird.
And one more thing, looking at the /etc/fstab from my host, the entry
for the drive looks like this:
/dev/hda2 /mnt/hda2 vfat,ext3,ext2 noauto,users,exec 0 0
The only problem with that is that I have to manually mount the drive
before starting Thunderbird, and I am trying to avoid that with my own
build.
Thanks again for the help.
rblythe
--
http://linuxfromscratch.org/mailman/listinfo/blfs-support
FAQ: http://www.linuxfromscratch.org/blfs/faq.html
Unsubscribe: See the above information page
Re: I need some help with Thunderbird please
Dan Nicholson wrote: On 7/2/06, rblythe <[EMAIL PROTECTED]> wrote: I googled around a bit and found where someone had posted that doing this kind of thing as a root user will make the Thunderbird package from the system I attempted this on (not BLFS specific, just in general) would be inoperable. I was able to access all of the e-mails as the root user, but not as the normal user. The only reason I was root user in the first place was per the BLFS instruction to run /user/bin/thunderbird as root "to create additional files in the usr/lib/thunderbird-1.5.0.2 directory. I'm not really sure what the problem is as you've described it. But if you just want to point your profile somewhere else, you can try this. $ cat ~/.thunderbird/profiles.ini [General] StartWithLastProfile=1 [Profile0] Name=default IsRelative=1 Path=1elsrg7x.default Try changing [Profile0] to "IsRelative=0" and "Path" to the location of the shared profile. As long as your user has write privelages, then it should work. Actually, I haven't tried this, but that's what seems like the correct thing to do. -- Dan When I try this, I get a message when I restart Thunderbird that says an instance of Thunderbird is already running and that I should close that one and try again. I then completely reboot the system and try again but I get the same message. I tried changing profiles.ini as a normal user and as root and I get the same results. I still have access to all of my e-mails (old and new) and I think I am just going to start over with a NEW shared e-mail storage area and try again. If I figure this out, I will definitely post my results. Thanks, rblythe -- http://linuxfromscratch.org/mailman/listinfo/blfs-support FAQ: http://www.linuxfromscratch.org/blfs/faq.html Unsubscribe: See the above information page
I need some help with Thuderbird please
using BLFS-SVN Kernel = 2.6.12.5 package = Thunderbird Installation status: OK Currently, I am able to share my one (1) e-mail account of 3 different boot systems (Mepis, Suse 10, Windows XP). I attempted to do this with BLFS-SVN, but I made some type of error. I googled around a bit and found where someone had posted that doing this kind of thing as a root user will make the Thunderbird package from the system I attempted this on (not BLFS specific, just in general) would be inoperable. I was able to access all of the e-mails as the root user, but not as the normal user. The only reason I was root user in the first place was per the BLFS instruction to run /user/bin/thunderbird as root "to create additional files in the usr/lib/thunderbird-1.5.0.2 directory. I appears that I went to far in trying to set my system up to share one e-mail account while logged in as root. Can this be fixed? I really don't want to have a separate e-mail location for my BLFS system if I can avoid it. All of my e-mails are in tact (they are located on a separate partition that is formatted vfat (for Windows access) so that is not the issue. I also backed them up BEFORE I tried this in case I made some error and accidentally deleted them. Any help in this matter will be greatly appreciated. Thanks, rblythe -- http://linuxfromscratch.org/mailman/listinfo/blfs-support FAQ: http://www.linuxfromscratch.org/blfs/faq.html Unsubscribe: See the above information page
MPlayer woes
This is exactly what happens when I try to use mplayer (gmplayer)
[EMAIL PROTECTED] gmplayer
MPlayer 1.0pre7try2-4.0.3 (C) 2000-2005 MPlayer Team
CPU: Advanced Micro Devices Athlon MP/XP/XP-M Barton (Family: 6,
Stepping: 0)
Detected cache-line size is 64 bytes
CPUflags: MMX: 1 MMX2: 1 3DNow: 1 3DNow2: 1 SSE: 1 SSE2: 0
Compiled for x86 CPU with extensions: MMX MMX2 3DNow 3DNowEx SSE
vo: X11 running at 1024x768 with depth 24 and 32 bpp (":0.0" => local
display)
85 audio & 196 video codecs
Playing /dev/DVD.
MPEG-PS file format detected.
MPEG: No audio stream found -> no sound.
VIDEO: MPEG2 720x480 (aspect 2) 29.970 fps 9800.0 kbps (1225.0
kbyte/s)
==
Trying to force video codec driver family libdv...
Opening video decoder: [mpegpes] MPEG 1/2 Video passthrough
VDec: vo config request - 720 x 480 (preferred csp: Mpeg PES)
Could not find matching colorspace - retrying with -vf scale...
Opening video filter: [scale]
The selected video_out device is incompatible with this codec.
VDecoder init failed :(
Opening video decoder: [libmpeg2] MPEG 1/2 Video decoder
libmpeg2-v0.4.0b
Selected video codec: [mpeg12] vfm:libmpeg2 (MPEG-1 or 2 (libmpeg2))
==
Audio: no sound
Starting playback...
VDec: vo config request - 720 x 480 (preferred csp: Planar YV12)
VDec: using Planar YV12 as output csp (no 0)
Movie-Aspect is 1.33:1 - prescaling to correct movie aspect.
VO: [xv] 720x480 => 720x540 Planar YV12
VDec: vo config request - 176 x 112 (preferred csp: Planar YV12)
VDec: using Planar YV12 as output csp (no 0)
Movie-Aspect is 1.33:1 - prescaling to correct movie aspect.
VO: [xv] 720x480 => 720x540 Planar YV12
V: 0.0 3/ 3 ??% ??% ??,?% 0 0
Exiting... (Exit)
---
The gui interface works just fine. It just won't play a movie. I
followed the instruction from BLFS-SVN to build & compile MPlayer but no
joy. I did not create the udev rule because I have a rule of my own for
this device:
BUS=="ide", KERNEL=="hdc",SYSFS{dev}=="22:0",SYMLINK="DVD", MODE="0777"
which allows XMMS to play my CD's (I was having trouble with that
before).
Now I am not sure what to do next. From the looks of it, I am having
some sort of trouble with the codecs (I am just guessing). I have the
following codecs installed (per BLFS-SVN instructions):
essential-20060501
all-20060501
win32codecs-20040703
I installed them in the order listed above. Is it possible that these
conflict in some way?
All help here would be greatly appreciated.
Thanks,
rblythe
--
http://linuxfromscratch.org/mailman/listinfo/blfs-support
FAQ: http://www.linuxfromscratch.org/blfs/faq.html
Unsubscribe: See the above information page
Trouble with Hplip
I found out that I need hplip in order for the system to see the scanner portion of my HP 1410v PSC. The shell script (called hplip) that comes with the source code works. When I moved to /etc/rc.d/init.d and created a symlink (ln -s ../init.d/hplip S23hplip) in rc3.d I get an error stating that I should not be reading an error and it points out that the error comes from S23hplip (sorry about not having the exact syntax of the error, I did not write it down and I am e-mailing from my host system). After viewing the hplip script, I realized that it does not look like the blfs bootscripts at all. I believe this is my problem. I have never written scripts like these before, but my thinking is this: I am going to use the cups bootscript as a template to create a blfs compatible hplip bootscript. My questions are: Has anyone had success with this before? Is there another method that someone has done to get the result of having hplip start before CUPS (it must start before cups)? Currently, after I boot the system, I have to issue these two commands as root: /etc/rc.d/init.d/hplip start # this starts hsiod and hpssd /etc/rc.d/init.d/cups restart # this makes cups aware of hplip Printing an scanning then function as normal. Any input on this would be greatly appreciated. rblythe -- http://linuxfromscratch.org/mailman/listinfo/blfs-support FAQ: http://www.linuxfromscratch.org/blfs/faq.html Unsubscribe: See the above information page
ALSA - can only use as root - How do I fix this?
I have ALSA installed and as root I am able to speaker-test and play default .wav files located in /user/share/sounds/alsa. When I try to do any of this as a normal user, I get a lot of errors (sorry, the error list was too long to write down by hand and I am e-mailing from my host distro). I visited http://www.alsa-project.org and they do suggest changing the permissions on the following like this: chmod a+rw /dev/dsp /dev/mixer /dev/sequencer /dev/midi I have not done this yet because I am very cautious about doing things in LFS/BLFS without confiming them first. Is the above command safe to do? Will it allow me to use ALSA as a normal user? Is there another method to get the same results? Thanks for the help rblythe -- http://linuxfromscratch.org/mailman/listinfo/blfs-support FAQ: http://www.linuxfromscratch.org/blfs/faq.html Unsubscribe: See the above information page
I am confused about udev, xmms, and playing cd's
Here's my info:
BLFS-SVN
KERNEL = 2.6.12.5
contents of /etc/udev/rules.d/10-blytheworks.rules
BUS=="ide", KERNEL=="hdc", SYSFS{dev}=="22:0", SYMLINK+="cdrom-%k",
GROUP=="cdrom"
BUS=="ide", KERNEL=="hdd", SYSFS{dev}=="22:64", SYMLINK+="cdrom-%k",
GROUP=="cdrom"
On my system hdc = DVD Drive, hdd = DVDRW Drive
I have recently install XMMS folowing the BLFS instructions. Everything
seems peachy. However, when I want to try to play an audio cd, I can't
pull up a playlist from anywhere. The SYMLINKs appear to work properly
(when I look at them in /dev, they show /dev/cdrom-hdc --> /dev/hdc and
/dev/cdrom-hdd --> /dev/hdd.
Do I need to create a mount point in /etc/fstab for my two devices? I
am confused because on my host distro, there are no moint points for
optical devices and I am able to play cd's. I am only trying to play
cd's (not dvd's) at this point because I haven't installed any other
media players yet.
How do I tell XMMS that /dev/cdrom-hdc actually contains the playlist
(audio cd)? What do I need to fix in my udev rules to get this working
correctly?
I am stumped on this one.
Thanks for the help.
rblythe
--
http://linuxfromscratch.org/mailman/listinfo/blfs-support
FAQ: http://www.linuxfromscratch.org/blfs/faq.html
Unsubscribe: See the above information page
Re: Lost mouse after kernel re-build -- Not sure what I did wrong
Andrey Voropaev wrote: Looks like you are getting wrong driver for your mouse. Which device uses GPM to read mouse events? /dev/input/mice? or /dev/input/event0 ? What type of device GPM expects it to be (this should be in /etc/sysconfig/mouse I believe) I think using udevinfo you can see which real driver sits behind the device your GMP is using. Also, very possible, you have now 2 drivers that pass information thru the same device for your mouse (multiplexing) I had it once on my laptop, with symilar symptoms. All of the above are just guesses. Without knowing exact configuration it's hard to say anything specific. I solved this problem. When I first compiled GPM, I had set my mouse to /dev/psaux and ps2. When I did the kernel recompile, something was changed (I still don't know what) and I needed to change my settings to /dev/mouse and ExplorerPS/2. Now everything is working great. Thanks for your reply. rblythe -- http://linuxfromscratch.org/mailman/listinfo/blfs-support FAQ: http://www.linuxfromscratch.org/blfs/faq.html Unsubscribe: See the above information page
Lost mouse after kernel re-build -- Not sure what I did wrong
Specs: Version: BLFS-SVN Kernel: 2.6.12.5 Mouse: Logitech Marble Mouse (Big Red Marble, 2 large mouse buttons, 2 smaller mouse buttons) I rebuilt the kernel (I followed the LFS instructions) because I was trying to get sound to work int this build (separate issue, another time). Anyway I had botched that whole thing, and mistakenly deleted both the config-2.6.12.5 and the lfskernel-2.6.12.5 files forcing me to do a complete rebuild of the kernel from scratch. After rebuilding the kernel, I lost use of my mouse. Actually, the little box pointer does appear and I can move it with the trackball, but then it automatically start to copy/paste without me clicking the mouse buttons. The screen also flashes multi-colors and then I get this message: psmouse.c: Explorer Mouse at isa0060/seri1/input0 lost sychronization, throwing 2 bytes away What I have done so far to try to fix this (none of it worked): Recompiled the kernel 4 times choosing different combinations of mouse support Recompiled gpm per the BLFS instructions (only once) I don't know how I messed this up, and I don't know how to fix it. I am looking at my host distro config file now, but if someone here could help me out, I would appreciate it. Thanks, rblythe -- http://linuxfromscratch.org/mailman/listinfo/blfs-support FAQ: http://www.linuxfromscratch.org/blfs/faq.html Unsubscribe: See the above information page
Re: Starting hplip before CUPS Server?
DJ Lucas wrote:
< snip >
Ideally, you should have S23hplip links in rc{3,4,5}.d/ to start in
multiuser, multiuser-test, and graphical user modes reps. You should
also have K35hplip links in rc{0,1,6}.d/ to stop on shutdown, single
user mode, and reboot repsectively. Runlevel 2 is questionable. You
should have one or the other but it depends on weather the service is
dependent on the network being up. RL2 is no network. I don't know
much about hplip. My question is, do you need it? Personally I need
only the IJS portion of it, so that is all that is built and installed.
-- DJ Lucas
DJ,
Thanks for the info. As for needing it, I am only working from what I
see on my host disto and from the hplip.sourceforge.net website. I have
seen where others only installed the IJS portion of it, but I decided
against that route. I will research it further for future builds but
now that it is on the system I'll work with it.
Thanks again,
rblythe
--
http://linuxfromscratch.org/mailman/listinfo/blfs-support
FAQ: http://www.linuxfromscratch.org/blfs/faq.html
Unsubscribe: See the above information page
Starting hplip before CUPS Server?
I have managed to install hplip and it's dependencies and I set the printer up with CUPS and printed a test page. Kernel = 2.6.12.5 Build = BLFS-SVN (20060411) when I started CUPS is installed and functioning (used BLFS bootscript) HPLIP-0.9.10 with dependencies listed at hplip.sourceforge.net installed I read somewhere (can't remember the link) that hplip must be started before cups. I have looked at the /etc/inittab that was created during LFS and found this line: id:3:initdefault: so if I understand this correctly, my system boots into rin level 3 (multi-user mode with networking) Looking at /etc/rc.d/rc3.d (command ls -l) I see (among other items) S25cups -> ../init.d/cups I managed to find the script supplied with the hplip-0.9.10 sources (hplip.sh) and I copied (not moved in case of error on my part): cp hplip.sh /etc/rc.d/init.d/hplip (done as root) and made it executable: chmod +x hplip Now here is where I am stuck: Do I only need to created this one link in rc3.d: ln -s ../init.d/hplip S23hplip (I am using S23 because it is before S25) - or - do I need to create simlinks in the other runlevels as well? I don't plan on installing the desktop environment until then end but when I do that and change the /etc/inittab to "5" instead of "3" am I right in thinking that I will have to create ln -s ../init.d/hplip S23hplip in rc5.d as well? My reason for asking is that I have viewed each one of the links in the various rc?.d directories and I noticed many of them appear more than once and I just want to make sure of doing this correctly. Also, if I have and "S" for start to I also need a corresponding "K" for kill (stop)? Thanks, rblythe -- http://linuxfromscratch.org/mailman/listinfo/blfs-support FAQ: http://www.linuxfromscratch.org/blfs/faq.html Unsubscribe: See the above information page
Re: Postfix, Thunderbird and separate shared e-mail partition
Ken Moffat wrote: On Wed, Jun 07, 2006 at 07:42:10PM -0400, rblythe wrote: Anyway, I have not installed these two packages yet (working my way up to it), but I want to know if installing Postfix may wreck my current e-mail set-up. Undoubtedly it _may_ wreck your current mail setup. You haven't really given enough detail for me to understand how your current setup works, but do you actually need postfix ? I've never used thunderbird, but if you are sending mail through comcast's mail server, I imagine thunderbird doesn't need a local mail server. Running postfix is very convenient, particularly if you have multiple machines, but for a single box without a static IP or its own proper domain it may be unnecessary. I run postfix on my desktop boxes and 'server' so that the output of cron jobs (backups, disk checks, etc) all ends up in my mailbox on the server. But as far as outgoing mail is concerned, mine now all goes through my isp's mail server. And I get external mail by pulling it from my isp's POP server with fetchmail. Ken Sorry fo being so vague. This is really only an experiment to increase my understanding. I will try to make some sense this time: LFS/BLFS build (Blytheworks) - No e-mail is currently set-up. I have not got that far in the compilation process yet (this is a learn as I go proposition). I was reading the BLFS web pages on the various type of servers and thought postfix might be cool to use once I did a little more research and tried to understand it better. Because my host distro is all gui based, I still consider myself a newbie when it comes to compiling software from scratch and providing the necessary configuration info. This was basically a thought to do something different then what I am used to, and try to learn something along the way. Host System Setup (Mepis) - I use Thunderbird to get my e-mail from Comcast without a problem. It is stored on a separate partition that I can also access from Suse 10.0 and Windows Xp (my machine is a multi-boot set-up). In the year I have been running this way, I have not had any problems. Mepis is Debian based (I have the non-Ubuntu version) and it is okay when Debian is not changing the repositories. Because of that, I was driven once again to try and successfully build a LFS/BLFS system from scratch. So far I have been successful and I understand what is going on finally. I want to make the complete transition if my own LFS/BLFS system for the freedom I believe that I have because of it. I think my desire may be true overkill for my situation where Blytheworks is concerned. What I would like to try is this (please provide alternate solutions if I am over-doing it): I am really interested in the various servers and security applications right now. For the mail, I wanted to install Postfix and configure it so that it would get the e-mail from the Comcast server and download it to a separate folder on my vfat partition. I woud then configure Thunderbird to get the e-mail from the Postfix downloaded area so that I could read, reply and compose new e-mails. My hope is that while using Thunderbird, my e-mail would then be stored on the original Thunderbird vfat partition so that if by chance I am booted into one of the other systems, I can still read, reply, and compose e-mail messages without messing something up. My other reason for wanting Postfix (or any MTA) is that right now, if I want to compose and e-mail, I have to boot back into a system with and e-mail program on it instead of staying in my build environment. I noticed that one of the optional packages where Lynx (my web browser of choice until I build Firefox) is an MTA. My thinking is that with an MTA, if I run into a compile problem or get confused about something while learning more about BLFS, I could just compose and e-mail to blfs-support and go on to something else until I receive a reply. My hope also is that if this can be done, then I could download the e-mail responses as outlined above and keep the integrity of my existing Thunderbird e-mail box on a separate partition. Ultimately, because this is a single user machine, I may not really need any kind of server. I am only looking at this as a possible solution, but again even though I have used Linux for close to two years, I am as green as they come with this type of things (having a do-it-all distro is not as beneficial as one might think, especially with a tinkerer like me). Plus in my research about Postfix, I have seen some interesting things (spamassassin, virus protection) that I would like to learn how to use as well. If you have any other possible solutions please let me know. I am not in a hurry. I am taking my time with this one. Thanks, rblythe -- http://linuxfromscratch.org/mailman/listinfo/blfs-support FAQ: http://www.linuxfromscratch.org/blfs/faq.html Unsubscribe: See the above information page
Postfix, Thunderbird and separate shared e-mail partition
I am venturing into the deep unknown with this one. I am going to be installing postfix and ultimately Thunderbird. I am doing this so I can understand how mail servers work and I like Thunderbird. My situation is this: Currently I multi-boot with Mepis, Suse, LFS/BLFS (working on now), and Windows XP (yeah I know, you don't have to say it). LFS/BLFS does not have any type of e-mail function yet (have not installed, configured, etc.). With the other distros, I have a shared fat32 partition that I use so that I can retrieve/read my e-mail from any booted situation. What I would like to do is preserve this set-up (until I can abandon the others for my own build) and I want to utilize Postfix and Thunderbird on the LFS/BLFS (I have named 'Blytheworks'). Anyway, I have not installed these two packages yet (working my way up to it), but I want to know if installing Postfix may wreck my current e-mail set-up. I promise to read the man pages and do my research. Also, I will not auto-start the mail server until I understand this more. My hope is that someone in the community who uses Postfix may shed some light on what I am going to do. I know I am fishing, but this is completely new and unknown territory for me. I will also make a back-up of my current e-mail file(s) in case I botch the whole thing. Any insights you may have will be greatly appreciated. Thanks, rblythe -- http://linuxfromscratch.org/mailman/listinfo/blfs-support FAQ: http://www.linuxfromscratch.org/blfs/faq.html Unsubscribe: See the above information page
Re: Problem with tests in NSS-3.11
Dan Nicholson wrote: as a valid address. Although, I'm confused about how a domain name is registered on a LAN. I think that's configured in the DHCP server (the router). If you have net-tools installed, what is the output of `dnsdomainname'? -- Dan Dan, Yes I am behind a router. when I issue the dnsdomainname command the output is: localdomain I never did put much thought on how this name is registered on a LAN. I guess now I have more to research and learn about. -- http://linuxfromscratch.org/mailman/listinfo/blfs-support FAQ: http://www.linuxfromscratch.org/blfs/faq.html Unsubscribe: See the above information page
Questions about openLDAP & Cyrus-SASL
This e-mail is for my learning purposes only. I have not installed openLDAP or Cyrus-SASL, but I would like to. I have been reading everything I could find on the Internet about these two packages, but I can't figure out if I need them for my situation: My computer is used only by me. I connect wirelessly to the internet via my Linksys router There are two other computers connected to the router (Windows machines), but I don't serve anything to them and I do not access them via the Linksys router (no samba, etc) This LFS/BLFS build I have decided is more for my learning than anything else. Could someone give me a couple of examples of what a user like myself could use these packages for? Please also keep in mind that I don't think learning about these packages is a waste of time, I am just looking for some direction. Thanks, rblythe -- http://linuxfromscratch.org/mailman/listinfo/blfs-support FAQ: http://www.linuxfromscratch.org/blfs/faq.html Unsubscribe: See the above information page
Re: Problem with tests in NSS-3.11
Randy McMurchy wrote: Dan Nicholson wrote these words on 06/04/06 09:26 CST: Is mcmurchy.prv an actual domain name? No, in the real world. Yes, in the 192.168 network behind a couple of Linksys/Dlink routers. The first router doing address translation from the ip address assigned by the broadband vendor to a 192.168 private network. I fixed it, thanks to Randy (we share the same first name). This is my old /etc/hosts: 127.0.0.1 localhost 192.168.1.1 blytheworks.example.org blytheworks It was my 'best guesss' from creating the /etc/hosts file in LFS (it appears I guessed wrong) This is my new /etc/hosts 127.0.0.1 localhost.localdomain localhost blytheworks blytheworks.rblythe.prv There is no 192 etc. line. This is actually a combination of what Randy McMurchy wrote in a previous post and what I read from my host distribution (Mepis) /etc/host file. I am only starting learn about this kind of stuff (using one distro of linux for almost 2 years but nothing like having to do so much CLI work. I like LFS/BLFS better because I get to learn so much). Anyway, I think my problem has to do with the 192.168.1.1 from my old /etc/hosts. I don't see where that IP is ever relevant in my system. I use dhcpcd and when I run ifconfig, the output is inet addr: 192.168.1.101 (has also been 102, 103, 104), bcast: 192.168.1.255, netmask: 255.255.255.0. I could be way off about this, so please feel free to comment and help me add to my knowledge and understanding. After I created the new /etc/hosts files, the testing output was 770 just like the instructions say it should be. BTW, I did do a complete configure/build/install from scratch by completely removing nss-3.11 and starting from step one. This is a long post simple to say thank you. rblythe -- http://linuxfromscratch.org/mailman/listinfo/blfs-support FAQ: http://www.linuxfromscratch.org/blfs/faq.html Unsubscribe: See the above information page
Re: Problem with tests in NSS-3.11
Dan Nicholson wrote: On 6/3/06, rblythe <[EMAIL PROTECTED]> wrote: Anyway, could this have something to do with openSSL? It is installed per the BLFS instructions, but the version I installed was openSSL-0.97 (I think it was"i" in early April when I installed it) not 0.98a or b. No, NSS itself is an alternative to OpenSSL. It provides the SSL and TLS encryption schemes as Netscape developed them back in the day. So, if SSL and TLS tests are failing, then that's pretty much the whole package. Could you pick one or two errors and post them here? We probably don't need to see the whole file since I'm guessing the failures all stem from the same basic issue. -- Dan Thank you for your reply. Here are some of the errors: CRL SSL Client Tests Test Case Result Wait for Server Failed TLS Request don't require client auth (client does not provide auth) (cert TestUser40 - revoked) Failed Wait for Server Failed TLS Require client auth (bad password) (cert TestUser40 - revoked) Failed Wait for Server Failed SSL3 Request don't require client auth (client does not provide auth) (cert TestUser40 - revoked) Failed Wait for Server Failed Wait for ServerFailed SSL2 RC4 128 WITH MD5 Failed SSL2 RC4 128 EXPORT40 WITH MD5 Failed SSL2 RC2 128 CBC WITH MD5 Failed SSL3 RSA WITH DES CBC SHA Failed SSL3 RSA EXPORT WITH RC4 40 MD5Failed SSL3 RSA EXPORT WITH RC2 CBC 40 MD5Failed This is just a small sampling taken from various places in the log file. The errors encompass TLS SSL2 and SSL3 and have the same types of failures in each category of the log file. Also, when I use DOMSUF=example.org my system freezes up and I have to do ctrl-c to stop the process (I waited over an hour befor doing this on three separate occasions. I mention this because in my /etc/hosts file I have 127.0.0.1 localhost 192.168.1.1 blytheworks.example.org blytheworks I thought example.org would be appropriate to use (please correct me if I am wrong) when I use DOMSUF=blytheworks.example.org I end up with 346 not 770. when I use DOMSUF= (blank) I end up with 346. When I use any other value in DOMSUF I get 346 (the exceptions was DOMSUF=mydoman.com, That yielded only 220). Only DOMSUF=example.org freezes. When I use DOMSUF=example.org. (that is an extra . at the end) I get 346 not 770. Thanks again. rblythe -- http://linuxfromscratch.org/mailman/listinfo/blfs-support FAQ: http://www.linuxfromscratch.org/blfs/faq.html Unsubscribe: See the above information page
Re: Problem with tests in NSS-3.11
rblythe wrote: rblythe wrote: I am trying to build NSS-3.11. The patch, configure, and make portions of the instruction work fine. When I attempt to perform the test (as unpriviledged user), I have had mixed results. When I have DOMSUF=blytheworks.com, I get 346 not 770. I removed everything (rm -Rf nss-3.11) and started over and used DOMSUF=mydomain.com. This time I get 222 not 770. I also made sure that I exited out of all sub-shells before starting over. I am working from blfs-svn-20060415 and I checked the pages I downloaded against what is on the site now, and the instructions are the same. I believe that I am not understaning what a is, in order to get this to test properly. Everything else on my system has compiled, tested, and built without a hitch, so I don't know what I am doing incorrectly. Is it 100% necessary to have 770 as an end result? I did md5sum nss-3.11.tar.gz and my output matched the book so it is a good download. I am stumped with this one. Any help will be appreciated. Also Thanks, rblythe After checking the test output .html, I notices that all of the failures had to do with TLS and SSL. The log is too large to post in the e-mail (I guess it is time for me to learn how to tar/compress files and not just unpack them). Anyway, could this have something to do with openSSL? It is installed per the BLFS instructions, but the version I installed was openSSL-0.97 (I think it was"i" in early April when I installed it) not 0.98a or b. If this turns out to be part (if not all) of the problem, can openSSL-0.98b (current BLFS-SVN) be installed without mucking up everything else? Thanks again, rblythe I tried this yet again, this time with DOMSUF=example.org (this is what I used in LFS for /etc/hosts). The tests start out fine, but then it just freezes. Three different times I tried this (starting over completely from scratch with nss-3.11) and still no success. I'm bummed, but not giving up. I really could use come input on this one. Thanks, rblythe -- http://linuxfromscratch.org/mailman/listinfo/blfs-support FAQ: http://www.linuxfromscratch.org/blfs/faq.html Unsubscribe: See the above information page
Re: Problem with tests in NSS-3.11
rblythe wrote: I am trying to build NSS-3.11. The patch, configure, and make portions of the instruction work fine. When I attempt to perform the test (as unpriviledged user), I have had mixed results. When I have DOMSUF=blytheworks.com, I get 346 not 770. I removed everything (rm -Rf nss-3.11) and started over and used DOMSUF=mydomain.com. This time I get 222 not 770. I also made sure that I exited out of all sub-shells before starting over. I am working from blfs-svn-20060415 and I checked the pages I downloaded against what is on the site now, and the instructions are the same. I believe that I am not understaning what a is, in order to get this to test properly. Everything else on my system has compiled, tested, and built without a hitch, so I don't know what I am doing incorrectly. Is it 100% necessary to have 770 as an end result? I did md5sum nss-3.11.tar.gz and my output matched the book so it is a good download. I am stumped with this one. Any help will be appreciated. Also Thanks, rblythe After checking the test output .html, I notices that all of the failures had to do with TLS and SSL. The log is too large to post in the e-mail (I guess it is time for me to learn how to tar/compress files and not just unpack them). Anyway, could this have something to do with openSSL? It is installed per the BLFS instructions, but the version I installed was openSSL-0.97 (I think it was"i" in early April when I installed it) not 0.98a or b. If this turns out to be part (if not all) of the problem, can openSSL-0.98b (current BLFS-SVN) be installed without mucking up everything else? Thanks again, rblythe -- http://linuxfromscratch.org/mailman/listinfo/blfs-support FAQ: http://www.linuxfromscratch.org/blfs/faq.html Unsubscribe: See the above information page
Problem with tests in NSS-3.11
I am trying to build NSS-3.11. The patch, configure, and make portions of the instruction work fine. When I attempt to perform the test (as unpriviledged user), I have had mixed results. When I have DOMSUF=blytheworks.com, I get 346 not 770. I removed everything (rm -Rf nss-3.11) and started over and used DOMSUF=mydomain.com. This time I get 222 not 770. I also made sure that I exited out of all sub-shells before starting over. I am working from blfs-svn-20060415 and I checked the pages I downloaded against what is on the site now, and the instructions are the same. I believe that I am not understaning what a is, in order to get this to test properly. Everything else on my system has compiled, tested, and built without a hitch, so I don't know what I am doing incorrectly. Is it 100% necessary to have 770 as an end result? I did md5sum nss-3.11.tar.gz and my output matched the book so it is a good download. I am stumped with this one. Any help will be appreciated. Also Thanks, rblythe -- http://linuxfromscratch.org/mailman/listinfo/blfs-support FAQ: http://www.linuxfromscratch.org/blfs/faq.html Unsubscribe: See the above information page
Re: General Topic: Trying to understand the ln (link) syntax
Peter B. Steiger wrote: On Sat, 2006-05-27 at 22:57 -0400, rblythe wrote: ln -s -v ../../path/to/target somename or directory Specifically what I don't understand is the ../../ portion. Well, do you understand the use of . and .. as directory names in *nix (and for that matter MSDOS) environments? The period by itself . is the current directory, and the double period .. is the parent directory. So if you're sitting at the console with the current directory (pwd) /foo/bar/baz/bat/ then .. would be /foo/bar/baz/ and ../.. would be /foo/bar/, right? You can use this to link a nearby directory tree. Say application 1 expects to find libraries in /foo/bar/lib1/app but application 2 expects to find those same libraries in /foo/bar/lib2/app. Rather than create two copies of an identical library, you might cd /foo/bar/lib2/app and write: ln -s -v ../../lib1/app/somelib.so This would create a symbolic link in /foo/bar/lib2/app called somelib.so that pointed back to /foo/bar/lib1/app/somelib.so . By using the grandparent (../../) folder name, you maintain relative positions even if the whole /foo path changes to /fue - /fue/bar/lib2/app/somelib.so would still successfully point back two directories then down into lib1/app regardless of the top level directory name change. Probably way more detail than you wanted, but that's the whole story :-) Thank you. This explanation was very clear and truly helpful. Now I understand. rblythe -- http://linuxfromscratch.org/mailman/listinfo/blfs-support FAQ: http://www.linuxfromscratch.org/blfs/faq.html Unsubscribe: See the above information page
Re: Is it possible to change the text size in a non-gui environment?
Dan Nicholson wrote: On 5/24/06, Stephen Liu <[EMAIL PROTECTED]> wrote: Hi Simon and folks, > With vga=ask you can get more information when you start the > computer. adding vga=ask on /boot/grub/grub.conf Colours 640x400 640x480 800x600 1024x768 1152x864 1280x1024 1600x1200 +-- 4 bits |? ? 0x302 ??? ? 8 bits | 0x300 0x301 0x3030x3050x1610x307 0x31C 15 bits |? 0x310 0x3130x3160x1620x319 0x31D 16 bits |? 0x311 0x3140x3170x1630x31A 0x31E 24 bits |? 0x312 0x3150x318 ? 0x31B 0x31F 32 bits |? ? ?? 0x164 ? This has always kind of confused me, but the above table should help you. Unfortunately, the numbers need to be converted out of hexadecimal into decimal. For instance, I use vga=789, which is 0x315 in the table above, or 800x600 with 24 bit color. This gives me over 100 columns horizontally with the font I'm using, which is plenty for me. Here's bc doing the brute force hexadecimal conversion from 0x315 (there may be a better way :) $ echo "5*(16^0) + 1*(16^1) + 3*(16^2)" | bc 789 If you see a letter above, it's because hex is base 16. That means that A is decimal 10, B is decimal 11, and so on. This also depends on your console being vesa. Look for CONFIG_FB_VESA in the graphics support section of the kernel config. I'm not the kernel guy, but that should get you going. -- Dan Thanks everyone for all of the replies. It turns out that I didn't have frame buffer support compiled into the kernel. I don't know how I missed that. Anyway, I recompiled the kernel with frame buffer support and now all is well. Thanks again. rblythe -- http://linuxfromscratch.org/mailman/listinfo/blfs-support FAQ: http://www.linuxfromscratch.org/blfs/faq.html Unsubscribe: See the above information page
General Topic: Trying to understand the ln (link) syntax
I have read the ln manpage. I have searched the internet. I hope someone can clear this up for me: ln -s -v ../../path/to/target somename or directory Specifically what I don't understand is the ../../ portion. I am used to using links that look like ln -s -v /my/file/one two # without ../../ which results in two -> /my/file/one Can someone please explain what this syntax means. I have seen it in various places in the BLFS book and I just want to understand what this does. Thanks. rblythe -- http://linuxfromscratch.org/mailman/listinfo/blfs-support FAQ: http://www.linuxfromscratch.org/blfs/faq.html Unsubscribe: See the above information page
Is it possible to change the text size in a non-gui environment?
After building LFS, I noticed that the text was large (don't know pixels, etc.). I was wondering if there is some way of changing this so that the text is smaller in a non-gui environment? I have not decided on a Window Manager / Desktop Environment yet and Xorg 6.9 is working perfectly. The reason I ask is that in my base system (Mepis 3.4.3) when it first boots up, the letters are small and legible and they remain that way until kdm starts and I log in. I would like to do the same thing with LFS/BLFS so that the characters are smaller but legible until I decide on an environment to build. Thanks, rblythe -- http://linuxfromscratch.org/mailman/listinfo/blfs-support FAQ: http://www.linuxfromscratch.org/blfs/faq.html Unsubscribe: See the above information page
Building BLFS-SVN-? (started 20060411) - cannot install expect-5.43.0 - trouble with patch
Here is another one. Again, I checked everything and had no problems. The package unpacked perfectly. This is what I typed in: patch -Np1 -i ../expect-5.43.0-spawn-2.patch && ./configure --prefix=/usr --with-tcl=/usr/lib --with-tclinclude=/usr/include/tcl8.4 --enable-shared && make This is the output I received: patching file exp_chan.c patching file exp_command.h Hunk #1 succeeded at 30 with fuzz 1. Hunk #2 FAILED at 104. 1 out of 2 hunks FAILED -- saving rejects to file exp_command.h.rej patching file expect.c Hunk #1 FAILED at 41. Hunk #3 succeeded at 1712 with fuzz 1. Hunk #4 FAILED at 1730. Hunk #5 FAILED at 1787. Hunk #6 succeeded at 2864 with fuzz 1. 3 out of 6 hunks FAILED -- saving rejects to file expect.c.rej -- It did not make it to the ./configure stage Again I say I do not know how to edit a patch file for it to work properly and I noticed the line breaks were different when I viewed it with vi(m) as opposed to lynx. Any help would be appreciated. Thank you. rblythe -- http://linuxfromscratch.org/mailman/listinfo/blfs-support FAQ: http://www.linuxfromscratch.org/blfs/faq.html Unsubscribe: See the above information page
Building BLFS-SVN-? (started @ 20060411) - Cannot build cpio-2.6 - problem with patch
I just now tried to install cpio-2.6. I checked the md5sum and it matched the information from the book. It unpacked without a problem. This is what it typed: sed -i "s/invalid_arg/argmatch_invalid/" src/mt.c && patch -Np1 -i ../cpio-2.6-security_fixes-1.patch && ./configure CPIO_MT_PROG=mt --prefix=/usr --bindir=/bin --libexecdir=/tmp --with-rmt=/usr/sbin/rmt && echo "#define HAVE_SETLOCALE 1" >> config.h && echo "#define HAVE_LSTAT 1" >> config.h && make This is the output I received: patching file doc/cpio.1 patching file doc/cpio.info Hunk #1 succeeded at 203 with fuzz 1. patching file src/copyin.c Hunk #1 succeeded at 25 with fuzz 2. Hunk #2 FAILED at 390. Hunk #4 FAILED at 594. Hunk #5 succeeded at 604 with fuzz 2. Hunk #7 FAILED at 1507. 3 out of 7 hunks FAILED -- saving rejects to file src/copyin.c.rej patching file src/copypass.c Hunk #1 FAILED at 181. 1 out of 1 hunk FAILED -- saving rejects to file src/copypass.c.rej patching file src/extern.h Hunk #2 succeeded at 91 with fuzz 1. patching file src/global.c patching file src/main.c patch: malformed patch at line 321: ng slashes from file names"), 210}, It never got to the ./configure stage and it didn't make I don't know how to edit a patch file to fix it. I did notice that when I looked at the patch with vi(m), the line breaks were different from when I viewed it with lynx. I am still in the non-gui environment while I am building the system. Any insight into fixing this would be greatly appreciated. Thanks. rblythe -- http://linuxfromscratch.org/mailman/listinfo/blfs-support FAQ: http://www.linuxfromscratch.org/blfs/faq.html Unsubscribe: See the above information page
XFree86 - I can't get this to compile either - please help
I tried compiling Xfree86-4.4.0 after I couldn't get Xorg to compile. I downloaded the seven separate files. I checked them (md5sums) and everything looked fine. I completely deleted the old xc and xcbuild stuff generated by my previous failed Xorg attempts. I recreated the config/cf/hosts.def file. I modified this taking out all of the commented lines that are normally built by the Xfree86 scripts or I just don't need on my system. I also only tried building the vga under XF86CardDrivers because I don't have any other cards. My motherboard does use the SiS architecture if that matters.
Here is the last of my xfree-compile log (it is different from my xorg failure):
-fomit-frame-pointer -march=i686 Attrib.c
Attrib.c: In function `CreateOldColorTable':
Attrib.c:51: error: argument "ncolors" doesn't match prototype
Attrib.c:38: error: prototype declaration
make[4]: *** [Attrib.o] Error 1
make[4]: Leaving directory `/blfs-sources/xc/lib/Xpm'
make[3]: *** [all] Error 2
make[3]: Leaving directory `/blfs-sources/xc/lib'
make[2]: *** [all] Error 2
make[2]: Leaving directory `/blfs-sources/xc'
make[1]: *** [World] Error 2
make[1]: Leaving directory `/blfs-sources/xc'
make: *** [World] Error 2
Here are the lines in Attrib.c that are giving me problems:
#include "XpmI.h"
/* 3.2 backward compatibility code */
---> (LINE 38) LFUNC(CreateOldColorTable, int, (XpmColor *ct, int ncolors,
XpmColor ***oldct));
LFUNC(FreeOldColorTable, void, (XpmColor **colorTable, int ncolors));
/*
* Create a colortable compatible with the old style colortable
*/
static int
CreateOldColorTable(ct, ncolors, oldct)
XpmColor *ct;
unsigned int ncolors;
XpmColor ***oldct;
---> (LINE 51) {
XpmColor **colorTable, **color;
int a;
if (ncolors >= SIZE_MAX / sizeof(XpmColor *))
return XpmNoMemory;
colorTable = (XpmColor **) XpmMalloc(ncolors * sizeof(XpmColor *));
I really don't know what to do with this now. I don't know what I could have done wrong. I do have all of the dependency packages installed (I have rechecked this several times during the Xorg fiasco). I was hoping this would turn out to be different. My build is worthless to me without the X window and the packages that rely on it. Any assistance with this will be greatly appreciated.
Thanks,
Randy Blythe
--
http://linuxfromscratch.org/mailman/listinfo/blfs-support
FAQ: http://www.linuxfromscratch.org/blfs/faq.html
Unsubscribe: See the above information page
Re: Can't compile X.org - missing float.h
On Saturday 10 September 2005 10:38 am, Ken Moffat wrote: > > Those sound like *raw* kernel headers, *not* the headers from > linux-libc-headers. Re-read what Peter said, then reinstall the > linux-libc-headers from the version that you used when you built the > system. Building userspace against "unsanitized" headers *will* give > you the sort of grief you seem to be experiencing. > > Ken Okay, I re-read and followed :) the instructions given by Peter and Ken. I re-installed freetype2-1.9 (prior to trying to build xorg) I got the same freetype errors as in my previous post. I am out of ideas, and I really don't want to start from scratch again if I don't have too. This is my third attempt at a build of BLFS (my first two also stalled in this exact spot, but I didn't ask for help then, I just scrapped them and started completely over from page 1 of LFS). This is a new machine that I built myself and the crowning achievement would be to have LFS/BLFS up and running (my distro, my rules kind of thing). I am hoping that I didn't make a mistake in configuring and installing this system. I have not deviated from the book at all (except for the raw kernel thing that I corrected) and everything up to now has built without error or warning. Here is the error again: ../../../../config/makedepend/makedepend: warning: psout.c (reading ../../../../extras/freetype2/include/freetype/config/ftconfig.h), line 82: #error "Unsupported size of `int' type!" ../../../../config/makedepend/makedepend: warning: psout.c (reading ../../../../extras/freetype2/include/freetype/config/ftconfig.h), line 91: #error "Unsupported size of `long' type!" make[5]: *** [depend] Floating point exception make[5]: Leaving directory `/blfs-sources/xcbuild/programs/Xserver/Xprint/ps' make[4]: *** [depend] Error 2 make[4]: Leaving directory `/blfs-sources/xcbuild/programs/Xserver/Xprint' make[3]: *** [depend] Error 2 make[3]: Leaving directory `/blfs-sources/xcbuild/programs/Xserver' make[2]: *** [depend] Error 2 make[2]: Leaving directory `/blfs-sources/xcbuild/programs' make[1]: *** [depend] Error 2 make[1]: Leaving directory `/blfs-sources/xcbuild' make: *** [World] Error 2 I also have been building this thing as root, not as user and then switching to root to 'make install'. Could this be one of my problems? I am no Linux guru and have only been using Linux for 1 year. There has got to be something I can do short of scrapping the whole thing and starting over again. Would I have these same issues in trying to install XFree86-4.4.0? Any help would be appreciated. Thanks. -- http://linuxfromscratch.org/mailman/listinfo/blfs-support FAQ: http://www.linuxfromscratch.org/blfs/faq.html Unsubscribe: See the above information page
Re: Can't compile X.org - missing float.h
On Saturday 10 September 2005 2:17 am, Peter B. Steiger wrote: > On Fri, 2005-09-09 at 23:25 -0400, rblythe wrote: > > error messages I recieved (listed below) are > > resulting from the missing float.h header file. > > Oooh, I know that one - because it happened to me. When you built lfs, > did you copy the headers over from /usr/src/linux, or did you install > the "sanitized" headers? > http://www.linuxfromscratch.org/lfs/view/development/chapter06/linux-libc-h >eaders.html > > I actually had to make that mistake two or three times (what, me > stubborn?) before I learned that the lfs team knows more than I ever > will :-) Even if you installed a newer kernel than they have in the > book, use the book's headers. > > Try reinstalling the headers per the lfs book and see if that doesn't > fix the float.h build problems. > > -- > Peter B. Steiger > Cheyenne, WY Here is the next bits of errors: (My xorg-compile.log has 1000's of lines in it now so it is too large to post) PsMisc.c PsSpans.c PsArea.c PsPixel.c PsLine.c PsPolygon.c PsArc.c PsText.c PsWindow.c PsFonts.c PsAttr.c PsAttVal.c PsColor.c PsPixmap.c psout.c PsCache.c ft.c pt1.c runt1asm.c ttf2pt1.c psout_ft.c psout_ftpstype1.c psout_ftpstype3.c PsFTFonts.c ../../../../config/makedepend/makedepend: warning: psout.c (reading ../../../../extras/freetype2/include/freetype/config/ftconfig.h), line 82: #error "Unsupported size of `int' type!" ../../../../config/makedepend/makedepend: warning: psout.c (reading ../../../../extras/freetype2/include/freetype/config/ftconfig.h), line 91: #error "Unsupported size of `long' type!" make[5]: *** [depend] Floating point exception make[5]: Leaving directory `/blfs-sources/xcbuild/programs/Xserver/Xprint/ps' make[4]: *** [depend] Error 2 make[4]: Leaving directory `/blfs-sources/xcbuild/programs/Xserver/Xprint' make[3]: *** [depend] Error 2 make[3]: Leaving directory `/blfs-sources/xcbuild/programs/Xserver' make[2]: *** [depend] Error 2 make[2]: Leaving directory `/blfs-sources/xcbuild/programs' make[1]: *** [depend] Error 2 make[1]: Leaving directory `/blfs-sources/xcbuild' make: *** [World] Error 2 What I did to get to here: I found the float.h right where Declan said it would be so I copied that to /usr/include. I don't know why my system was not finding it where is was so if anyone can shed som light on that I would appreciate it. Other headers (stdarg.h, stddef.h) were in the linux-2.6.8.1/include/linux directory. I saved this when I first compiled the kernel for this build (thankfully) and copied those to /usr/include. I checked to see if I installed FreeType2-1.9 as Randy McMurchy suggested and it is in fact installed, but It seems the errors are related in some way to FreeType2-1.9 (I am not sure on this, so enlightenment will be appreciated). I was also having problems with (limit.h) that I failed to mention before. I found this out because I started over (removed xc, xcbuild directories) and followed the commands again and received errors relating to the limits.h. This was also in linux-2.6.8.1/include/linux directory so I copied it to /usr include and this allowed the following command from the book to work: pushd config/util && make -f Makefile.ini lndir Previously it would error and not make lndir This is where I am now. I am trying to be as thorough as possible. Any and all answers/suggestions will be greeatly appreciated. Thank you, Randy Blythe -- http://linuxfromscratch.org/mailman/listinfo/blfs-support FAQ: http://www.linuxfromscratch.org/blfs/faq.html Unsubscribe: See the above information page
Can't build/compile Xorg-6.8.2. Need help please
Thanks in advance for the help. Since copying this into my mail, I have found two typos. I will correct those, and try again, but in the meantime, if you can see anything that isn't right please let me know because once I get X, Fluxbox, Firefox, etc. installed and working on this machine, I will be able to dump my host distro for good. I fixed the typos and now the xorg-compile file is to large to copy here. Is it okay to attach files? Here is the last eleven lines of my second xorg-compile.log file after I fixed the typos (note: there is a total of 13,395 lines in the file now): make[3]: Entering directory `/usr/src/xcbuild/lib/Xt' rm -f StringDefs.c StringDefs.h Shell.h ../../config/util/makestrs < util/string.list > StringDefs.c /bin/sh: line 1: 18654 Illegal instruction ../../config/util/makestrs StringDefs.c make[3]: *** [Shell.h] Error 132 make[3]: Leaving directory `/usr/src/xcbuild/lib/Xt' make[2]: *** [includes] Error 2 make[2]: Leaving directory `/usr/src/xcbuild/lib' make[1]: *** [includes] Error 2 make[1]: Leaving directory `/usr/src/xcbuild' make: *** [World] Error 2 Here is what I have from xorg-compile.log (first time before fixing typos): Building Release 6.8.2. I hope you checked the configuration parameters in ./config/cf to see if you need to pass BOOTSTRAPCFLAGS. Mon Jul 4 11:07:00 EDT 2005 cd ./config/imake && make -f Makefile.ini BOOTSTRAPCFLAGS="" CC="cc" clean make[1]: Entering directory `/usr/src/xc/config/imake' rm -f ccimake imake.o imake rm -f *.CKP *.ln *.BAK *.bak *.o core errs ,* *~ *.a tags TAGS make.log \#* rm -f -r Makefile.proto Makefile Makefile.dep bootstrap rm -f imakemdep_cpp.h make[1]: Leaving directory `/usr/src/xc/config/imake' make Makefile.boot make[1]: Entering directory `/usr/src/xc' cd ./config/imake && make -w -f Makefile.ini BOOTSTRAPCFLAGS="" CC="cc" make[2]: Entering directory `/usr/src/xc/config/imake' making imake with BOOTSTRAPCFLAGS= and CROSSCOMPILEFLAGS=-DCROSSCOMPILEDIR="" in config/imake cc -o ccimake -DCROSSCOMPILEDIR=\"\" -O -I../../include -I../../imports/x11/include/X11 ccimake.c if [ -n "" ] ; then \ /cc -E `./ccimake` \ -DCROSSCOMPILE_CPP imakemdep.h > imakemdep_cpp.h; \ else touch imakemdep_cpp.h; fi cc -c -O -I../../include -I../../imports/x11/include/X11 `./ccimake` imake.c cc -o imake -O -I../../include -I../../imports/x11/include/X11 imake.o make[2]: Leaving directory `/usr/src/xc/config/imake' rm -f ./config/makedepend/Makefile.proto ./config/imake/imake -I./config/cf -s ./config/makedepend/Makefile.proto -f ./config/makedepend/Imakefile -DTOPDIR=../.. -DCURDIR=./config/makedepend cd ./config/makedepend && rm -f -r Makefile Makefile.dep makedepend *.o bootstrap cd ./config/makedepend && make -f Makefile.proto bootstrap make[2]: Entering directory `/usr/src/xc/config/makedepend' make[2]: Nothing to be done for `bootstrap'. make[2]: Leaving directory `/usr/src/xc/config/makedepend' ./config/imake/imake -I./config/cf -s ./config/imake/Makefile.proto -f ./config/imake/Imakefile -DTOPDIR=../.. -DCURDIR=./config/imake -DBootStrap cd ./config/imake && make -f Makefile.proto bootstrapdepend make[2]: Entering directory `/usr/src/xc/config/imake' make[2]: Nothing to be done for `bootstrapdepend'. make[2]: Leaving directory `/usr/src/xc/config/imake' cd ./config/imake && make -w -f Makefile.ini BOOTSTRAPCFLAGS="" CC="cc" bootstrap make[2]: Entering directory `/usr/src/xc/config/imake' + mkdir bootstrap mv *.o bootstrap + mv imake bootstrap make[2]: Leaving directory `/usr/src/xc/config/imake' cd ./config/imake && make -f Makefile.proto imakeonly make[2]: Entering directory `/usr/src/xc/config/imake' gcc -O2 -fno-strength-reduce -fno-strict-aliasing -march=i686 */-I../../include -I../../exports/include/X11 -I../.. -I../../exports/include -Dlinux -D__i386__ -D_POSIX_C_SOURCE=199309L-D_POSIX_SOURCE -D_XOPEN_SOURCE -D_BSD_SOURCE -D_SVID_SOURCE -D_GNU_SOURCE-DFUNCPROTO=15 -DNARROWPROTO -DCPP_PROGRAM="\"cpp\"" -DHAS_MERGE_CONSTANTS=`if gcc -fmerge-constants -xc /dev/null -S -o /dev/null 2> /dev/null 1> /dev/null; then echo 1; else echo 0; fi`-c -o imake.o imake.c gcc: bootstrap/: linker input file unused because linking not done gcc: CVS/: linker input file unused because linking not done cc1: bootstrap/: No such file or directory cc1: internal compiler error: Segmentation fault Please submit a full bug report, with preprocessed source if appropriate. See http://gcc.gnu.org/bugs.html> for instructions. make[2]: *** [imake.o] Error 1 make[2]: Leaving directory `/usr/src/xc/config/imake' make[1]: *** [imake.bootstrap] Error 2 make[1]: Leaving directory `/usr/src/xc' make: *** [World] Error 2 Here is my host.def: /* Begin Xorg host.def file */ /* System related Information. If you read and configure only on * section then it should be this one. The Intel architecture defaults * are set
