Re: [gentoo-user] grub help (hda1=winxp, hda3=linux)

[Ajay Sharma]

On Fri, 7 Mar 2003, Gary Davis wrote:

> I have almost exactly the same setup.  What you do is follow the grub
> configure in the Installation Instructions and simply put (hd0,2)
> wherever it lists (hd0,0).  Everything else should remain the same,
> including setup (hd0) which puts grub in the MBR.

Arg, now you've got me thinking again...  when I run grub and type in
"root (hd0,2)", it responds with "unrecognized file system".  Then if I 
type in "setup (hd0)" after that it says something like "can't mount 
filesystem".  I tried it in all kinds of configurations and the only 
thing that worked was:

root (hd0,2)
setup (hd0,2)

Maybe it can't write to the bootloader since the Windows XP one is 
already there?  The last time I installed my sistem, I installed 
gentoo/grub first, then Windows XP second.   

later,
ajay


Satyajot (Ajay) Sharma 
[EMAIL PROTECTED]



--
[EMAIL PROTECTED] mailing list

RE: [gentoo-user] grub help (hda1=winxp, hda3=linux)

[Ajay Sharma]

Just so I can wrap up this thread for the archives...   

I got it working with Kevin's method (well, his description of someone
else's method).  I know I had it working before with just grub but for
the life of me I can't remember what I did.  Anyway, thanks for all your
help!

later,
ajay

On Thu, 6 Mar 2003, Kevin J. Anderson wrote:

> You could also install grub into your root partition, use dd to copy the
> first 512k out into a file, transfer that file to your xp c:\ root, and then
> use XP's boot.ini to give you the option to boot linux instead of xp.  This
> is the way I have always done dual boots w/ nt/2k/xp and I find it to be
> much less "dangerous" esp since windows seems to like the situtation much
> better.
> 
> I think setup (hd0,2)  root (hd0,2) would be the grub commands (please
> correct me if I am wrong)
> 
> then you can use this command in linux dd if=/dev/hda3 of=/bootsect.lnx
> bs=512 count=1
> transfer that file somehow over to your c:\ (floppy, ftp it somewhere else,
> bring it back down, etc)
> add c:\Bootsect.lnx="Gentoo Linux" to your boot.ini
> 
> google for "boot.ini dd linux dual boot" for more info on the subject, as my
> explanation is very cut and dry.
> 
> kev
> 
> 
> 
> ->-Original Message-
> ->From: Jeremy Workman [mailto:[EMAIL PROTECTED]
> ->Sent: Thursday, March 06, 2003 12:14 PM
> ->To: [EMAIL PROTECTED]
> ->Subject: Re: [gentoo-user] grub help (hda1=winxp, hda3=linux)
> ->
> ->
> ->I notice you said you tried "setup (hd0,0)". That would be the first
> ->partition on the first hard drive (/dev/hda1). To install Grub into the
> ->master boot record the command would be "setup (hd0)".
> ->
> ->Hope this helps.
> ->
> ->
> ->On Thu, 2003-03-06 at 13:04, Ajay Sharma wrote:
> ->> hey everyone,
> ->>
> ->> I just finished rebuilding my machine and I repartitioned my
> ->box so that
> ->> it looks like:
> ->>
> ->> /dev/hda1 = windows xp (ntfs)
> ->> /dev/hda2 = linux swap
> ->> /dev/hda3 = gentoo (ext3)
> ->>
> ->> no, I don't have a separate /boot partition.  Anyway, I followed the
> ->> installation instructions (excellent btw!) and I'm having problems
> ->> getting grub to install on my MBR.  In the grub prompt, when I type in
> ->> "root(hd0,0)" it would come back with something like "unrecognized file
> ->> type".  After that it's pretty obvious that the "setup(hd0,0)" wasn't
> ->> going to work.  I then ran root(hd0,2) and that went over fine but now
> ->> when I boot up it goes right into windows xp and not the grub boot
> ->> loader because it's not installed on the MBR.
> ->>
> ->> I'd appreciate any help.
> ->>
> ->> Thanks,
> ->> Ajay
> ->>
> ->> 
> ->> Satyajot (Ajay) Sharma
> ->> [EMAIL PROTECTED]
> ->> 
> ->>
> ->>
> ->> --
> ->> [EMAIL PROTECTED] mailing list
> ->>
> ->>
> ->
> ->
> ->--
> ->[EMAIL PROTECTED] mailing list
> ->
> 
> 
> --
> [EMAIL PROTECTED] mailing list
> 


Satyajot (Ajay) Sharma 
[EMAIL PROTECTED]



--
[EMAIL PROTECTED] mailing list

Re: [gentoo-user] Opinion Poll: KDE Vs Gnome

[Stanislav Karchebny]

> > Well, while we're all flopping it out in an unresolvable argument, I'd
> > just like to chip in about my favourite.. FVWM! (http://www.fvwm.org/)
> > , which as it happens is one of the oldest WM's still in common use.
>

OpenBox/Fluxbox and other *boxen are nice too. Some would say ion is better, 
because of lots of keyboard shortcuts. Dunno.

I stuck with KDE since month ago. My cpu power is enough for all its Xp-like 
bells and whistles, it also does rather tight integration of apps (e.g. sim 
can use Kaddressbook for its im contacts, cervisia and kbear just easily 
integrate into konqueror file view, i can do previews on files from konqueror 
and all this integrity is what i'm striving for).

So to choose, you'd better add some more characteristics that are important to 
you and decide on them. Good luck.

-- 
keep in touch. berk. -- http://odin-os.sf.net -- http://nasm.sf.net


--
[EMAIL PROTECTED] mailing list

Re: [gentoo-user] Portage errors

[Matthew Kennedy]

This may be a bug.

Ric Messier <[EMAIL PROTECTED]> writes:

> No luck. I'm still getting ACCESS VIOLATION errors attempting to
> open_wr /var/lib/scrollkeeper/scrollkeeper_docs

-- 
Matthew Kennedy
Gentoo Linux Developer
Bugs go to http://bugs.gentoo.org!


--
[EMAIL PROTECTED] mailing list

Re: [gentoo-user] "emerge clean" confusion

[Sami Näätänen]

On Friday 07 March 2003 08:17, Sami Näätänen wrote:
> On Thursday 06 March 2003 23:11, [EMAIL PROTECTED] wrote:
> > My guess would be Gentoo's support for multiple versions of gcc.
> > That would make it very difficult to determine what is removable
> > and what is not.
> >
> > * Toby Dickenson <[EMAIL PROTECTED]>
>
> [2003-03-06 20:30:10 +]:
> > > I understand emerge clean is supposed to remove old versions of a
> > > package, keeping one per slot. I cant understand why on this
> > > system it doesnt want to clean up the two old versions of gcc:
> > >
> > >
> > > vrumpet root # emerge clean gcc
> > >
> > >  sys-devel/gcc
> > > selected: none
> > >protected: 2.95.3-r5 2.95.3-r7 2.95.3-r8
> > >  omitted: none
> > >
> > > >>> clean: No packages selected for removal.
> > >
> > > Any ideas? Thanks in advance
>
> Some how portage wont clean old versions although the files in it
> have been overwriten. To original poster you can delete those older
> GCC's like I did.
>
> emerge -Cp \
> should show that it will remove the older versions.

And now to the ML. ;)

--
[EMAIL PROTECTED] mailing list

[gentoo-user] grub help (hda1=winxp, hda3=linux)

[Gary Davis]

I have almost exactly the same setup.  What you do is follow the grub configure in the Installation Instructions and simply put (hd0,2) wherever it lists (hd0,0).  Everything else should remain the same, including setup (hd0) which puts grub in the MBR.

Hope this helps,
-Gary
I just finished rebuilding my machine and I repartitioned my box so that 
it looks like:

/dev/hda1 = windows xp (ntfs)
/dev/hda2 = linux swap
/dev/hda3 = gentoo (ext3)
no, I don't have a separate /boot partition.  Anyway, I followed the
installation instructions (excellent btw!) and I'm having problems 
getting grub to install on my MBR.  In the grub prompt, when I type in 
"root(hd0,0)" it would come back with something like "unrecognized file 
type".  After that it's pretty obvious that the "setup(hd0,0)" wasn't 
going to work.  I then ran root(hd0,2) and that went over fine but now 
when I boot up it goes right into windows xp and not the grub boot 
loader because it's not installed on the MBR.


--
[EMAIL PROTECTED] mailing list

Re: [gentoo-user] Installation problem

[Joshua J. Berry]

Oh geez...did I send that encrypted?

Sorry. Encryption is on by default in my mail client...

Anyway, try using the 8139too driver...it seems to work for most Realtek-based 
cards.

On Thursday 06 March 2003 19:19, Brett I. Holcomb wrote:
> It maybe able to be loaded using modprobe modulename where modulename is
> the name of the module.  You'll have to find out what that is for you
> realtek card.
>
> > Hi all,
> >
> > I've been downloading the 2 latest releases of gentoo (rc2 and rc3) as it
> > was highly recommended to me by people around me.
> >
> > Unfortunately, the kernel that is included on the liveCD doesn't seem to
> > include the driver for my ethernet card (Realtek 8100B that comes with
> > Shuttle SS51G.) This driver comes on the debian potato 2.2 cd though...
> >
> > Is there a "simple" way to overcome this problem without recompiling a
> > kernel or anything similar. I was hoping that the kernel on the liveCD
> > would include as much hw support as possible in order to render the
> > installation task as easy as possible on various platforms. Am I missing
> > something here?
> >
> > Thanks for letting me know,
> >
> > LdS

-- 

-
Joshua J. Berry

"I haven't lost my mind -- it's backed up on tape somewhere."
-- /usr/games/fortune

PGP Key: http://deneb.condordes.net/node/16/view

NOTE: Please do not submit this email address to any mailing
lists or websites without prior permission.  Thank you.


pgp0.pgp
Description: signature

Re: [gentoo-user] Installation problem

[Joshua J. Berry]

bin0.bin
Description: version code


msg.asc
Description: encrypted data

[gentoo-user] Re: thai fonts?

[Steve Hutton]

In article <[EMAIL PROTECTED]>, Steve Hutton wrote:
> Has anyone installed thai fonts for gentoo?
> I'm having a hard time finding documentation
> on how to do this.
> 
In case anyone else is interested - I got this
working.  First I downloaded a hard to find rpm of
free thai fonts -
http://trolls.troll.no/lars/fonts/thai-ttf-0.1-1.noarch.rpm
then I:
- unpacked it using rpm2targz
- copied the *.ttf files to /usr/X11R6/lib/X11/fonts/truetype/

ttmkfdir -o fonts.scale
mkfontdir -e /usr/X11R6/lib/X11/fonts/encodings 

(restarted X)

Steve



--
[EMAIL PROTECTED] mailing list

[gentoo-user] Samba and Swat

[Matthew Daubenspeck]

I have Samba installed and working perfectly, but am having problems
getting Swat to work. I'm getting lazy these days and I would like to
see if Swat can save me any time...

Does it HAVE to be run from xinetd? I tried running the command by
itself, and I cannot connect to http://localhost:901/
-- 
:wq!

  Matthew Daubenspeck
  http://www.oddprocess.org

--
[EMAIL PROTECTED] mailing list

Re: [gentoo-user] Installation problem

[Brett I. Holcomb]

It maybe able to be loaded using modprobe modulename where modulename is the 
name of the module.  You'll have to find out what that is for you realtek 
card.

> Hi all,
>
> I've been downloading the 2 latest releases of gentoo (rc2 and rc3) as it
> was highly recommended to me by people around me.
>
> Unfortunately, the kernel that is included on the liveCD doesn't seem to
> include the driver for my ethernet card (Realtek 8100B that comes with
> Shuttle SS51G.) This driver comes on the debian potato 2.2 cd though...
>
> Is there a "simple" way to overcome this problem without recompiling a
> kernel or anything similar. I was hoping that the kernel on the liveCD
> would include as much hw support as possible in order to render the
> installation task as easy as possible on various platforms. Am I missing
> something here?
>
> Thanks for letting me know,
>
> LdS

-- 

Brett I. Holcomb
AKA Grunt <><

--
[EMAIL PROTECTED] mailing list

[gentoo-user] Installation problem

[Laurent de Segur]

Hi all,

I've been downloading the 2 latest releases of gentoo (rc2 and rc3) as it
was highly recommended to me by people around me.

Unfortunately, the kernel that is included on the liveCD doesn't seem to
include the driver for my ethernet card (Realtek 8100B that comes with
Shuttle SS51G.) This driver comes on the debian potato 2.2 cd though...

Is there a "simple" way to overcome this problem without recompiling a
kernel or anything similar. I was hoping that the kernel on the liveCD would
include as much hw support as possible in order to render the installation
task as easy as possible on various platforms. Am I missing something here?

Thanks for letting me know,

LdS



--
[EMAIL PROTECTED] mailing list

Re: [gentoo-user] qt compiles fail

[Oleg Letsinsky]

On Thu, Mar 06, 2003 at 02:05:02PM +, Joel Wright wrote:

> Okay, turns out libxft was the problem, i removed it and can now compile qt 
> apps happily again.
> 
> But what I want to know is what package installed libxft as a dependency in 
> the first place? (i.e. have I now broken something).
> 
> Still, at least I can upgrade my system again :)

Congratulations :) But IMO you'd better emerge x11-libs/xft (the package
which provides the libXft.so.2 library, so you can get antialiased fonts
in your compiled applications again - AA won't work w/o this library, as
I understand it.

-- 
  May the bluebird of happiness twiddle your bits.

--
[EMAIL PROTECTED] mailing list

Re: [gentoo-user] Building a compilation server

[Alec Berryman]

>  Now:
> - I don't like to install all packages on my compilation server.
> - I want it to compile all new packages automatically for all defined 
> platforms.
> 
> Any idea how I can set up such a beast?

I spent some time looking through the forums, but could not find the
script I was looking for - it would compile and make packages without
actually installing the programs.  You might have better luck.

If not, check out the man page for ebuild - you ought to be able to go
through the steps emerge automates and skip the actual installation.

-- 

Alec Berryman <[EMAIL PROTECTED]>


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

[gentoo-user] Winex problems

[Kirtis Bakalarczyk]

Hello,

I've been having some problems with winex that i'm not familiar with.  When  i try to 
start winex with any game i get the message: 

"wine: lstat /home/kirtis/.transgaming/wineserver-3jane/socket : No such file or 
directory"  

This is with Gentoo 1.4 and XFree 4.3.. Other than that there's no "unstable" ebuilds 
on my system.

I've checked, and it's true that socket isn't there so i tried wiping out my 
.transgaming directory to no avail.  I'm really not sure how to solve this one.. Can 
anyone shed some light on this?

KIRT

-- 
Perception != Reality
http://csplinux.saultc.on.ca/~kbakalar/index.htm

--
[EMAIL PROTECTED] mailing list

Re: [gentoo-user] dragging and dropping on kicker

[nealbirch]

Robert Cole wrote:
So absolutely NO ONE has tried to drag an application icon from the
KDE 3.1 desktop to the kicker toolbar and insert a new app? I really
find this hard to believe. Why am I being ignored anyway? No one does
this type of thing?


Sorry, but I have never done that, and I have been using kde for years.
Maybe I misunderstand the question, though; if I want to add an item to
the kick bar, I just right click on it.
neal



--
[EMAIL PROTECTED] mailing list

[gentoo-user] Building a compilation server

[Thomas Schweikle]

Hi!

I would like to build a compilation server for gentoo. I have to maintain 
some clients. Compilation takes time. Installing binaries is fast

 Now:
- I don't like to install all packages on my compilation server.
- I want it to compile all new packages automatically for all defined 
platforms.

Any idea how I can set up such a beast?

-- 
Thomas


--
[EMAIL PROTECTED] mailing list

Re: [gentoo-user] XFree Unmasked?

[Mike Williams]

On Thu, 2003-03-06 at 10:16, Paul de Vrieze wrote:
> The portage manpage is allways lagging behind on the features. Even the help 
> funcion is, although that is most times less behind. The --help function can 
> be seen as the authorative source of information about portage though.

Not trying to sound like a moan, but --help should contain a *short* (i.e. 1 
line) description, and the man page the long description. Same goes for init 
scripts.
I hate getting spamed with 5-6 pages of information when all I want to know is 
the options available :)

IMO the documentation is excellent (when fully upto date), just slightly in the
wrong place.

-- 
Mike Williams <[EMAIL PROTECTED]>


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

Re: [gentoo-user] Installing unstable/testing packages

[Norberto Bensa]

On Thursday 06 March 2003 08:08 pm, Louis C. Candell wrote:
> Jason Giangrande <[EMAIL PROTECTED]> writes:
> > What's the difference between the two?
>
> One of them ( the one I suggested ) lets you unmask individual ebuilds as
> opposed to ACCEPT_KEY... which unmasks everything which is masked.

Hmm, you're doing it the hard way, no doubts :-)

# ACCEPT_KEYWORDS="~x86" emerge whateverpackage


Regards,
Norberto

pgp0.pgp
Description: signature

Re: [gentoo-user] quicktime movies?

[Greg]

On Mon, 2003-03-03 at 22:24, Susie wrote:

> That is masked for x86 and ppc I think.  It also depends on a version of
> mplayer that seems to also be masked.

Well I figured I'd try the plugin for mozilla. Installed the mplayer to
go with it as well without troubles by:

ACCEPT_KEYWORDS="~x86" emerge mplayerplug-in

and installed:

net-www/mplayerplug-in-0.40
media-video/mplayer-0.90_rc4

mplayer and the plugin are both working fine. I did have to edit the
/etc/mplayerplug-in.conf and set use-gui option to mini to get the video
to embed rather than play full screen. 

Greg


--
[EMAIL PROTECTED] mailing list

Re: [gentoo-user] Installing unstable/testing packages

[Mike Williams]

On Thu, 2003-03-06 at 23:08, Louis C. Candell wrote:
> > What's the difference between the two?
> 
> One of them ( the one I suggested ) lets you unmask individual ebuilds as opposed to 
> ACCEPT_KEY... which unmasks everything which is masked. Be warned, some
> ebuilds will not work without some tweaking, so do not use unless you are
> capable of fixing the dependencies / problems on your own.

And be prepared to change the KEYWORD's everytime you sync...

-- 
Mike Williams <[EMAIL PROTECTED]>


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

[gentoo-user] Upgrade to X4.3, keyboard layout chaged.

[Norberto Bensa]

Hello,

I've upgraded to X 4.3 (from 4.2.1) and now my keyboard has changed its 
layout. I use Latin American layout, most of the keys work, but @ and euro 
don't.

@ is now located in AltGr+2 (before upgrade it was AltGr+Q.)
I can't get euro (it was AltGr+E.)  

This is the section of XF86Config:

Section "InputDevice"
Identifier  "Keyboard0"
Driver  "keyboard"
Option  "AutoRepeat""250 30"
Option  "XkbModel"  "microsoftpro"
Option  "XkbLayout" "la"
EndSection


Oh btw,

I really hate when someone breaks something that WAS working.


Thanks for any advice,
Norberto



pgp0.pgp
Description: signature

Re: [gentoo-user] Installing unstable/testing packages

[Ajay Sharma]

On Thu, 6 Mar 2003, Jason Giangrande wrote:

> If I install some ebuilds from the unstable/testing group using
> "ACCEPT_KEYWORDS="~x86" can I then go back and remove the
> accept_keywords setting and install the stable package builds or will
> this cause a problem?  If this is a problem, will I be able to install
> older ebuilds (say apache 1.3.27 instead of 2.0.44) if I normally use
> the "ACCEPT_KEYWORDS="~x86" option when installing packages?

switching back and forth between unstable/stable is alright for a few 
packages but I wouldn't suggest building your entire system as 
'unstable' and then try to go back.  But upgrading and downgrading stuff 
like apache shouldn't be a problem.

later,
ajay


Satyajot (Ajay) Sharma 
[EMAIL PROTECTED]



--
[EMAIL PROTECTED] mailing list

Re: [gentoo-user] Installing unstable/testing packages

[Louis C. Candell]

Jason Giangrande <[EMAIL PROTECTED]> writes:

> What's the difference between the two?

One of them ( the one I suggested ) lets you unmask individual ebuilds as opposed to 
ACCEPT_KEY... which unmasks everything which is masked. Be warned, some
ebuilds will not work without some tweaking, so do not use unless you are
capable of fixing the dependencies / problems on your own.

> 
> Jason
> 
> Louis C. Candell wrote:
> 
> >Jason Giangrande <[EMAIL PROTECTED]> writes:
> >
> >  
> >>If I install some ebuilds from the unstable/testing group using
> >>"ACCEPT_KEYWORDS="~x86" can I then go back and remove the
> >>accept_keywords setting and install the stable package builds or will
> >>this cause a problem?  If this is a problem, will I be able to install
> >>older ebuilds (say apache 1.3.27 instead of 2.0.44) if I normally use
> >>the "ACCEPT_KEYWORDS="~x86" option when installing packages?
> >>
> >
> >I dont know if I do it the hard way, but I just open up a masked ebuild and
> >change
> >
> >KEYWORDS="~x86"
> >to
> >KEYWORDS="x86"
> >
> >I suggest using the above method instead of using ACCEPT_KEYWORDS="~x86"
> >
> >  
> >>Thanks,
> >> -- 
> >>-Jason Giangrande
> >>  giangrande.org - http://www.giangrande.org 
> >>  Dog's I View - http://www.dogsiview.com 
> >>
> >>
> >>--
> >>[EMAIL PROTECTED] mailing list
> >>
> >>
> >>
> >
> >  
> 
> -- 
> -Jason Giangrande
>   giangrande.org - http://www.giangrande.org 
>   Dog's I View - http://www.dogsiview.com 
> 
> 
> --
> [EMAIL PROTECTED] mailing list
> 
> 

-- 
Louis C. Candell

--
[EMAIL PROTECTED] mailing list

Re: [gentoo-user] Installing unstable/testing packages

[Jason Giangrande]

What's the difference between the two?

Jason

Louis C. Candell wrote:

Jason Giangrande <[EMAIL PROTECTED]> writes:

 

If I install some ebuilds from the unstable/testing group using
"ACCEPT_KEYWORDS="~x86" can I then go back and remove the
accept_keywords setting and install the stable package builds or will
this cause a problem?  If this is a problem, will I be able to install
older ebuilds (say apache 1.3.27 instead of 2.0.44) if I normally use
the "ACCEPT_KEYWORDS="~x86" option when installing packages?
   

I dont know if I do it the hard way, but I just open up a masked ebuild and
change
KEYWORDS="~x86"
to
KEYWORDS="x86"
I suggest using the above method instead of using ACCEPT_KEYWORDS="~x86"

 

Thanks,
--
-Jason Giangrande
 giangrande.org - http://www.giangrande.org 
 Dog's I View - http://www.dogsiview.com 
--
[EMAIL PROTECTED] mailing list
   

 

--
-Jason Giangrande
 giangrande.org - http://www.giangrande.org 
 Dog's I View - http://www.dogsiview.com 
--
[EMAIL PROTECTED] mailing list

Re: [gentoo-user] Installing unstable/testing packages

[Louis C. Candell]

Jason Giangrande <[EMAIL PROTECTED]> writes:

> If I install some ebuilds from the unstable/testing group using
> "ACCEPT_KEYWORDS="~x86" can I then go back and remove the
> accept_keywords setting and install the stable package builds or will
> this cause a problem?  If this is a problem, will I be able to install
> older ebuilds (say apache 1.3.27 instead of 2.0.44) if I normally use
> the "ACCEPT_KEYWORDS="~x86" option when installing packages?

I dont know if I do it the hard way, but I just open up a masked ebuild and
change

KEYWORDS="~x86"
to
KEYWORDS="x86"

I suggest using the above method instead of using ACCEPT_KEYWORDS="~x86"

> 
> Thanks,
> -- 
> -Jason Giangrande
>   giangrande.org - http://www.giangrande.org 
>   Dog's I View - http://www.dogsiview.com 
> 
> 
> --
> [EMAIL PROTECTED] mailing list
> 
> 

-- 
Louis C. Candell

--
[EMAIL PROTECTED] mailing list

Re: [gentoo-user] Warning to hardware purchasers

[Ted Goodridge, Jr]

On Thu, 6 Mar 2003 17:44:39 -0500, Ernie Schroder <[EMAIL PROTECTED]> 
wrote:

On Thursday 06 March 2003 14:45, [EMAIL PROTECTED] wrote:
Quoting Ernie Schroder <[EMAIL PROTECTED]>:
>I would whole heartedly recommend Monarch to anyone in the market
> for computer hardware. Sorry about the commercial here, I realize
> that this is OT but I would hate for anyone else to go through the
> garbage I endured.
Newegg is also excellent in this regard (I've returned two items and
they were very gracious and accomodating in the process).  CDW,
though expensive, was also very good with the return when my Firegl
X1 wouldn't work with the binary Linux drivers ATI ships.
Perhaps we should have a hardware rating page, aimed at Gentoo (or
perhaps more generally GNU/Linux) folks, where one can check against
other's experiences before purchasing from a particular vendor.
Jean.
I'd vote for that... I could have saved myself a lot of hastle had I 
consulted a page like that instead of shopping for price alone.
There is a site, www.resellerratings.com that rates computer hardware 
dealers.
Tufshop.com;
http://www.resellerratings.com/seller1924.html
Monarchcomputer.com:
http://www.resellerratings.com/seller_info.pl?seller_id=2079
newegg.com:
http://www.resellerratings.com/seller2121.html
and CDW:
http://www.resellerratings.com/seller1971.html
also multiwave
http://www.mwave.com
compuplus
http://www.compuplus.com
They are much much cheaper than CDW and the big boys, and they have 
excellent service.

Ted

--
Using M2, Opera's revolutionary e-mail client: http://www.opera.com/m2/
--
[EMAIL PROTECTED] mailing list

[gentoo-user] Installing unstable/testing packages

[Jason Giangrande]

If I install some ebuilds from the unstable/testing group using 
"ACCEPT_KEYWORDS="~x86" can I then go back and remove the 
accept_keywords setting and install the stable package builds or will 
this cause a problem?  If this is a problem, will I be able to install 
older ebuilds (say apache 1.3.27 instead of 2.0.44) if I normally use 
the "ACCEPT_KEYWORDS="~x86" option when installing packages?

Thanks,
--
-Jason Giangrande
 giangrande.org - http://www.giangrande.org 
 Dog's I View - http://www.dogsiview.com 
--
[EMAIL PROTECTED] mailing list

Re: [gentoo-user] Warning to hardware purchasers

[Ernie Schroder]

On Thursday 06 March 2003 14:45, [EMAIL PROTECTED] wrote:
> Quoting Ernie Schroder <[EMAIL PROTECTED]>:
> > I would whole heartedly recommend Monarch to anyone in the market
> > for computer hardware. Sorry about the commercial here, I realize
> > that this is OT but I would hate for anyone else to go through the
> > garbage I endured.
>
> Newegg is also excellent in this regard (I've returned two items and
> they were very gracious and accomodating in the process).  CDW,
> though expensive, was also very good with the return when my Firegl
> X1 wouldn't work with the binary Linux drivers ATI ships.
>
> Perhaps we should have a hardware rating page, aimed at Gentoo (or
> perhaps more generally GNU/Linux) folks, where one can check against
> other's experiences before purchasing from a particular vendor.
>
> Jean.

I'd vote for that... I could have saved myself a lot of hastle had I 
consulted a page like that instead of shopping for price alone.
There is a site, www.resellerratings.com that rates computer hardware 
dealers.
Tufshop.com;
http://www.resellerratings.com/seller1924.html
Monarchcomputer.com:
http://www.resellerratings.com/seller_info.pl?seller_id=2079
newegg.com:
http://www.resellerratings.com/seller2121.html
and CDW:
http://www.resellerratings.com/seller1971.html
-- 
Regards, Ernie
100% Microsoft and Intel free

--
[EMAIL PROTECTED] mailing list

Re: [gentoo-user] Kernel Re-Compilation Question

[Doug Gorley]

Alrighty then; I'll experiment with it and post my results.

> Unfortunately, I'm not sure about this, as I have never used kernel
> module autoloading :-\ good luck! :)
>
> (My understanding is that this means that MANY module needs can be
> automatically detected and loaded and unlaoded as needed in this manner,
> but I'm not real clear on it)
>
> --Brandon
>
> On Thu, 03/06/03 at 14:12:53 -0800, Doug Gorley wrote:
>> Thanks; one more question before I go experiment.  If I've got
>>
>> CONFIG_KMOD=y
>>
>> in my .config file after configuring my kernel, how do I take
>> advantage of it?  Documentation/kmod.txt says to use the following
>> command to set the path to modprobe:
>>
>> echo "/sbin/modprobe" > /proc/sys/kernel/modprobe
>>
>> and the following crontab entry to remove unused modules.
>>
>> 0-59/5 * * * * /sbin/rmmod -a
>>
>> If I do this, do I still need to list my modules in
>> /etc/modules.autoload, or will kmod handle everything correctly?
>>
>> > Yeah, quite correct... been a while since I did two different
>> .configs same source that I wanted to use.. :)
>> >
>> > --B
>> >
>> > On Thu, 03/06/03 at 13:38:34 -0800, Doug Gorley wrote:
>> >> Thanks for your reply.
>> >>
>> >> So, not only is this OK, but it's a good idea if I want to
>> experiment with different kernels from the same source, yes?
>> Otherwise, if I compiled two kernels from the vanilla-sources,
>> wouldn't the modules from the second clobber those from the first?
>> >>
>> >> > The kernel 'knows' it's name, and all packages that install
>> kernel
>> >> modules also look at the same version that the kernel in
>> >> /usr/src/linux knows (/usr/src/linux/include/version.h (or
>> something of the sort)), so changing the name and recompiling works
>> just fine and dandy.
>> >> >
>> >> > --Brandon
>> >> >
>> >> > On Thu, 03/06/03 at 13:16:43 -0800, Doug Gorley wrote:
>> >> >> Good afternoon (PST) list,
>> >> >>
>> >> >> I'm trying to get a little more comfortable with kernel
>> >> configuration, and have just compiled a 2.4.20 kernel from the
>> vanilla-sources that uses modules wherever possible.  Following the
>> Kernel HOWTO at http://www.tldp.org/, I changed my Makefile to read
>> >> >>
>> >> >> EXTRAVERSION = -Carnage_2003-03-06
>> >> >>
>> >> >> Now, what I've noticed is that in /lib/modules, I now have
>> separate
>> >> directories called 2.4.20 and 2.4.20-Carnage_2003-03-06.  Will the
>> new kernel automatically know that it's modules are in the
>> >> directory bearing it's name, or do I now have to somehow tell my
>> kernel where to look?
>> >> >>
>> >> >> Thanks,
>> >> >>
>> >> >> --
>> >> >> Doug Gorley | [EMAIL PROTECTED]
>> >> >>
>> >> >>
>> >> >>
>> >> >>
>> >> >> --
>> >> >> [EMAIL PROTECTED] mailing list
>> >> >
>> >> > --
>> >> > [EMAIL PROTECTED] mailing list
>> >>
>> >>
>> >> --
>> >> Doug Gorley | [EMAIL PROTECTED]
>> >>
>> >>
>> >
>> > --
>> > [EMAIL PROTECTED] mailing list
>>
>>
>> --
>> Doug Gorley | [EMAIL PROTECTED]
>>
>>
>
> --
> [EMAIL PROTECTED] mailing list


-- 
Doug Gorley | [EMAIL PROTECTED]




--
[EMAIL PROTECTED] mailing list

Re: [gentoo-user] Kernel Re-Compilation Question

[Brandon Low]

Unfortunately, I'm not sure about this, as I have never used kernel module autoloading 
:-\ good luck! :)

(My understanding is that this means that MANY module needs can be automatically 
detected and loaded and unlaoded as needed in this manner, but I'm not real clear on 
it)

--Brandon

On Thu, 03/06/03 at 14:12:53 -0800, Doug Gorley wrote:
> Thanks; one more question before I go experiment.  If I've got
> 
> CONFIG_KMOD=y
> 
> in my .config file after configuring my kernel, how do I take advantage of
> it?  Documentation/kmod.txt says to use the following command to set the
> path to modprobe:
> 
> echo "/sbin/modprobe" > /proc/sys/kernel/modprobe
> 
> and the following crontab entry to remove unused modules.
> 
> 0-59/5 * * * * /sbin/rmmod -a
> 
> If I do this, do I still need to list my modules in /etc/modules.autoload,
> or will kmod handle everything correctly?
> 
> > Yeah, quite correct... been a while since I did two different .configs
> > same source that I wanted to use.. :)
> >
> > --B
> >
> > On Thu, 03/06/03 at 13:38:34 -0800, Doug Gorley wrote:
> >> Thanks for your reply.
> >>
> >> So, not only is this OK, but it's a good idea if I want to experiment
> >> with different kernels from the same source, yes?  Otherwise, if I
> >> compiled two kernels from the vanilla-sources, wouldn't the modules
> >> from the second clobber those from the first?
> >>
> >> > The kernel 'knows' it's name, and all packages that install kernel
> >> modules also look at the same version that the kernel in
> >> /usr/src/linux knows (/usr/src/linux/include/version.h (or something
> >> of the sort)), so changing the name and recompiling works just fine
> >> and dandy.
> >> >
> >> > --Brandon
> >> >
> >> > On Thu, 03/06/03 at 13:16:43 -0800, Doug Gorley wrote:
> >> >> Good afternoon (PST) list,
> >> >>
> >> >> I'm trying to get a little more comfortable with kernel
> >> configuration, and have just compiled a 2.4.20 kernel from the
> >> vanilla-sources that uses modules wherever possible.  Following the
> >> Kernel HOWTO at http://www.tldp.org/, I changed my Makefile to read
> >> >>
> >> >> EXTRAVERSION = -Carnage_2003-03-06
> >> >>
> >> >> Now, what I've noticed is that in /lib/modules, I now have separate
> >> directories called 2.4.20 and 2.4.20-Carnage_2003-03-06.  Will the
> >> new kernel automatically know that it's modules are in the
> >> directory bearing it's name, or do I now have to somehow tell my
> >> kernel where to look?
> >> >>
> >> >> Thanks,
> >> >>
> >> >> --
> >> >> Doug Gorley | [EMAIL PROTECTED]
> >> >>
> >> >>
> >> >>
> >> >>
> >> >> --
> >> >> [EMAIL PROTECTED] mailing list
> >> >
> >> > --
> >> > [EMAIL PROTECTED] mailing list
> >>
> >>
> >> --
> >> Doug Gorley | [EMAIL PROTECTED]
> >>
> >>
> >
> > --
> > [EMAIL PROTECTED] mailing list
> 
> 
> -- 
> Doug Gorley | [EMAIL PROTECTED]
> 
> 

--
[EMAIL PROTECTED] mailing list

Re: [gentoo-user] etc-update

[Brandon Low]

Check bugs.gentoo.org for more information on the new dispatch-conf script, post any 
etc-update type feature requests there, as dispatch conf is currently slated to 
replace etc-update pending matching etc-update's feature set.  Since dispatch-conf is 
written in python, it has much more flexibility looking forward in terms of being able 
to integrate into portage, and to have various UIs integrated.  

--Brandon

On Thu, 03/06/03 at 11:57:05 -0500, Jason Giangrande wrote:
> Is there a way to use etc-update to update all config files at once, 
> either merge or replace?  The manpage didn't have anything and doing 
> them one at a time is a huge pain.
> 
> Thanks,
> -- 
> -Jason Giangrande
>  giangrande.org - http://www.giangrande.org 
>  Dog's I View - http://www.dogsiview.com 
> 
> 
> --
> [EMAIL PROTECTED] mailing list

--
[EMAIL PROTECTED] mailing list

Re: [gentoo-user] Kernel Re-Compilation Question

[Doug Gorley]

Thanks; one more question before I go experiment.  If I've got

CONFIG_KMOD=y

in my .config file after configuring my kernel, how do I take advantage of
it?  Documentation/kmod.txt says to use the following command to set the
path to modprobe:

echo "/sbin/modprobe" > /proc/sys/kernel/modprobe

and the following crontab entry to remove unused modules.

0-59/5 * * * * /sbin/rmmod -a

If I do this, do I still need to list my modules in /etc/modules.autoload,
or will kmod handle everything correctly?

> Yeah, quite correct... been a while since I did two different .configs
> same source that I wanted to use.. :)
>
> --B
>
> On Thu, 03/06/03 at 13:38:34 -0800, Doug Gorley wrote:
>> Thanks for your reply.
>>
>> So, not only is this OK, but it's a good idea if I want to experiment
>> with different kernels from the same source, yes?  Otherwise, if I
>> compiled two kernels from the vanilla-sources, wouldn't the modules
>> from the second clobber those from the first?
>>
>> > The kernel 'knows' it's name, and all packages that install kernel
>> modules also look at the same version that the kernel in
>> /usr/src/linux knows (/usr/src/linux/include/version.h (or something
>> of the sort)), so changing the name and recompiling works just fine
>> and dandy.
>> >
>> > --Brandon
>> >
>> > On Thu, 03/06/03 at 13:16:43 -0800, Doug Gorley wrote:
>> >> Good afternoon (PST) list,
>> >>
>> >> I'm trying to get a little more comfortable with kernel
>> configuration, and have just compiled a 2.4.20 kernel from the
>> vanilla-sources that uses modules wherever possible.  Following the
>> Kernel HOWTO at http://www.tldp.org/, I changed my Makefile to read
>> >>
>> >> EXTRAVERSION = -Carnage_2003-03-06
>> >>
>> >> Now, what I've noticed is that in /lib/modules, I now have separate
>> directories called 2.4.20 and 2.4.20-Carnage_2003-03-06.  Will the
>> new kernel automatically know that it's modules are in the
>> directory bearing it's name, or do I now have to somehow tell my
>> kernel where to look?
>> >>
>> >> Thanks,
>> >>
>> >> --
>> >> Doug Gorley | [EMAIL PROTECTED]
>> >>
>> >>
>> >>
>> >>
>> >> --
>> >> [EMAIL PROTECTED] mailing list
>> >
>> > --
>> > [EMAIL PROTECTED] mailing list
>>
>>
>> --
>> Doug Gorley | [EMAIL PROTECTED]
>>
>>
>
> --
> [EMAIL PROTECTED] mailing list


-- 
Doug Gorley | [EMAIL PROTECTED]




--
[EMAIL PROTECTED] mailing list

Re: [gentoo-user] NIS: exporting groups?

[Harald Kümmerle]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

On Thursday 06 March 2003 01:52, you wrote:
> Have a looking /var/yp/Makefile and adjust the MINGID setting which
> tells YP what is the minimum group ID to export.

Thanks, now everything works.
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.0.7 (GNU/Linux)

iD8DBQE+Z8dLpy/Gjo6hsHYRArL5AJ9ENsEjr6nDKk1hcaC2Sm1ujm+4PgCeLv5w
gYuefzztpdAWNCYTQEAHTjc=
=3Upx
-END PGP SIGNATURE-


--
[EMAIL PROTECTED] mailing list

[gentoo-user] gnome-base/gconf-2.2.0 failed.

[Voicu Liviu]

Any one knows why does this failes?
Thanks in advance,
Liviu


mkdir /var/tmp/portage/gconf-2.2.0/image/usr/share/gtk-doc
mkdir /var/tmp/portage/gconf-2.2.0/image/usr/share/gtk-doc/html
mkdir /var/tmp/portage/gconf-2.2.0/image/usr/share/gtk-doc/html/gconf
(installfiles=`echo ./html/*`; \
if test "$installfiles" = './html/*'; \
then echo '-- Nothing to install' ; \
else \
  for i in $installfiles; do \
echo '-- Installing '$i ; \
/bin/install -c -m 644 $i 
/var/tmp/portage/gconf-2.2.0/image//usr/share/gtk-
doc/html/gconf; \
  done; \
  echo '-- Installing ./html/index.sgml' ; \
  /bin/install -c -m 644 ./html/index.sgml 
/var/tmp/portage/gconf-2.2.0/image//u
sr/share/gtk-doc/html/gconf; \
fi)
-- Installing ./html/home.png
-- Installing ./html/left.png
-- Installing ./html/right.png
-- Installing ./html/up.png
-- Installing ./html/index.sgml
/bin/install: cannot stat `./html/index.sgml': No such file or directory
make[3]: *** [install-data-local] Error 1
make[3]: Leaving directory 
`/var/tmp/portage/gconf-2.2.0/work/GConf-2.2.0/doc/gc
onf'
make[2]: *** [install-am] Error 2
make[2]: Leaving directory 
`/var/tmp/portage/gconf-2.2.0/work/GConf-2.2.0/doc/gc
onf'
make[1]: *** [install-recursive] Error 1
make[1]: Leaving directory `/var/tmp/portage/gconf-2.2.0/work/GConf-2.2.0/doc'
make: *** [install-recursive] Error 1

!!! ERROR: gnome-base/gconf-2.2.0 failed.
!!! Function einstall, Line 278, Exitcode 2
!!! einstall failed



--
[EMAIL PROTECTED] mailing list

Re: [gentoo-user] Kernel Re-Compilation Question

[Brandon Low]

Yeah, quite correct... been a while since I did two different .configs same source 
that I wanted to use.. :)

--B

On Thu, 03/06/03 at 13:38:34 -0800, Doug Gorley wrote:
> Thanks for your reply.
> 
> So, not only is this OK, but it's a good idea if I want to experiment with
> different kernels from the same source, yes?  Otherwise, if I compiled two
> kernels from the vanilla-sources, wouldn't the modules from the second
> clobber those from the first?
> 
> > The kernel 'knows' it's name, and all packages that install kernel
> > modules also look at the same version that the kernel in /usr/src/linux
> > knows (/usr/src/linux/include/version.h (or something of the sort)), so
> > changing the name and recompiling works just fine and dandy.
> >
> > --Brandon
> >
> > On Thu, 03/06/03 at 13:16:43 -0800, Doug Gorley wrote:
> >> Good afternoon (PST) list,
> >>
> >> I'm trying to get a little more comfortable with kernel configuration,
> >> and have just compiled a 2.4.20 kernel from the vanilla-sources that
> >> uses modules wherever possible.  Following the Kernel HOWTO at
> >> http://www.tldp.org/, I changed my Makefile to read
> >>
> >> EXTRAVERSION = -Carnage_2003-03-06
> >>
> >> Now, what I've noticed is that in /lib/modules, I now have separate
> >> directories called 2.4.20 and 2.4.20-Carnage_2003-03-06.  Will the new
> >> kernel automatically know that it's modules are in the directory
> >> bearing it's name, or do I now have to somehow tell my kernel where to
> >> look?
> >>
> >> Thanks,
> >>
> >> --
> >> Doug Gorley | [EMAIL PROTECTED]
> >>
> >>
> >>
> >>
> >> --
> >> [EMAIL PROTECTED] mailing list
> >
> > --
> > [EMAIL PROTECTED] mailing list
> 
> 
> -- 
> Doug Gorley | [EMAIL PROTECTED]
> 
> 

--
[EMAIL PROTECTED] mailing list

Re: [gentoo-user] OT hardware ?

[Alan]

> I guess I can expect to see a pretty healthy increase in speed with the 
> UDMA/133. If there was a great advantege to the larger cache, I'd be 
> inclined to go that way, after all, who needs 120 gigs? Of course 2 
> years ago, I asked who needs 20 gigs?

If you're interested in gobs of space you can always do like this guy did 
(link from /. a couple of days ago):
http://anthrax.ds.pg.gda.pl/server/
-- 
Alan <[EMAIL PROTECTED]> - http://arcterex.net
-
"The only thing that experience teaches us is that experience teaches 
us nothing. -- Andre Maurois (Emile Herzog)

--
[EMAIL PROTECTED] mailing list

Re: [gentoo-user] Kernel Re-Compilation Question

[Doug Gorley]

Thanks for your reply.

So, not only is this OK, but it's a good idea if I want to experiment with
different kernels from the same source, yes?  Otherwise, if I compiled two
kernels from the vanilla-sources, wouldn't the modules from the second
clobber those from the first?

> The kernel 'knows' it's name, and all packages that install kernel
> modules also look at the same version that the kernel in /usr/src/linux
> knows (/usr/src/linux/include/version.h (or something of the sort)), so
> changing the name and recompiling works just fine and dandy.
>
> --Brandon
>
> On Thu, 03/06/03 at 13:16:43 -0800, Doug Gorley wrote:
>> Good afternoon (PST) list,
>>
>> I'm trying to get a little more comfortable with kernel configuration,
>> and have just compiled a 2.4.20 kernel from the vanilla-sources that
>> uses modules wherever possible.  Following the Kernel HOWTO at
>> http://www.tldp.org/, I changed my Makefile to read
>>
>> EXTRAVERSION = -Carnage_2003-03-06
>>
>> Now, what I've noticed is that in /lib/modules, I now have separate
>> directories called 2.4.20 and 2.4.20-Carnage_2003-03-06.  Will the new
>> kernel automatically know that it's modules are in the directory
>> bearing it's name, or do I now have to somehow tell my kernel where to
>> look?
>>
>> Thanks,
>>
>> --
>> Doug Gorley | [EMAIL PROTECTED]
>>
>>
>>
>>
>> --
>> [EMAIL PROTECTED] mailing list
>
> --
> [EMAIL PROTECTED] mailing list


-- 
Doug Gorley | [EMAIL PROTECTED]




--
[EMAIL PROTECTED] mailing list

Re: [gentoo-user] Kernel Re-Compilation Question

[Brandon Low]

The kernel 'knows' it's name, and all packages that install kernel modules also look 
at the same version that the kernel in /usr/src/linux knows 
(/usr/src/linux/include/version.h (or something of the sort)), so changing the name 
and recompiling works just fine and dandy.

--Brandon

On Thu, 03/06/03 at 13:16:43 -0800, Doug Gorley wrote:
> Good afternoon (PST) list,
> 
> I'm trying to get a little more comfortable with kernel configuration, and
> have just compiled a 2.4.20 kernel from the vanilla-sources that uses
> modules wherever possible.  Following the Kernel HOWTO at
> http://www.tldp.org/, I changed my Makefile to read
> 
> EXTRAVERSION = -Carnage_2003-03-06
> 
> Now, what I've noticed is that in /lib/modules, I now have separate
> directories called 2.4.20 and 2.4.20-Carnage_2003-03-06.  Will the new
> kernel automatically know that it's modules are in the directory bearing
> it's name, or do I now have to somehow tell my kernel where to look?
> 
> Thanks,
> 
> -- 
> Doug Gorley | [EMAIL PROTECTED]
> 
> 
> 
> 
> --
> [EMAIL PROTECTED] mailing list

--
[EMAIL PROTECTED] mailing list

Re: [gentoo-user] OT hardware ?

[Ernie Schroder]

On Thursday 06 March 2003 14:46, Christian Herzyk wrote:
> Hi there,
>
> Am Donnerstag, 6. März 2003 20:48 schrieb Alan:
> > On Thu, Mar 06, 2003 at 02:41:20PM -0500, Ernie Schroder wrote:
> > > I'm getting ready to take Gentoo_2 off line and replace mobo and
> > > processor I was looking at hdd's this morning and saw Western
> > > Digital 80 and 120 gig udma 133 drives at both the same pricethe
> > > difference is that the 80 gig has 8 megs of cache and the 120
> > > 2gigs.both claim a seek

 ooops 2 megs
> >
> >   ^ Gigs???
> >
> > > time of 9.1ms. could some one voice an opinion here.(as if I have
> > > to look far on this list for an opinion :) ) Which would you buy?
> >
> > More cache is always better performance, but if it means you can
> > get the extra 40G  Kinda hard to say.  I got myself some of the
> > WD 80G/8m drives a couple of months ago and have been quite happy. 
> > There was a bigger price difference between the 80 and 120 at that
> > point though :)
> >
> > Tough decision, I guess it really comes down to space or
> > performance. Granted, I can't give you quantitative evidence of the
> > performance of the 8m vs 2m cache.
>
> I have the WD 80G drive with 2MB Cache, so I will just post my
> values.
>
> > My hdparm results:
> >
> > phoenix alan # hdparm /dev/hdf
> >
> > /dev/hdf:
> >  multcount= 16 (on)
> >  IO_support   =  0 (default 16-bit)
> >  unmaskirq=  0 (off)
> >  using_dma=  1 (on)
> >  keepsettings =  0 (off)
> >  readonly =  0 (off)
> >  readahead=  8 (on)
> >  geometry = 155061/16/63, sectors = 156301488, start = 0
>
> /dev/hda:/dev/hda:
 multcount=  8 (on)
 IO_support   =  0 (default 16-bit)
 unmaskirq=  0 (off)
 using_dma=  1 (on)
 keepsettings =  0 (off)
 readonly =  0 (off)
 readahead=  8 (on)
 geometry = 1247/255/63, sectors = 20044080, start = 0
Ernie root # hdparm -Tt /dev/hda

/dev/hda:
 Timing buffer-cache reads:   128 MB in  0.86 seconds =148.84 MB/sec
 Timing buffered disk reads:  64 MB in  3.13 seconds = 20.45 MB/sec
>  multcount= 16 (on)
>  IO_support   =  1 (32-bit)
>  unmaskirq=  1 (on)
>  using_dma=  1 (on)
>  keepsettings =  0 (off)
>  readonly =  0 (off)
>  readahead=  8 (on)
>  geometry = 9729/255/63, sectors = 156301488, start = 0
>
> > phoenix alan # hdparm -Tt /dev/hdf
> > /dev/hdf:
> >  Timing buffer-cache reads:   128 MB in  0.80 seconds =159.60
> > MB/sec Timing buffered disk reads:  64 MB in  2.07 seconds = 30.86
> > MB/sec
>
> /dev/hda:
>  Timing buffer-cache reads:   128 MB in  0.50 seconds =256.00 MB/sec
>  Timing buffered disk reads:  64 MB in  2.04 seconds = 31.37 MB/sec
>
> Quite strange that my cache reads are much faster, probably it's teh
> controller.
>
> > phoenix alan # dmesg | grep hdf
> > hdf: WDC WD800JB-00CRA1, ATA DISK drive
> > hdf: 156301488 sectors (80026 MB) w/8192KiB Cache,
> > CHS=155061/16/63, UDMA(100)
>
> Kernel command line: root=/dev/hda8 hdd=ide-scsi
> ide0: BM-DMA at 0xd400-0xd407, BIOS settings: hda:DMA, hdb:DMA
> hda: WDC WD800BB-53BSA0, ATA DISK drive
> hda: 156301488 sectors (80026 MB) w/2048KiB Cache, CHS=9729/255/63,
> UDMA(100)
>
> So it seems that in this small "benchmark" the cache doesn't score
> too much perhaps there are some better things to run like bonnie or
> dbench.
>
> Greetings
>
> Christian

Seems that the larger cache doesn't make much difference.
This is my UDMA/66 WD 20 gig/2megs cache:

/dev/hda:
 multcount=  8 (on)
 IO_support   =  0 (default 16-bit)
 unmaskirq=  0 (off)
 using_dma=  1 (on)
 keepsettings =  0 (off)
 readonly =  0 (off)
 readahead=  8 (on)
 geometry = 1247/255/63, sectors = 20044080, start = 0
Ernie root # hdparm -Tt /dev/hda

/dev/hda:
 Timing buffer-cache reads:   128 MB in  0.86 seconds =148.84 MB/sec
 Timing buffered disk reads:  64 MB in  3.13 seconds = 20.45 MB/sec

I guess I can expect to see a pretty healthy increase in speed with the 
UDMA/133. If there was a great advantege to the larger cache, I'd be 
inclined to go that way, after all, who needs 120 gigs? Of course 2 
years ago, I asked who needs 20 gigs?
-- 
Regards, Ernie
100% Microsoft and Intel free

--
[EMAIL PROTECTED] mailing list

Re: [gentoo-user] OT hardware ?

[Daniel A. Segel]

> Which would you buy?

I'd go for the 120. There are other factors than cache that can have a
bigger impact on speed, and from what I've seen the speed difference
between a 2MB and 8MB cache is fairly small to begin with.

My WD 60GB drive with 2MB cache:

/home/daniel 2>hdparm /dev/hda

/dev/hda:
 multcount= 16 (on)
 IO_support   =  0 (default 16-bit)
 unmaskirq=  1 (on)
 using_dma=  1 (on)
 keepsettings =  0 (off)
 readonly =  0 (off)
 readahead= 64 (on)
 geometry = 7297/255/63, sectors = 117231408, start = 0

/home/daniel 3>hdparm -Tt /dev/hda

/dev/hda:
 Timing buffer-cache reads:   128 MB in  0.43 seconds =297.67 MB/sec
 Timing buffered disk reads:  64 MB in  1.38 seconds = 46.38 MB/sec



--
[EMAIL PROTECTED] mailing list

[gentoo-user] Kernel Re-Compilation Question

[Doug Gorley]

Good afternoon (PST) list,

I'm trying to get a little more comfortable with kernel configuration, and
have just compiled a 2.4.20 kernel from the vanilla-sources that uses
modules wherever possible.  Following the Kernel HOWTO at
http://www.tldp.org/, I changed my Makefile to read

EXTRAVERSION = -Carnage_2003-03-06

Now, what I've noticed is that in /lib/modules, I now have separate
directories called 2.4.20 and 2.4.20-Carnage_2003-03-06.  Will the new
kernel automatically know that it's modules are in the directory bearing
it's name, or do I now have to somehow tell my kernel where to look?

Thanks,

-- 
Doug Gorley | [EMAIL PROTECTED]




--
[EMAIL PROTECTED] mailing list

Re: [gentoo-user] rc3 install - /lib/cpp failure

[brett holcomb]

Yup, I saw this.  I did a -ncurses in the USE variable and 
it worked.  Finding that solution was due to a mistake.  I 
had filed a bug and gotten a reply that I thought said but 
--ncurses in USE so I did and it worked.  That wasn't what 
the person meant.  However, the bug was closed but I never 
got a response as to what was done to fix it.  I even 
emailed the bug maintainer.



On Thu, 6 Mar 2003 11:57:00 -0800
 [EMAIL PROTECTED] wrote:
This is my second gentoo install, and on the same 
machine. Last time, I
did a stage 1. Though it took quite some time to get it 
right. The
system got munged, and I decided to correct things by 
doing a fresh
install, a Stage 3 install. It's been going fine until...

section 14:

during 'emerge -u world', compiling of ncurses-5.3 (the 
first item)
failed with "/lib/cpp" fails sanity check.

Interestingly, during the compiling "checks"...

first it says we ARE using 'gcc', then later, it says no.

libc.h no

how to run the C++ preprocessor... /lib/cpp

configure: error: C++ preprocesssor "/lib/cpp" fails 
sanity check

!!! ERROR: sys-libs/ncurses-5.3-r1 failed.
!!! Function econf, Line 262, Exitcode 1
!!! econf failed
In irc channel #gentoo, it was suggested I do a 'emerge 
gcc-config'. I
did, to no avail.

This is very frustrating, because, upto this point, all 
that's really
been going on is simple hardware setup, with the 
exception of editing
/etc/make.conf.  So, to have a show stopping error right 
at the
beginning is bad.

Additionally, I'm not seeing any other reports of this 
problem from
anyone else.  What is it that _I_ could have done that's 
different from
everyone else?

The only place for _me_ to make a configuration error and 
get this kind of
problem at this point would be, I assume, in 
/etc/make.conf.  I've tried
numerous variations, including no changes, no P4, 
whatever.  I always
get the same error.

I just don't understand how this could be happening.

Any ideas?

Thank you.
--
- Martin J. Brown, Jr. -
- [EMAIL PROTECTED] -
  Public PGP Key ID: 0xDB184F7B keyserver: 
http://certserver.pgp.com/

--
[EMAIL PROTECTED] mailing list


--
[EMAIL PROTECTED] mailing list

Re: [gentoo-user] OT hardware ?

[brett holcomb]

Knowing only what you gave me here - I'd go for the bigger 
drive and get more storage space for the money.

On Thu, 6 Mar 2003 14:41:20 -0500
 Ernie Schroder <[EMAIL PROTECTED]> wrote:
I'm getting ready to take Gentoo_2 off line and replace 
mobo and 
processor I was looking at hdd's this morning and saw 
Western Digital 
80 and 120 gig udma 133 drives at both the same pricethe 
difference is 
that the 80 gig has 8 megs of cache and the 120 
2gigs.both claim a seek 
time of 9.1ms. could some one voice an opinion here.(as 
if I have to 
look far on this list for an opinion :) ) Which would you 
buy?
--
Regards, Ernie
100% Microsoft and Intel free

--
[EMAIL PROTECTED] mailing list


--
[EMAIL PROTECTED] mailing list

Re: [gentoo-user] etc-update

[brett holcomb]

Another option - some apps have /etc/*.conf files that the 
system sets up but have an include /etc/someotherconfigdir 
file.  For example /etc/devfsd.conf has an include 
/etc/devfs.d and includes any files from there that are 
NOT hidden.

On Thu, 6 Mar 2003 11:31:23 -0800 
 Balaji Srinivasan <[EMAIL PROTECTED]> 
wrote:
One way these conflicts could be reduced is by separating 
out sections in
config files that will most probably be modified by the 
user and those which
are not. For example the USE directive and the CFLAGS 
directive from
make.conf could be moved to a separate file. That way 
whenever portage
changes, they wouldnt need to update those flags (or even 
if they did it
would be easy to merge). This is in ofcourse be in 
addition to having a way
for the user to indicate which files he is interested in 
and hence those
files should not be auto updated. Also maintaining a 
history of updates in a
separate directory would also
help. This way in case things do go wrong we still have 
access to the old
files.

Balaji

-Original Message-
From: Jason Giangrande [mailto:[EMAIL PROTECTED]
Sent: Thursday, March 06, 2003 11:30 AM
To: [EMAIL PROTECTED]
Subject: Re: [gentoo-user] etc-update
Ian, that was my point exactly. 

I'm not saying that, by default, etc-update should update 
all files 
automatically but being able to tell it to update all 
files that I know 
I haven't manually changed would be a good thing.

Jason

Ian Truelsen wrote:

On Thu, 6 Mar 2003 13:53:00 -0500
"Todd Punderson" <[EMAIL PROTECTED]> wrote:
 

Blindly merging new config files is extremely dangerous. 
As much of a
pain as it is, it is of utmost importance that you take 
the time to
review the changes and be sure they make sense for your 
system. You
surely wouldn't want your make.conf to just be 
overwritten. Not
updating or blindly updating the files in /etc is 
probably one of the
top reasons for system problems. There is a nice new 
feature that will
automatically update insignificant changes to the config 
files. Check
out /etc/etc-update.conf. Todd

   

True enough, but isn't there some way that we could track 
whether a
config file has been altered? It seems to me that, except 
in very odd
cases, a config file that has not been altered could just 
be updated
automatically. 

Of course, I may be biased by having just updated to X 
4.3 and had to
cruise through 75 config files in etc-update. I think I 
wore out the
colon and q keys :)

 

--
-Jason Giangrande
  giangrande.org - http://www.giangrande.org 

  Dog's I View - http://www.dogsiview.com 


--
[EMAIL PROTECTED] mailing list
--
[EMAIL PROTECTED] mailing list


--
[EMAIL PROTECTED] mailing list

Re: [gentoo-user] dragging and dropping on kicker

[Ernie Schroder]

On Thursday 06 March 2003 14:27, Robert Cole wrote:
> So absolutely NO ONE has tried to drag an application icon from the
> KDE 3.1 desktop to the kicker toolbar and insert a new app? I really
> find this hard to believe. Why am I being ignored anyway? No one does
> this type of thing?
>
> Robert

I guess no one bothered to try it I just did and I find the same 
thing you did. I can however drag desktop icons to the right side of 
the kicker (right of the pager) or right click in an empty spot over on 
the left and select panel menu -->add --> application button. I think 
its a "feature".
In the future, if no one answers a question, ask again. Hostility will 
be met by the same
-- 
Regards, Ernie
100% Microsoft and Intel free

--
[EMAIL PROTECTED] mailing list

Re: [gentoo-user] "emerge clean" confusion

[gentoo]

My guess would be Gentoo's support for multiple versions of gcc.
That would make it very difficult to determine what is removable
and what is not.


* Toby Dickenson <[EMAIL PROTECTED]> [2003-03-06 20:30:10 +]:
> I understand emerge clean is supposed to remove old versions of a package, 
> keeping one per slot. I cant understand why on this system it doesnt want to 
> clean up the two old versions of gcc:
> 
> 
> vrumpet root # emerge clean gcc
> 
>  sys-devel/gcc
> selected: none
>protected: 2.95.3-r5 2.95.3-r7 2.95.3-r8
>  omitted: none
> 
> >>> clean: No packages selected for removal.
> 
> Any ideas? Thanks in advance

-- 
Thomas M. Beaudry
k8la / ys1ztm
[EMAIL PROTECTED]

--
[EMAIL PROTECTED] mailing list

Re: [gentoo-user] Which stage tarball should I choose?

[Joshua J. Berry]

It's been a while since I actually did any Gentoo installations, so all of 
this is IIRC. If I screw up, somebody please correct me.

On Thursday 06 March 2003 11:31, Daniel Carrera wrote:
> Hello,
> I've read the Gentoo x86 installation instructions.  I have them in front
> of me right now.  I don't entirely understand the difference between
> stage1, stage2 and stage3 tarballs.  I have some questions that might
> clear things up for me:
>
> * What is the bootstrap process (in this context)?

The bootstrap process gives you a working set of basic tools, and a compiler, 
so you can compile your own base system (i.e. the compiler, and system 
tools).

> I know that the difference between stage1 and stage2 is that stage2 skips
> this.  Why would I want to go through the bootstrap process?  What does it
> compile?

You might want to go through the bootstrap process if you have special CPU 
optimizations you want to use. Although, I've found the stageX tarballs which 
are compiled for your specific CPU tend to give pretty reasonable defaults.

>
> * I get the impression that stage3 doesn't compile anything.  Is this
> correct?

Sort of. Stage3 is basically a precompiled base system for your CPU, so you 
don't have to go through bootstrapping and compiling glibc and whatnot 
yourself. This tends to be a pretty big time-saver.

After you have a stage3 system setup, you'll have basic system utilities like 
a shell, cp, ls, fsck, etc. You still need to install programs like a cron 
daemon, a syslog daemon, and any other programs you want to use (X, mozilla, 
KDE, GNOME, etc). And, you'll also still need to compile your own kernel.

Hope this helps.

-- Josh

-
Joshua J. Berry

"I haven't lost my mind -- it's backed up on tape somewhere."
-- /usr/games/fortune

PGP Key: http://deneb.condordes.net/node/16/view

NOTE: Please do not submit this email address to any mailing
lists or websites without prior permission.  Thank you.


pgp0.pgp
Description: signature

Re: [gentoo-user] Portage errors

[Ric Messier]

No luck. I'm still getting ACCESS VIOLATION errors attempting to open_wr 
/var/lib/scrollkeeper/scrollkeeper_docs 

Ric

--
[EMAIL PROTECTED] mailing list

Re: [gentoo-user] Which stage tarball should I choose?

[brett holcomb]

Stage 1 is just a base - well not even a system.  It's a 
tarball with enough stuff in it to allow you to emerge 
sync and then build a system.  It builds everything - the 
entire system, the whole thing!  After untaring stage1 you 
have a very minimal system - just enough to start building 
the system.  That's what bootstrap does (it's an old 
computer term).  Bootstrap takes what you untarred, builds 
the system and eventually gets you to the point where you 
can then add the system packages.
  

Stage 2 has more built into it and less compiling - it's 
closer to the end product.

Stage 3 is even closer to being ready to run.

Why use them:

Stage 1 - it builds everything - the entire system, the 
whole thing - hence it is optimized for your machine.  The 
downside is it can take a long time to build this way.

Stage 2 and 3 - These have a lot of the system already 
compiled.  However, the compiles were done for a "generic" 
machine. The downside is that many of the programs are not 
optimized for your system.  The upside is that it takes 
less time to get a system up and running.

I've only done stage 1 installs but the two systems I've 
done them on have been a dual Athlon 1.9 gig and a dual 
PIII 933 so it didn't take tha
On Thu, 6 Mar 2003 14:31:03 -0500
 Daniel Carrera <[EMAIL PROTECTED]> wrote:
Hello,
I've read the Gentoo x86 installation instructions.  I 
have them in front 
of me right now.  I don't entirely understand the 
difference between 
stage1, stage2 and stage3 tarballs.  I have some 
questions that might 
clear things up for me:

* What is the bootstrap process (in this context)?
I know that the difference between stage1 and stage2 is 
that stage2 skips 
this.  Why would I want to go through the bootstrap 
process?  What does it 
compile?

* I get the impression that stage3 doesn't compile 
anything.  Is this 
correct?
So with stage2 you compile your system, but with stage3 
you just install 
pre-compiled binaries just like all the other Linux 
distributions.
Am I right?

Thanks for the help.

--
Daniel Carrera
Graduate Teaching Assistant.  Math Dept.
University of Maryland.  (301) 405-5137
--
[EMAIL PROTECTED] mailing list


--
[EMAIL PROTECTED] mailing list

Re: [gentoo-user] etc-update

[gentoo]

Still not a good thing.  There could be some new option in the config
file that needs setting.  Anyway, with the exception of the XFree
upgrade , we're not talking about a lot of time reviewing config
files.  The longest I've ever spent other than the XFree upgrade was
five minutes.  And that was after I had been in the hospital over a
month and a lot of stuff had changed.

So with the present system, we're looking at 15-20 minutes once or twice
a year for XFree.  Add in 5 minutes monthly or 2-3 minutes if you
upgrade more often than monthly (nightly for me).  Does not seem like a
hardship to me for the peace of mind of not having any changes made
unknowingly to my system.

Maybe you can get the maintaner of etc-update to add such a feature.
I'll never use it.  Just one screwed up config file will cost me more
time debugging than it looks like I would spend in a year reviewing
updated config files.

* Jason Giangrande <[EMAIL PROTECTED]> [2003-03-06 14:29:58 -0500]:
> Ian, that was my point exactly. 
> 
> I'm not saying that, by default, etc-update should update all files 
> automatically but being able to tell it to update all files that I know 
> I haven't manually changed would be a good thing.
> 
> >>Blindly merging new config files is extremely dangerous. As much of a
> >>pain as it is, it is of utmost importance that you take the time to
> >>review the changes and be sure they make sense for your system. You
> >>surely wouldn't want your make.conf to just be overwritten. Not
> >>updating or blindly updating the files in /etc is probably one of the
> >>top reasons for system problems. There is a nice new feature that will
> >>automatically update insignificant changes to the config files. Check
> >>out /etc/etc-update.conf. Todd
> >>
> >True enough, but isn't there some way that we could track whether a
> >config file has been altered? It seems to me that, except in very odd
> >cases, a config file that has not been altered could just be updated
> >automatically. 
> >
> >Of course, I may be biased by having just updated to X 4.3 and had to
> >cruise through 75 config files in etc-update. I think I wore out the
> >colon and q keys :)

-- 
Thomas M. Beaudry
k8la / ys1ztm
[EMAIL PROTECTED]

--
[EMAIL PROTECTED] mailing list

Re: [gentoo-user] Re: Opinion Poll: KDE Vs Gnome

[Joshua J. Berry]

FWIW, I know that Qt is thread-safe (if compiled with threading enabled), and 
I'm relatively sure that KDE makes use of that.

On Thursday 06 March 2003 10:39, Daniel Carrera wrote:
> > > The question is ill-posed in the sense that KDE and Gnome have nothing
> > > to do with threads.  There is nothing they can do about threading.
> > > The kernel is encharged of that.  You can make a GUI that hangs or one
> > > that doesn't hang just as easily in each.
> >
> > Not entirely true.  The toolkits can be implemented to use threading
> > "well", "poorly", or not at all.
>
> Alright, that's not the question I thought he was asking.  Af for the
> toolkits themselves, I think that they are both good, but I don't have any
> detailed information.

-- 

-
Joshua J. Berry

"I haven't lost my mind -- it's backed up on tape somewhere."
-- /usr/games/fortune

PGP Key: http://deneb.condordes.net/node/16/view

NOTE: Please do not submit this email address to any mailing
lists or websites without prior permission.  Thank you.


pgp0.pgp
Description: signature

Re: [gentoo-user] dragging and dropping on kicker

[brett holcomb]

I don't think it's a case of anyone ignoring you. 
Probably no one has seen the problem either because they 
haven't tried it or have done it and it worked so no one 
has any useful input for you.  The rest of us don't run 
KDE so we can't help .

Did you check bugzilla both for Gentoo and KDE?

On Thu, 6 Mar 2003 11:27:12 -0800
 Robert Cole <[EMAIL PROTECTED]> wrote:
So absolutely NO ONE has tried to drag an application 
icon from the KDE 3.1 
desktop to the kicker toolbar and insert a new app? I 
really find this hard 
to believe. Why am I being ignored anyway? No one does 
this type of thing?

Robert

--
[EMAIL PROTECTED] mailing list


--
[EMAIL PROTECTED] mailing list

RE: [gentoo-user] Re: etc-update

[Balaji Srinivasan]

Exactly...which is why we need a way for the user to mark files as being
protected...
(some character in the comment is one way...)
Balaji

-Original Message-
From: Tyler Trafford [mailto:[EMAIL PROTECTED]
Sent: Thursday, March 06, 2003 12:30 PM
To: [EMAIL PROTECTED]
Subject: [gentoo-user] Re: etc-update


On Thu, Mar 06, 2003 at 03:04:29PM -0500, Ian Truelsen wrote:
> On Thu, 6 Mar 2003 11:31:23 -0800 
> Balaji Srinivasan <[EMAIL PROTECTED]> wrote:
> 
> > One way these conflicts could be reduced is by separating out sections
> > in config files that will most probably be modified by the user and
> > those which are not. For example the USE directive and the CFLAGS
> > directive from make.conf could be moved to a separate file. That way
> > whenever portage changes, they wouldnt need to update those flags (or
> > even if they did it would be easy to merge). This is in ofcourse be in
> > addition to having a way for the user to indicate which files he is
> > interested in and hence those files should not be auto updated. Also
> > maintaining a history of updates in a separate directory would also
> > help. This way in case things do go wrong we still have access to the
> > old files.
> > 
> Could we not just check the last modified date vs. the create date and
> if they match, assume the file has not been altered? Or am I not as
> clear as I thought I was on what those two represent?

Some people aren't eager to change a working configuration whether
they've made any changes or not.
-- 
Tyler Trafford

--
[EMAIL PROTECTED] mailing list

--
[EMAIL PROTECTED] mailing list

Re: [gentoo-user] Re: etc-update

[Ian Truelsen]

On Thu, 6 Mar 2003 15:30:01 -0500
Tyler Trafford <[EMAIL PROTECTED]> wrote:

> On Thu, Mar 06, 2003 at 03:04:29PM -0500, Ian Truelsen wrote:
> > On Thu, 6 Mar 2003 11:31:23 -0800 
> > Balaji Srinivasan <[EMAIL PROTECTED]> wrote:
> > 
> > > One way these conflicts could be reduced is by separating out
> > > sections in config files that will most probably be modified by
> > > the user and those which are not. For example the USE directive
> > > and the CFLAGS directive from make.conf could be moved to a
> > > separate file. That way whenever portage changes, they wouldnt
> > > need to update those flags (or even if they did it would be easy
> > > to merge). This is in ofcourse be in addition to having a way for
> > > the user to indicate which files he is interested in and hence
> > > those files should not be auto updated. Also maintaining a history
> > > of updates in a separate directory would also help. This way in
> > > case things do go wrong we still have access to the old files.
> > > 
> > Could we not just check the last modified date vs. the create date
> > and if they match, assume the file has not been altered? Or am I not
> > as clear as I thought I was on what those two represent?
> 
> Some people aren't eager to change a working configuration whether
> they've made any changes or not.

Okay. But might we consider an option in etc-update.conf or something
like that so that those who don't mind it could try it?

-- 
Ian Truelsen
Email: [EMAIL PROTECTED]
AIM: ihtruelsen
Homepage: http://www.ihtruelsen.dyndns.org

--
[EMAIL PROTECTED] mailing list

Re: [gentoo-user] grub help (hda1=winxp, hda3=linux)

[brett holcomb]

There is an excellent tutorial by D Robbins on grub. 
Someone posted the URL a day or so ago but I'm not at a 
machine where I can get to it but it is recommended 
reading.

On Thu, 6 Mar 2003 12:36:21 -0500
 "Kevin J. Anderson" <[EMAIL PROTECTED]> wrote:
You could also install grub into your root partition, use 
dd to copy the
first 512k out into a file, transfer that file to your xp 
c:\ root, and then
use XP's boot.ini to give you the option to boot linux 
instead of xp.  This
is the way I have always done dual boots w/ nt/2k/xp and 
I find it to be
much less "dangerous" esp since windows seems to like the 
situtation much
better.

I think setup (hd0,2)  root (hd0,2) would be the grub 
commands (please
correct me if I am wrong)

then you can use this command in linux dd if=/dev/hda3 
of=/bootsect.lnx
bs=512 count=1
transfer that file somehow over to your c:\ (floppy, ftp 
it somewhere else,
bring it back down, etc)
add c:\Bootsect.lnx="Gentoo Linux" to your boot.ini

google for "boot.ini dd linux dual boot" for more info on 
the subject, as my
explanation is very cut and dry.

kev



->-Original Message-
->From: Jeremy Workman [mailto:[EMAIL PROTECTED]
->Sent: Thursday, March 06, 2003 12:14 PM
->To: [EMAIL PROTECTED]
->Subject: Re: [gentoo-user] grub help (hda1=winxp, 
hda3=linux)
->
->
->I notice you said you tried "setup (hd0,0)". That would 
be the first
->partition on the first hard drive (/dev/hda1). To 
install Grub into the
->master boot record the command would be "setup (hd0)".
->
->Hope this helps.
->
->
->On Thu, 2003-03-06 at 13:04, Ajay Sharma wrote:
->> hey everyone,
->>
->> I just finished rebuilding my machine and I 
repartitioned my
->box so that
->> it looks like:
->>
->> /dev/hda1 = windows xp (ntfs)
->> /dev/hda2 = linux swap
->> /dev/hda3 = gentoo (ext3)
->>
->> no, I don't have a separate /boot partition.  Anyway, 
I followed the
->> installation instructions (excellent btw!) and I'm 
having problems
->> getting grub to install on my MBR.  In the grub 
prompt, when I type in
->> "root(hd0,0)" it would come back with something like 
"unrecognized file
->> type".  After that it's pretty obvious that the 
"setup(hd0,0)" wasn't
->> going to work.  I then ran root(hd0,2) and that went 
over fine but now
->> when I boot up it goes right into windows xp and not 
the grub boot
->> loader because it's not installed on the MBR.
->>
->> I'd appreciate any help.
->>
->> Thanks,
->> Ajay
->>
->> 

->> Satyajot (Ajay) Sharma
->> [EMAIL PROTECTED]
->> 

->>
->>
->> --
->> [EMAIL PROTECTED] mailing list
->>
->>
->
->
->--
->[EMAIL PROTECTED] mailing list
->

--
[EMAIL PROTECTED] mailing list


--
[EMAIL PROTECTED] mailing list

Re: [gentoo-user] My gentoo grub does not work properly (Was installing gentoo on Thinkpad R31)

[Tim Sutton]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Yes, I have the menu.lst file created.

Tim

On Thursday 06 March 2003 12:13, Sigurd Stordal wrote:
> On Thursday 06 March 2003 12:57, Tim Sutton wrote:
> > Hi all :-)
> >
> > Well after a fairly late night I managed to get gentoo installed by
> > booting with gentoo 1.2 and swapping the cd out for gentoo 1.4 afterwards
> > for the stage3 tarball (many thanks for suggesting that approach
> > Gwendolyn and for the people who provided me with links to the gentoo 1.2
> > download area!).
> >
> > After recompiling my kernel a few times (ACPI caused the boot to hang the
> > first kernel, and I forgot to compile in devfs on the second :-( ), I
> > managed to get it to boot up. However my grub is not 'playing nice'. When
> > I boot, I get no splash screen / menu and I have to manually type
>
> Have you created the /boot/grub/menu.lst file.
>
> --
> Sigurd Stordal
> President of GOGS
> Experimental Petrologist

- -- 
Tim Sutton
BDWorld Middleware Programmer
- ---
BiodiversityWorld Project
Centre for Plant Diversity & Systematics
School of Plant Sciences
The University of Reading
Reading, RG66AS, UK

Web :   www.bdworld.org
Phone  :   +44-(0)118-378-6052  
Email:   t.sutton [at] reading.ac.uk
(email preferred method of correspondence)
- ---
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.2.1 (GNU/Linux)

iD8DBQE+Z1qFZxb+GYjbYHkRAjw0AKCE78xz22yzkT3znMNB3d5NVTPjywCgo4Rq
/VEuBa0gk4goHDtzfDQyW1I=
=LWZZ
-END PGP SIGNATURE-


--
[EMAIL PROTECTED] mailing list

[gentoo-user] Re: etc-update

[Tyler Trafford]

On Thu, Mar 06, 2003 at 03:04:29PM -0500, Ian Truelsen wrote:
> On Thu, 6 Mar 2003 11:31:23 -0800 
> Balaji Srinivasan <[EMAIL PROTECTED]> wrote:
> 
> > One way these conflicts could be reduced is by separating out sections
> > in config files that will most probably be modified by the user and
> > those which are not. For example the USE directive and the CFLAGS
> > directive from make.conf could be moved to a separate file. That way
> > whenever portage changes, they wouldnt need to update those flags (or
> > even if they did it would be easy to merge). This is in ofcourse be in
> > addition to having a way for the user to indicate which files he is
> > interested in and hence those files should not be auto updated. Also
> > maintaining a history of updates in a separate directory would also
> > help. This way in case things do go wrong we still have access to the
> > old files.
> > 
> Could we not just check the last modified date vs. the create date and
> if they match, assume the file has not been altered? Or am I not as
> clear as I thought I was on what those two represent?

Some people aren't eager to change a working configuration whether
they've made any changes or not.
-- 
Tyler Trafford

--
[EMAIL PROTECTED] mailing list

[gentoo-user] "emerge clean" confusion

[Toby Dickenson]

I understand emerge clean is supposed to remove old versions of a package, 
keeping one per slot. I cant understand why on this system it doesnt want to 
clean up the two old versions of gcc:


vrumpet root # emerge clean gcc

 sys-devel/gcc
selected: none
   protected: 2.95.3-r5 2.95.3-r7 2.95.3-r8
 omitted: none

>>> clean: No packages selected for removal.

Any ideas? Thanks in advance

--
[EMAIL PROTECTED] mailing list

Re: [gentoo-user] OT hardware ?

[Christian Herzyk]

Hi there,

Am Donnerstag, 6. März 2003 20:48 schrieb Alan:
> On Thu, Mar 06, 2003 at 02:41:20PM -0500, Ernie Schroder wrote:
> > I'm getting ready to take Gentoo_2 off line and replace mobo and
> > processor I was looking at hdd's this morning and saw Western Digital
> > 80 and 120 gig udma 133 drives at both the same pricethe difference is
> > that the 80 gig has 8 megs of cache and the 120 2gigs.both claim a seek
>
>   ^ Gigs???
>
> > time of 9.1ms. could some one voice an opinion here.(as if I have to
> > look far on this list for an opinion :) ) Which would you buy?
>
> More cache is always better performance, but if it means you can get the
> extra 40G  Kinda hard to say.  I got myself some of the WD 80G/8m
> drives a couple of months ago and have been quite happy.  There was a
> bigger price difference between the 80 and 120 at that point though :)
>
> Tough decision, I guess it really comes down to space or performance.
> Granted, I can't give you quantitative evidence of the performance of
> the 8m vs 2m cache.

I have the WD 80G drive with 2MB Cache, so I will just post my values.

>
> My hdparm results:
>
> phoenix alan # hdparm /dev/hdf
>
> /dev/hdf:
>  multcount= 16 (on)
>  IO_support   =  0 (default 16-bit)
>  unmaskirq=  0 (off)
>  using_dma=  1 (on)
>  keepsettings =  0 (off)
>  readonly =  0 (off)
>  readahead=  8 (on)
>  geometry = 155061/16/63, sectors = 156301488, start = 0

/dev/hda:
 multcount= 16 (on)
 IO_support   =  1 (32-bit)
 unmaskirq=  1 (on)
 using_dma=  1 (on)
 keepsettings =  0 (off)
 readonly =  0 (off)
 readahead=  8 (on)
 geometry = 9729/255/63, sectors = 156301488, start = 0


> phoenix alan # hdparm -Tt /dev/hdf
> /dev/hdf:
>  Timing buffer-cache reads:   128 MB in  0.80 seconds =159.60 MB/sec
>  Timing buffered disk reads:  64 MB in  2.07 seconds = 30.86 MB/sec

/dev/hda:
 Timing buffer-cache reads:   128 MB in  0.50 seconds =256.00 MB/sec
 Timing buffered disk reads:  64 MB in  2.04 seconds = 31.37 MB/sec

Quite strange that my cache reads are much faster, probably it's teh 
controller. 

> phoenix alan # dmesg | grep hdf
> hdf: WDC WD800JB-00CRA1, ATA DISK drive
> hdf: 156301488 sectors (80026 MB) w/8192KiB Cache, CHS=155061/16/63,
> UDMA(100)

Kernel command line: root=/dev/hda8 hdd=ide-scsi
ide0: BM-DMA at 0xd400-0xd407, BIOS settings: hda:DMA, hdb:DMA
hda: WDC WD800BB-53BSA0, ATA DISK drive
hda: 156301488 sectors (80026 MB) w/2048KiB Cache, CHS=9729/255/63, UDMA(100)

So it seems that in this small "benchmark" the cache doesn't score too much 
perhaps there are some better things to run like bonnie or dbench.

Greetings

Christian



--
[EMAIL PROTECTED] mailing list

Re: [gentoo-user] etc-update

[Ian Truelsen]

On Thu, 6 Mar 2003 11:31:23 -0800 
Balaji Srinivasan <[EMAIL PROTECTED]> wrote:

> One way these conflicts could be reduced is by separating out sections
> in config files that will most probably be modified by the user and
> those which are not. For example the USE directive and the CFLAGS
> directive from make.conf could be moved to a separate file. That way
> whenever portage changes, they wouldnt need to update those flags (or
> even if they did it would be easy to merge). This is in ofcourse be in
> addition to having a way for the user to indicate which files he is
> interested in and hence those files should not be auto updated. Also
> maintaining a history of updates in a separate directory would also
> help. This way in case things do go wrong we still have access to the
> old files.
> 
Could we not just check the last modified date vs. the create date and
if they match, assume the file has not been altered? Or am I not as
clear as I thought I was on what those two represent?

-- 
Ian Truelsen
Email: [EMAIL PROTECTED]
AIM: ihtruelsen
Homepage: http://www.ihtruelsen.dyndns.org
PGP key available at: http://www.ihtruelsen.dyndns.org/pgp.html 
and http://pgp.mit.edu (search 'ihtruelsen')

--
[EMAIL PROTECTED] mailing list

RE: [gentoo-user] OT hardware ?

[Graham, Steve]

The 8mb cache really comes into play with sustained transfer rates of large
files.

The 2mb cache is similar to a drive I'm using, it's only 60gb though, and I
have no complaints.

It all comes down to what the drive is going to be used for. If the drive
will primarily be used with lots of small files and small file transfers,
then the 2mb will suffice, BUT if you do any graphics work, lots of movie
downloads :0) or are planning to run a server get the 8mb drive.

~Steve

--
[EMAIL PROTECTED] mailing list

Re: [gentoo-user] Warning to hardware purchasers

[jsmith]

Quoting Ernie Schroder <[EMAIL PROTECTED]>:

>   I would whole heartedly recommend Monarch to anyone in the market for 
> computer hardware. Sorry about the commercial here, I realize that this 
> is OT but I would hate for anyone else to go through the garbage I 
> endured.

Newegg is also excellent in this regard (I've returned two items and they were
very gracious and accomodating in the process).  CDW, though expensive, was also
very good with the return when my Firegl X1 wouldn't work with the binary Linux
drivers ATI ships.

Perhaps we should have a hardware rating page, aimed at Gentoo (or perhaps more
generally GNU/Linux) folks, where one can check against other's experiences
before purchasing from a particular vendor.

Jean.


--
[EMAIL PROTECTED] mailing list

[gentoo-user] rc3 install - /lib/cpp failure

[mjbjr]

This is my second gentoo install, and on the same machine. Last time, I
did a stage 1. Though it took quite some time to get it right. The
system got munged, and I decided to correct things by doing a fresh
install, a Stage 3 install. It's been going fine until...

section 14:

during 'emerge -u world', compiling of ncurses-5.3 (the first item)
failed with "/lib/cpp" fails sanity check.

Interestingly, during the compiling "checks"...

first it says we ARE using 'gcc', then later, it says no.

libc.h no

how to run the C++ preprocessor... /lib/cpp

configure: error: C++ preprocesssor "/lib/cpp" fails sanity check

!!! ERROR: sys-libs/ncurses-5.3-r1 failed.
!!! Function econf, Line 262, Exitcode 1
!!! econf failed

In irc channel #gentoo, it was suggested I do a 'emerge gcc-config'. I
did, to no avail.

This is very frustrating, because, upto this point, all that's really
been going on is simple hardware setup, with the exception of editing
/etc/make.conf.  So, to have a show stopping error right at the
beginning is bad.

Additionally, I'm not seeing any other reports of this problem from
anyone else.  What is it that _I_ could have done that's different from
everyone else?

The only place for _me_ to make a configuration error and get this kind of
problem at this point would be, I assume, in /etc/make.conf.  I've tried
numerous variations, including no changes, no P4, whatever.  I always
get the same error.

I just don't understand how this could be happening.

Any ideas?

Thank you.
-- 
- Martin J. Brown, Jr. -
- [EMAIL PROTECTED] -

  Public PGP Key ID: 0xDB184F7B keyserver: http://certserver.pgp.com/


--
[EMAIL PROTECTED] mailing list

Re: [gentoo-user] Which stage tarball should I choose?

[Steve Juranich]

Steve Juranich wrote:
Yes.  You can still choose stuff built specifically for your 
architecture (depending on which iso you downloaded), but you won't be 
compiling anything with a stage3 install. Thus, you will be unable to 
tweak optimizer flags and USE variable stuff.
Spoke too soon.  You still have the option to build all of your 
"user-level" apps.  What you'll be skipping is the "emerge -u system" 
step.  To get a list of what this will do, you can do an 'emerge -up 
system' and see what it looks like (if you already have a stage1 system 
running).  I'm pretty sure this is mostly stuff like awk, sed, perl, 
python, etc.  All of the stuff that you need to have a "functioning" 
gentoo box.

With the stage3 setup, you'd install most of your user-level apps like 
mozilla, gnome/kde, .

HTH

--
Stephen W. Juranich [EMAIL PROTECTED]
Electrical Engineering http://students.washington.edu/sjuranic
University of Washingtonhttp://ssli.ee.washington.edu/ssli
--
[EMAIL PROTECTED] mailing list

Re: [gentoo-user] OT hardware ?

[Alan]

On Thu, Mar 06, 2003 at 02:41:20PM -0500, Ernie Schroder wrote:
> I'm getting ready to take Gentoo_2 off line and replace mobo and 
> processor I was looking at hdd's this morning and saw Western Digital 
> 80 and 120 gig udma 133 drives at both the same pricethe difference is 
> that the 80 gig has 8 megs of cache and the 120 2gigs.both claim a seek 
  ^ Gigs???
> time of 9.1ms. could some one voice an opinion here.(as if I have to 
> look far on this list for an opinion :) ) Which would you buy?

More cache is always better performance, but if it means you can get the
extra 40G  Kinda hard to say.  I got myself some of the WD 80G/8m
drives a couple of months ago and have been quite happy.  There was a
bigger price difference between the 80 and 120 at that point though :) 

Tough decision, I guess it really comes down to space or performance.
Granted, I can't give you quantitative evidence of the performance of
the 8m vs 2m cache.  

My hdparm results:

phoenix alan # hdparm /dev/hdf
 
/dev/hdf:
 multcount= 16 (on)
 IO_support   =  0 (default 16-bit)
 unmaskirq=  0 (off)
 using_dma=  1 (on)
 keepsettings =  0 (off)
 readonly =  0 (off)
 readahead=  8 (on)
 geometry = 155061/16/63, sectors = 156301488, start = 0
phoenix alan # hdparm -Tt /dev/hdf
/dev/hdf:
 Timing buffer-cache reads:   128 MB in  0.80 seconds =159.60 MB/sec
 Timing buffered disk reads:  64 MB in  2.07 seconds = 30.86 MB/sec
phoenix alan # dmesg | grep hdf
hdf: WDC WD800JB-00CRA1, ATA DISK drive
hdf: 156301488 sectors (80026 MB) w/8192KiB Cache, CHS=155061/16/63, UDMA(100)

Alan


-- 
Alan <[EMAIL PROTECTED]> - http://arcterex.net
-
"The only thing that experience teaches us is that experience teaches 
us nothing. -- Andre Maurois (Emile Herzog)

--
[EMAIL PROTECTED] mailing list

Re: [gentoo-user] dragging and dropping on kicker

[Hemmann, Volker Armin]

Hi,

On Thursday 06 March 2003 20:27, Robert Cole wrote:
> So absolutely NO ONE has tried to drag an application icon from the KDE 3.1
> desktop to the kicker toolbar and insert a new app? I really find this hard
> to believe. Why am I being ignored anyway? No one does this type of thing?
>
> Robert
>
> --
> [EMAIL PROTECTED] mailing list

 I just created a programm link on the desktop, dragged it onto the kicker, 
where it was added, klicked on the button and licq is started fine.

So, what's your problem?

Glück Auf
Volker

--
[EMAIL PROTECTED] mailing list

Re: [gentoo-user] Which stage tarball should I choose?

[Steve Juranich]

Daniel Carrera wrote:
Hello,
I've read the Gentoo x86 installation instructions.  I have them in front 
of me right now.  I don't entirely understand the difference between 
stage1, stage2 and stage3 tarballs.  I have some questions that might 
clear things up for me:

* What is the bootstrap process (in this context)?
I know that the difference between stage1 and stage2 is that stage2 skips 
this.  Why would I want to go through the bootstrap process?  What does it 
compile?
http://www.gentoo.org/doc/en/gentoo-x86-install.xml#doc_chap12>

bootstrap.sh will build binutils, gcc, gettext, and glibc, rebuilding 
binutils, gcc, and gettext  after glibc. Needless to say, this process 
takes a while. Once this process completes, your system will be 
equivalent to a "stage2" system, which means you can now move on to the 
stage2 instructions.


This took about 3 hours on my Athlon XP 2100+, IIRC.

* I get the impression that stage3 doesn't compile anything.  Is this 
correct?
So with stage2 you compile your system, but with stage3 you just install 
pre-compiled binaries just like all the other Linux distributions.
Am I right?
Yes.  You can still choose stuff built specifically for your 
architecture (depending on which iso you downloaded), but you won't be 
compiling anything with a stage3 install. Thus, you will be unable to 
tweak optimizer flags and USE variable stuff.

HTH

--
Stephen W. Juranich [EMAIL PROTECTED]
Electrical Engineering http://students.washington.edu/sjuranic
University of Washingtonhttp://ssli.ee.washington.edu/ssli
--
[EMAIL PROTECTED] mailing list

[gentoo-user] OT hardware ?

[Ernie Schroder]

I'm getting ready to take Gentoo_2 off line and replace mobo and 
processor I was looking at hdd's this morning and saw Western Digital 
80 and 120 gig udma 133 drives at both the same pricethe difference is 
that the 80 gig has 8 megs of cache and the 120 2gigs.both claim a seek 
time of 9.1ms. could some one voice an opinion here.(as if I have to 
look far on this list for an opinion :) ) Which would you buy?
-- 
Regards, Ernie
100% Microsoft and Intel free

--
[EMAIL PROTECTED] mailing list

RE: [gentoo-user] etc-update

[Balaji Srinivasan]

One way these conflicts could be reduced is by separating out sections in
config files that will most probably be modified by the user and those which
are not. For example the USE directive and the CFLAGS directive from
make.conf could be moved to a separate file. That way whenever portage
changes, they wouldnt need to update those flags (or even if they did it
would be easy to merge). This is in ofcourse be in addition to having a way
for the user to indicate which files he is interested in and hence those
files should not be auto updated. Also maintaining a history of updates in a
separate directory would also
help. This way in case things do go wrong we still have access to the old
files.

Balaji

-Original Message-
From: Jason Giangrande [mailto:[EMAIL PROTECTED]
Sent: Thursday, March 06, 2003 11:30 AM
To: [EMAIL PROTECTED]
Subject: Re: [gentoo-user] etc-update


Ian, that was my point exactly. 

I'm not saying that, by default, etc-update should update all files 
automatically but being able to tell it to update all files that I know 
I haven't manually changed would be a good thing.

Jason

Ian Truelsen wrote:

>On Thu, 6 Mar 2003 13:53:00 -0500
>"Todd Punderson" <[EMAIL PROTECTED]> wrote:
>
>  
>
>>Blindly merging new config files is extremely dangerous. As much of a
>>pain as it is, it is of utmost importance that you take the time to
>>review the changes and be sure they make sense for your system. You
>>surely wouldn't want your make.conf to just be overwritten. Not
>>updating or blindly updating the files in /etc is probably one of the
>>top reasons for system problems. There is a nice new feature that will
>>automatically update insignificant changes to the config files. Check
>>out /etc/etc-update.conf. Todd
>>
>>
>>
>True enough, but isn't there some way that we could track whether a
>config file has been altered? It seems to me that, except in very odd
>cases, a config file that has not been altered could just be updated
>automatically. 
>
>Of course, I may be biased by having just updated to X 4.3 and had to
>cruise through 75 config files in etc-update. I think I wore out the
>colon and q keys :)
>
>  
>

-- 
-Jason Giangrande
  giangrande.org - http://www.giangrande.org 
  Dog's I View - http://www.dogsiview.com 


--
[EMAIL PROTECTED] mailing list

--
[EMAIL PROTECTED] mailing list

Re: [gentoo-user] dragging and dropping on kicker

[Peter Ruskin]

On Thursday 06 Mar 2003 19:27, Robert Cole wrote:
> So absolutely NO ONE has tried to drag an application icon from the KDE
> 3.1 desktop to the kicker toolbar and insert a new app? I really find
> this hard to believe. Why am I being ignored anyway? No one does this
> type of thing?
>
Works fine here.
-- 
Gentoo-1.4.2.8 Unstable. KDE: 3.1.0 Qt: 3.1.0
AMD Athlon(tm) XP 1900+ 768MB.  Kernel: 2.4.20-xfs_pre6.GCC 3.2.2


--
[EMAIL PROTECTED] mailing list

[gentoo-user] Which stage tarball should I choose?

[Daniel Carrera]

Hello,
I've read the Gentoo x86 installation instructions.  I have them in front 
of me right now.  I don't entirely understand the difference between 
stage1, stage2 and stage3 tarballs.  I have some questions that might 
clear things up for me:

* What is the bootstrap process (in this context)?
I know that the difference between stage1 and stage2 is that stage2 skips 
this.  Why would I want to go through the bootstrap process?  What does it 
compile?

* I get the impression that stage3 doesn't compile anything.  Is this 
correct?
So with stage2 you compile your system, but with stage3 you just install 
pre-compiled binaries just like all the other Linux distributions.
Am I right?

Thanks for the help.

-- 
Daniel Carrera
Graduate Teaching Assistant.  Math Dept.
University of Maryland.  (301) 405-5137

--
[EMAIL PROTECTED] mailing list

Re: [gentoo-user] Re: Opinion Poll: KDE Vs Gnome

[Ben Ricker]

On Thu, 2003-03-06 at 12:35, Tyler Trafford wrote:
> On Thu, Mar 06, 2003 at 10:55:50AM -0500, Daniel Carrera wrote:
> > On Thu, Mar 06, 2003 at 08:59:30PM +0530, Raghuram Rajagopal wrote:
> > 
> >> And I think I have some reasons behind asking this question and not
> >> just "ill-posed".
> > 
> > The question is ill-posed in the sense that KDE and Gnome have nothing
> > to do with threads.  There is nothing they can do about threading.
> > The kernel is encharged of that.  You can make a GUI that hangs or one
> > that doesn't hang just as easily in each.
> 
> Not entirely true.  The toolkits can be implemented to use threading
> "well", "poorly", or not at all.

Good point. However, I don't know about you, but I am not the guru who
can judge the respective toolkits and I would expect that most users of
gentoo are not initmately familiar with the threading implementations of
the toolkits. At least, one would have to do much testing and profiling
to judge such a request. Expecting users of gentoo to be intimately
aware of threading implementation of such HUGE "apps" is, well,
ill-posed (no to mention the fact that no one, as far as I can tell, has
been able to answer the question). 

If there is a difference that tips the scale towards one or the other, I
can say that, as an educated guess, it is most probably not in the
threading implementation of the respective thread libraries. Others have
brought up more important issues (i.e., ease-of-use, memory hogginess,
etc.).

Ben Ricker

-- 
Ben Ricker <[EMAIL PROTECTED]>
Wellinx.com


--
[EMAIL PROTECTED] mailing list

Re: [gentoo-user] Back up advice. Computer trade in.

[Craig Barrett]

On Thu, 6 Mar 2003 17:22:20 +1100
"LoJack80" <[EMAIL PROTECTED]> wrote:

> Or, specifly request that they replace the screen. Searches for 
> replacement screens show me that 14 inch screens can be had for $100 to 
> $150 or less. This is certainly cheaper than sending you a new laptop

Why not send the laptop back without the hard drive, and request that they do the same 
on their end?  then you pop your drive back in, and have no worries about your data at 
all.

Saludos
Craig

--
[EMAIL PROTECTED] mailing list

[gentoo-user] dragging and dropping on kicker

[Robert Cole]

So absolutely NO ONE has tried to drag an application icon from the KDE 3.1 
desktop to the kicker toolbar and insert a new app? I really find this hard 
to believe. Why am I being ignored anyway? No one does this type of thing?

Robert

--
[EMAIL PROTECTED] mailing list

Re: [gentoo-user] etc-update

[Jason Giangrande]

Ian, that was my point exactly. 

I'm not saying that, by default, etc-update should update all files 
automatically but being able to tell it to update all files that I know 
I haven't manually changed would be a good thing.

Jason

Ian Truelsen wrote:

On Thu, 6 Mar 2003 13:53:00 -0500
"Todd Punderson" <[EMAIL PROTECTED]> wrote:
 

Blindly merging new config files is extremely dangerous. As much of a
pain as it is, it is of utmost importance that you take the time to
review the changes and be sure they make sense for your system. You
surely wouldn't want your make.conf to just be overwritten. Not
updating or blindly updating the files in /etc is probably one of the
top reasons for system problems. There is a nice new feature that will
automatically update insignificant changes to the config files. Check
out /etc/etc-update.conf. Todd
   

True enough, but isn't there some way that we could track whether a
config file has been altered? It seems to me that, except in very odd
cases, a config file that has not been altered could just be updated
automatically. 

Of course, I may be biased by having just updated to X 4.3 and had to
cruise through 75 config files in etc-update. I think I wore out the
colon and q keys :)
 

--
-Jason Giangrande
 giangrande.org - http://www.giangrande.org 
 Dog's I View - http://www.dogsiview.com 
--
[EMAIL PROTECTED] mailing list

Re: [gentoo-user] etc-update

[Ian Truelsen]

On Thu, 6 Mar 2003 13:53:00 -0500
"Todd Punderson" <[EMAIL PROTECTED]> wrote:

> Blindly merging new config files is extremely dangerous. As much of a
> pain as it is, it is of utmost importance that you take the time to
> review the changes and be sure they make sense for your system. You
> surely wouldn't want your make.conf to just be overwritten. Not
> updating or blindly updating the files in /etc is probably one of the
> top reasons for system problems. There is a nice new feature that will
> automatically update insignificant changes to the config files. Check
> out /etc/etc-update.conf. Todd
> 
True enough, but isn't there some way that we could track whether a
config file has been altered? It seems to me that, except in very odd
cases, a config file that has not been altered could just be updated
automatically. 

Of course, I may be biased by having just updated to X 4.3 and had to
cruise through 75 config files in etc-update. I think I wore out the
colon and q keys :)

-- 
Ian Truelsen
Email: [EMAIL PROTECTED]
AIM: ihtruelsen
Homepage: http://www.ihtruelsen.dyndns.org
PGP key available at: http://www.ihtruelsen.dyndns.org/pgp.html 
and http://pgp.mit.edu (search 'ihtruelsen')

--
[EMAIL PROTECTED] mailing list

[gentoo-user] libbonoboui broke evolution.

[Spundun Bhatt]

I dont know if this is worth posting on this list... but when I emerged
the latest libbonoboui-2.2.0.1 , It broke evolution such that I could
not compose any mails. I reemrged evolution and now it works fine. Some
API problem I guess... 
Spundun


--
[EMAIL PROTECTED] mailing list

Re: [gentoo-user] System screwed up

[Peter Dijkstra]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Hi there,

When you run the following commands you can use portage (ander other things) 
again (replace the i586 thing whit your own):

LD_LIBRARY_PATH=/usr/lib/gcc-lib/i586-pc-linux-gnu/3.2.2 env-update
source /etc/profile

I have remerged glibc to avoid typing the above each time I restart the 
system.

Regards, peter Dijkstra

On Thursday 06 March 2003 13:13, Ming-Che Lee wrote:
> Hi Cedric!
>
> On Thursday, 6. March 2003 08:52 Cedric Veilleux wrote:
> > # emerge
> > python2.2: error while loading shared libraries: libstdc++.so.5:
> > cannot open shared object file: No such file or directory
>
> Have you recently upgraded gcc? Then maybe this bug report at
> http://bugs.gentoo.org/show_bug.cgi?id=15288 is what you are
> looking for!
>
> > Many programs refuses to run, giving the previous error,
> > including portage and python... Seems like I would need to
> > re-install GCC but how can I do this without a working portage?
> > bootstrap.sh?
>
> A broken portage can be fixed according to this:
> /usr/portage/sys-apps/portage/files/README.RESCUE
>
> Best regards,
>
> Ming-Che
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.2.1 (GNU/Linux)

iD8DBQE+Z6kdlH43C52z1d0RAgr6AJ9HnwjB6R+EPfkZ9DJSnqJPvj1D8ACfcfT6
+sBJWFBvnu8DtoiAJcHDxCY=
=+uTL
-END PGP SIGNATURE-


--
[EMAIL PROTECTED] mailing list

Re: [gentoo-user] dev-php/mod_php-4.3.1 emerge fails when testing for libjpeg

[Louis C. Candell]

"Charles H. Leggett" <[EMAIL PROTECTED]> writes:

> I recently tried to "emerge -u mod_php" and it fails in the middle of
> configure with:

I had the same problem emerging mod_php and fixed it by using

USE="-jpeg"

in /etc/make.conf. 

Hope that helps,

> checking for jpeg_read_header in -ljpeg... no
> configure: error: libjpeg not found!
> I have tried to "emerge /media-libs/jpeg" thinking that this would satisfy
> the dependancy but no dice.

-- 
Louis C. Candell

--
[EMAIL PROTECTED] mailing list

Re: [gentoo-user] etc-update

[Todd Punderson]

Blindly merging new config files is extremely dangerous. As much of a pain
as it is, it is of utmost importance that you take the time to review the
changes and be sure they make sense for your system. You surely wouldn't
want your make.conf to just be overwritten. Not updating or blindly updating
the files in /etc is probably one of the top reasons for system problems.
There is a nice new feature that will automatically update insignificant
changes to the config files. Check out /etc/etc-update.conf.
Todd

- Original Message -
From: "Jason Giangrande" <[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>
Sent: Thursday, March 06, 2003 11:57 AM
Subject: [gentoo-user] etc-update


> Is there a way to use etc-update to update all config files at once,
> either merge or replace?  The manpage didn't have anything and doing
> them one at a time is a huge pain.
>
> Thanks,
> --
> -Jason Giangrande
>   giangrande.org - http://www.giangrande.org 
>   Dog's I View - http://www.dogsiview.com 
>
>
> --
> [EMAIL PROTECTED] mailing list
>
>


--
[EMAIL PROTECTED] mailing list

[gentoo-user] glibc vmware problem and partial solution

[valeanu . a]

After upgrading my gentoo to Xfree 4.3.0 and glibc 2.3.2_pre1
vmware bails out with
'XIO:  fatal IO error 0 (Success) on X server ":0.0"'
it seem that our brothers/sisters from vmware.for-linux.experimental
already know about this problem. Following the instructions
from 
http://groups.google.com/groups?hl=en&lr=&ie=UTF-8&threadm=m3znptdse7.fsf%40fischman.org&rnum=1&prev=/groups%3Fq%3Dvmware%2BXIO:%2Bglibc%26hl%3Den%26lr%3D%26ie%3DUTF-8%26selm%3Dm3znptdse7.fsf%2540fischman.org%26rnum%3D1
now i can start vmware as root.


pgp0.pgp
Description: PGP signature

Re: [gentoo-user] Re: Opinion Poll: KDE Vs Gnome

[Daniel Carrera]

> > The question is ill-posed in the sense that KDE and Gnome have nothing
> > to do with threads.  There is nothing they can do about threading.
> > The kernel is encharged of that.  You can make a GUI that hangs or one
> > that doesn't hang just as easily in each.
> 
> Not entirely true.  The toolkits can be implemented to use threading
> "well", "poorly", or not at all.

Alright, that's not the question I thought he was asking.  Af for the 
toolkits themselves, I think that they are both good, but I don't have any 
detailed information.

-- 
Daniel Carrera
Graduate Teaching Assistant.  Math Dept.
University of Maryland.  (301) 405-5137

--
[EMAIL PROTECTED] mailing list

[gentoo-user] Re: Opinion Poll: KDE Vs Gnome

[Tyler Trafford]

On Thu, Mar 06, 2003 at 10:55:50AM -0500, Daniel Carrera wrote:
> On Thu, Mar 06, 2003 at 08:59:30PM +0530, Raghuram Rajagopal wrote:
> 
>> And I think I have some reasons behind asking this question and not
>> just "ill-posed".
> 
> The question is ill-posed in the sense that KDE and Gnome have nothing
> to do with threads.  There is nothing they can do about threading.
> The kernel is encharged of that.  You can make a GUI that hangs or one
> that doesn't hang just as easily in each.

Not entirely true.  The toolkits can be implemented to use threading
"well", "poorly", or not at all.
-- 
Tyler Trafford

--
[EMAIL PROTECTED] mailing list

[gentoo-user] ckhfontpath command

[Kurt Bechstein]

Anyone know off hand what package the chkfontpath command is apart of? 
On a red hat box there is actually a chkfontpath package, but I can't
seem to locate anything like this in portage.  I know can just manually
edit the /etc/X11/fs/config file but I kind of like having the
chkfontpath command around.  Thanks in advance.


--
[EMAIL PROTECTED] mailing list

Re: [gentoo-user] Problem compiling KDE multimedia

[Ernie Schroder]

On Thursday 06 March 2003 05:07, Jonathan Hunt wrote:
> Hi,
>
> I was in the process of upgrading from Kde 3.0.3 to KDE 3.1 when this
> happened. I have gcc 3.2.2 on my system. I'm guessing that the
> linking errors are a result of some library being complied before I
> upgraded gcc but I've reemerged glibc, qt, kdelibs and kdebase to no
> effect. 


I have 2 Gento boxes here, one running KDE 3.0.5 and the other has been 
updated to 3.1, so I have built kdemultimedia 3 times. The first 
machine I built had the same problem you're seeing. (I can't be sure 
that I got the exact same errors though build died on multimedia)
Someone on the list, (thanks whoever it was) advized me to emerge arts 
first. I did that and the problem was solved. I repeated tis on the 
second box and all went well as did an upgrade on one box that is going 
down for a brain transplant this evening. It will get an athlon XP2100 
on an nForce2 based board so I will be building from scratch. Though 
I'm inclined to try out the GRP packages for X and KDE.
HTH
-- 
Regards, Ernie
100% Microsoft and Intel free

--
[EMAIL PROTECTED] mailing list

Re: [gentoo-user] Opinion Poll: KDE Vs Gnome

[Mark Hazell]

On Wed, 05 Mar 2003 16:09:45 + Andy typed:

> [EMAIL PROTECTED] wrote:
> |> Which desktop (KDE/Gnome) suits the best based on
> |>
> |>   1. Performance
> |>2. Efficiency
> |>3. Multithreaded application handling process
> |>4. Ease of use
> |>5. Best look and feel
> 
> Well, while we're all flopping it out in an unresolvable argument, I'd
> just like to chip in about my favourite.. FVWM! (http://www.fvwm.org/)
> , which as it happens is one of the oldest WM's still in common use.

Excellent, somebody else who still runs FVWM :-)

> It's got the first 3 points above completely nailed. I would say it is
> very strong on Ease of Use, though it will not win any awards for Ease
> of Learning (there is a difference...). As for 5, aesthetics are
> probably FVWM's weak point, but that doesn't mean it has to look *bad*
> just that it doesn't look as good as some of the others.

My reasoning for going with FVWM was configurability and an easy config
file to change, as well as a lack of any sort of toolbar and icons. I
love the clean desktop look - the only things I have are a 'sticky'
gkrellm2 and pager that are on each of my 6 workspaces.

And that's all I have - lots of keyboard shortcuts for switching screens
and opening applications, and a couple of menus with just the
applications I use.
I couldn't ask for a more minimalist WM that just does it's job with no
bloat. It can be a bit ugly if you use a lot of buttons and panels and
stuff, but I don't, and I love it!

Cheers,
Mark.

--
[EMAIL PROTECTED] mailing list

RE: [gentoo-user] grub help (hda1=winxp, hda3=linux)

[Kevin J. Anderson]

You could also install grub into your root partition, use dd to copy the
first 512k out into a file, transfer that file to your xp c:\ root, and then
use XP's boot.ini to give you the option to boot linux instead of xp.  This
is the way I have always done dual boots w/ nt/2k/xp and I find it to be
much less "dangerous" esp since windows seems to like the situtation much
better.

I think setup (hd0,2)  root (hd0,2) would be the grub commands (please
correct me if I am wrong)

then you can use this command in linux dd if=/dev/hda3 of=/bootsect.lnx
bs=512 count=1
transfer that file somehow over to your c:\ (floppy, ftp it somewhere else,
bring it back down, etc)
add c:\Bootsect.lnx="Gentoo Linux" to your boot.ini

google for "boot.ini dd linux dual boot" for more info on the subject, as my
explanation is very cut and dry.

kev



->-Original Message-
->From: Jeremy Workman [mailto:[EMAIL PROTECTED]
->Sent: Thursday, March 06, 2003 12:14 PM
->To: [EMAIL PROTECTED]
->Subject: Re: [gentoo-user] grub help (hda1=winxp, hda3=linux)
->
->
->I notice you said you tried "setup (hd0,0)". That would be the first
->partition on the first hard drive (/dev/hda1). To install Grub into the
->master boot record the command would be "setup (hd0)".
->
->Hope this helps.
->
->
->On Thu, 2003-03-06 at 13:04, Ajay Sharma wrote:
->> hey everyone,
->>
->> I just finished rebuilding my machine and I repartitioned my
->box so that
->> it looks like:
->>
->> /dev/hda1 = windows xp (ntfs)
->> /dev/hda2 = linux swap
->> /dev/hda3 = gentoo (ext3)
->>
->> no, I don't have a separate /boot partition.  Anyway, I followed the
->> installation instructions (excellent btw!) and I'm having problems
->> getting grub to install on my MBR.  In the grub prompt, when I type in
->> "root(hd0,0)" it would come back with something like "unrecognized file
->> type".  After that it's pretty obvious that the "setup(hd0,0)" wasn't
->> going to work.  I then ran root(hd0,2) and that went over fine but now
->> when I boot up it goes right into windows xp and not the grub boot
->> loader because it's not installed on the MBR.
->>
->> I'd appreciate any help.
->>
->> Thanks,
->> Ajay
->>
->> 
->> Satyajot (Ajay) Sharma
->> [EMAIL PROTECTED]
->> 
->>
->>
->> --
->> [EMAIL PROTECTED] mailing list
->>
->>
->
->
->--
->[EMAIL PROTECTED] mailing list
->


--
[EMAIL PROTECTED] mailing list

Re: [gentoo-user] Warning to hardware purchasers

[Ernie Schroder]

On Thursday 27 February 2003 08:14, Stephen Varga wrote:
> You should not have to process a refund, since it is not legal to
> charge your credit card until the product ships. If they charged your
> card tell them to reverse it immediately and cancel the order, then
> call your credit card company and start the process to dispute the
> charges.
>
> Do you really want to keep dealing with this company that has been
> giving you the run around? How are they going to react if you have a
> problem with one of the components? Waiting two weeks to deal with a
> comapnay that is going to provide you the service you want is a small
> price to pay compared to the pain of keep dealing with this company.
>
> Steve
>
> On Wed, 2003-02-26 at 12:29, Ernie Schroder wrote:
> > I know that this is way off topic and probably belongs on the
> > forum, but I figured that I could reach the maximum # of people
> > this way. A couple of weeks ago, I decided to upgrade one of my
> > computers with a new motherboard and processor. I searched the web
> > for top rated boards at a good price and decided on the Leadtek
> > K7NCR18D-pro. My focus then changed to obtaining the best price
> > possible.
> > My search led me to decide on a company called PCRapids. Although
> > their shipping charges were on the high side, their prices were
> > still the best. My step-son and 3 of his co-workers decided to join
> > in and I placed my order on Feb. 15th for 5 Leadtek boards and 5
> > AthlonXP 2100+ processors. (at a fair price of $1.113.00) This sum
> > was immediately charged to my account and I was able to access a
> > page that reported my order status as "IN PROCESS".
> > Now, after 11 days, this page still says "IN PROCESS" `10 or 12
> > emails inquiring about my order have gone un answered and most
> > telephone calls are not picked up on their end. When a call is
> > actually answered, I am given an answer designed to placate me.
> > Last Thursday, I was told that my order had been shipped and that I
> > could expect an email shortly with a UPS tracking #. The email
> > never appeared. Several emails and phone calles were all
> > unanswered. The following day, at 11:00 A.M. EST, I was able to
> > contact someone there by phone. He was very apolligetic and
> > informed me that in fact, the order was still in shipping and I
> > could expect an email shortly with my tracking number.
> > Once again, no email was received. Several phone calls over the
> > weekend went unanswered (their site does state that they are closed
> > Saturday and Sunday) Monday morning, once again at 11:00 AM I was
> > able to reach someone by phone. This person, Miguele, informed me
> > that for some unknown reason, my order had not shipped as promised
> > he assured me that he would check into my situation and email me
> > immediately with more information. Again, I received no email.
> > Still more phone calls and emails were ignored.
> > Yesterday,(2/26) I once again reached Miguele at 11:00 AM. He
> > explained that once again, my order had not shipped and again
> > appologised to the point where he actually sounded somewhat
> > sincere. He assured me that the order was packaged and waiting for
> > their 4:00 PM UPS pickup and that shipping had been upgraded to UPS
> > 2nd day to partially make up for any inconvenience. He also assured
> > me that soon after the order left the building,I would receive an
> > email with the tracking number. Instead of a tracking number I
> > received my first email response from PCRapids durring this whole
> > saga. I have pasted it below:
> >
> >  i appologize for the delayed update. Your order with PCR for 5
> > leadtek motherboards, was not shipped today. the reason for this
> > delay is
> > because we only have limited stock of the Leadtek board that you
> > wanted to
> > purchase. We cannot send you an incomplete order that is why your
> > order has
> > not left our warehouse. I sincerely appologize for the
> > inconvenience caused
> > and understand excactly how you are feeling at this time. At this
> > point however i will not be able to get your order sent out today
> > and will probably have this out the door tomorrow, thought this is
> > not also not guaranteed. We do want your business but due to the
> > circumstance, we wont be
> > able to fulfill your request at this point. Again we appologize. If
> > you are
> > still interested with this order please reply to this email and i
> > will try
> > my best to expedite your order right away.
> >
> > Well, I am pretty flippin' mad here, but as it will take about a
> > week for my charge account to process a refund and another week to
> > process an order with another company, I am pretty much stuck with
> > these people. I intend to take what action I can against this
> > company as soon as I have received either my order or a refund, but
> > I wanted to start here with a warning to my friends at Gentoo.
> > This same company also does b

Re: [gentoo-user] grub help (hda1=winxp, hda3=linux)

[Jeremy Workman]

I notice you said you tried "setup (hd0,0)". That would be the first
partition on the first hard drive (/dev/hda1). To install Grub into the
master boot record the command would be "setup (hd0)".

Hope this helps.


On Thu, 2003-03-06 at 13:04, Ajay Sharma wrote:
> hey everyone,
> 
> I just finished rebuilding my machine and I repartitioned my box so that 
> it looks like:
> 
> /dev/hda1 = windows xp (ntfs)
> /dev/hda2 = linux swap
> /dev/hda3 = gentoo (ext3)
> 
> no, I don't have a separate /boot partition.  Anyway, I followed the
> installation instructions (excellent btw!) and I'm having problems 
> getting grub to install on my MBR.  In the grub prompt, when I type in 
> "root(hd0,0)" it would come back with something like "unrecognized file 
> type".  After that it's pretty obvious that the "setup(hd0,0)" wasn't 
> going to work.  I then ran root(hd0,2) and that went over fine but now 
> when I boot up it goes right into windows xp and not the grub boot 
> loader because it's not installed on the MBR.
> 
> I'd appreciate any help.
> 
> Thanks,
> Ajay
> 
> 
> Satyajot (Ajay) Sharma 
> [EMAIL PROTECTED]
> 
> 
> 
> --
> [EMAIL PROTECTED] mailing list
> 
> 


--
[EMAIL PROTECTED] mailing list

Re: [gentoo-user] grub help (hda1=winxp, hda3=linux)

[Ing. Bernardo Lopez]

If you like to try a nice (dangerous? i dont know, to me works fine with w98)

from dos's fdisk:

fdisk /mbr

This clear the mbr and then you go to grub and try again.

Good luck

On Thursday 06 March 2003 18:04, Ajay Sharma wrote:
> hey everyone,
>
> I just finished rebuilding my machine and I repartitioned my box so that
> it looks like:
>
> /dev/hda1 = windows xp (ntfs)
> /dev/hda2 = linux swap
> /dev/hda3 = gentoo (ext3)
>
> no, I don't have a separate /boot partition.  Anyway, I followed the
> installation instructions (excellent btw!) and I'm having problems
> getting grub to install on my MBR.  In the grub prompt, when I type in
> "root(hd0,0)" it would come back with something like "unrecognized file
> type".  After that it's pretty obvious that the "setup(hd0,0)" wasn't
> going to work.  I then ran root(hd0,2) and that went over fine but now
> when I boot up it goes right into windows xp and not the grub boot
> loader because it's not installed on the MBR.
>
> I'd appreciate any help.
>
> Thanks,
> Ajay
>
> 
> Satyajot (Ajay) Sharma
> [EMAIL PROTECTED]
> 


--
[EMAIL PROTECTED] mailing list

[gentoo-user] grub help (hda1=winxp, hda3=linux)

[Ajay Sharma]

hey everyone,

I just finished rebuilding my machine and I repartitioned my box so that 
it looks like:

/dev/hda1 = windows xp (ntfs)
/dev/hda2 = linux swap
/dev/hda3 = gentoo (ext3)

no, I don't have a separate /boot partition.  Anyway, I followed the
installation instructions (excellent btw!) and I'm having problems 
getting grub to install on my MBR.  In the grub prompt, when I type in 
"root(hd0,0)" it would come back with something like "unrecognized file 
type".  After that it's pretty obvious that the "setup(hd0,0)" wasn't 
going to work.  I then ran root(hd0,2) and that went over fine but now 
when I boot up it goes right into windows xp and not the grub boot 
loader because it's not installed on the MBR.

I'd appreciate any help.

Thanks,
Ajay


Satyajot (Ajay) Sharma 
[EMAIL PROTECTED]



--
[EMAIL PROTECTED] mailing list

[gentoo-user] etc-update

[Jason Giangrande]

Is there a way to use etc-update to update all config files at once, 
either merge or replace?  The manpage didn't have anything and doing 
them one at a time is a huge pain.

Thanks,
--
-Jason Giangrande
 giangrande.org - http://www.giangrande.org 
 Dog's I View - http://www.dogsiview.com 
--
[EMAIL PROTECTED] mailing list

Re: [gentoo-user] Opinion Poll: KDE Vs Gnome

[Daniel Carrera]

On Thu, Mar 06, 2003 at 08:59:30PM +0530, Raghuram Rajagopal wrote:
> Yeah ofcourse, GUI itself is a process and it is handled
> concurrently by the OS. 
> 
> But, my question was, does GUI itself have some built-in
> functionalities to enhance the way of handling multiple
> process at a time.
> 
> I came across some GUI applications, where if a single
> process gets somewhere hanged or dead, then all its
> parent processes are also killed or even sometime leads  to 
> system crash. So, I just wanted to know , how 
> effectively does KDE/Gnome handles this type of situation.
> 
> And I think I have some reasons behind asking this 
> question and not just "ill-posed".

The question is ill-posed in the sense that KDE and Gnome have nothing to 
do with threads.  There is nothing they can do about threading.  The 
kernel is encharged of that.  You can make a GUI that hangs or one that 
doesn't hang just as easily in each.

-- 
Daniel Carrera
Graduate Teaching Assistant.  Math Dept.
University of Maryland.  (301) 405-5137

--
[EMAIL PROTECTED] mailing list

Re: [gentoo-user] tetex build fails

[Peter Gantner]

Quoting Heino Herrlich from Mar 6

> Peter wrote:
> 
> PG> I get this error with emergeing tetex-2.0-r1:
> 
> PG> /usr/lib/gcc-lib/i686-pc-linux-gnu/3.2.2/../../../../i686-pc-linux-gnu/bin/ld:
> PG> warning: libmysqlclient.so.10, needed by /usr/lib/libwwwxml.so, not 
> PG> found (try using -rpath or -rpath-link)
> PG> /usr/lib/libwwwsql.so: undefined reference to `mysql_connect'
> PG> /usr/lib/libwwwsql.so: undefined reference to `mysql_create_db'
> 
> Did you add mysql to the USE variables in your make.conf?
> net-libs/libwww (needed by tetex) has mysql as a package dependency
> and so i think it should be compiled with mysql support. Maybe
> recompiling it with mysql enabled will do it.
> Hope that helps.

Yes, thanks.

I got it (tetex) installed by removing mysql-4, because that 
conflicts with libwww, which is needed by tetex.

To accomplish this, I had to _remove_ mysql from the USE flags (it
causes an upgrade o -4.0.11a), and manually emerge
/usr/portage/dev-db/mysql/mysql-3.23.55.ebuild.

This in turn has a bug (#16767), so I first had to set:
export CFLAGS="${CFLAGS} -DHAVE_ERRNO_AS_DEFINE=1" 
export CXXFLAGS="${CXXFLAGS} -DHAVE_ERRNO_AS_DEFINE=1"
to get it to compile.

So it runs now.

But the question now is: does this qualify as a portage or an mysql 
ebuild bug?
As I see it, the mysql-4 ebuilds should check for the libwww 
incompatibility (the libwww ebuild does!), and warn about it, no?
Or mysql 3.x/4.x be put in different SLOTS so they can be installed 
in parallel, which might get messy?


-- 
"The Empire never ended."
Tractates: Cryptica Scriptura, no. 6

--
[EMAIL PROTECTED] mailing list

Re: [gentoo-user] Opinion Poll: KDE Vs Gnome

[Raghuram Rajagopal]

Yeah ofcourse, GUI itself is a process and it is handled
concurrently by the OS. 

But, my question was, does GUI itself have some built-in
functionalities to enhance the way of handling multiple
process at a time.

I came across some GUI applications, where if a single
process gets somewhere hanged or dead, then all its
parent processes are also killed or even sometime leads  to 
system crash. So, I just wanted to know , how 
effectively does KDE/Gnome handles this type of situation.

And I think I have some reasons behind asking this 
question and not just "ill-posed".

Cheers




- Original Message - 
From: "Ben Ricker" <[EMAIL PROTECTED]>
To: "Gentoo Mailing List" <[EMAIL PROTECTED]>
Sent: Thursday, March 06, 2003 8:40 PM
Subject: Re: [gentoo-user] Opinion Poll: KDE Vs Gnome


> On Wed, 2003-03-05 at 19:08, Daniel Carrera wrote:
> > On Wed, Mar 05, 2003 at 12:27:13PM -0300, Norberto Bensa wrote:
> 
> > > > 3. Multithreaded application handling process
> > > 
> > > No idea.
> > Me neither.
> > 
> 
> I think what he is asking is how the GUI affects application handling.
> The answer is that is does not make a difference. The OS does the
> multithreading, not the GUI. The question is just ill-posed.
> 
> Ben Ricker
> 
> -- 
> Ben Ricker <[EMAIL PROTECTED]>
> Wellinx.com
> 
> 
> --
> [EMAIL PROTECTED] mailing list

--
[EMAIL PROTECTED] mailing list

Re: [gentoo-user] My gentoo grub does not work properly (Wasinstalling gentoo on Thinkpad R31)

[brett holcomb]

Grub will boot without it.  The file is just a way of 
avoiding you having to type in commands.  If it's not 
there Grub drops back and you have to type the commands 
in.

On Thu, 6 Mar 2003 14:08:59 +0100
 Heino Herrlich <[EMAIL PROTECTED]> wrote:
HH> I do not know what Sigurd means with menu.lst as I do 
not know this
HH> file.

Damn, I took a look into the installation guide and found 
the
following:

Important: To ensure backwards compatibility with GRUB, 
make sure to make a link from grub.conf to menu.lst. You 
can do this by doing ln -s /boot/grub/grub.conf 
/boot/grub/menu.lst .

I should really clean my glasses :(

Am I a lucky guy that my Gentoo works without this link 
or is it only
important under certain circumstances?

__

Gesendet von Yahoo! Mail - http://mail.yahoo.de
Bis zu 100 MB Speicher bei http://premiummail.yahoo.de
--
[EMAIL PROTECTED] mailing list


--
[EMAIL PROTECTED] mailing list

Re: [gentoo-user] sendmail wants to talk to docserver.cac.washington.edu?

[gabriel]

On March 6, 2003 10:12 am, you wrote:
> Where you using pine?  Was it the first time you ran pine?  When you first
> run pine there's a little intro message and the program asks if it can
> send a message to the washington.edu to add to the stats of people who are
> using pine. That host name is the host that receieves that message.
>
> -Esteban


thanks
that's exactly what was going on!


-- 
why does any advanced civilization seek to destroy a less advanced one? 
because the land is strategically valuable, because there are resources that 
can be cultivated and exploited, but most of all, simply because they can.
- g'kar, babylon 5 "and now for a word"


--
[EMAIL PROTECTED] mailing list

Re: [gentoo-user] Opinion Poll: KDE Vs Gnome

[Ben Ricker]

On Wed, 2003-03-05 at 19:08, Daniel Carrera wrote:
> On Wed, Mar 05, 2003 at 12:27:13PM -0300, Norberto Bensa wrote:

> > > 3. Multithreaded application handling process
> > 
> > No idea.
> Me neither.
> 

I think what he is asking is how the GUI affects application handling.
The answer is that is does not make a difference. The OS does the
multithreading, not the GUI. The question is just ill-posed.

Ben Ricker

-- 
Ben Ricker <[EMAIL PROTECTED]>
Wellinx.com


--
[EMAIL PROTECTED] mailing list

[gentoo-user] sendmail wants to talk to docserver.cac.washington.edu?

[gabriel]

the oddest thing has been happening on my box and i don't understand.  i 
emerged sendmail and started it... everything seems to be working fine, but 
watching the lights on my cable modem, i notice that every time i restart 
sendmail, the send/receive light goes on

this made me do some digging

doing a "netstat -c" while i restart sendmail shows that it does a dns lookup 
for "docserver.cac.washington.edu" and upon finding it does whatever this 
means:


  tcp0  1 alexandria.localh:34544 docserver.cac.wash:smtp SYN_SENT


that line stays on netstat until i shut down sendmail again  what is it?  
is it supposed to be there?

thanks.


-- 
i would not be a capitalist, i would be a man;
you cannot be both at the same time.
- eugene debs


--
[EMAIL PROTECTED] mailing list