Re: [gentoo-user] blocking package isn't really there.

[Neil Bothwick]

Hello Ernie Schroder,

> > Kino 1.1.0 on the other hand is satisfied by any version later than
> > 0.4.9_p20061016.  
> 
> Not so.
> 
> # emerge -p kino
> 
> These are the packages that would be merged, in order:
> 
> Calculating dependencies... done!
> [ebuild  N] media-video/ffmpeg-0.4.9_p20070616-r1  USE="X encode
> mmx ogg oss sdl truetype vorbis zlib -a52 -aac 
> (-altivec) -amr -debug -doc -ieee1394 -imlib -network -test -theora
> -threads -v4l -x264 -xvid" [ebuild  N] media-libs/libdv-1.0.0-r2
> USE="sdl xv -debug" [ebuild  N] media-libs/libiec61883-1.1.0
> USE="-examples" [ebuild  N] media-video/dvdauthor-0.6.11
> [ebuild  N] media-sound/rawrec-0.9.98
> [ebuild  N] media-video/kino-1.1.0  USE="alsa dvdr quicktime 
> vorbis -gpac -sox"

It is so, kino 1.1.0 pulls in ffmpeg-0.4.9_p20070616, which doesn't cause
any blocks. The facts that the block disappears when you switch to 1.1.0
should be enough to convince you.

Incidentally this problem is caused by changes in the ffmpeg API, making
Kino, and other programs, break each time the API changes. Kino 1.1.0 is
fixed to work with the updated API.


-- 
Neil Bothwick

MACINTOSH: Most Applications Crash; If Not, The Operating System Hangs


signature.asc
Description: PGP signature

Re: [gentoo-user] blocking package isn't really there.

[Abraham Marín Pérez]

Neil Bothwick escribió:

Hello Ernie Schroder,

  

Kino 1.1.0 on the other hand is satisfied by any version later than
0.4.9_p20061016.  
  

Not so.

# emerge -p kino

These are the packages that would be merged, in order:

Calculating dependencies... done!
[ebuild  N] media-video/ffmpeg-0.4.9_p20070616-r1  USE="X encode
mmx ogg oss sdl truetype vorbis zlib -a52 -aac 
(-altivec) -amr -debug -doc -ieee1394 -imlib -network -test -theora

-threads -v4l -x264 -xvid" [ebuild  N] media-libs/libdv-1.0.0-r2
USE="sdl xv -debug" [ebuild  N] media-libs/libiec61883-1.1.0
USE="-examples" [ebuild  N] media-video/dvdauthor-0.6.11
[ebuild  N] media-sound/rawrec-0.9.98
[ebuild  N] media-video/kino-1.1.0  USE="alsa dvdr quicktime 
vorbis -gpac -sox"



It is so, kino 1.1.0 pulls in ffmpeg-0.4.9_p20070616, which doesn't cause
any blocks. The facts that the block disappears when you switch to 1.1.0
should be enough to convince you.
  
The block isn't caused by ffmpeg, but by kino's ebuild. Kino 1.1.0 is 
still marked as unstable, and I guess devs have a good reason to do such 
a thing. See my other post for the workaround.


BTW, ffmpeg-0.4.9_p20070525 and above are also marked as unstable (or so 
they were last Monday, haven't checked since then).


Abraham

--
[EMAIL PROTECTED] mailing list

Re: [gentoo-user] blocking package isn't really there.

[Neil Bothwick]

On Thu, 23 Aug 2007 10:04:30 +0200, Abraham Marín Pérez wrote:

> > It is so, kino 1.1.0 pulls in ffmpeg-0.4.9_p20070616, which doesn't
> > cause any blocks. The facts that the block disappears when you switch
> > to 1.1.0 should be enough to convince you.
 
> The block isn't caused by ffmpeg, but by kino's ebuild.

Which contains those lines because Kino 1.0.0. doesn't work with later
versions of ffmpeg. The ebuild is responding to the situation, not
causing it.

> Kino 1.1.0 is 
> still marked as unstable, and I guess devs have a good reason to do
> such a thing. See my other post for the workaround.

Yes, the reason is that policy is that ebuilds should remain in testing
for around thirty days. This doesn't mean the software is unstable or
unfit for use, only that the ebuild had not received sufficient testing
to meet the criteria for inclusion i the stable tree.


-- 
Neil Bothwick

The fact that no one understands you doesn't mean you're an artist.


signature.asc
Description: PGP signature

Re: [gentoo-user] blocking package isn't really there.

[Bo Ørsted Andresen]

On Thursday 23 August 2007 08:15:58 Abraham Marín Pérez wrote:
> What I wonder now is, shouldn't Portage itself have detected this issue
> and choose a version between 0.4.9_p20061016 and 0.4.9_p20070525? Would
> this be a Portage bug?

Sure. And it's a very new bug. Hence the high bug number... ;)

https://bugs.gentoo.org/show_bug.cgi?id=1343

-- 
Bo Andresen


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

Re: [gentoo-user] blocking package isn't really there.

[Abraham Marín Pérez]

Bo Ørsted Andresen escribió:

On Thursday 23 August 2007 08:15:58 Abraham Marín Pérez wrote:
  

What I wonder now is, shouldn't Portage itself have detected this issue
and choose a version between 0.4.9_p20061016 and 0.4.9_p20070525? Would
this be a Portage bug?



Sure. And it's a very new bug. Hence the high bug number... ;)

https://bugs.gentoo.org/show_bug.cgi?id=1343

  
Wow, now that's too big for me, I'll leave the business to those closer 
to God :-P.


Abraham

--
[EMAIL PROTECTED] mailing list

Re: [gentoo-user] blocking package isn't really there.

[Ernie Schroder]

On Thursday 23 August 2007, a tiny voice compelled Abraham Marín Pérez to 
write:
> Ernie Schroder escribió:
> > On Wednesday 22 August 2007, a tiny voice compelled Ernie Schroder to 
write:
> >> On Wednesday 22 August 2007, a tiny voice compelled James Ausmus to 
write:
> >>> emerge -ptv kino
> >>
> >> nothing that helps me here that I see
> >> $ emerge -ptv kino
> >>
> >> These are the packages that would be merged, in reverse order:
> >>
> >> Calculating dependencies... done!
> >> [ebuild  N] media-video/kino-1.0.0  USE="alsa dvdr quicktime
> >> vorbis -gpac -sox" 5,301 kB
> >> [ebuild  N]  media-sound/rawrec-0.9.98  60 kB
> >> [ebuild  N]  media-video/dvdauthor-0.6.11  282 kB
> >> [ebuild  N]  media-libs/libiec61883-1.1.0  USE="-examples" 359 kB
> >> [ebuild  N]  media-libs/libdv-1.0.0-r2  USE="sdl xv -debug" 571 kB
> >> [ebuild  N]  media-video/ffmpeg-0.4.9_p20070616-r1  USE="X encode
> >> mmx ogg oss sdl truetype vorbis zlib -a52 -aac
> >> (-altivec) -amr -debug -doc -ieee1394 -imlib -network -test -theora
> >> -threads -v4l -x264 -xvid" 0 kB
> >> [blocks B ] >media-video/ffmpeg-0.4.9_p20070525 (is blocking
> >> media-video/kino-1.0.0)
> >>
> >> Total: 6 packages (6 new, 1 block), Size of downloads: 6,571 kB
> >>
> >>
> >> --
> >> Regards, Ernie
> >
> > So exactly how do I convince portage that the blocking ffmpeg is NOT
> > installed
>
> Taking a look at the ebuild I saw this in the DEPEND list:
> >=media-video/ffmpeg-0.4.9_p20061016
>
> !>media-video/ffmpeg-0.4.9_p20070525
>
> Portage seems to look for a version later than 0.4.9_p20061016 in order
> to satisfy the first condition and finds 0.4.9_p20070616-r1; however,
> then it sees version must not be older than 0.4.9_p20070525 and it
>
> fails... simply add the following line to /etc/portage/packages.mask:
>  >media-video/ffmpeg-0.4.9_p20070525

Very nice work there Abraham!
>
> and you'll be able to install kino-1.0.0 with no problem.
>
> What I wonder now is, shouldn't Portage itself have detected this issue
> and choose a version between 0.4.9_p20061016 and 0.4.9_p20070525? Would
> this be a Portage bug?
maybe portage is dyslexic, like me and recognised  0.4.9_p20061016 as 
0.4.9_p20070616.  One would think it's either a bug or a syntax problem in 
the ebuild.
Thanks for your detective work!
>
> Abraham



-- 
Regards, Ernie
--
[EMAIL PROTECTED] mailing list

[gentoo-user] i586 install

[James]

Hello,

I seem to be having trouble installing
grub on i586 systems.

What I use is the 
"install-x86-minmal-2006.0.iso" cd to boot the system
and follow the hadbook install guide. But I use the stage3 tarfile:
stage3-i586-2006.1.tar.bz2

Everything runs fine until I try to reboot off the hard drive.
I get this

Grub loading stage1.5

Grub loading, please waite...
Error 16

It's repeatable on (2) different (amd k6) systems.

Googling mostly reveals suggestions about file system
being corrupt. I ran file system chekcs (reiserfsck)
per previous instuctions and /boot and / come back
clean and happy.

The ide disk setup is very simple:
fdisk /dev/hda
#   Device BootStartEnd  Blocks   Id  System
# /dev/hda1   *  1 50  401593+  83  Linux
# /dev/hda2 51 185 1084387+  82  Linux swap
# /dev/hda3   * 186 243418065092+  83  Linux

and grub

default 0
fallback 1
timeout 30

#0
title=kernel-2.6.21-gentoo-r4
root(hd0,1)
kernel /boot/kernel-2.6.21-gentoo-r4 root=/dev/hda3
#0
title=kernel-2.6.21-gentoo-r4
root(hd0,1)
kernel /boot/kernel-2.6.21-gentoo-r4 root=/dev/hda1


(Note I use 2 entries in case one does not work)

Then run:
grep -v rootfs /proc/mounts > /etc/mtab
grub-install --no-floppy /dev/hda

exit chroot and unmount and reboot.


What am I missing ?

Is there a better way to install minimalistic gentoo
firewalls? If somebody was to make a HD image
that worked, I'd just dd it directly onto 
a 8G HD.

I have about a dozen of these firewall to build and using
gentoo has worked very well, once I get a minimal (no X)
system installed.

ideas?

James




-- 
[EMAIL PROTECTED] mailing list

Re: [gentoo-user] i586 install

[Neil Bothwick]

Hello James,

> Googling mostly reveals suggestions about file system
> being corrupt. I ran file system chekcs (reiserfsck)
> per previous instuctions and /boot and / come back
> clean and happy.

If you must use reiserfs on /boot (it's a complete waste of space, ext2
is more sensible), you must mount it with the notail option. GRUB cannot
read files from a tail-packed Reiser filesystem.


-- 
Neil Bothwick

WinErr 002: No Error - Yet


signature.asc
Description: PGP signature

Re: [gentoo-user] Rolling upgrades, are you making sure everything gets restarted?

[Dan Johansson]

On Thursday 23 August 2007, Steen Eugen Poulsen wrote:
> > BUT! It runs on four out of five boxes here. This is the error I get on
> > the fifth box (I'm not a phyton programmer).
>
> Neither am I, but that wont stop me. o.O
>
> > # bin/checkrestart
> > Found 54 processes using old versions of upgraded files
> > (32 distinct programs)
> > Traceback (most recent call last):
> >   File "bin/checkrestart", line 321, in ?
> > main()
> >   File "bin/checkrestart", line 116, in main
> > packages[packagename].processes.extend(programs[program])
> > KeyError: '/lib/nut/genericups'
> >
> > Could you please try to explain whats wrong?
>
> The code tried to insert the key in the key hole, but no key hole was
> found.
>
> Strange thing is, line 115 creates the keyhole, so this shouldn't
> happen. Reading the documentation for python, there is no indication
> that key creation can fail without terminating the program, so I'm
> afraid we have a mystery here, doesn't seem like upstreams bugzilla has
> this error, so maybe it is created by me somehow.
>
> I don't think the new version is likely to fix this issue, but it's new
> layout (from upstream) and improved detection engine (from me), should
> at least make it more fun to use on the four working machines and it's
> available at the same URL as the old one.

Still a no-go :-(
I put in some dubugging lines (three print-lines), and the result is that the 
first time the program reaches the following program segment (line 118 ansd 
119) everything works (no error).
packages.setdefault(packagename,Package(packagename))
packages[packagename].processes.extend(programs[program])

But the second time the program comes here only the first line work and it 
bails out on the second.

-- 
Dan Johansson, 
***
This message is printed on 100% recycled electrons!
***


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

Re: [gentoo-user] i586 install

[Florian Philipp]

Am Donnerstag 23 August 2007 19:14:12 schrieb Neil Bothwick:
> Hello James,
>
> > Googling mostly reveals suggestions about file system
> > being corrupt. I ran file system chekcs (reiserfsck)
> > per previous instuctions and /boot and / come back
> > clean and happy.
>
> If you must use reiserfs on /boot (it's a complete waste of space, ext2
> is more sensible), you must mount it with the notail option. GRUB cannot
> read files from a tail-packed Reiser filesystem.

You do not only need to mount it with notail, you need to write all files with 
notail in the first place.

So ...
1. boot from CD 
2. mount -o notail /dev/hda1 /mount/gentoo
3. tar czf /tmp/boot.tar.gz /mount/gentoo/*
4. rm -rf /mount/gentoo/*
5. tar xzf /tmp/boot.tar.gz -C /mount/gentoo/

Or you switch to ext2 because you do not need a filesystem with journal on a 
partition that doesn't need to be bigger than 50MB.


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

Re: [gentoo-user] "Treason uncloaked!" solution?

[Dan Farrell]

On Wed, 22 Aug 2007 23:48:55 +0200
Hans-Werner Hilse <[EMAIL PROTECTED]> wrote:

> Hi,
> 
> On Wed, 22 Aug 2007 12:18:16 -0700
> Grant <[EMAIL PROTECTED]> wrote:
> 
> > Sometimes I get "Treason uncloaked!" in dmesg when running
> > bittorrent. The solution here:
> > 
> > http://www.linuxquestions.org/questions/showthread.php?t=127984
> > 
> > is:
> > 
> > You'd best set iptables to block all packets from BOGON networks
> > (nets that shouldn't exist) so you can avoid this type of attack.
> > You may find a list of bogon nets here. Note: unallocated nets
> > change from time to time! Just in November IANA allocated two more
> > blocks to RIPE, so you really need to pay attention if you're
> > blocking all bogon IPs.
> > 
> > Which doesn't sound great.  What would you guys recommend I do?  I
> > use a Gentoo router.
> 
> Hm, I don't think that those "attacks" (which do no harm to Linux
> systems since some 1.x version of the kernel -- the warning is a
> reminiscence) will always come from wrong nets. I have those
> occasionally on all my larger server installs and never really
> bothered about them. It usually means that the other side of the TCP
> connection reduced the window to zero size, thus leading stupid TCP
> stacks to save information on a basically starved connection. The
> kernel just sends an information to the log, so in case if you
> recognize the IP and are in charge of the sender, you'll know that it
> has a very broken TCP stack. Essentially: Just ignore it, if the
> sender IP doesn't belong to one of your own networks.
> 
> -hwh
I found a line in my Treason-related output that pointed to an internal
IP on a distcc port.  Should I be worried about this computer?  It's
running a brand new gentoo install and is solely for the purpose of
distcc.  
-- 
[EMAIL PROTECTED] mailing list

Re: [gentoo-user] i586 install

[Neil Bothwick]

Hello Florian Philipp,

> > If you must use reiserfs on /boot (it's a complete waste of space,
> > ext2 is more sensible), you must mount it with the notail option.
> > GRUB cannot read files from a tail-packed Reiser filesystem.  
> 
> You do not only need to mount it with notail, you need to write all
> files with notail in the first place.

That's what I was trying to say, but thanks for clarifying it.


-- 
Neil Bothwick

"One world, one web, one program"  -- Microsoft promotional ad
"Ein Volk, ein Reich, ein Fuehrer"  -- Adolf Hitler


signature.asc
Description: PGP signature

[gentoo-user] removing old kernels from system

[Denis]

My "eselect kernel list" was getting too long for my liking, so I
decided to remove some of the older kernel versions.  I did

emerge -C sys-kernel/gentoo-sources-VERSION

for the versions I wanted to get rid of but soon discovered that
"eselect kernel list" hasn't gotten any shorter.  Indeed, the old
kernel directories were still in /usr/src, and they were still filled
with executables.  The modules for the old kernels in /lib/modules
were also still there.  I ended up removing all those directories by
hand.

Is there a tool for automatically removing the old kernels/modules and
all their respective directories?
-- 
[EMAIL PROTECTED] mailing list

[gentoo-user] Re: i586 install

[James]

Florian Philipp  addcom.de> writes:


> You do not only need to mount it with notail, you need to write all files 
> with 
> notail in the first place.

Hmm,

I have several system where I use this for the fstab with reiser:
/dev/hda2/boot   reiserfsdefaults1 2

so the defaults include notial?

> So ...
> 1. boot from CD 
> 2. mount -o notail /dev/hda1 /mount/gentoo
> 3. tar czf /tmp/boot.tar.gz /mount/gentoo/*
> 4. rm -rf /mount/gentoo/*
> 5. tar xzf /tmp/boot.tar.gz -C /mount/gentoo/

I'm going to try this and even edit the fstab to reflect notail.
I did not see this anywhere in the Handbook section on grub, although
they do talk about notail in the fstab stuff


> Or you switch to ext2 because you do not need a filesystem with journal on a 
> partition that doesn't need to be bigger than 50MB.


I like reiserfs, even for boot partitions. It is wonderful.
ext2 may be superior () but, I really, really like reiserfs
and use it on lots of boot partitions for gentoo systems.
I also use larger /boot partitions to keep multiple 
kernels and related files. A few hundered megabytes does not
significantly hurt, when using a HD. If I go to 2 gig compact
flash cards, then I'll revisit these issues.  






-- 
[EMAIL PROTECTED] mailing list

Re: [gentoo-user] Re: i586 install

[Volker Armin Hemmann]

On Donnerstag, 23. August 2007, James wrote:
> Florian Philipp  addcom.de> writes:
> > You do not only need to mount it with notail, you need to write all files
> > with notail in the first place.
>
> Hmm,
>
> I have several system where I use this for the fstab with reiser:
> /dev/hda2/boot   reiserfsdefaults1 2
>
> so the defaults include notial?
>
> > So ...
> > 1. boot from CD
> > 2. mount -o notail /dev/hda1 /mount/gentoo
> > 3. tar czf /tmp/boot.tar.gz /mount/gentoo/*
> > 4. rm -rf /mount/gentoo/*
> > 5. tar xzf /tmp/boot.tar.gz -C /mount/gentoo/
>
> I'm going to try this and even edit the fstab to reflect notail.
> I did not see this anywhere in the Handbook section on grub, although
> they do talk about notail in the fstab stuff
>
> > Or you switch to ext2 because you do not need a filesystem with journal
> > on a partition that doesn't need to be bigger than 50MB.
>
> I like reiserfs, even for boot partitions. It is wonderful.
> ext2 may be superior () but, I really, really like reiserfs
> and use it on lots of boot partitions for gentoo systems.
> I also use larger /boot partitions to keep multiple
> kernels and related files. A few hundered megabytes does not
> significantly hurt, when using a HD. If I go to 2 gig compact
> flash cards, then I'll revisit these issues.

well, I love reiserfs too - but for /boot it is just... stupid. Even 15(!)mb 
is enough for several kernels - and since boot is only mounted to copy the 
new kernel onto it or through early boot, journaling is just a waste of space 
and time.

About the notail - maybe grub has learned to deal with reiserfs even without 
notail ...
-- 
[EMAIL PROTECTED] mailing list

Re: [gentoo-user] i586 install

[Benno Schulenberg]

Neil Bothwick wrote:
> GRUB cannot read files from a tail-packed Reiser filesystem.

Huh?  How is this machine able to boot then?  Here /boot isn't a 
separate partition, but just a subdir of /, which is mounted 
without notail, and has been so for years.

# mount | grep " / "
/dev/hda9 on / type reiserfs (rw,noatime)

Benno
-- 
[EMAIL PROTECTED] mailing list

Re: [gentoo-user] i586 install

[Neil Bothwick]

Hello Benno Schulenberg,

> > GRUB cannot read files from a tail-packed Reiser filesystem.  
> 
> Huh?  How is this machine able to boot then?  Here /boot isn't a 
> separate partition, but just a subdir of /, which is mounted 
> without notail, and has been so for years.

As Volker said, maybe things have changed recently, but the requirement
was always to use notail when running reiserfs on /boot. that's why I use
ext3 on my root partition (no /boot partition), but I'd be happy to
switch it to reiser.

The latest baselayout-1 still has this warning in fstab
# NOTE: If your BOOT partition is ReiserFS, add the notail option to opts.


-- 
Neil Bothwick

And all the Borg left was this copy of OS/2...


signature.asc
Description: PGP signature

[gentoo-user] Re: i586 install

[James]

Neil Bothwick  digimed.co.uk> writes:

> 
> Hello Benno Schulenberg,
> 
> > > GRUB cannot read files from a tail-packed Reiser filesystem.  
> > 
> > Huh?  How is this machine able to boot then?  Here /boot isn't a 
> > separate partition, but just a subdir of /, which is mounted 
> > without notail, and has been so for years.
> 
> As Volker said, maybe things have changed recently, but the requirement
> was always to use notail when running reiserfs on /boot. that's why I use
> ext3 on my root partition (no /boot partition), but I'd be happy to
> switch it to reiser.
> 
> The latest baselayout-1 still has this warning in fstab
> # NOTE: If your BOOT partition is ReiserFS, add the notail option to opts.


Well all of this is interesting. I just booted off of the 2006.0 minimal
cd for x86 (the 2006.1 does not boot the system)
and here's the fstab I have been using:

/dev/hda1   /boot reiserfs   notail,defaults  1 2
/dev/hda2   none  swapsw  0 0
/dev/hda3   / reiserfsnoatime 0 0

I reemerge grub (sys-boot/grub-0.97-r3)

just for grins and then:

grep -v rootfs /proc/mounts > /etc/mtab
grub-install --no-floppy /dev/hda

Installation finished. No error reported.
This is the contents of the device map /boot/grub/device.map.
Check if this is correct or not. If any of the lines is incorrect,
fix it and re-run the script `grub-install'.
(hd0)   /dev/hda


then:
exit
cd
umount /mnt/gentoo/boot /mnt/gentoo/dev /mnt/gentoo/proc /mnt/gentoo
reboot;exit
exit


It should boot, or give a kernel error...

I get the same errors as before.

I wanted to re emerge grub again to see if that is the problem.
(it did not fix the problem). So now I am going to try
Florian's early suggestion, later on tonight.

If I have notail in the fstab, and that is not sufficient,
then I have ask when was I support to use 'notail' following
the command syntax in the handbook?


James


James





-- 
[EMAIL PROTECTED] mailing list

Re: [gentoo-user] i586 install

[Sarpy Sam]

On 8/23/07, James <[EMAIL PROTECTED]> wrote:
> Hello,
>
> I seem to be having trouble installing
> grub on i586 systems.
>
> What I use is the
> "install-x86-minmal-2006.0.iso" cd to boot the system
> and follow the hadbook install guide. But I use the stage3 tarfile:
> stage3-i586-2006.1.tar.bz2
>
> Everything runs fine until I try to reboot off the hard drive.
> I get this
>
> Grub loading stage1.5
>
> Grub loading, please waite...
> Error 16
>
> It's repeatable on (2) different (amd k6) systems.
>
> Googling mostly reveals suggestions about file system
> being corrupt. I ran file system chekcs (reiserfsck)
> per previous instuctions and /boot and / come back
> clean and happy.
>
> The ide disk setup is very simple:
> fdisk /dev/hda
> #   Device BootStartEnd  Blocks   Id  System
> # /dev/hda1   *  1 50  401593+  83  Linux
> # /dev/hda2 51 185 1084387+  82  Linux swap
> # /dev/hda3   * 186 243418065092+  83  Linux
>
> and grub
>
> default 0
> fallback 1
> timeout 30
>
> #0
> title=kernel-2.6.21-gentoo-r4
> root(hd0,1)
> kernel /boot/kernel-2.6.21-gentoo-r4 root=/dev/hda3
> #0
> title=kernel-2.6.21-gentoo-r4
> root(hd0,1)
> kernel /boot/kernel-2.6.21-gentoo-r4 root=/dev/hda1
>
>
Grub method of numbering has always confused me but I thought that
root(hd0,1) would point to the hda2 partition which is the swap
partition.  Shouldn't it be root(hd0,0) to point it to the hda1
partition?  I could easily be wrong but that is what I would try.
-- 
[EMAIL PROTECTED] mailing list

Re: [gentoo-user] [OT] advices about motherboard+cpu+fan(+soundcard) combo?

[Iain Buchanan]

On Wed, 2007-08-22 at 21:01 +0200, b.n. wrote:
> Florian Philipp ha scritto:
> >> Are they silent too? I really need silent things.
> >>
> > Oh, you want silence? I hope you know that this is a real science? 
> 
> Yes :) But my current setup is silent enough for my needs (don't
> remember what cpu fan is it, nor I want to open it now), and I'd like to
> stay it that way. It is not completely silent but it gets almost unnoticed.

you might find that your current setup is silent enough because it
didn't need the cooling that a modern system will.  10 years ago, my
chipsets didn't even have heatsinks!  I had a cpu with a heatsink only
and no fan!

Anyway, what I wanted to say is this:  First try the system as is.  Then
if you have heat problems (from hard-drives, motherboard, powersupply,
etc) then add some case fans.  A larger, slower case fan can move as
much air as a smaller faster one, but slower generally means quieter, so
go for a 10cm low speed fan instead of a 8cm high speed if you can.  And
remember that good design can be more important that fans.  Try and get
your airflow right - calculate the same amount of air being blown in as
out.  Block off extra holes in the case, and use round ide cables, or
sata, or at least cable tie up your ribbon cables so they don't block
the airflow.

> > I'd say if you stick with AMD, try the boxed cooler and test it. Since it's 
> > not a dedicated silent system, it might be good enough. If you switch to 
> > Intel, buy a good cooler from the beginning.

my intel cooler sounds like a jet starting!

-- 
Iain Buchanan 

In the days of Old Terra there were experts in poisons, deviously clever 
persons who dealt in what were known as "the powders of inheritance."

  -- Filmbook excerpt, Royal Library of Kaitain

-- 
[EMAIL PROTECTED] mailing list

[gentoo-user] Re: i586 install

[James]

Sarpy Sam  gmail.com> writes:


> > #0
> > title=kernel-2.6.21-gentoo-r4
> > root(hd0,1)
> > kernel /boot/kernel-2.6.21-gentoo-r4 root=/dev/hda3
> > #0
> > title=kernel-2.6.21-gentoo-r4
> > root(hd0,1)
> > kernel /boot/kernel-2.6.21-gentoo-r4 root=/dev/hda1

> Grub method of numbering has always confused me but I thought that
> root(hd0,1) would point to the hda2 partition which is the swap
> partition.  Shouldn't it be root(hd0,0) to point it to the hda1
> partition?  I could easily be wrong but that is what I would try.



actually the above is what I tried first,
the last few attempts have been:

 root(hd0,0)

After I made that change, I even reemerged  grub
and even used the grub shell..

It's dame frustrating

The i586 has been orphaned, and it's a dam shame because
lots of them are cheap and they make a good firewall

When I finally get it to work, I'm going to dd off the
entire install and just dd the firewalls in the future
to get a working i586 firewall

Somehow, someway, the devs need to fix this install nightmare.
It's not just me, if you google, that what tons of folks
say about gentoo. Nightmare installation, easy to upgrade
and maintain.

I'm about ready to just use another distro to bootstrap the 
install.

There has to be an easier method

James



-- 
[EMAIL PROTECTED] mailing list

[gentoo-user] Can't install Gentoo on Dell Inspiron 530

[Walter Dnes]

  I got a shiny new Dell Inspiron from the "PC fairy".  Windows Vista
works OK (at least good enough for Windows).  It does not want to be
formatted.  I insert the latest minimal install CD, and things start
off OK at the beginning of the boot process.  *THE USB KEYBOARD WORKS OK
AT THE BEGINNING*  I can type in "gentoo" or "gentoo-nofb".  A bunch of
modules load, and eventually we get to the "default keymap" stage, at
which point, things fall apart.

  1) The keyboard (USB) now types each character twice.  If I attempt to
type "41", it comes out as "4411", and the install program complains
about an invalid keymap number.

  2) Even though the CD managed to boot initially, it now complains
about not being able to find any bootable medium.  The CMOS setup shows...

SATA-0   SAMSUNG HD501LJ
SATA-1   HL-DT-ST CD-RW/DVD-R
SATA-2   port not present
SATA-3   port not present
SATA-4   HL-DT-STDVD+-RW GSA-
SATA-5   None

Drive A  1.44M, 3.5 in

  The install asks me to type in either the boot device or "shell" for a
shell.  Of course, with a b0rk3n keyboard, I can't do either.

-- 
Walter Dnes <[EMAIL PROTECTED]> In linux /sbin/init is Job #1
Q. Mr. Ghandi, what do you think of Microsoft security?
A. I think it would be a good idea.
-- 
[EMAIL PROTECTED] mailing list

Re: [gentoo-user] [OT] illegal instruction ... in modf?

[Iain Buchanan]

On Sat, 2007-08-18 at 02:16 +0200, Alex Schuster wrote:
> Graham Murray writes:
> 
> > Iain Buchanan <[EMAIL PROTECTED]> writes:
> > > I'm compiling a program blah.cpp with g++ 4.1.2.  I've rebuilt lots
> > > of my system with 4.1.2.  I'm copying the binary onto another gentoo
> > > system which has gcc 4.1.1 as the default compiler.  When I run the
> > > binary on that system, I get this:
> > >
> > > $ ./blah
> > > Illegal instruction
> [...]
> > That sounds like you do not have the same processor in both systems and
> > the processor on the 2nd system is not the same architecture as the
> > -march= with which the program was compiled. eg if you compiled on a P4
> > and used -march=pentium4 then tried running the program on a Pentium.
> 
> At least the system was probably compiles with -march. Even if the actual 
> program then was compiled without, some static system library was, and 
> this gives the error.

hm, my system has always been compiled with the following options:

CFLAGS="-O2 -march=pentium4 -pipe ${DEBUG}"
CHOST="i686-pc-linux-gnu"

and the target system is
CFLAGS="-O2 -march=i686 -pipe"
CHOST="i686-pc-linux-gnu"

The problematic program in question is compiled with -static.

But what confuses me is why did this same setup work with gcc3.4.6?

thanks,
-- 
Iain Buchanan 

"It is hard to overstate the debt that we owe to men and women of genius."
-- Robert G. Ingersoll

-- 
[EMAIL PROTECTED] mailing list