Re: [gentoo-user] xmms alternative

2006-10-28 Thread Myk Taylor
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

I had my music collection mounted over nfs and I ran into a similar
problem: amarok wouldn't be able to read the entire thing and the nfs
shares would stop working.

The problem turned out to be that amarok keeps all these files /open/
and the remote share runs out of resources.  I'm using audacious now
without problems.

- --Myk

Sarah wrote:
 All our music (about 80 Gb, I think, mainly (but not all) OGG files) is
 stored on a server, accessed on a mounted Samba share.  The share would
 mount correctly - and I tried mounting it with a variety of permissions
 set - but I couldn't get Amarok to 'read' the music collection
 successfully.
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.5 (GNU/Linux)

iD8DBQFFQwi7BOPsJyAQkeARAu+NAJ99BK6kVPvBk0iMYrwKpoeEQD/R+QCfc75B
ZcX9RL9HDtvHRjoQ72BI9JI=
=YJlw
-END PGP SIGNATURE-
-- 
gentoo-user@gentoo.org mailing list



Re: [gentoo-user] [OT?] ram question

2006-01-14 Thread Myk Taylor
memtest86+ ( http://www.memtest.org/ ) works well for detecting bad RAM, 
and comes standard on the x86 Gentoo install CD, I believe.


Ryan Sims wrote:

only the new stick is in.  Is this indicative of something?  I'm
already pretty convinced that it's a bad stick, but I wondered if
anyone could shed further light for me.

--
gentoo-user@gentoo.org mailing list



Re: [gentoo-user] Removing Specific Kernel Sources

2005-12-18 Thread Myk Taylor

Mariusz Pękala wrote:
 Additionally emerge unmerge leaves the remnants of compilation
 process, so you still have to do rm -r on the sources.

I've always liked to keep /usr as read-only as possible.  I set a few 
variables in /etc/make.conf that (I find) make everything a little more 
maintainable:


--- from /etc/make.conf --

PORTAGE_TMPDIR=/var/portage
DISTDIR=${PORTAGE_TMPDIR}/distfiles
PKGDIR=${PORTAGE_TMPDIR}/portage-pkg
DISTCC_DIR=${PORTAGE_TMPDIR}/distcc
KBUILD_OUTPUT=${PORTAGE_TMPDIR}/kernel

 end --

then I have a script that builds the kernel, setting the 'O' (oh, not 
zero) parameter so that build output goes into the ${KBUILD_OUTPUT} 
directory, and leaves /usr/src/kernel version pristine:


--- excerpt from makekernel script -

sourceDir=/usr/src/linux
workDir=${KBUILD_OUTPUT}
binDir=/boot
target=kernel.hostname.version.date

(cd ${sourceDir}  sudo -u portage make O=${workDir} menuconfig)
(cd ${workDir}/include  sudo -u portage lndir ${sourceDir}/include)
(cd ${sourceDir}  sudo -u portage make O=${workDir})
(cd ${sourceDir}  make O=${workDir} modules_install)
cp ${workDir}/arch/${arch}/boot/bzImage ${binDir}/${target}

-- end ---

I've been using this script (full script inlined below) for more than a 
year without problems.  The lndir line is there to appease packages that 
build kernel modules, and expect to find ${KBUILD_OUTPUT}/include 
populated with header files.  make is run as user 'portage' since some 
packages like to run make O=${KBUILD_OUTPUT} mrproper, so user portage 
needs write access.


--myk


-- makekernel script 

#!/bin/sh
#
# makekernel
#
# written by Myk Taylor, 2004

# 'makekernel' without parameters will build and install the custom 
kernel for
#   the local machine (that is, it will use the config file named 
HOSTNAME.config

#   where HOSTNAME is the hostname of the local host in caps)  For example,
#   TUX.config would be the config file for host 'tux'.
# 'makekernel CONF_ID' will build and install only that specific 
kernel, that is,

#   it will used the file named CONF_ID.config
# 'makekernel all' will build and install all kernel confs in the current
#   directory (a kernel conf is identified as a file with an all-caps 
stem and a

#   '.config' extension)

[ -f /etc/make.conf ]  source /etc/make.conf

# script parameters
sourceDir=/usr/src/linux
workDir=${KBUILD_OUTPUT:-/tmp/kernel}
binDir=/boot

# get a list of config stems
configNames=$*

# if 'all' is one of the config names, set configNames to be all the
#   config files in the current directory
# elif no parameters specified, use local kernel config
if [ ! -z `echo ${configNames} | grep -w 'all'` ]; then
configNames=`ls | grep '^[A-Z0-9 _-]*\.config$' | sed 
's/\.config//'`

elif [ -z ${configNames} ]; then
configNames=`hostname | sed 's/\..*//' | tr a-z A-Z`
fi

# do some sanity checking
versionFile=${sourceDir}/Makefile

if [ ! -f ${versionFile} ]; then
echo ${versionFile} not found
exit 1
fi
if [ `echo ${configNames} | grep '[^A-Z0-9 _-]'` ]; then
echo config file name stems should be in all caps
exit 1
fi

echo Making kernel(s): `echo -n ${configNames}`

osvString=^VERSION = 
osplString=^PATCHLEVEL = 
osslString=^SUBLEVEL = 
osevString=^EXTRAVERSION = 

osv=`grep ${osvString} ${versionFile} | sed s/${osvString}//`
ospl=`grep ${osplString} ${versionFile} | sed s/${osplString}//`
ossl=`grep ${osslString} ${versionFile} | sed s/${osslString}//`
osev=`grep ${osevString} ${versionFile} | sed s/${osevString}//`

if [ ! -z ${osv} -a ! -z ${ospl} -a ! -z ${ossl} -a ! -z ${osev} 
]; then

osr=${osv}.${ospl}.${ossl}${osev}
else
osr=${osr:-'Unknown'}
fi

kerPrefix=kernel.
dateStr=`date +%Y%m%d`
kerSuffix=-${osr}.${dateStr}

for configName in ${configNames}; do
configFileName=${configName}.config

# add more architectures as required
if [ ! -z `grep '^CONFIG_X86_64=y' \${configFileName}\` ]; then
arch=x86_64
elif [ ! -z `grep '^CONFIG_X86=y' \${configFileName}\` ]; then
arch=i386
else
echo cannot determine target architecture in 
${configFileName}

echo the script probably needs to be updated
exit 1
fi

target=${kerPrefix}${configName}${kerSuffix}
echo == building ${binDir}/${target} from ${configFileName}

mkdir -p ${workDir}  chown portage:portage ${workDir}
cp ${configFileName} ${workDir}/.config  chown 
portage:portage ${workDir}/.config || exit 1
(cd ${sourceDir}  sudo -u portage make O=${workDir} 
menuconfig) || exit 1
(cd ${workDir}/include  sudo -u portage lndir 
${sourceDir}/include  /dev/null) || exit 1
(cd ${sourceDir}  sudo -u portage make O=${workDir}) || 
exit 1
(cd ${sourceDir}  make O=${workDir} modules_install) || 
exit 1


mkdir -p ${binDir

Re: [gentoo-user] Removing Specific Kernel Sources

2005-12-17 Thread Myk Taylor

emerge -C gentoo-sources-2.6.12-r10
or
emerge -C =gentoo-sources-2.6.12-r10

-C is short for --unmerge

--myk

C. Beamer wrote:

I would like to remove the oldest one - specifically linux-2.6.12-gentoo-r10

--
gentoo-user@gentoo.org mailing list



Re: [gentoo-user] Removing Specific Kernel Sources

2005-12-17 Thread Myk Taylor
just 'rm -rf'ing the source dirs gets rid of the files, true, but not 
formally unmerging packages desynchronizes the package management 
system, which will think you still have the older versions installed. 
In practice, I don't think it causes any real problems (maybe it will 
make emerge(1) take a little longer when calculating dependencies...), 
but it will allow cruft to build in your system, and with it, possible 
troubleshooting difficulties.


--myk

Samir Faci wrote:
is there an advantage to doing that as opposed to rm -fr 
/usr/src/linux-2.6.12-r10  (or whatever the dir is called)?  Just 
curious, I always just used the rm -fr


Myk Taylor wrote:

emerge -C gentoo-sources-2.6.12-r10
or
emerge -C =gentoo-sources-2.6.12-r10

-C is short for --unmerge

C. Beamer wrote:
I would like to remove the oldest one - specifically 
linux-2.6.12-gentoo-r10

--
gentoo-user@gentoo.org mailing list



Re: [gentoo-user] browser,news,mail

2005-08-27 Thread Myk Taylor
default gnupg (Enigmail) integration with Thunderbird was removed 
recently because of trouble with the build.  from the ebuild:


ewarn Enigmail Support has been dropped since it doesn't work on 
fresh install.
ewarn The Gentoo Mozilla team is working on making enigmail its 
own build,
ewarn sorry for the inconvenience.  For now, you can download 
enigmail from

ewarn http://enigmail.mozdev.org;

--myk

John Dangler wrote:

Brett~
Thanks for the reply.  I did find some additional information about these
that tells me I should be using Firefox and Thunderbird...
The USE flags on portage for thunderbird don't require gnupg, but I noticed
in Mozilla mail that in order to use encrypted mail, Mozilla mail wanted it.
Is there a gnupg USE flag that will emerge Thunderbird with this feature
built-in?

John D

--
gentoo-user@gentoo.org mailing list



[gentoo-user] root and portage alternately own distcc lock files

2005-07-11 Thread Myk Taylor
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

After getting ccache to work, I set up distcc with another computer on
my lan and compiled a few packages to test it out.  It worked great
until I tried to compile busybox, where it displayed the error message:

distcc[12251] (dcc_open_lockfile) ERROR: failed to creat
/var/portage/distcc/lock/cpu_localhost_0: Permission denied

the file was owned by root:portage, so I deleted the lock directory and
tried to compile busybox again.  This time the error messages didn't
show up and when I checked, the files were owned by portage:portage.  I
deleted the directory again and compiled nano.  The lock files were
owned by root:portage again.  What's going on?  Why are distcc lock
files sometimes owned by root and sometimes by portage?
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.1 (GNU/Linux)

iD8DBQFC0hnmBOPsJyAQkeARAlevAJ0dVwXRXClpLdeZeK1e9UntmsNJgwCgqAzo
VZvYcM8fgwl+nlPALwTAW3Q=
=qmBX
-END PGP SIGNATURE-
-- 
gentoo-user@gentoo.org mailing list



Re: [gentoo-user] what's new with all those new packages?

2005-07-11 Thread Myk Taylor
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

I think if you're using Gentoo, there is an assumption that if it's new
(and acceptably stable), I want it.  I agree that a comprehensive list
of fixed bugs and new features would be nice for each updated package,
but I imagine that it would be unwelcome extra work for the port
maintainers.

Especially handy would be if I were updating from version x to version y
of a particular package and emerge would tell me the functional delta
between those two versions.

If there's a package I'm particularly interested in (or dubious about),
 I'll venture over to its homepage to check out the changelist when it
gets updated, but I usually defer to the wisdom of the portage and
package maintainers.

- --myk

[EMAIL PROTECTED] wrote:
 portage reminds me of updating packages frequently,
 i have to decide what to update, and if it's worth downloading?
 
 is there a easy way to see what's new of those packages to be
 updated? the ChangeLog within /usr/portage/xxx-xxx/ make no sense to me.
 
 thanks.
 
 regards 
 daniel..
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.1 (GNU/Linux)

iD8DBQFC0pxMBOPsJyAQkeARAsvEAJ9RWQ9LBWOdXJpBsesgfJP2wOBtBgCbBU0m
7LtGA+VdldKj/sxxrjlMtzg=
=NhMW
-END PGP SIGNATURE-
-- 
gentoo-user@gentoo.org mailing list



Re: [gentoo-user] Another question on mailing cron-job output

2005-06-29 Thread Myk Taylor
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

This sounds perfectly reasonable.  cron mails you and output the command
sends to stdout, so yes, redirecting its output to a file would also
prevent cron's email.  the crontab lines would look something like:

0 * * * * command  /dev/null  echo commandTitle  done_deals.txt
1 3 * * * cat done_deals.txt  rm done_deals.txt

man bash for more info on commandline syntax.  you could get even
sneakier with an 'if' statement (or a combination of  and ||),
outputting different messages for success and failure of the cron job.

compiling successful jobs into a list can be useful, but the
modifications to cron to get it done are pretty minimal--I doubt there's
much call for an external tool in this case.  the default behavior for
cron accomplishes something functionally equivalent anyway: it informs
you by mail if a job fails.  So you'd still know (by process of
elimination) which jobs succeeded and which failed.

- --myk

Holly Bostick wrote:
 If I:
 
 Changed the individual job that I don't want mailed, but want
 notification of to send it's output to /dev/null (is it only /dev/null,
 or will cron still mail me if I output to a text file instead?) *and
 then to* echo the name of the job to (let's say) done_deals.txt after
 the job completed, and then
 
 Made a new cron job that runs last in that time period which sent me the
 output of cat done_deals.txt (which would hopefully contain the names of
 all the jobs that completed, but whose actual output was sent to
 /dev/null or whatever)
 
 would that work?
 
 It sounds reasonable, although I don't know how to do it (I am so not a
 command-line jockey), and it also sounds easily repeatable (for cron
 jobs of all time periods, or if new jobs are added to any time period),
 which supports it being doable.
 
 Does this sound flakey? Does this sound useful? And if it does sound
 useful, is there any tool that would allow me to create a more global
 setup in case I wanted to provide this solution to others (i.e. post it
 on the forums as some kind of script, or package it in some way for
 Bugzilla)?
 
 Holly
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.1 (GNU/Linux)

iD8DBQFCwsrxBOPsJyAQkeARArh+AKCyPhrkqNFdQV4RovfmzBzmjjdr8wCfS6W6
cM+/Ji3FIVOMXNb/djN787g=
=SlBP
-END PGP SIGNATURE-
-- 
gentoo-user@gentoo.org mailing list



Re: [gentoo-user] root block device unspecified error on boot

2005-06-03 Thread Myk Taylor
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

ah, sounds similar to the discussion on
http://www.linuxforums.org/forum/topic-23080.html

short version: make sure the drivers for the hdd device and filesystem
are compiled into the kernel and doubld check your grub.conf kernel= line.

- --myk

Gentoo Linux Mailing List Client wrote:
 yeah I did that as well.
 grub is not the problem as far as I can tell.  
 I get the boot menu just fine.
 my system stops when it tries to (re)mount the root partition.
 It seems to think I didn't specify it.
 I thought that's what fstab was for ???
 
 Gentoo Linux Mailing List Client wrote:
 
I had to move my linux partition from /dev/hda3 to /dev/hda1
I have altered /etc/fstab to reflect the move, along with /boot/grub/menu.lst
yet everytime I boot I still get the following error.
 
   The root block device is unspecified or not detected.
   Please specify a device or shell for a shell.
   BOOT () ::
 
at this point if I enter /dev/hda1 the system continues to boot fine.
what am I forgetting to fix?
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.1 (GNU/Linux)
Comment: Using GnuPG with Thunderbird - http://enigmail.mozdev.org

iD8DBQFCoBhSBOPsJyAQkeARAn9bAJ4kizZmArNUONC0+FjoSoR+0PLJiACfe3v8
lU22SEfXXLzhyx3de5U0Y7c=
=OkNf
-END PGP SIGNATURE-
-- 
gentoo-user@gentoo.org mailing list



[gentoo-user] When is Gentoo CVS synched to the trunk?

2005-06-02 Thread Myk Taylor
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

I've noticed a few Gentoo bugs have been closed with remarks stating:
'fixed in CVS'.  However, weeks later, I still haven't seen the changes
reflected in the web view of Gentoo CVS.  Are the maintainers committing
to a different repository?  And if so, how often are the two
repositories synched?

- --myk
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.1 (GNU/Linux)
Comment: Using GnuPG with Thunderbird - http://enigmail.mozdev.org

iD8DBQFCnsHmBOPsJyAQkeARAiUpAKCse2OBIxCHgDbZIupeb0AWWPzQQACffcZi
RhXbTix3EV9lBvKzAQ1kzSM=
=SZGm
-END PGP SIGNATURE-
-- 
gentoo-user@gentoo.org mailing list



Re: [gentoo-user] root block device unspecified error on boot

2005-06-02 Thread Myk Taylor
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Have you directed the first-stage bootloader to find root on hda1?

from
http://www.gentoo.org/doc/en/handbook/2005.0/handbook-x86.xml?part=1chap=10#doc_chap2

try 'grub-install /dev/hda'
or 'grub' and
grub root (hd0,0)  (Specify where your /boot partition resides)
grub setup (hd0)   (Install GRUB in the MBR)
grub quit  (Exit the GRUB shell)

- --myk

Gentoo Linux Mailing List Client wrote:
 I had to move my linux partition from /dev/hda3 to /dev/hda1
 I have altered /etc/fstab to reflect the move, along with /boot/grub/menu.lst
 yet everytime I boot I still get the following error.
 
The root block device is unspecified or not detected. 
Please specify a device or shell for a shell.
BOOT () ::
 
 at this point if I enter /dev/hda1 the system continues to boot fine.
 what am I forgetting to fix?
 
 TIA,
 Tomoki Taniguchi
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.1 (GNU/Linux)
Comment: Using GnuPG with Thunderbird - http://enigmail.mozdev.org

iD8DBQFCnsxnBOPsJyAQkeARAvGqAKDAFeCbyTVjPqezDwaGr3lITAiE4ACfTxku
RwFWxRzeKggA0rutmGBTBaQ=
=3UIJ
-END PGP SIGNATURE-
-- 
gentoo-user@gentoo.org mailing list



Re: [gentoo-user] When is Gentoo CVS synched to the trunk?

2005-06-02 Thread Myk Taylor
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

D'oh.  I see my mistake.  The bug was fixed in an eclass, not the ebuild
where I was expecting it.  Sorry for the noise.

Ciaran McCreesh wrote:
 On Thu, 02 Jun 2005 01:23:02 -0700 Myk Taylor [EMAIL PROTECTED] wrote:
 | I've noticed a few Gentoo bugs have been closed with remarks stating:
 | 'fixed in CVS'.  However, weeks later, I still haven't seen the
 | changes reflected in the web view of Gentoo CVS.  Are the maintainers
 | committing to a different repository?  And if so, how often are the
 | two repositories synched?
 
 viewcvs updates pretty quickly. We're talking hours, not weeks.
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.1 (GNU/Linux)
Comment: Using GnuPG with Thunderbird - http://enigmail.mozdev.org

iD8DBQFCn0bcBOPsJyAQkeARAuUPAJ9V/tgbkGBsjzjgY8pZBz7pZoEAkwCeN/Hb
YMk+fqwKOGPaZxAiV3yV3T8=
=/MrV
-END PGP SIGNATURE-
-- 
gentoo-user@gentoo.org mailing list



Re: [gentoo-user] Back up with no root

2005-05-27 Thread Myk Taylor
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Run sudo after you ssh.  On my network, I backup my servers by setting
up sudoers on the server I want to backup and running the following
command from my workstation:

ssh flags user@hostname sudo dump -udumplevelf- filesystem |
gzip  hostname.dumplevel.date.gz

Pupeno wrote:
 Hello,
 I'm trying to set up my computers so I make backups of my server from my 
 workstation (we don't have a backup server). The thing is that I have a 
 normal user on that server and I'm on the sudoers file to perform any 
 root-task.
 Now, to back up, I'm running an rsync thru ssh to the server, but that runs 
 as 
 a my user because I can't ssh as root, how can I achieve root privileges to 
 be able to perform the back up ?
 Thank you.
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.1 (GNU/Linux)
Comment: Using GnuPG with Thunderbird - http://enigmail.mozdev.org

iD8DBQFCl2JIBOPsJyAQkeARAglpAKChjai/SSVils4LLAhvBHFw4GPF0gCgx50O
4JBov/gZcdie8jtIhnBLvGw=
=1cGc
-END PGP SIGNATURE-
-- 
gentoo-user@gentoo.org mailing list



Re: [gentoo-user] console switching wierdness

2005-05-18 Thread Myk Taylor
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Which framebuffer module are you using?  I have an nVidia card, and
found that using the nVidia framebuffer resulted in ~1 sec latency when
switching between virtual consoles.  Removing the nVidia framebuffer and
using vesafb-tng was the solution for me.

- --myk

maxim wexler wrote:
 Hello everybody,
 
 Don't even know how to google for this one.
 
 Having giving up for now getting a Radeon9250 to work
 with gentoo. I replaced it with an ATI Rage128 AGP
 card and referenced the appropriate module, aty128fb,
 in modules.autoload. Now, even before configuring X
 something bizarre happens: I'll be working away in one
 console then decide to ctlaltf2 to the next one,
 do some stuff there and ctlaltf1 back to the
 first; but then suddenly I lose ~80% of screen
 brightness! Horrified,I reboot and find everything
 back to normal, switch consoles back and forth, same
 thing happens.
 
 Anybody know what's going on?
 
 -mw
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.1 (GNU/Linux)
Comment: Using GnuPG with Thunderbird - http://enigmail.mozdev.org

iD8DBQFCjCWtBOPsJyAQkeARApZDAJ9dfmAfQf7TTP/IlBcwlGM5fXGKMgCguCrS
nvtqwtAzXE8C2gK2RQfBDao=
=Cb3q
-END PGP SIGNATURE-
-- 
gentoo-user@gentoo.org mailing list



Re: [gentoo-user] Sometimes emerges fail

2005-05-10 Thread Myk Taylor
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

I find that sometimes MAKEOPTS=-j2 is the culprit.  Make sometimes tries
to link an executable before all of its constituent object files are
compiled (this shouldn't normally happen, but maybe the structure of the
Makefile isn't exactly what make expects or something).

After modifying the MAKEOPTS line in /etc/make.conf to explicityly allow
only one thread: MAKEOPTS=-j1 (and patching
kernel-mod.eclass so that it respects KBUILD_OUTPUT
http://bugs.gentoo.org/show_bug.cgi?id=89142), all my compilations have
gone smoothly.

- ---make.conf---
#
# make.conf

CHOST=i686-pc-linux-gnu
CFLAGS=-O3 -march=i686 -fforce-addr -ftracer -pipe
CXXFLAGS=${CFLAGS} -fvisibility-inlines-hidden

PORTAGE_NICENESS=3
MAKEOPTS=-j1
AUTOCLEAN=yes
FEATURES=ccache distlocks sandbox buildpkg userpriv fixpackages
CCACHE_SIZE=4G

PORTAGE_TMPDIR=/var/portage
DISTDIR=${PORTAGE_TMPDIR}/distfiles
PKGDIR=${PORTAGE_TMPDIR}/packages
PORT_LOGDIR=/var/log/portage
PORTDIR_OVERLAY=/usr/local/portage
PORTAGE_TMPFS=/dev/shm

GENTOO_MIRRORS=http://mirror.datapipe.net/gentoo
ftp://ftp.ucsb.edu/pub/mirrors/linux/gentoo;

USE=3dnow Xaw3d acpi audiofile bash-completion cdparanoia cjk dvd dvdr
dvdread esd flac immqt java junit kdeenablefinal mmx mozilla nptl nvidia
offensive samba sse unicode usb win32codecs xine xscreensaver

# for KDE and OpenOffice.org
LINGUAS=en zh_TW

# for gtk 1.x
#LINGUAS=en zh_TW.Big5

# for packages that build kernel modules
KBUILD_OUTPUT=/tmp/kernel
- --end make.conf-

- --myk

Grant wrote:
 All of my machines will sometimes fail to emerge a package
 successfully, but the next try will be successful.  I've read through
 this metabug:
 
 http://bugs.gentoo.org/show_bug.cgi?id=20600
 
 but it can't be bad hardware because it happens on 4 different systems
 including a hosted remote server.  Here are my CFLAGS:
 
 CFLAGS=-O2 -march=pentium4 -pipe -fomit-frame-pointer
 
 Could the -O2 be the problem?
 
 - Grant
 
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.1 (GNU/Linux)
Comment: Using GnuPG with Thunderbird - http://enigmail.mozdev.org

iD8DBQFCgNsVBOPsJyAQkeARArBCAJ9w92jf52vx9GzrHi9lItJHX8K+GwCguBo3
Q1JXYYLzd6J6T+DSEKl1YI4=
=T/+H
-END PGP SIGNATURE-
-- 
gentoo-user@gentoo.org mailing list