Re: [gentoo-user] Is there and Alternative to compiling kde?

2005-12-10 Thread Spider (D.m.D. Lj.)
On Fri, 2005-12-09 at 15:55 -0600, Harry Putnam wrote:
 I'll probably need the asbestos drawers here shortly:
 
 I've burned up several hours here with a grindingly slow compile of
 kde. It is an older machine ( a few years) but is a P4 2Ghz and 500MB
 ram. Is there an alternative to this?  I mean aside from using a
 lighter, faster compiling, X setup.

 Is there some burning important reason why we need to throw away hours
 and hours compiling kde?  Wouldn't a binary distribution of kde serve
 as well in most ways?



Okay, selfplugging a bit here, but thats ok, I've got a permit.  Or
wait, no I don't, so it seems that this is still unofficial and
unsupported by others than me and then only at best effort:

http://chinstrap.alternating.net/





-- 
begin  .signature
Tortured users / Laughing in pain
See Microsoft KB Article Q265230 for more information.
end



signature.asc
Description: This is a digitally signed message part


Re: [gentoo-user] GCC only for priviliged users?

2005-12-09 Thread Spider (D.m.D. Lj.)
On Fri, 2005-12-09 at 18:21 +0100, Jesús García Crespo wrote:
 Hi! I thought that GCC could means a risk if all of the users of my
 system are able to run it! I talked this with a friend and he propossed
 to create a new group, compiler, for example, where all the users
 who will be able to run gcc must belong to it!
 
 Wouldn't be interesting to implement this into Gentoo gcc ebuild as an
 USE?


Exactly what risk is there from an end-user running a compiler?   A
compiler doesn't access any kind of restricted environment, doesn't
auytomatically create binaries with other rights than its own and is
about as safe a product as there can be.

And if you think that users running their own programs is a risk, simply
mount /home as noexec,  ( make sure to impose the same limitations
on /tmp and /var/tmp as well,  since users have write-access there)


And.. really. python, perl, awk, bash ... All of those are fully capable
of creating and running programs. And no, I do not think you can limit
the use thereof from user accounts.: )


If you're really paranoid about execution and so on, start reading the
SELinux FAQ and create a ruleset.. The default one is probably more
lenient than you want it ;)

//Spider
-- 
begin  .signature
Tortured users / Laughing in pain
See Microsoft KB Article Q265230 for more information.
end



signature.asc
Description: This is a digitally signed message part


Re: [gentoo-user] dmesg yes fdisk no - new HDD

2005-12-07 Thread Spider (D.m.D. Lj.)
On Tue, 2005-12-06 at 17:34 -0600, Harry Putnam wrote:
 I'm installing a new HDD on an older (in P4 terms) P4 2.0 GB.
 
 On bootup I see it noticed in dmesg or but doesn't get an IRQ.  (See
 snippet from dmesg) And once booted up, fdisk doesn't know about it.
 
 I pulled out the ribbon to two cd drives and connected This new drive
 by itself as master with no slave.  (Just to start tinkering).
 
 Anyone that can spot something or has a nifty idea or question please respond.
 
 DMESG:

This is one of the fun ones.

try cfdisk and partedit, I think cfdisk will be more verbose and might
tell you it is refused to seek() the drive...


if thats the case, its probably the controller (or driver) being odd on
you. Had this recently when a 250Gb disk was recognized as
655535Mb .   Which is wrong ;) 


However,  I got around it by formatting it in another machine, the drive
works perfectly, (it seems, guess I'll know more later;)  Though there
may be some obscure config option in the IDE/ATA part of the kernel
config to avoid this kind of situation.

//Spider

-- 
begin  .signature
Tortured users / Laughing in pain
See Microsoft KB Article Q265230 for more information.
end



signature.asc
Description: This is a digitally signed message part


Re: [gentoo-user] Need drive space, what to delete?

2005-12-04 Thread Spider (D.m.D. Lj.)
On Sun, 2005-12-04 at 04:42 -0600, Dale wrote:
 LOL  It helped a little bit, but not much.
 
  swifty / # df
  Filesystem   1K-blocks  Used Available Use% Mounted on
  /dev/hda6  3564108   3505584 58524  99% /
  udev12738880127308   1% /dev
  /dev/hda148312 37412 10900  78% /boot
  none127388 0127388   0% /dev/shm
  swifty / #
 
 Any more ideas?  I would hate to have to remove KDE from that thing.


Little known things that may help:

app-admin/localepurge
 Handy tool.  Wipes locales that you don't use.
( 262 Mb here )

make sure you strip binaries, build them with -O2 or -Os instead of -O3.
(debug info alone on my system is  490 Mb)

Wipe old kernels.
make clean in the one kernel dir you have left.

/lib/modules : clean out things you don't have left.



cd /usr ;
du -ab |sort -n   


Look at the results,  then use equery ( or qfile, qpkg, epm or any other
tool)  to look them up.

emerge --prune ( handle with care... .)


Remove tetex if you have it installed. ( Also make sure you set
USE=-doc  unless you really want API documentations )



Logs?  Logrotate + compression.

KDE:  perhaps using split builds and only installing the pieces you
need/want?

//Spider



-- 
begin  .signature
Tortured users / Laughing in pain
See Microsoft KB Article Q265230 for more information.
end



signature.asc
Description: This is a digitally signed message part


Re: [gentoo-user] how to use EXTRA_ECONF?

2005-10-18 Thread Spider (D.m.D. Lj.)
On Tue, 2005-10-18 at 16:50 -0400, Dave Nebinger wrote:
 On Tuesday 18 October 2005 04:34 pm, Holly Bostick wrote:
   For Holly's case, I'm wondering if she's syncing against a system
   that doesn't mirror that file from upstream?  Just a guess.
 
  I sync against the Netherlands rsync pool,
  SYNC=rsync://rsync.nl.gentoo.org/gentoo-portage.
 
  I suppose what you say is possible, but does not seem to be the case.
 
 I figured you sync'd daily Holly; I wouldn't have expected any less. ;-)
 
 The question is, however, if you and I both sync daily and, although file 
 times suggest they have been updated, but the file contents are different, 
 where would the problem lie?
 
 The only guess I could come up with is the upstream mirror.  I sync against 
 http://gentoo.osuosl.org/, and you're syncing against the netherlands pool.  
 Either one themselves could be sync'd against another mirror which is sync'd 
 against another mirror...
 
 Somewhere along the line I'm guessing that perhaps this particular file is 
 not 
 fetched/updated for some reason which would leave one of us with an outdated 
 copy.
 
 I know mine comes out of CVS with the header 
 $Header: /var/cvsroot/gentoo-x86/profiles/use.local.desc,v 1.1502 2005/10/18 
 00:03:00 vapier Exp $, so I'm guessing that I have the later file.
 
[EMAIL PROTECTED] cvs up use.local.desc ; grep vim-with-x use.local.desc

app-editors/vim:vim-with-x - Linking console vim against X11 libraries
to enable title and clipboard features in xterm


[EMAIL PROTECTED] head -n 4 use.local.desc |tail -n 2
# $Header: /var/cvsroot/gentoo-x86/profiles/use.local.desc,v 1.1505
2005/10/18 20:18:45 agriffis Exp $


//Spider

-- 
begin  .signature
Tortured users / Laughing in pain
See Microsoft KB Article Q265230 for more information.
end



signature.asc
Description: This is a digitally signed message part


Re: [gentoo-user] Simple command line stuff

2005-10-11 Thread Spider (D.m.D. Lj.)
On Tue, 2005-10-11 at 00:25 -0700, Drew Tomlinson wrote:
 
 Oh, and another thought.  The find command can do this for you as 
 well, IIRC.



find . -type f -iname '*.wav' -exec command {} {}.foo \;

   is the syntax, IIRC.   Note the \  that exists to escape the
semicolon, therefore telling find to end processing.

find . -type f -iname '*.wav' |while read LIST; do command ${LIST}
${LIST/old/new} ; done

is another possibility.Adding limiters to find can prevent it from
recursing too deeply.


//Spider

-- 
begin  .signature
Tortured users / Laughing in pain
See Microsoft KB Article Q265230 for more information.
end



signature.asc
Description: This is a digitally signed message part


Re: [gentoo-user] Simple command line stuff

2005-10-10 Thread Spider (D.m.D. Lj.)
On Sun, 2005-10-09 at 19:06 -0700, Mark Knecht wrote:
 Hi,
I don't have a single book on Linux. (Amazing...) Can someone
 recommend a simple book on command line stuff, or better yet a good
 web site on this topic?
 
For instance, I wanted to run a specific command on every file in a
 directory which will create a new file, so I need to do
 
 commandfile1.wavfile1-convert.wav
 
 I need to take each name, create a new name to build the actual
 command that gets run and then do that for every file in the
 directory, or even in a hierarchy of directories.
 
 Thanks,
 Mark



For bash / zsh and other advanced(?-) shells: 

for f in *.wav; do command $f ${f/.wav/-convert.wav};done

The   are there to prevent files with spaces in them (evil!) from
becoming too annoying and appearing as multiple commandline arguments.


//Spider

-- 
begin  .signature
Tortured users / Laughing in pain
See Microsoft KB Article Q265230 for more information.
end



signature.asc
Description: This is a digitally signed message part


Re: [gentoo-user] possible defective memory

2005-10-08 Thread Spider (D.m.D. Lj.)
On Sat, 2005-10-08 at 21:08 -0400, bruce harding wrote:
 I've got 2 sticks of Kingston HyperX 3200 DDR Registered  ECC.  Is it
 possible that this if I ran memtest86+ for 2 days straight and found no
 error that the memory could still be defective?
 
 I ask because I can't get a complete compile of glibc.  I have to
 restart the process at lease 3 times before the compile will complete.
 
 
 Let me know what you think.

Well, here's a good list to go through.:

http://people.redhat.com/davej/hardware-problems.txt



//Spider

-- 
begin  .signature
Tortured users / Laughing in pain
See Microsoft KB Article Q265230 for more information.
end



signature.asc
Description: This is a digitally signed message part