From stefan at mdy.univie.ac.at  Sat Nov  1 07:02:32 2008
From: stefan at mdy.univie.ac.at (Stefan Boresch)
Date: Sat, 1 Nov 2008 15:02:32 +0100
Subject: [cAos] Some feedback/problems concerning caos-nsa-1.0rc2.i386
In-Reply-To: <571f1a060810311656i173f6758j17aa29cac9e86e18@mail.gmail.com>
References: <20081030130252.GM28976@loop.mdy.univie.ac.at>
	<571f1a060810300801u557fef05v975cfa920e827086@mail.gmail.com>
	<20081030155109.GP28976@loop.mdy.univie.ac.at>
	<571f1a060810301509r7ce70654n8507f8c32022f2a0@mail.gmail.com>
	<20081031094801.GR28976@loop.mdy.univie.ac.at>
	<571f1a060810311656i173f6758j17aa29cac9e86e18@mail.gmail.com>
Message-ID: <20081101140232.GA19635@loop.mdy.univie.ac.at>

Greg,

thanks for your efforts!

On Fri, Oct 31, 2008 at 04:56:42PM -0700, Greg Kurtzer wrote:
> On Fri, Oct 31, 2008 at 2:48 AM, Stefan Boresch <stefan at mdy.univie.ac.at> wrote:
> > On Thu, Oct 30, 2008 at 03:09:26PM -0700, Greg Kurtzer wrote:
> 
> Can you give this kernel a spin?
> 
> http://www.runlevelzero.net/tmp/caos-kernel-2.6.27.4-1.nsa1.src.rpm
> 

Unfortunately, rpmbuild --rebuild caos-kernel-2.6.27.4-1.nsa1.src.rpm

doesn't compile for me (on ia32) -- these are the
last output messages:

...
  LD      drivers/built-in.o
  LD      vmlinux.o
  MODPOST vmlinux.o
WARNING: modpost: Found 1 section mismatch(es).
To see full details build your kernel with:
'make CONFIG_DEBUG_SECTION_MISMATCH=y'
  GEN     .version
  CHK     include/linux/compile.h
  UPD     include/linux/compile.h
  CC      init/version.o
  LD      init/built-in.o
  LD      .tmp_vmlinux1
  KSYM    .tmp_kallsyms1.S
No valid symbol.
make: *** [.tmp_kallsyms1.S] Error 1
error: Bad exit status from /var/tmp/rpm-tmp.25272 (%build)

> >
> > I think that's all I can report for the moment ...
> 
> Do you happen to have your xorg.conf from your ubuntu install that we can test?
> 

I hope that there is no misunderstanding: On this Intel G45 based 
machine I don't have Ubuntu running, since 8.04s kernel is too old to 
even recognize the ethernet (which is fun when trying a network install 
;-). I could try to throw 8.10 on it, which supposedly supports the 
network card, not sure about the X server though ... (I haven't done 
this since for our Ubuntu setup I had planned to stay with 8.04 for a 
while ...) Normally, the xorg.conf that ubuntu provides is a bunch of 
dummy entries, i.e., all is left to the X server anyways ... I am happy 
to send this, but I doubt you'll gain any insights.

Hopefully you can help me get the kernel.rpm compiled, then I could 
test it on Monday (i.e., reboot into the new kernel to see what 
happens). IF the X problems remain with the new kernel, then I guess I 
could try how far Ubuntu 8.10 succeeds on this machine and (if it works) 
mail configuration files etc.

Best regards,

Stefan

-- 
Stefan Boresch
Institute for Computational Biological Chemistry
University of Vienna, Waehringerstr. 17       A-1090 Vienna, Austria
Phone: -43-1-427752715                        Fax:   -43-1-427752790


From stefan at mdy.univie.ac.at  Sun Nov  2 07:05:05 2008
From: stefan at mdy.univie.ac.at (stefan at mdy.univie.ac.at)
Date: Sun, 2 Nov 2008 16:05:05 +0100 (CET)
Subject: [cAos] Some feedback/problems concerning caos-nsa-1.0rc2.i386
In-Reply-To: <571f1a060810311656i173f6758j17aa29cac9e86e18@mail.gmail.com>
References: <20081030130252.GM28976@loop.mdy.univie.ac.at>
	<571f1a060810300801u557fef05v975cfa920e827086@mail.gmail.com>
	<20081030155109.GP28976@loop.mdy.univie.ac.at>
	<571f1a060810301509r7ce70654n8507f8c32022f2a0@mail.gmail.com>
	<20081031094801.GR28976@loop.mdy.univie.ac.at>
	<571f1a060810311656i173f6758j17aa29cac9e86e18@mail.gmail.com>
Message-ID: <10160.62.47.156.76.1225638305.squirrel@www.mdy.univie.ac.at>

Greg,

> On Fri, Oct 31, 2008 at 2:48 AM, Stefan Boresch <stefan at mdy.univie.ac.at>
> wrote:
>> On Thu, Oct 30, 2008 at 03:09:26PM -0700, Greg Kurtzer wrote:
> http://www.runlevelzero.net/tmp/caos-kernel-2.6.27.4-1.nsa1.src.rpm
>

was trying to work around the point of failure

KSYM .top_kallsyms1.S
No valid symbol

by hand, but what I found leaves me more confused than ever ... Basically,
(as you presumably know perfectly) the KSYM hides

 nm -n .tmp_vmlinux1 | scripts/kallsyms --all-symbols > .tmp_kallsyms1.S

and indeed the nm segfaults! However, whatever .tmp_vmlinux1 is, it is
a mighty strange file since e.g.,

file .tmp_vmlinux1

produces hundreds of output lines containing stuff like ...
...
bad note name size 0xc0834b03, bad note name size 0xc0a6f734, bad note
name size 0xc0834a45, bad note description size 0xc09751d7, bad note name
size 0xc0b60364,
...

I guess what I am unsure about is whether this is a problem of the rpm
(kernel config file), a kernel problem, a
nm (binutil?) problem, or whether this is  indicative of hardware issues
(bad memory). Could it be that installing the ia32 version on a quadcore
(hence em64t) causes this ...

Concerning video itself, there seem to be some pertinent patches to
xf86-video-intel-2.4.2 which I have compiled; however, testing has to
wait till tomorrow.

Best,

Stefan Boresch



From gmkurtzer at gmail.com  Sun Nov  2 08:27:20 2008
From: gmkurtzer at gmail.com (Greg Kurtzer)
Date: Sun, 2 Nov 2008 09:27:20 -0700
Subject: [cAos] Some feedback/problems concerning caos-nsa-1.0rc2.i386
In-Reply-To: <10160.62.47.156.76.1225638305.squirrel@www.mdy.univie.ac.at>
References: <20081030130252.GM28976@loop.mdy.univie.ac.at>
	<571f1a060810300801u557fef05v975cfa920e827086@mail.gmail.com>
	<20081030155109.GP28976@loop.mdy.univie.ac.at>
	<571f1a060810301509r7ce70654n8507f8c32022f2a0@mail.gmail.com>
	<20081031094801.GR28976@loop.mdy.univie.ac.at>
	<571f1a060810311656i173f6758j17aa29cac9e86e18@mail.gmail.com>
	<10160.62.47.156.76.1225638305.squirrel@www.mdy.univie.ac.at>
Message-ID: <571f1a060811020827n6c2e5c7xe5c01bac2dff8a29@mail.gmail.com>

Stefan,

Thanks for the info. I too have the same issue when compiling the
32bit kernel (problem doesn't exist with x86_64). I have tried several
patches to head_32.S which is where I thought the problem is, but none
fixed that error without introducing others. I have also tried to
disable grsecurity, but the kernel still fails to build.

At this point, it could either be our .config but more likely it is a
bug in the kernel itself and hopefully the next patch release will fix
it. I will play with it a bit more today and let you know.

Greg



On Sun, Nov 2, 2008 at 8:05 AM,  <stefan at mdy.univie.ac.at> wrote:
> Greg,
>
>> On Fri, Oct 31, 2008 at 2:48 AM, Stefan Boresch <stefan at mdy.univie.ac.at>
>> wrote:
>>> On Thu, Oct 30, 2008 at 03:09:26PM -0700, Greg Kurtzer wrote:
>> http://www.runlevelzero.net/tmp/caos-kernel-2.6.27.4-1.nsa1.src.rpm
>>
>
> was trying to work around the point of failure
>
> KSYM .top_kallsyms1.S
> No valid symbol
>
> by hand, but what I found leaves me more confused than ever ... Basically,
> (as you presumably know perfectly) the KSYM hides
>
>  nm -n .tmp_vmlinux1 | scripts/kallsyms --all-symbols > .tmp_kallsyms1.S
>
> and indeed the nm segfaults! However, whatever .tmp_vmlinux1 is, it is
> a mighty strange file since e.g.,
>
> file .tmp_vmlinux1
>
> produces hundreds of output lines containing stuff like ...
> ...
> bad note name size 0xc0834b03, bad note name size 0xc0a6f734, bad note
> name size 0xc0834a45, bad note description size 0xc09751d7, bad note name
> size 0xc0b60364,
> ...
>
> I guess what I am unsure about is whether this is a problem of the rpm
> (kernel config file), a kernel problem, a
> nm (binutil?) problem, or whether this is  indicative of hardware issues
> (bad memory). Could it be that installing the ia32 version on a quadcore
> (hence em64t) causes this ...
>
> Concerning video itself, there seem to be some pertinent patches to
> xf86-video-intel-2.4.2 which I have compiled; however, testing has to
> wait till tomorrow.
>
> Best,
>
> Stefan Boresch
>
> _______________________________________________
> cAos mailing list
> cAos at caoslinux.org
> http://lists.caosity.org/mailman/listinfo/caos
>



-- 
Greg Kurtzer
http://www.infiscale.com/
http://www.runlevelzero.net/
http://www.perceus.org/
http://www.caoslinux.org/


From landman at scalableinformatics.com  Sun Nov  2 10:59:14 2008
From: landman at scalableinformatics.com (Joe Landman)
Date: Sun, 02 Nov 2008 13:59:14 -0500
Subject: [cAos] Some feedback/problems concerning caos-nsa-1.0rc2.i386
In-Reply-To: <571f1a060811020827n6c2e5c7xe5c01bac2dff8a29@mail.gmail.com>
References: <20081030130252.GM28976@loop.mdy.univie.ac.at>	<571f1a060810300801u557fef05v975cfa920e827086@mail.gmail.com>	<20081030155109.GP28976@loop.mdy.univie.ac.at>	<571f1a060810301509r7ce70654n8507f8c32022f2a0@mail.gmail.com>	<20081031094801.GR28976@loop.mdy.univie.ac.at>	<571f1a060810311656i173f6758j17aa29cac9e86e18@mail.gmail.com>	<10160.62.47.156.76.1225638305.squirrel@www.mdy.univie.ac.at>
	<571f1a060811020827n6c2e5c7xe5c01bac2dff8a29@mail.gmail.com>
Message-ID: <490DF882.5090901@scalableinformatics.com>

Greg Kurtzer wrote:
> Stefan,
> 
> Thanks for the info. I too have the same issue when compiling the
> 32bit kernel (problem doesn't exist with x86_64). I have tried several
> patches to head_32.S which is where I thought the problem is, but none
> fixed that error without introducing others. I have also tried to
> disable grsecurity, but the kernel still fails to build.
> 
> At this point, it could either be our .config but more likely it is a
> bug in the kernel itself and hopefully the next patch release will fix
> it. I will play with it a bit more today and let you know.

FWIW:  our 2.6.27.4 kernel builds w/o problem.  See 
http://downloads.scalableinformatics.com/downloads/kernel-2.6.27/ .  We 
are testing this for our next kernel update (beyond the 2.6.23.14 we are 
using now as our stable kernel).  The 64 bit version of this is working 
well for our systems and at that of our customers.

What additional patches are you using?

We have found a somewhat increased sensitivity of the build to compiler 
versions last week.

> 
> Greg
> 
> 
> 
> On Sun, Nov 2, 2008 at 8:05 AM,  <stefan at mdy.univie.ac.at> wrote:
>> Greg,
>>
>>> On Fri, Oct 31, 2008 at 2:48 AM, Stefan Boresch <stefan at mdy.univie.ac.at>
>>> wrote:
>>>> On Thu, Oct 30, 2008 at 03:09:26PM -0700, Greg Kurtzer wrote:
>>> http://www.runlevelzero.net/tmp/caos-kernel-2.6.27.4-1.nsa1.src.rpm
>>>
>> was trying to work around the point of failure
>>
>> KSYM .top_kallsyms1.S
>> No valid symbol
>>
>> by hand, but what I found leaves me more confused than ever ... Basically,
>> (as you presumably know perfectly) the KSYM hides
>>
>>  nm -n .tmp_vmlinux1 | scripts/kallsyms --all-symbols > .tmp_kallsyms1.S
>>
>> and indeed the nm segfaults! However, whatever .tmp_vmlinux1 is, it is
>> a mighty strange file since e.g.,
>>
>> file .tmp_vmlinux1
>>
>> produces hundreds of output lines containing stuff like ...
>> ...
>> bad note name size 0xc0834b03, bad note name size 0xc0a6f734, bad note
>> name size 0xc0834a45, bad note description size 0xc09751d7, bad note name
>> size 0xc0b60364,
>> ...
>>
>> I guess what I am unsure about is whether this is a problem of the rpm
>> (kernel config file), a kernel problem, a
>> nm (binutil?) problem, or whether this is  indicative of hardware issues
>> (bad memory). Could it be that installing the ia32 version on a quadcore
>> (hence em64t) causes this ...
>>
>> Concerning video itself, there seem to be some pertinent patches to
>> xf86-video-intel-2.4.2 which I have compiled; however, testing has to
>> wait till tomorrow.
>>
>> Best,
>>
>> Stefan Boresch
>>
>> _______________________________________________
>> cAos mailing list
>> cAos at caoslinux.org
>> http://lists.caosity.org/mailman/listinfo/caos
>>
> 
> 
> 


-- 
Joseph Landman, Ph.D
Founder and CEO
Scalable Informatics LLC,
email: landman at scalableinformatics.com
web  : http://www.scalableinformatics.com
        http://jackrabbit.scalableinformatics.com
phone: +1 734 786 8423 x121
fax  : +1 866 888 3112
cell : +1 734 612 4615


From gmkurtzer at gmail.com  Sun Nov  2 11:17:22 2008
From: gmkurtzer at gmail.com (Greg Kurtzer)
Date: Sun, 2 Nov 2008 12:17:22 -0700
Subject: [cAos] Some feedback/problems concerning caos-nsa-1.0rc2.i386
In-Reply-To: <490DF882.5090901@scalableinformatics.com>
References: <20081030130252.GM28976@loop.mdy.univie.ac.at>
	<571f1a060810300801u557fef05v975cfa920e827086@mail.gmail.com>
	<20081030155109.GP28976@loop.mdy.univie.ac.at>
	<571f1a060810301509r7ce70654n8507f8c32022f2a0@mail.gmail.com>
	<20081031094801.GR28976@loop.mdy.univie.ac.at>
	<571f1a060810311656i173f6758j17aa29cac9e86e18@mail.gmail.com>
	<10160.62.47.156.76.1225638305.squirrel@www.mdy.univie.ac.at>
	<571f1a060811020827n6c2e5c7xe5c01bac2dff8a29@mail.gmail.com>
	<490DF882.5090901@scalableinformatics.com>
Message-ID: <571f1a060811021117h489e9c5at9f5aee3c4e5211e5@mail.gmail.com>

Interesting.

Our config is at:

   http://www.runlevelzero.net/tmp/ia32.config

Would you mind trying a build 2.6.27.4 against your toolchain?

I have tried both with and without patches applied, but for the most
part, we just patch for grsecurity.

We use the gcc-4.3 compiler to build the kernel which has built the
previous kernels no problem.

Thanks,
Greg



On Sun, Nov 2, 2008 at 11:59 AM, Joe Landman
<landman at scalableinformatics.com> wrote:
> Greg Kurtzer wrote:
>> Stefan,
>>
>> Thanks for the info. I too have the same issue when compiling the
>> 32bit kernel (problem doesn't exist with x86_64). I have tried several
>> patches to head_32.S which is where I thought the problem is, but none
>> fixed that error without introducing others. I have also tried to
>> disable grsecurity, but the kernel still fails to build.
>>
>> At this point, it could either be our .config but more likely it is a
>> bug in the kernel itself and hopefully the next patch release will fix
>> it. I will play with it a bit more today and let you know.
>
> FWIW:  our 2.6.27.4 kernel builds w/o problem.  See
> http://downloads.scalableinformatics.com/downloads/kernel-2.6.27/ .  We
> are testing this for our next kernel update (beyond the 2.6.23.14 we are
> using now as our stable kernel).  The 64 bit version of this is working
> well for our systems and at that of our customers.
>
> What additional patches are you using?
>
> We have found a somewhat increased sensitivity of the build to compiler
> versions last week.
>
>>
>> Greg
>>
>>
>>
>> On Sun, Nov 2, 2008 at 8:05 AM,  <stefan at mdy.univie.ac.at> wrote:
>>> Greg,
>>>
>>>> On Fri, Oct 31, 2008 at 2:48 AM, Stefan Boresch <stefan at mdy.univie.ac.at>
>>>> wrote:
>>>>> On Thu, Oct 30, 2008 at 03:09:26PM -0700, Greg Kurtzer wrote:
>>>> http://www.runlevelzero.net/tmp/caos-kernel-2.6.27.4-1.nsa1.src.rpm
>>>>
>>> was trying to work around the point of failure
>>>
>>> KSYM .top_kallsyms1.S
>>> No valid symbol
>>>
>>> by hand, but what I found leaves me more confused than ever ... Basically,
>>> (as you presumably know perfectly) the KSYM hides
>>>
>>>  nm -n .tmp_vmlinux1 | scripts/kallsyms --all-symbols > .tmp_kallsyms1.S
>>>
>>> and indeed the nm segfaults! However, whatever .tmp_vmlinux1 is, it is
>>> a mighty strange file since e.g.,
>>>
>>> file .tmp_vmlinux1
>>>
>>> produces hundreds of output lines containing stuff like ...
>>> ...
>>> bad note name size 0xc0834b03, bad note name size 0xc0a6f734, bad note
>>> name size 0xc0834a45, bad note description size 0xc09751d7, bad note name
>>> size 0xc0b60364,
>>> ...
>>>
>>> I guess what I am unsure about is whether this is a problem of the rpm
>>> (kernel config file), a kernel problem, a
>>> nm (binutil?) problem, or whether this is  indicative of hardware issues
>>> (bad memory). Could it be that installing the ia32 version on a quadcore
>>> (hence em64t) causes this ...
>>>
>>> Concerning video itself, there seem to be some pertinent patches to
>>> xf86-video-intel-2.4.2 which I have compiled; however, testing has to
>>> wait till tomorrow.
>>>
>>> Best,
>>>
>>> Stefan Boresch
>>>
>>> _______________________________________________
>>> cAos mailing list
>>> cAos at caoslinux.org
>>> http://lists.caosity.org/mailman/listinfo/caos
>>>
>>
>>
>>
>
>
> --
> Joseph Landman, Ph.D
> Founder and CEO
> Scalable Informatics LLC,
> email: landman at scalableinformatics.com
> web  : http://www.scalableinformatics.com
>        http://jackrabbit.scalableinformatics.com
> phone: +1 734 786 8423 x121
> fax  : +1 866 888 3112
> cell : +1 734 612 4615
> _______________________________________________
> cAos mailing list
> cAos at caoslinux.org
> http://lists.caosity.org/mailman/listinfo/caos
>



-- 
Greg Kurtzer
http://www.infiscale.com/
http://www.runlevelzero.net/
http://www.perceus.org/
http://www.caoslinux.org/


From landman at scalableinformatics.com  Sun Nov  2 13:48:45 2008
From: landman at scalableinformatics.com (Joe Landman)
Date: Sun, 02 Nov 2008 16:48:45 -0500
Subject: [cAos] Some feedback/problems concerning caos-nsa-1.0rc2.i386
In-Reply-To: <571f1a060811021117h489e9c5at9f5aee3c4e5211e5@mail.gmail.com>
References: <20081030130252.GM28976@loop.mdy.univie.ac.at>	<571f1a060810300801u557fef05v975cfa920e827086@mail.gmail.com>	<20081030155109.GP28976@loop.mdy.univie.ac.at>	<571f1a060810301509r7ce70654n8507f8c32022f2a0@mail.gmail.com>	<20081031094801.GR28976@loop.mdy.univie.ac.at>	<571f1a060810311656i173f6758j17aa29cac9e86e18@mail.gmail.com>	<10160.62.47.156.76.1225638305.squirrel@www.mdy.univie.ac.at>	<571f1a060811020827n6c2e5c7xe5c01bac2dff8a29@mail.gmail.com>	<490DF882.5090901@scalableinformatics.com>
	<571f1a060811021117h489e9c5at9f5aee3c4e5211e5@mail.gmail.com>
Message-ID: <490E203D.8060801@scalableinformatics.com>

Greg Kurtzer wrote:
> Interesting.
> 
> Our config is at:
> 
>    http://www.runlevelzero.net/tmp/ia32.config
> 
> Would you mind trying a build 2.6.27.4 against your toolchain?
> 
> I have tried both with and without patches applied, but for the most
> part, we just patch for grsecurity.

Ok,  which grsecurity patch are you using (2.1.11)?  Does it apply 
cleanly against 2.6.27?

Do you want to see a diff of the .configs?

trying it now ...

> 
> We use the gcc-4.3 compiler to build the kernel which has built the
> previous kernels no problem.
> 
> Thanks,
> Greg
> 
> 
> 
> On Sun, Nov 2, 2008 at 11:59 AM, Joe Landman
> <landman at scalableinformatics.com> wrote:
>> Greg Kurtzer wrote:
>>> Stefan,
>>>
>>> Thanks for the info. I too have the same issue when compiling the
>>> 32bit kernel (problem doesn't exist with x86_64). I have tried several
>>> patches to head_32.S which is where I thought the problem is, but none
>>> fixed that error without introducing others. I have also tried to
>>> disable grsecurity, but the kernel still fails to build.
>>>
>>> At this point, it could either be our .config but more likely it is a
>>> bug in the kernel itself and hopefully the next patch release will fix
>>> it. I will play with it a bit more today and let you know.
>> FWIW:  our 2.6.27.4 kernel builds w/o problem.  See
>> http://downloads.scalableinformatics.com/downloads/kernel-2.6.27/ .  We
>> are testing this for our next kernel update (beyond the 2.6.23.14 we are
>> using now as our stable kernel).  The 64 bit version of this is working
>> well for our systems and at that of our customers.
>>
>> What additional patches are you using?
>>
>> We have found a somewhat increased sensitivity of the build to compiler
>> versions last week.
>>
>>> Greg
>>>
>>>
>>>
>>> On Sun, Nov 2, 2008 at 8:05 AM,  <stefan at mdy.univie.ac.at> wrote:
>>>> Greg,
>>>>
>>>>> On Fri, Oct 31, 2008 at 2:48 AM, Stefan Boresch <stefan at mdy.univie.ac.at>
>>>>> wrote:
>>>>>> On Thu, Oct 30, 2008 at 03:09:26PM -0700, Greg Kurtzer wrote:
>>>>> http://www.runlevelzero.net/tmp/caos-kernel-2.6.27.4-1.nsa1.src.rpm
>>>>>
>>>> was trying to work around the point of failure
>>>>
>>>> KSYM .top_kallsyms1.S
>>>> No valid symbol
>>>>
>>>> by hand, but what I found leaves me more confused than ever ... Basically,
>>>> (as you presumably know perfectly) the KSYM hides
>>>>
>>>>  nm -n .tmp_vmlinux1 | scripts/kallsyms --all-symbols > .tmp_kallsyms1.S
>>>>
>>>> and indeed the nm segfaults! However, whatever .tmp_vmlinux1 is, it is
>>>> a mighty strange file since e.g.,
>>>>
>>>> file .tmp_vmlinux1
>>>>
>>>> produces hundreds of output lines containing stuff like ...
>>>> ...
>>>> bad note name size 0xc0834b03, bad note name size 0xc0a6f734, bad note
>>>> name size 0xc0834a45, bad note description size 0xc09751d7, bad note name
>>>> size 0xc0b60364,
>>>> ...
>>>>
>>>> I guess what I am unsure about is whether this is a problem of the rpm
>>>> (kernel config file), a kernel problem, a
>>>> nm (binutil?) problem, or whether this is  indicative of hardware issues
>>>> (bad memory). Could it be that installing the ia32 version on a quadcore
>>>> (hence em64t) causes this ...
>>>>
>>>> Concerning video itself, there seem to be some pertinent patches to
>>>> xf86-video-intel-2.4.2 which I have compiled; however, testing has to
>>>> wait till tomorrow.
>>>>
>>>> Best,
>>>>
>>>> Stefan Boresch
>>>>
>>>> _______________________________________________
>>>> cAos mailing list
>>>> cAos at caoslinux.org
>>>> http://lists.caosity.org/mailman/listinfo/caos
>>>>
>>>
>>>
>>
>> --
>> Joseph Landman, Ph.D
>> Founder and CEO
>> Scalable Informatics LLC,
>> email: landman at scalableinformatics.com
>> web  : http://www.scalableinformatics.com
>>        http://jackrabbit.scalableinformatics.com
>> phone: +1 734 786 8423 x121
>> fax  : +1 866 888 3112
>> cell : +1 734 612 4615
>> _______________________________________________
>> cAos mailing list
>> cAos at caoslinux.org
>> http://lists.caosity.org/mailman/listinfo/caos
>>
> 
> 
> 


-- 
Joseph Landman, Ph.D
Founder and CEO
Scalable Informatics LLC,
email: landman at scalableinformatics.com
web  : http://www.scalableinformatics.com
        http://jackrabbit.scalableinformatics.com
phone: +1 734 786 8423 x121
fax  : +1 866 888 3112
cell : +1 734 612 4615


From landman at scalableinformatics.com  Sun Nov  2 14:01:06 2008
From: landman at scalableinformatics.com (Joe Landman)
Date: Sun, 02 Nov 2008 17:01:06 -0500
Subject: [cAos] Some feedback/problems concerning caos-nsa-1.0rc2.i386
In-Reply-To: <490E203D.8060801@scalableinformatics.com>
References: <20081030130252.GM28976@loop.mdy.univie.ac.at>	<571f1a060810300801u557fef05v975cfa920e827086@mail.gmail.com>	<20081030155109.GP28976@loop.mdy.univie.ac.at>	<571f1a060810301509r7ce70654n8507f8c32022f2a0@mail.gmail.com>	<20081031094801.GR28976@loop.mdy.univie.ac.at>	<571f1a060810311656i173f6758j17aa29cac9e86e18@mail.gmail.com>	<10160.62.47.156.76.1225638305.squirrel@www.mdy.univie.ac.at>	<571f1a060811020827n6c2e5c7xe5c01bac2dff8a29@mail.gmail.com>	<490DF882.5090901@scalableinformatics.com>	<571f1a060811021117h489e9c5at9f5aee3c4e5211e5@mail.gmail.com>
	<490E203D.8060801@scalableinformatics.com>
Message-ID: <490E2322.5080501@scalableinformatics.com>

grsecurity 2.1.11 doesn't patch 2.6.27 stock kernel correctly.  Are you 
using a different patch?  From CVS?

  patch -p1 < ../grsecurity-2.1.11-2.6.24.5-200804211829.patch
patching file arch/alpha/kernel/module.c
Hunk #1 succeeded at 182 (offset 6 lines).
patching file arch/alpha/kernel/osf_sys.c
Hunk #1 succeeded at 1232 (offset -56 lines).
patching file arch/alpha/kernel/ptrace.c
patching file arch/alpha/mm/fault.c
patching file arch/arm/mm/mmap.c
patching file arch/avr32/mm/fault.c
patching file arch/ia64/ia32/binfmt_elf32.c
patching file arch/ia64/ia32/ia32priv.h
Hunk #1 succeeded at 296 (offset -7 lines).
patching file arch/ia64/kernel/module.c
Hunk #1 FAILED at 321.
Hunk #2 succeeded at 491 (offset -8 lines).
Hunk #4 succeeded at 848 with fuzz 2 (offset -8 lines).
1 out of 4 hunks FAILED -- saving rejects to file 
arch/ia64/kernel/module.c.rej
patching file arch/ia64/kernel/ptrace.c
Hunk #1 FAILED at 17.
Hunk #2 FAILED at 1452.
2 out of 2 hunks FAILED -- saving rejects to file 
arch/ia64/kernel/ptrace.c.rej
patching file arch/ia64/kernel/sys_ia64.c
patching file arch/ia64/mm/fault.c
patching file arch/ia64/mm/init.c
Hunk #2 succeeded at 122 (offset -6 lines).
patching file arch/mips/kernel/binfmt_elfn32.c
patching file arch/mips/kernel/binfmt_elfo32.c
patching file arch/mips/kernel/syscall.c
Hunk #1 succeeded at 100 (offset 7 lines).
patching file arch/mips/mm/fault.c
patching file arch/parisc/kernel/module.c
Hunk #1 succeeded at 75 (offset 2 lines).
Hunk #3 succeeded at 354 (offset 2 lines).
Hunk #5 succeeded at 390 (offset 2 lines).
Hunk #7 succeeded at 783 (offset 2 lines).
patching file arch/parisc/kernel/sys_parisc.c
Hunk #1 succeeded at 98 (offset -13 lines).
patching file arch/parisc/kernel/traps.c
Hunk #1 succeeded at 732 (offset 19 lines).
patching file arch/parisc/mm/fault.c
patching file arch/powerpc/kernel/module_32.c
Hunk #1 succeeded at 158 (offset 32 lines).
patching file arch/powerpc/kernel/signal_32.c
Hunk #1 succeeded at 857 (offset 126 lines).
patching file arch/powerpc/kernel/signal_64.c
Hunk #1 succeeded at 434 (offset 65 lines).
patching file arch/powerpc/kernel/vdso.c
Hunk #1 succeeded at 212 (offset 1 line).
patching file arch/powerpc/mm/fault.c
Hunk #3 succeeded at 495 (offset -25 lines).
Hunk #4 succeeded at 719 (offset -1 lines).
patching file arch/powerpc/mm/mmap.c
can't find file to patch at input line 1529
Perhaps you used the wrong -p or --strip option?
The text leading up to this was:
--------------------------
|diff -urNp linux-2.6.24.5/arch/ppc/mm/fault.c 
linux-2.6.24.5/arch/ppc/mm/fault.c
|--- linux-2.6.24.5/arch/ppc/mm/fault.c	2008-03-24 14:49:18.000000000 -0400
|+++ linux-2.6.24.5/arch/ppc/mm/fault.c	2008-03-26 20:21:07.000000000 -0400
--------------------------
File to patch:


-- 
Joseph Landman, Ph.D
Founder and CEO
Scalable Informatics LLC,
email: landman at scalableinformatics.com
web  : http://www.scalableinformatics.com
        http://jackrabbit.scalableinformatics.com
phone: +1 734 786 8423 x121
fax  : +1 866 888 3112
cell : +1 734 612 4615


From gmkurtzer at gmail.com  Sun Nov  2 15:20:34 2008
From: gmkurtzer at gmail.com (Greg Kurtzer)
Date: Sun, 2 Nov 2008 15:20:34 -0800
Subject: [cAos] Some feedback/problems concerning caos-nsa-1.0rc2.i386
In-Reply-To: <490E2322.5080501@scalableinformatics.com>
References: <20081030130252.GM28976@loop.mdy.univie.ac.at>
	<571f1a060810301509r7ce70654n8507f8c32022f2a0@mail.gmail.com>
	<20081031094801.GR28976@loop.mdy.univie.ac.at>
	<571f1a060810311656i173f6758j17aa29cac9e86e18@mail.gmail.com>
	<10160.62.47.156.76.1225638305.squirrel@www.mdy.univie.ac.at>
	<571f1a060811020827n6c2e5c7xe5c01bac2dff8a29@mail.gmail.com>
	<490DF882.5090901@scalableinformatics.com>
	<571f1a060811021117h489e9c5at9f5aee3c4e5211e5@mail.gmail.com>
	<490E203D.8060801@scalableinformatics.com>
	<490E2322.5080501@scalableinformatics.com>
Message-ID: <571f1a060811021520k729ff990l68567bf42790f86@mail.gmail.com>

Here ya go, give this a try:

http://www.grsecurity.org/test/grsecurity-2.1.12-2.6.27.4-200810272314.patch

Thanks!

On Sun, Nov 2, 2008 at 2:01 PM, Joe Landman
<landman at scalableinformatics.com> wrote:
> grsecurity 2.1.11 doesn't patch 2.6.27 stock kernel correctly.  Are you
> using a different patch?  From CVS?
>
>  patch -p1 < ../grsecurity-2.1.11-2.6.24.5-200804211829.patch
> patching file arch/alpha/kernel/module.c
> Hunk #1 succeeded at 182 (offset 6 lines).
> patching file arch/alpha/kernel/osf_sys.c
> Hunk #1 succeeded at 1232 (offset -56 lines).
> patching file arch/alpha/kernel/ptrace.c
> patching file arch/alpha/mm/fault.c
> patching file arch/arm/mm/mmap.c
> patching file arch/avr32/mm/fault.c
> patching file arch/ia64/ia32/binfmt_elf32.c
> patching file arch/ia64/ia32/ia32priv.h
> Hunk #1 succeeded at 296 (offset -7 lines).
> patching file arch/ia64/kernel/module.c
> Hunk #1 FAILED at 321.
> Hunk #2 succeeded at 491 (offset -8 lines).
> Hunk #4 succeeded at 848 with fuzz 2 (offset -8 lines).
> 1 out of 4 hunks FAILED -- saving rejects to file
> arch/ia64/kernel/module.c.rej
> patching file arch/ia64/kernel/ptrace.c
> Hunk #1 FAILED at 17.
> Hunk #2 FAILED at 1452.
> 2 out of 2 hunks FAILED -- saving rejects to file
> arch/ia64/kernel/ptrace.c.rej
> patching file arch/ia64/kernel/sys_ia64.c
> patching file arch/ia64/mm/fault.c
> patching file arch/ia64/mm/init.c
> Hunk #2 succeeded at 122 (offset -6 lines).
> patching file arch/mips/kernel/binfmt_elfn32.c
> patching file arch/mips/kernel/binfmt_elfo32.c
> patching file arch/mips/kernel/syscall.c
> Hunk #1 succeeded at 100 (offset 7 lines).
> patching file arch/mips/mm/fault.c
> patching file arch/parisc/kernel/module.c
> Hunk #1 succeeded at 75 (offset 2 lines).
> Hunk #3 succeeded at 354 (offset 2 lines).
> Hunk #5 succeeded at 390 (offset 2 lines).
> Hunk #7 succeeded at 783 (offset 2 lines).
> patching file arch/parisc/kernel/sys_parisc.c
> Hunk #1 succeeded at 98 (offset -13 lines).
> patching file arch/parisc/kernel/traps.c
> Hunk #1 succeeded at 732 (offset 19 lines).
> patching file arch/parisc/mm/fault.c
> patching file arch/powerpc/kernel/module_32.c
> Hunk #1 succeeded at 158 (offset 32 lines).
> patching file arch/powerpc/kernel/signal_32.c
> Hunk #1 succeeded at 857 (offset 126 lines).
> patching file arch/powerpc/kernel/signal_64.c
> Hunk #1 succeeded at 434 (offset 65 lines).
> patching file arch/powerpc/kernel/vdso.c
> Hunk #1 succeeded at 212 (offset 1 line).
> patching file arch/powerpc/mm/fault.c
> Hunk #3 succeeded at 495 (offset -25 lines).
> Hunk #4 succeeded at 719 (offset -1 lines).
> patching file arch/powerpc/mm/mmap.c
> can't find file to patch at input line 1529
> Perhaps you used the wrong -p or --strip option?
> The text leading up to this was:
> --------------------------
> |diff -urNp linux-2.6.24.5/arch/ppc/mm/fault.c
> linux-2.6.24.5/arch/ppc/mm/fault.c
> |--- linux-2.6.24.5/arch/ppc/mm/fault.c 2008-03-24 14:49:18.000000000 -0400
> |+++ linux-2.6.24.5/arch/ppc/mm/fault.c 2008-03-26 20:21:07.000000000 -0400
> --------------------------
> File to patch:
>
>
> --
> Joseph Landman, Ph.D
> Founder and CEO
> Scalable Informatics LLC,
> email: landman at scalableinformatics.com
> web  : http://www.scalableinformatics.com
>        http://jackrabbit.scalableinformatics.com
> phone: +1 734 786 8423 x121
> fax  : +1 866 888 3112
> cell : +1 734 612 4615
> _______________________________________________
> cAos mailing list
> cAos at caoslinux.org
> http://lists.caosity.org/mailman/listinfo/caos
>



-- 
Greg Kurtzer
http://www.infiscale.com/
http://www.runlevelzero.net/
http://www.perceus.org/
http://www.caoslinux.org/


From landman at scalableinformatics.com  Sun Nov  2 15:23:34 2008
From: landman at scalableinformatics.com (Joe Landman)
Date: Sun, 02 Nov 2008 18:23:34 -0500
Subject: [cAos] Some feedback/problems concerning caos-nsa-1.0rc2.i386
In-Reply-To: <571f1a060811021520k729ff990l68567bf42790f86@mail.gmail.com>
References: <20081030130252.GM28976@loop.mdy.univie.ac.at>	
	<571f1a060810301509r7ce70654n8507f8c32022f2a0@mail.gmail.com>	
	<20081031094801.GR28976@loop.mdy.univie.ac.at>	
	<571f1a060810311656i173f6758j17aa29cac9e86e18@mail.gmail.com>	
	<10160.62.47.156.76.1225638305.squirrel@www.mdy.univie.ac.at>	
	<571f1a060811020827n6c2e5c7xe5c01bac2dff8a29@mail.gmail.com>	
	<490DF882.5090901@scalableinformatics.com>	
	<571f1a060811021117h489e9c5at9f5aee3c4e5211e5@mail.gmail.com>	
	<490E203D.8060801@scalableinformatics.com>	
	<490E2322.5080501@scalableinformatics.com>
	<571f1a060811021520k729ff990l68567bf42790f86@mail.gmail.com>
Message-ID: <490E3676.2080105@scalableinformatics.com>

Greg Kurtzer wrote:
> Here ya go, give this a try:
> 
> http://www.grsecurity.org/test/grsecurity-2.1.12-2.6.27.4-200810272314.patch
> 

getting a 404 error on that URL.  If you have it handy, please email it.

-- 
Joseph Landman, Ph.D
Founder and CEO
Scalable Informatics LLC,
email: landman at scalableinformatics.com
web  : http://www.scalableinformatics.com
        http://jackrabbit.scalableinformatics.com
phone: +1 734 786 8423 x121
fax  : +1 866 888 3112
cell : +1 734 612 4615


From gmkurtzer at gmail.com  Sun Nov  2 15:54:35 2008
From: gmkurtzer at gmail.com (Greg Kurtzer)
Date: Sun, 2 Nov 2008 15:54:35 -0800
Subject: [cAos] Some feedback/problems concerning caos-nsa-1.0rc2.i386
In-Reply-To: <490E3676.2080105@scalableinformatics.com>
References: <20081030130252.GM28976@loop.mdy.univie.ac.at>
	<571f1a060810311656i173f6758j17aa29cac9e86e18@mail.gmail.com>
	<10160.62.47.156.76.1225638305.squirrel@www.mdy.univie.ac.at>
	<571f1a060811020827n6c2e5c7xe5c01bac2dff8a29@mail.gmail.com>
	<490DF882.5090901@scalableinformatics.com>
	<571f1a060811021117h489e9c5at9f5aee3c4e5211e5@mail.gmail.com>
	<490E203D.8060801@scalableinformatics.com>
	<490E2322.5080501@scalableinformatics.com>
	<571f1a060811021520k729ff990l68567bf42790f86@mail.gmail.com>
	<490E3676.2080105@scalableinformatics.com>
Message-ID: <571f1a060811021554v7d556470u8f38c1373386ff2b@mail.gmail.com>

Haha, sorry. They just updated it. ;)

   http://www.grsecurity.org/test/grsecurity-2.1.12-2.6.27.4-200811011834.patch



On Sun, Nov 2, 2008 at 3:23 PM, Joe Landman
<landman at scalableinformatics.com> wrote:
> Greg Kurtzer wrote:
>>
>> Here ya go, give this a try:
>>
>>
>> http://www.grsecurity.org/test/grsecurity-2.1.12-2.6.27.4-200810272314.patch
>>
>
> getting a 404 error on that URL.  If you have it handy, please email it.
>
> --
> Joseph Landman, Ph.D
> Founder and CEO
> Scalable Informatics LLC,
> email: landman at scalableinformatics.com
> web  : http://www.scalableinformatics.com
>       http://jackrabbit.scalableinformatics.com
> phone: +1 734 786 8423 x121
> fax  : +1 866 888 3112
> cell : +1 734 612 4615
>



-- 
Greg Kurtzer
http://www.infiscale.com/
http://www.runlevelzero.net/
http://www.perceus.org/
http://www.caoslinux.org/


From landman at scalableinformatics.com  Sun Nov  2 16:23:06 2008
From: landman at scalableinformatics.com (Joe Landman)
Date: Sun, 02 Nov 2008 19:23:06 -0500
Subject: [cAos] Some feedback/problems concerning caos-nsa-1.0rc2.i386
In-Reply-To: <571f1a060811021554v7d556470u8f38c1373386ff2b@mail.gmail.com>
References: <20081030130252.GM28976@loop.mdy.univie.ac.at>	<571f1a060810311656i173f6758j17aa29cac9e86e18@mail.gmail.com>	<10160.62.47.156.76.1225638305.squirrel@www.mdy.univie.ac.at>	<571f1a060811020827n6c2e5c7xe5c01bac2dff8a29@mail.gmail.com>	<490DF882.5090901@scalableinformatics.com>	<571f1a060811021117h489e9c5at9f5aee3c4e5211e5@mail.gmail.com>	<490E203D.8060801@scalableinformatics.com>	<490E2322.5080501@scalableinformatics.com>	<571f1a060811021520k729ff990l68567bf42790f86@mail.gmail.com>	<490E3676.2080105@scalableinformatics.com>
	<571f1a060811021554v7d556470u8f38c1373386ff2b@mail.gmail.com>
Message-ID: <490E446A.2020001@scalableinformatics.com>

Greg Kurtzer wrote:
> Haha, sorry. They just updated it. ;)
> 
>    http://www.grsecurity.org/test/grsecurity-2.1.12-2.6.27.4-200811011834.patch

Still not perfect ...


patching file arch/x86/kernel/acpi/sleep.c
Hunk #1 succeeded at 36 (offset -1 lines).
Hunk #2 FAILED at 102.
1 out of 2 hunks FAILED -- saving rejects to file 
arch/x86/kernel/acpi/sleep.c.rej

This one inserts a pax_open_kernel call before assigning a gdt 
descriptor.  Might be able to insert this by hand.


patching file fs/proc/task_mmu.c
Hunk #2 FAILED at 209.
Hunk #3 succeeded at 237 (offset 6 lines).
Hunk #5 succeeded at 403 with fuzz 2 (offset -10 lines).
Hunk #6 succeeded at 438 (offset 2 lines).
Hunk #7 succeeded at 782 (offset -12 lines).
1 out of 7 hunks FAILED -- saving rejects to file fs/proc/task_mmu.c.rej

This is a PAX address randomization bit.

Our toolchain is Makefile based, so I have to force this by hand to see 
if I can get it to build.  It could collide with some of our other patches.

-- 
Joseph Landman, Ph.D
Founder and CEO
Scalable Informatics LLC,
email: landman at scalableinformatics.com
web  : http://www.scalableinformatics.com
        http://jackrabbit.scalableinformatics.com
phone: +1 734 786 8423 x121
fax  : +1 866 888 3112
cell : +1 734 612 4615


From gmkurtzer at gmail.com  Sun Nov  2 16:24:43 2008
From: gmkurtzer at gmail.com (Greg Kurtzer)
Date: Sun, 2 Nov 2008 16:24:43 -0800
Subject: [cAos] Some feedback/problems concerning caos-nsa-1.0rc2.i386
In-Reply-To: <490E446A.2020001@scalableinformatics.com>
References: <20081030130252.GM28976@loop.mdy.univie.ac.at>
	<571f1a060811020827n6c2e5c7xe5c01bac2dff8a29@mail.gmail.com>
	<490DF882.5090901@scalableinformatics.com>
	<571f1a060811021117h489e9c5at9f5aee3c4e5211e5@mail.gmail.com>
	<490E203D.8060801@scalableinformatics.com>
	<490E2322.5080501@scalableinformatics.com>
	<571f1a060811021520k729ff990l68567bf42790f86@mail.gmail.com>
	<490E3676.2080105@scalableinformatics.com>
	<571f1a060811021554v7d556470u8f38c1373386ff2b@mail.gmail.com>
	<490E446A.2020001@scalableinformatics.com>
Message-ID: <571f1a060811021624o28dc0e4fh7e61cc5863d0a223@mail.gmail.com>

I tried it with this patch disabled, and we still got errors. Building
your kernel now and crossing fingers. ;)

Thanks,
Greg

On Sun, Nov 2, 2008 at 4:23 PM, Joe Landman
<landman at scalableinformatics.com> wrote:
> Greg Kurtzer wrote:
>> Haha, sorry. They just updated it. ;)
>>
>>    http://www.grsecurity.org/test/grsecurity-2.1.12-2.6.27.4-200811011834.patch
>
> Still not perfect ...
>
>
> patching file arch/x86/kernel/acpi/sleep.c
> Hunk #1 succeeded at 36 (offset -1 lines).
> Hunk #2 FAILED at 102.
> 1 out of 2 hunks FAILED -- saving rejects to file
> arch/x86/kernel/acpi/sleep.c.rej
>
> This one inserts a pax_open_kernel call before assigning a gdt
> descriptor.  Might be able to insert this by hand.
>
>
> patching file fs/proc/task_mmu.c
> Hunk #2 FAILED at 209.
> Hunk #3 succeeded at 237 (offset 6 lines).
> Hunk #5 succeeded at 403 with fuzz 2 (offset -10 lines).
> Hunk #6 succeeded at 438 (offset 2 lines).
> Hunk #7 succeeded at 782 (offset -12 lines).
> 1 out of 7 hunks FAILED -- saving rejects to file fs/proc/task_mmu.c.rej
>
> This is a PAX address randomization bit.
>
> Our toolchain is Makefile based, so I have to force this by hand to see
> if I can get it to build.  It could collide with some of our other patches.
>
> --
> Joseph Landman, Ph.D
> Founder and CEO
> Scalable Informatics LLC,
> email: landman at scalableinformatics.com
> web  : http://www.scalableinformatics.com
>        http://jackrabbit.scalableinformatics.com
> phone: +1 734 786 8423 x121
> fax  : +1 866 888 3112
> cell : +1 734 612 4615
> _______________________________________________
> cAos mailing list
> cAos at caoslinux.org
> http://lists.caosity.org/mailman/listinfo/caos
>



-- 
Greg Kurtzer
http://www.infiscale.com/
http://www.runlevelzero.net/
http://www.perceus.org/
http://www.caoslinux.org/


From gmkurtzer at gmail.com  Sun Nov  2 22:04:44 2008
From: gmkurtzer at gmail.com (Greg Kurtzer)
Date: Sun, 2 Nov 2008 23:04:44 -0700
Subject: [cAos] Some feedback/problems concerning caos-nsa-1.0rc2.i386
In-Reply-To: <571f1a060811021624o28dc0e4fh7e61cc5863d0a223@mail.gmail.com>
References: <20081030130252.GM28976@loop.mdy.univie.ac.at>
	<490DF882.5090901@scalableinformatics.com>
	<571f1a060811021117h489e9c5at9f5aee3c4e5211e5@mail.gmail.com>
	<490E203D.8060801@scalableinformatics.com>
	<490E2322.5080501@scalableinformatics.com>
	<571f1a060811021520k729ff990l68567bf42790f86@mail.gmail.com>
	<490E3676.2080105@scalableinformatics.com>
	<571f1a060811021554v7d556470u8f38c1373386ff2b@mail.gmail.com>
	<490E446A.2020001@scalableinformatics.com>
	<571f1a060811021624o28dc0e4fh7e61cc5863d0a223@mail.gmail.com>
Message-ID: <571f1a060811022204j16fe9d50s6128a6ee4565c178@mail.gmail.com>

Joe, your kernel doesn't build either due to SPEC requirements (looks
like it requires a custom build environment or --define options).

Stefan, I updated our grsecurity, and worked around all of the build
errors so give this package a shot:

http://www.runlevelzero.net/tmp/caos-kernel-2.6.27.4-1.nsa1.i386.rpm

Please let me know if this kernel helps with the DRI support for the
new Intel graphics card.

What were the intel driver updates you implemented?

Thanks to you both, and let me know how your testing turns out!

Greg



On Sun, Nov 2, 2008 at 5:24 PM, Greg Kurtzer <gmkurtzer at gmail.com> wrote:
> I tried it with this patch disabled, and we still got errors. Building
> your kernel now and crossing fingers. ;)
>
> Thanks,
> Greg
>
> On Sun, Nov 2, 2008 at 4:23 PM, Joe Landman
> <landman at scalableinformatics.com> wrote:
>> Greg Kurtzer wrote:
>>> Haha, sorry. They just updated it. ;)
>>>
>>>    http://www.grsecurity.org/test/grsecurity-2.1.12-2.6.27.4-200811011834.patch
>>
>> Still not perfect ...
>>
>>
>> patching file arch/x86/kernel/acpi/sleep.c
>> Hunk #1 succeeded at 36 (offset -1 lines).
>> Hunk #2 FAILED at 102.
>> 1 out of 2 hunks FAILED -- saving rejects to file
>> arch/x86/kernel/acpi/sleep.c.rej
>>
>> This one inserts a pax_open_kernel call before assigning a gdt
>> descriptor.  Might be able to insert this by hand.
>>
>>
>> patching file fs/proc/task_mmu.c
>> Hunk #2 FAILED at 209.
>> Hunk #3 succeeded at 237 (offset 6 lines).
>> Hunk #5 succeeded at 403 with fuzz 2 (offset -10 lines).
>> Hunk #6 succeeded at 438 (offset 2 lines).
>> Hunk #7 succeeded at 782 (offset -12 lines).
>> 1 out of 7 hunks FAILED -- saving rejects to file fs/proc/task_mmu.c.rej
>>
>> This is a PAX address randomization bit.
>>
>> Our toolchain is Makefile based, so I have to force this by hand to see
>> if I can get it to build.  It could collide with some of our other patches.
>>
>> --
>> Joseph Landman, Ph.D
>> Founder and CEO
>> Scalable Informatics LLC,
>> email: landman at scalableinformatics.com
>> web  : http://www.scalableinformatics.com
>>        http://jackrabbit.scalableinformatics.com
>> phone: +1 734 786 8423 x121
>> fax  : +1 866 888 3112
>> cell : +1 734 612 4615
>> _______________________________________________
>> cAos mailing list
>> cAos at caoslinux.org
>> http://lists.caosity.org/mailman/listinfo/caos
>>
>
>
>
> --
> Greg Kurtzer
> http://www.infiscale.com/
> http://www.runlevelzero.net/
> http://www.perceus.org/
> http://www.caoslinux.org/
>



-- 
Greg Kurtzer
http://www.infiscale.com/
http://www.runlevelzero.net/
http://www.perceus.org/
http://www.caoslinux.org/


From landman at scalableinformatics.com  Sun Nov  2 22:47:33 2008
From: landman at scalableinformatics.com (Joe Landman)
Date: Mon, 03 Nov 2008 01:47:33 -0500
Subject: [cAos] Some feedback/problems concerning caos-nsa-1.0rc2.i386
In-Reply-To: <571f1a060811022204j16fe9d50s6128a6ee4565c178@mail.gmail.com>
References: <20081030130252.GM28976@loop.mdy.univie.ac.at>	<490DF882.5090901@scalableinformatics.com>	<571f1a060811021117h489e9c5at9f5aee3c4e5211e5@mail.gmail.com>	<490E203D.8060801@scalableinformatics.com>	<490E2322.5080501@scalableinformatics.com>	<571f1a060811021520k729ff990l68567bf42790f86@mail.gmail.com>	<490E3676.2080105@scalableinformatics.com>	<571f1a060811021554v7d556470u8f38c1373386ff2b@mail.gmail.com>	<490E446A.2020001@scalableinformatics.com>	<571f1a060811021624o28dc0e4fh7e61cc5863d0a223@mail.gmail.com>
	<571f1a060811022204j16fe9d50s6128a6ee4565c178@mail.gmail.com>
Message-ID: <490E9E85.1060900@scalableinformatics.com>

Greg Kurtzer wrote:
> Joe, your kernel doesn't build either due to SPEC requirements (looks
> like it requires a custom build environment or --define opti

Yeah, we have a makefile based build environment.  Lets us build .debs 
from exactly the same sources and Makefiles as we can build RPMs.


-- 
Joseph Landman, Ph.D
Founder and CEO
Scalable Informatics LLC,
email: landman at scalableinformatics.com
web  : http://www.scalableinformatics.com
        http://jackrabbit.scalableinformatics.com
phone: +1 734 786 8423 x121
fax  : +1 866 888 3112
cell : +1 734 612 4615


From stefan at mdy.univie.ac.at  Mon Nov  3 00:54:04 2008
From: stefan at mdy.univie.ac.at (Stefan Boresch)
Date: Mon, 3 Nov 2008 09:54:04 +0100
Subject: [cAos] Some feedback/problems concerning caos-nsa-1.0rc2.i386
In-Reply-To: <571f1a060811022204j16fe9d50s6128a6ee4565c178@mail.gmail.com>
References: <490DF882.5090901@scalableinformatics.com>
	<571f1a060811021117h489e9c5at9f5aee3c4e5211e5@mail.gmail.com>
	<490E203D.8060801@scalableinformatics.com>
	<490E2322.5080501@scalableinformatics.com>
	<571f1a060811021520k729ff990l68567bf42790f86@mail.gmail.com>
	<490E3676.2080105@scalableinformatics.com>
	<571f1a060811021554v7d556470u8f38c1373386ff2b@mail.gmail.com>
	<490E446A.2020001@scalableinformatics.com>
	<571f1a060811021624o28dc0e4fh7e61cc5863d0a223@mail.gmail.com>
	<571f1a060811022204j16fe9d50s6128a6ee4565c178@mail.gmail.com>
Message-ID: <20081103085404.GW28976@loop.mdy.univie.ac.at>

Greg,

quick update:

First, I am glad that you could resolve the kernel issue -- honestly,
I thought yesterday I was hallucinating ... But, let's take things
in order:

xf86-video-intel: Sometimes it pays to be lazy and to read, e.g.,
kernelplanet.org instead of finishing a paper that should have been
submitted ages ago ...  Jesse Barnes blogs about 2.5.0 of
xf86-video-intel, referencing a list of bugs fixed:

https://bugs.freedesktop.org/showdependencytree.cgi?id=16926&hide_resolved=0

Well, one of them is exactly my problem, so I incorporated the suggested patch
into your rpm and rebuilt -- voila, X not only works, one can even get out
of it again ;-) [The site is down at the moment, so I can't point you
to the exact links] However, if you read through the above buglist, it
becomes evident that 2.4.2 probably is almost broken on the newest Intel
chipsets; hence 2.5 is probably the way to go. Unfortunately, this requires
a newer libdrm (2.4 rather than 2.3), and I was not that adventurous...

Still no 3D acceleration though, even with the 2.6.27
kernel. Unfortunately, I don't have much time at the moment to play
with this further, also the requirements for 3D according to
http://intellinuxgraphics.org/download.html are a bit daunting.  (Note
towards the bottom: "an additional AGP patch for G4X stolen memory
counting fix") 

BTW, noticed this in the syslog:
[drm] Initialized i915 1.6.0 20060119 on minor 0
mtrr: type mismatch for c0000000,10000000 old: write-back new: write-combining
set status page addr 0x02000000

The mtrr error seems to reoccur at every X restart (and X takes quite
some time to start > 5seconds)

Finally, two little X-related things/questions: First, e.g., from the
graphical login prompt, how does one get into a console window,
CTRL-ALT-F? doesn't do anything.

Second, when I started working with the newer video-intel driver, I
went to default runlevel 3. There I noted that only root could do
startx (basically, normal users could/can not move
/var/log/Xorg.log.0)

Thanks for all your help,

Stefan

-- 
Stefan Boresch
Institute for Computational Biological Chemistry
University of Vienna, Waehringerstr. 17       A-1090 Vienna, Austria
Phone: -43-1-427752715                        Fax:   -43-1-427752790


From ve3wsz at netscape.net  Thu Nov  6 13:49:11 2008
From: ve3wsz at netscape.net (Werner)
Date: Thu, 6 Nov 2008 16:49:11 -0500
Subject: [cAos] NAS rc2 install...
Message-ID: <20081106164911.6fcb113c@sysv.homeip.net>


Hi all. I just installed the most recent NSA iso and I set the
filesystem to ext2 (install fs=ext2). Everything else was defaults. When
I booted up after the install, the filesystem is in an error state even
though I shutdown properly and the fsck fails with an error, no other
information, just "error". When I try to ctr-d or put in my root
password, nothing happens. The system is frozen. I turn off the system
manually and reboot and it works fine without checking the filesystem!
Get this, I reboot again properly to make sure it's ok but again the
same problem. 

I am thinking that this may be something with the system clock being set
to local time. I had a similar issue with debian a while back. How do I
set the system clock to UTC? This is what fixed it in debian.

Werner Z.


From stefan at mdy.univie.ac.at  Fri Nov  7 00:53:02 2008
From: stefan at mdy.univie.ac.at (Stefan Boresch)
Date: Fri, 7 Nov 2008 09:53:02 +0100
Subject: [cAos] caos-nsa-1.0rc2 again
Message-ID: <20081107085302.GH5928@loop.mdy.univie.ac.at>


Some more observations (this time the x64 image, and a different machine,
but I believe that all of the following is hardware independent)

1) Since this is a server/clusterhead, I was very much inclined to use
layout=manual. Aren't your dire warnings right ...  

a) Could parted be made available in addition to fdisk? (I mean in the
initrd; I realize it's available for caos itself)

b) Since my relationship with vi is nonexistent, having to edit
/etc/fstab with vi sent me scrambling to look for my very first Unix
book ;-) (Yes, I know, real men use ex, but idiots like me are so much
more comfortable with nano ...)

c) Admittedly, a) and b) are whining. The next, however, was annoying:
Since I never understood the rationale for a separate /boot partition,
I didn't make one; only to find out upon the first reboot that grub
didn't work ...  I didn't debug this and I may have
forgotten to set /dev/sda1=/ bootable, but since I didn't arrive at a
grub prompt at all, I really think that grub wasn't installed properly
...

2) X Server was initially unusable. On this hardware VESA *is* the only
option, but the server guesses the screen resolution wrong (and on
a low quality TFT this results in completely blurred fonts). 
X -configure goes wrong as well, had to force 1280x1024 by hand.
This is not necessarily bad, but somehow Ubuntu 8.04 gets it right
automatically -- no clue how they do it, since in 8.04 they basically
also rely on the hardware recognition of the X Server ...

3) In some sidekick dialog I noted 'lunch' instead of 'lauch' ;-)

Best regards,

Stefan


-- 
Stefan Boresch
Institute for Computational Biological Chemistry
University of Vienna, Waehringerstr. 17       A-1090 Vienna, Austria
Phone: -43-1-427752715                        Fax:   -43-1-427752790


From stefan at mdy.univie.ac.at  Fri Nov  7 05:58:31 2008
From: stefan at mdy.univie.ac.at (Stefan Boresch)
Date: Fri, 7 Nov 2008 14:58:31 +0100
Subject: [cAos] Pinning, e.g., eth2 to a specific Ethernet card
Message-ID: <20081107135831.GK5928@loop.mdy.univie.ac.at>

Maybe that's all obvious to everyone else. Just for the record, the
new(?), standard(?) udev based method to pin ethernet devices to a
particular MAC address seems to work for CAOS-NSA as well. By
contrast, the (very) old(?) Redhat way of doing things (adding a
HWADDR line to /etc/sysconfig/nics/eth?)  does not work (anymore).

By putting something like the following

# This file was automatically generated by the /lib/udev/write_net_rules
# program run by the persistent-net-generator.rules rules file.
#
# You can modify it, as long as you keep each rule on a single line.

# tg3
SUBSYSTEM=="net", ACTION=="add", DRIVERS=="?*", ATTR{address}=="00:1e:8c:e0:f2:af", ATTR{type}=="1", KERNEL=="eth*", NAME="eth0"
SUBSYSTEM=="net", ACTION=="add", DRIVERS=="?*", ATTR{address}=="00:1e:8c:e0:f4:f0", ATTR{type}=="1", KERNEL=="eth*", NAME="eth1"
SUBSYSTEM=="net", ACTION=="add", DRIVERS=="?*", ATTR{address}=="00:1e:8c:e0:f6:33", ATTR{type}=="1", KERNEL=="eth*", NAME="eth2"
SUBSYSTEM=="net", ACTION=="add", DRIVERS=="?*", ATTR{address}=="00:1e:8c:e0:f8:8e", ATTR{type}=="1", KERNEL=="eth*", NAME="eth3"

into  /etc/udev/rules.d/70-persistent-net.rules
I get my network cards into the order I want them .... (Info taken from
Ubuntu documentation ...)  Ubuntu >= 8.04 generate this file during install,
so there you only have to do some editing. By contrast, the file is missing
in Caos.

Best regards,

Stefan

-- 
Stefan Boresch
Institute for Computational Biological Chemistry
University of Vienna, Waehringerstr. 17       A-1090 Vienna, Austria
Phone: -43-1-427752715                        Fax:   -43-1-427752790


From laytonjb at att.net  Fri Nov  7 06:09:59 2008
From: laytonjb at att.net (Jeff Layton)
Date: Fri, 7 Nov 2008 06:09:59 -0800 (PST)
Subject: [cAos] Pinning, e.g., eth2 to a specific Ethernet card
Message-ID: <921910.82033.qm@web80708.mail.mud.yahoo.com>

Maybe that's all obvious to everyone else. Just for the record, the

new(?), standard(?) udev based method to pin ethernet devices to a
particular MAC address seems to work for CAOS-NSA as well. By
contrast, the (very) old(?) Redhat way of doing things (adding a
HWADDR line to /etc/sysconfig/nics/eth?)  does not work (anymore).



Very Nice! I've always been wondering how to do this. I tried the RHEL way
for a while and it never worked 100% of the time. Is this tied to a specific
version of the kernel?

Thanks!

Jeff
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.infiscale.org/pipermail/caos/attachments/20081107/e557b6c1/attachment.html 

From stefan at mdy.univie.ac.at  Fri Nov  7 06:36:54 2008
From: stefan at mdy.univie.ac.at (Stefan Boresch)
Date: Fri, 7 Nov 2008 15:36:54 +0100
Subject: [cAos] Pinning, e.g., eth2 to a specific Ethernet card
In-Reply-To: <921910.82033.qm@web80708.mail.mud.yahoo.com>
References: <921910.82033.qm@web80708.mail.mud.yahoo.com>
Message-ID: <20081107143654.GL5928@loop.mdy.univie.ac.at>

On Fri, Nov 07, 2008 at 06:09:59AM -0800, Jeff Layton wrote:
> Maybe that's all obvious to everyone else. Just for the record, the
> 
> new(?), standard(?) udev based method to pin ethernet devices to a
> particular MAC address seems to work for CAOS-NSA as well. By
> contrast, the (very) old(?) Redhat way of doing things (adding a
> HWADDR line to /etc/sysconfig/nics/eth?)  does not work (anymore).
> 
> 
> 
> Very Nice! I've always been wondering how to do this. I tried the RHEL way

Glad to learn that I could tell someone something new ;-)

> for a while and it never worked 100% of the time. Is this tied to a specific
> version of the kernel?
> 

I don't know -- I think it's mostly tied to the udev subsystem
becoming more and more "pervasive" ...  Up to Ubuntu 7.10 (2.6.22
based kernel, I believe), Ubuntu used a Redhat/HWADDR like method with
a small config file, /etc/iftab). I learned the hard way with 8.04
(2.6.24 based kernel) that things had changed ... But whether this
change of how to do things was really due to the newer kernel, or whether
they started exploiting a capability of udev that had been there for
some time, I don't know ... However, as long as udev doesn't undergo
a major overhaul, I think that's how it should be handled (how the
udev people think that it should be handled ...)

Best regards,

Stefan

-- 
Stefan Boresch
Institute for Computational Biological Chemistry
University of Vienna, Waehringerstr. 17       A-1090 Vienna, Austria
Phone: -43-1-427752715                        Fax:   -43-1-427752790


From gmkurtzer at gmail.com  Fri Nov  7 13:39:09 2008
From: gmkurtzer at gmail.com (Greg Kurtzer)
Date: Fri, 7 Nov 2008 13:39:09 -0800
Subject: [cAos] caos-nsa-1.0rc2 again
In-Reply-To: <20081107085302.GH5928@loop.mdy.univie.ac.at>
References: <20081107085302.GH5928@loop.mdy.univie.ac.at>
Message-ID: <571f1a060811071339k6d400910j47004c0f1e38f1cd@mail.gmail.com>

I don't know why but your email's always go to my SPAM box. Gmail,... go figure!

On Fri, Nov 7, 2008 at 12:53 AM, Stefan Boresch <stefan at mdy.univie.ac.at> wrote:
>
> Some more observations (this time the x64 image, and a different machine,
> but I believe that all of the following is hardware independent)
>
> 1) Since this is a server/clusterhead, I was very much inclined to use
> layout=manual. Aren't your dire warnings right ...
>
> a) Could parted be made available in addition to fdisk? (I mean in the
> initrd; I realize it's available for caos itself)

Will check into this. :)

>
> b) Since my relationship with vi is nonexistent, having to edit
> /etc/fstab with vi sent me scrambling to look for my very first Unix
> book ;-) (Yes, I know, real men use ex, but idiots like me are so much
> more comfortable with nano ...)

I am not sure if that will fit in the installer bootstrap OS, but I can check.

>
> c) Admittedly, a) and b) are whining. The next, however, was annoying:
> Since I never understood the rationale for a separate /boot partition,
> I didn't make one; only to find out upon the first reboot that grub
> didn't work ...  I didn't debug this and I may have
> forgotten to set /dev/sda1=/ bootable, but since I didn't arrive at a
> grub prompt at all, I really think that grub wasn't installed properly

There are historic and file system reasons why /boot should exist, but
as you mentioned, depending on the configuration of the system it may
not be required.

> ...
>
> 2) X Server was initially unusable. On this hardware VESA *is* the only
> option, but the server guesses the screen resolution wrong (and on
> a low quality TFT this results in completely blurred fonts).
> X -configure goes wrong as well, had to force 1280x1024 by hand.
> This is not necessarily bad, but somehow Ubuntu 8.04 gets it right
> automatically -- no clue how they do it, since in 8.04 they basically
> also rely on the hardware recognition of the X Server ...

What video card is it? do you happen to have a copy of their xorg.conf?

>
> 3) In some sidekick dialog I noted 'lunch' instead of 'lauch' ;-)

Fixed. Thanks!

>
> Best regards,
>
> Stefan
>
>
> --
> Stefan Boresch
> Institute for Computational Biological Chemistry
> University of Vienna, Waehringerstr. 17       A-1090 Vienna, Austria
> Phone: -43-1-427752715                        Fax:   -43-1-427752790
> _______________________________________________
> cAos mailing list
> cAos at caoslinux.org
> http://lists.caosity.org/mailman/listinfo/caos
>



-- 
Greg Kurtzer
http://www.infiscale.com/
http://www.runlevelzero.net/
http://www.perceus.org/
http://www.caoslinux.org/


From gmkurtzer at gmail.com  Fri Nov  7 15:17:13 2008
From: gmkurtzer at gmail.com (Greg Kurtzer)
Date: Fri, 7 Nov 2008 15:17:13 -0800
Subject: [cAos] Some feedback/problems concerning caos-nsa-1.0rc2.i386
In-Reply-To: <20081103085404.GW28976@loop.mdy.univie.ac.at>
References: <490DF882.5090901@scalableinformatics.com>
	<490E203D.8060801@scalableinformatics.com>
	<490E2322.5080501@scalableinformatics.com>
	<571f1a060811021520k729ff990l68567bf42790f86@mail.gmail.com>
	<490E3676.2080105@scalableinformatics.com>
	<571f1a060811021554v7d556470u8f38c1373386ff2b@mail.gmail.com>
	<490E446A.2020001@scalableinformatics.com>
	<571f1a060811021624o28dc0e4fh7e61cc5863d0a223@mail.gmail.com>
	<571f1a060811022204j16fe9d50s6128a6ee4565c178@mail.gmail.com>
	<20081103085404.GW28976@loop.mdy.univie.ac.at>
Message-ID: <571f1a060811071517w65e41e12rc5ac29fad5eec1e9@mail.gmail.com>

On Mon, Nov 3, 2008 at 12:54 AM, Stefan Boresch <stefan at mdy.univie.ac.at> wrote:
> Greg,
>
> quick update:
>
> First, I am glad that you could resolve the kernel issue -- honestly,
> I thought yesterday I was hallucinating ... But, let's take things
> in order:

haha,...

>
> xf86-video-intel: Sometimes it pays to be lazy and to read, e.g.,
> kernelplanet.org instead of finishing a paper that should have been
> submitted ages ago ...  Jesse Barnes blogs about 2.5.0 of
> xf86-video-intel, referencing a list of bugs fixed:
>
> https://bugs.freedesktop.org/showdependencytree.cgi?id=16926&hide_resolved=0
>
> Well, one of them is exactly my problem, so I incorporated the suggested patch
> into your rpm and rebuilt -- voila, X not only works, one can even get out
> of it again ;-) [The site is down at the moment, so I can't point you
> to the exact links] However, if you read through the above buglist, it
> becomes evident that 2.4.2 probably is almost broken on the newest Intel
> chipsets; hence 2.5 is probably the way to go. Unfortunately, this requires
> a newer libdrm (2.4 rather than 2.3), and I was not that adventurous...

Yes, I noticed that too about the 2.5 update. We should be able to get
this into the update stream sometime between the 1.0 release and the
1.1 release. ;)

>
> Still no 3D acceleration though, even with the 2.6.27
> kernel. Unfortunately, I don't have much time at the moment to play
> with this further, also the requirements for 3D according to
> http://intellinuxgraphics.org/download.html are a bit daunting.  (Note
> towards the bottom: "an additional AGP patch for G4X stolen memory
> counting fix")

Will these fixes be in 2.6.28?

>
> BTW, noticed this in the syslog:
> [drm] Initialized i915 1.6.0 20060119 on minor 0
> mtrr: type mismatch for c0000000,10000000 old: write-back new: write-combining
> set status page addr 0x02000000
>
> The mtrr error seems to reoccur at every X restart (and X takes quite
> some time to start > 5seconds)
>
> Finally, two little X-related things/questions: First, e.g., from the
> graphical login prompt, how does one get into a console window,
> CTRL-ALT-F? doesn't do anything.

# chvt should do it. ctrl-alt-fX doesn't work we think it is a change
in Xorg 7. If someone is familiar with the fix, or can isolate the
patch from a different distro, please let us know so we can implement.

>
> Second, when I started working with the newer video-intel driver, I
> went to default runlevel 3. There I noted that only root could do
> startx (basically, normal users could/can not move
> /var/log/Xorg.log.0)

Yes, we didn't package X as setuid root. I am not sure if there is
another way around this yet.

>
> Thanks for all your help,

Sure thing. :)

-- 
Greg Kurtzer
http://www.infiscale.com/
http://www.runlevelzero.net/
http://www.perceus.org/
http://www.caoslinux.org/


From stefan at mdy.univie.ac.at  Mon Nov 10 08:28:28 2008
From: stefan at mdy.univie.ac.at (Stefan Boresch)
Date: Mon, 10 Nov 2008 17:28:28 +0100
Subject: [cAos] caos-nsa-1.0rc2 again
In-Reply-To: <571f1a060811071339k6d400910j47004c0f1e38f1cd@mail.gmail.com>
References: <20081107085302.GH5928@loop.mdy.univie.ac.at>
	<571f1a060811071339k6d400910j47004c0f1e38f1cd@mail.gmail.com>
Message-ID: <20081110162828.GS5928@loop.mdy.univie.ac.at>

On Fri, Nov 07, 2008 at 01:39:09PM -0800, Greg Kurtzer wrote:
> I don't know why but your email's always go to my SPAM box. Gmail,... go figure!

Strange ... Probably because I still use a local mail server for outbound
mail, which is not the MX for incoming for (mdy.)univie.ac.at (i.e., I have
no SPF(?) record ...)

> > 2) X Server was initially unusable. On this hardware VESA *is* the only
> > option, but the server guesses the screen resolution wrong (and on
> > a low quality TFT this results in completely blurred fonts).
> > X -configure goes wrong as well, had to force 1280x1024 by hand.
> > This is not necessarily bad, but somehow Ubuntu 8.04 gets it right
> > automatically -- no clue how they do it, since in 8.04 they basically
> > also rely on the hardware recognition of the X Server ...
> 
> What video card is it? do you happen to have a copy of their xorg.conf?
> 

I attach the Ubuntu xorg.conf below. As you can see, it's a complete dummy,
but it seems to work (slightly) better than caos' no xorg policy. It's
also strange that Ubuntu got the screen resolution right, whereas in the
caos case It guessed wrong, which led to the "fuzzy" fonts ... So I am
not sure that it's going to help you!

Best regards,

Stefan

-----------------------------

xorg.conf, following a default Ubuntu 8.04 install. The graphics
"card" is an on-board

06:02.0 VGA compatible controller: ATI Technologies Inc ES1000 (rev 02)

# xorg.conf (X.Org X Window System server configuration file)
#
# This file was generated by dexconf, the Debian X Configuration tool, using
# values from the debconf database.
#
# Edit this file with caution, and see the xorg.conf manual page.
# (Type "man xorg.conf" at the shell prompt.)
#
# This file is automatically updated on xserver-xorg package upgrades *only*
# if it has not been modified since the last upgrade of the xserver-xorg
# package.
#
# If you have edited this file but would like it to be automatically updated
# again, run the following command:
#   sudo dpkg-reconfigure -phigh xserver-xorg

Section "InputDevice"
        Identifier      "Generic Keyboard"
        Driver          "kbd"
        Option          "XkbRules"      "xorg"
        Option          "XkbModel"      "pc105"
        Option          "XkbLayout"     "us"
EndSection

Section "InputDevice"
        Identifier      "Configured Mouse"
        Driver          "mouse"
        Option          "CorePointer"
EndSection

Section "Device"
        Identifier      "Configured Video Device"
EndSection

Section "Monitor"
        Identifier      "Configured Monitor"
EndSection

Section "Screen"
        Identifier      "Default Screen"
        Monitor         "Configured Monitor"
        Device          "Configured Video Device"
EndSection

Section "ServerLayout"
        Identifier      "Default Layout"
        Screen          "Default Screen"
EndSection

-----------------------------

-- 
Stefan Boresch
Institute for Computational Biological Chemistry
University of Vienna, Waehringerstr. 17       A-1090 Vienna, Austria
Phone: -43-1-427752715                        Fax:   -43-1-427752790


From stefan at mdy.univie.ac.at  Mon Nov 10 08:48:21 2008
From: stefan at mdy.univie.ac.at (Stefan Boresch)
Date: Mon, 10 Nov 2008 17:48:21 +0100
Subject: [cAos] Automated perceus configuration
Message-ID: <20081110164821.GT5928@loop.mdy.univie.ac.at>

I noted that the automated perceus configuration want's to set the
total nodes number to something ridiculously high (>> 1000000, at
least on x86-64).  Whatever the configuration then does, this takes
*really* long....

For some reasons, sidekick always wants to download vnfs-s from
mirror.caoslinux.org, which is (deliberately) slow. Could one not tell
it where it can find an existing *vnfs ...?

Best regards,

Stefan Boresch

-- 
Stefan Boresch
Institute for Computational Biological Chemistry
University of Vienna, Waehringerstr. 17       A-1090 Vienna, Austria
Phone: -43-1-427752715                        Fax:   -43-1-427752790


From gmkurtzer at gmail.com  Mon Nov 10 09:30:13 2008
From: gmkurtzer at gmail.com (Greg Kurtzer)
Date: Mon, 10 Nov 2008 09:30:13 -0800
Subject: [cAos] Automated perceus configuration
In-Reply-To: <20081110164821.GT5928@loop.mdy.univie.ac.at>
References: <20081110164821.GT5928@loop.mdy.univie.ac.at>
Message-ID: <571f1a060811100930m7d533cbft67dffe2e93c5e7cb@mail.gmail.com>

On Mon, Nov 10, 2008 at 8:48 AM, Stefan Boresch <stefan at mdy.univie.ac.at> wrote:
> I noted that the automated perceus configuration want's to set the
> total nodes number to something ridiculously high (>> 1000000, at
> least on x86-64).  Whatever the configuration then does, this takes
> *really* long....

This is based on the netmask of the decided on Perceus device. We can
of course give it a max. ;)

>
> For some reasons, sidekick always wants to download vnfs-s from
> mirror.caoslinux.org, which is (deliberately) slow. Could one not tell
> it where it can find an existing *vnfs ...?

This should be a yes/no dialog. Are you not seeing that?

If selected no, it makes sense to provide a URL so I agree.

BTW, I obviously got this email, did you change anything?

Thanks,

Greg


-- 
Greg Kurtzer
http://www.infiscale.com/
http://www.runlevelzero.net/
http://www.perceus.org/
http://www.caoslinux.org/


From benjamin at py-soft.co.uk  Mon Nov 10 09:30:37 2008
From: benjamin at py-soft.co.uk (Benjamin Donnachie)
Date: Mon, 10 Nov 2008 17:30:37 +0000
Subject: [cAos] caos-nsa-1.0rc2 again
In-Reply-To: <20081110162828.GS5928@loop.mdy.univie.ac.at>
References: <20081107085302.GH5928@loop.mdy.univie.ac.at>
	<571f1a060811071339k6d400910j47004c0f1e38f1cd@mail.gmail.com>
	<20081110162828.GS5928@loop.mdy.univie.ac.at>
Message-ID: <732076a80811100930x4a0ca130l53d9cdb1eae247b8@mail.gmail.com>

2008/11/10 Stefan Boresch <stefan at mdy.univie.ac.at>

> Strange ... Probably because I still use a local mail server for outbound
> mail, which is not the MX for incoming for (mdy.)univie.ac.at (i.e., I
> have
> no SPF(?) record ...)
>

I had exactly the same set up and ended up relaying all my mail through my
ISP as too many messages were getting rejected.

Ben
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.infiscale.org/pipermail/caos/attachments/20081110/8fd1edcf/attachment.html 

From gmkurtzer at gmail.com  Mon Nov 10 09:32:13 2008
From: gmkurtzer at gmail.com (Greg Kurtzer)
Date: Mon, 10 Nov 2008 09:32:13 -0800
Subject: [cAos] caos-nsa-1.0rc2 again
In-Reply-To: <20081110162828.GS5928@loop.mdy.univie.ac.at>
References: <20081107085302.GH5928@loop.mdy.univie.ac.at>
	<571f1a060811071339k6d400910j47004c0f1e38f1cd@mail.gmail.com>
	<20081110162828.GS5928@loop.mdy.univie.ac.at>
Message-ID: <571f1a060811100932m38be206dt1fa14e518fb7441c@mail.gmail.com>

What versions of the intel drivers, kernel, dri, and Xorg does ubuntu use?

Any ideas on what else might be different?

Thanks!

On Mon, Nov 10, 2008 at 8:28 AM, Stefan Boresch <stefan at mdy.univie.ac.at> wrote:
> On Fri, Nov 07, 2008 at 01:39:09PM -0800, Greg Kurtzer wrote:
>> I don't know why but your email's always go to my SPAM box. Gmail,... go figure!
>
> Strange ... Probably because I still use a local mail server for outbound
> mail, which is not the MX for incoming for (mdy.)univie.ac.at (i.e., I have
> no SPF(?) record ...)
>
>> > 2) X Server was initially unusable. On this hardware VESA *is* the only
>> > option, but the server guesses the screen resolution wrong (and on
>> > a low quality TFT this results in completely blurred fonts).
>> > X -configure goes wrong as well, had to force 1280x1024 by hand.
>> > This is not necessarily bad, but somehow Ubuntu 8.04 gets it right
>> > automatically -- no clue how they do it, since in 8.04 they basically
>> > also rely on the hardware recognition of the X Server ...
>>
>> What video card is it? do you happen to have a copy of their xorg.conf?
>>
>
> I attach the Ubuntu xorg.conf below. As you can see, it's a complete dummy,
> but it seems to work (slightly) better than caos' no xorg policy. It's
> also strange that Ubuntu got the screen resolution right, whereas in the
> caos case It guessed wrong, which led to the "fuzzy" fonts ... So I am
> not sure that it's going to help you!
>
> Best regards,
>
> Stefan
>
> -----------------------------
>
> xorg.conf, following a default Ubuntu 8.04 install. The graphics
> "card" is an on-board
>
> 06:02.0 VGA compatible controller: ATI Technologies Inc ES1000 (rev 02)
>
> # xorg.conf (X.Org X Window System server configuration file)
> #
> # This file was generated by dexconf, the Debian X Configuration tool, using
> # values from the debconf database.
> #
> # Edit this file with caution, and see the xorg.conf manual page.
> # (Type "man xorg.conf" at the shell prompt.)
> #
> # This file is automatically updated on xserver-xorg package upgrades *only*
> # if it has not been modified since the last upgrade of the xserver-xorg
> # package.
> #
> # If you have edited this file but would like it to be automatically updated
> # again, run the following command:
> #   sudo dpkg-reconfigure -phigh xserver-xorg
>
> Section "InputDevice"
>        Identifier      "Generic Keyboard"
>        Driver          "kbd"
>        Option          "XkbRules"      "xorg"
>        Option          "XkbModel"      "pc105"
>        Option          "XkbLayout"     "us"
> EndSection
>
> Section "InputDevice"
>        Identifier      "Configured Mouse"
>        Driver          "mouse"
>        Option          "CorePointer"
> EndSection
>
> Section "Device"
>        Identifier      "Configured Video Device"
> EndSection
>
> Section "Monitor"
>        Identifier      "Configured Monitor"
> EndSection
>
> Section "Screen"
>        Identifier      "Default Screen"
>        Monitor         "Configured Monitor"
>        Device          "Configured Video Device"
> EndSection
>
> Section "ServerLayout"
>        Identifier      "Default Layout"
>        Screen          "Default Screen"
> EndSection
>
> -----------------------------
>
> --
> Stefan Boresch
> Institute for Computational Biological Chemistry
> University of Vienna, Waehringerstr. 17       A-1090 Vienna, Austria
> Phone: -43-1-427752715                        Fax:   -43-1-427752790
> _______________________________________________
> cAos mailing list
> cAos at caoslinux.org
> http://lists.caosity.org/mailman/listinfo/caos
>



-- 
Greg Kurtzer
http://www.infiscale.com/
http://www.runlevelzero.net/
http://www.perceus.org/
http://www.caoslinux.org/


From stefan at mdy.univie.ac.at  Mon Nov 10 09:40:35 2008
From: stefan at mdy.univie.ac.at (Stefan Boresch)
Date: Mon, 10 Nov 2008 18:40:35 +0100
Subject: [cAos] Automated perceus configuration
In-Reply-To: <571f1a060811100930m7d533cbft67dffe2e93c5e7cb@mail.gmail.com>
References: <20081110164821.GT5928@loop.mdy.univie.ac.at>
	<571f1a060811100930m7d533cbft67dffe2e93c5e7cb@mail.gmail.com>
Message-ID: <20081110174035.GU5928@loop.mdy.univie.ac.at>

First, I didn't change anything with the email -- maybe gmail just
got smarter ...?

On Mon, Nov 10, 2008 at 09:30:13AM -0800, Greg Kurtzer wrote:
> On Mon, Nov 10, 2008 at 8:48 AM, Stefan Boresch <stefan at mdy.univie.ac.at> wrote:
> > I noted that the automated perceus configuration want's to set the
> > total nodes number to something ridiculously high (>> 1000000, at
> > least on x86-64).  Whatever the configuration then does, this takes
> > *really* long....
> 
> This is based on the netmask of the decided on Perceus device. We can
> of course give it a max. ;)

OK! But not paying attention and using a netmask of 255.0.0.0 gives
even a high end machine quite a lot to do ... (after about two screens
of dots, I gave up ...)

> > For some reasons, sidekick always wants to download vnfs-s from
> > mirror.caoslinux.org, which is (deliberately) slow. Could one not tell
> > it where it can find an existing *vnfs ...?
> 
> This should be a yes/no dialog. Are you not seeing that?

Yes, that I see
 
> If selected no, it makes sense to provide a URL so I agree.

but if I select no, the operation basically aborts ...

I just ran into a different, more serious problem though. As I told you, the
Intel G45 boards (with the onboard e1000e network card) are intended
as my cluster nodes. Caos-NSA boots fine on them, in particular the
network was automatically recognized. When I try to boot
perceus (albeit x86-64 rather than i386!), however, I end up with ...

ERROR: No supported network interface cards found!

I got so far as unpacking the initramfs.img that perceus provides (the
e1000e.ko seems to be there), and I located the line printing the
undesired message above in rc.perceus, but I am a bit at a loss how to
proceed ... I have to run now, but I'd appreciate suggestions what to
do / try do. Unfortunately, I don't see an easy way to force loading
of the e1000e module ...

Best regards,

Stefan

-- 
Stefan Boresch
Institute for Computational Biological Chemistry
University of Vienna, Waehringerstr. 17       A-1090 Vienna, Austria
Phone: -43-1-427752715                        Fax:   -43-1-427752790


From gmkurtzer at gmail.com  Mon Nov 10 10:34:54 2008
From: gmkurtzer at gmail.com (Greg Kurtzer)
Date: Mon, 10 Nov 2008 10:34:54 -0800
Subject: [cAos] Automated perceus configuration
In-Reply-To: <20081110174035.GU5928@loop.mdy.univie.ac.at>
References: <20081110164821.GT5928@loop.mdy.univie.ac.at>
	<571f1a060811100930m7d533cbft67dffe2e93c5e7cb@mail.gmail.com>
	<20081110174035.GU5928@loop.mdy.univie.ac.at>
Message-ID: <571f1a060811101034u1844deabw8fdfcfbde2e7ef51@mail.gmail.com>

Can you get me the vendor/model ID for that card (lspci;lspci -n) and
I will check on that.

We are on the other things. ;)

Thanks!

On Mon, Nov 10, 2008 at 9:40 AM, Stefan Boresch <stefan at mdy.univie.ac.at> wrote:
> First, I didn't change anything with the email -- maybe gmail just
> got smarter ...?
>
> On Mon, Nov 10, 2008 at 09:30:13AM -0800, Greg Kurtzer wrote:
>> On Mon, Nov 10, 2008 at 8:48 AM, Stefan Boresch <stefan at mdy.univie.ac.at> wrote:
>> > I noted that the automated perceus configuration want's to set the
>> > total nodes number to something ridiculously high (>> 1000000, at
>> > least on x86-64).  Whatever the configuration then does, this takes
>> > *really* long....
>>
>> This is based on the netmask of the decided on Perceus device. We can
>> of course give it a max. ;)
>
> OK! But not paying attention and using a netmask of 255.0.0.0 gives
> even a high end machine quite a lot to do ... (after about two screens
> of dots, I gave up ...)
>
>> > For some reasons, sidekick always wants to download vnfs-s from
>> > mirror.caoslinux.org, which is (deliberately) slow. Could one not tell
>> > it where it can find an existing *vnfs ...?
>>
>> This should be a yes/no dialog. Are you not seeing that?
>
> Yes, that I see
>
>> If selected no, it makes sense to provide a URL so I agree.
>
> but if I select no, the operation basically aborts ...
>
> I just ran into a different, more serious problem though. As I told you, the
> Intel G45 boards (with the onboard e1000e network card) are intended
> as my cluster nodes. Caos-NSA boots fine on them, in particular the
> network was automatically recognized. When I try to boot
> perceus (albeit x86-64 rather than i386!), however, I end up with ...
>
> ERROR: No supported network interface cards found!
>
> I got so far as unpacking the initramfs.img that perceus provides (the
> e1000e.ko seems to be there), and I located the line printing the
> undesired message above in rc.perceus, but I am a bit at a loss how to
> proceed ... I have to run now, but I'd appreciate suggestions what to
> do / try do. Unfortunately, I don't see an easy way to force loading
> of the e1000e module ...
>
> Best regards,
>
> Stefan
>
> --
> Stefan Boresch
> Institute for Computational Biological Chemistry
> University of Vienna, Waehringerstr. 17       A-1090 Vienna, Austria
> Phone: -43-1-427752715                        Fax:   -43-1-427752790
> _______________________________________________
> cAos mailing list
> cAos at caoslinux.org
> http://lists.caosity.org/mailman/listinfo/caos
>



-- 
Greg Kurtzer
http://www.infiscale.com/
http://www.runlevelzero.net/
http://www.perceus.org/
http://www.caoslinux.org/


From stefan at mdy.univie.ac.at  Tue Nov 11 00:27:44 2008
From: stefan at mdy.univie.ac.at (Stefan Boresch)
Date: Tue, 11 Nov 2008 09:27:44 +0100
Subject: [cAos] Automated perceus configuration
In-Reply-To: <571f1a060811101034u1844deabw8fdfcfbde2e7ef51@mail.gmail.com>
References: <20081110164821.GT5928@loop.mdy.univie.ac.at>
	<571f1a060811100930m7d533cbft67dffe2e93c5e7cb@mail.gmail.com>
	<20081110174035.GU5928@loop.mdy.univie.ac.at>
	<571f1a060811101034u1844deabw8fdfcfbde2e7ef51@mail.gmail.com>
Message-ID: <20081111082744.GV5928@loop.mdy.univie.ac.at>


On Mon, Nov 10, 2008 at 10:34:54AM -0800, Greg Kurtzer wrote:
> Can you get me the vendor/model ID for that card (lspci;lspci -n) and
> I will check on that.

>From the "sister" machine that has caos-nsa-1.0rc2/i386 installed,
plus the 2.6.27.4-1.nsa1 kernel that you provided:

[boresch at eta detect]$ lspci
-bash: lspci: command not found
[boresch at eta detect]$ /sbin/lspci
00:00.0 Host bridge: Intel Corporation: Unknown device 2e20 (rev 03)
00:02.0 VGA compatible controller: Intel Corporation: Unknown device 2e22 (rev 03)
00:02.1 Display controller: Intel Corporation: Unknown device 2e23 (rev 03)
00:03.0 Communication controller: Intel Corporation: Unknown device 2e24 (rev 03)
00:19.0 Ethernet controller: Intel Corporation: Unknown device 10cd
00:1a.0 USB Controller: Intel Corporation: Unknown device 3a37
00:1a.1 USB Controller: Intel Corporation: Unknown device 3a38
00:1a.2 USB Controller: Intel Corporation: Unknown device 3a39
00:1a.7 USB Controller: Intel Corporation: Unknown device 3a3c
00:1b.0 Audio device: Intel Corporation: Unknown device 3a3e
00:1d.0 USB Controller: Intel Corporation: Unknown device 3a34
00:1d.1 USB Controller: Intel Corporation: Unknown device 3a35
00:1d.2 USB Controller: Intel Corporation: Unknown device 3a36
00:1d.7 USB Controller: Intel Corporation: Unknown device 3a3a
00:1e.0 PCI bridge: Intel Corporation 82801 PCI Bridge (rev 90)
00:1f.0 ISA bridge: Intel Corporation: Unknown device 3a16
00:1f.2 IDE interface: Intel Corporation: Unknown device 3a20
00:1f.3 SMBus: Intel Corporation: Unknown device 3a30
00:1f.5 IDE interface: Intel Corporation: Unknown device 3a26
01:01.0 FireWire (IEEE 1394): Agere Systems FW323 (rev 70)

[boresch at eta detect]$ /sbin/lspci -n
00:00.0 Class 0600: 8086:2e20 (rev 03)
00:02.0 Class 0300: 8086:2e22 (rev 03)
00:02.1 Class 0380: 8086:2e23 (rev 03)
00:03.0 Class 0780: 8086:2e24 (rev 03)
00:19.0 Class 0200: 8086:10cd
00:1a.0 Class 0c03: 8086:3a37
00:1a.1 Class 0c03: 8086:3a38
00:1a.2 Class 0c03: 8086:3a39
00:1a.7 Class 0c03: 8086:3a3c
00:1b.0 Class 0403: 8086:3a3e
00:1d.0 Class 0c03: 8086:3a34
00:1d.1 Class 0c03: 8086:3a35
00:1d.2 Class 0c03: 8086:3a36
00:1d.7 Class 0c03: 8086:3a3a
00:1e.0 Class 0604: 8086:244e (rev 90)
00:1f.0 Class 0601: 8086:3a16
00:1f.2 Class 0101: 8086:3a20
00:1f.3 Class 0c05: 8086:3a30
00:1f.5 Class 0101: 8086:3a26
01:01.0 Class 0c00: 11c1:5811 (rev 70)

[boresch at eta detect]$ /sbin/detect
AUDIO intel-agp
AUDIO e1000e
AUDIO uhci-hcd
AUDIO uhci-hcd
AUDIO uhci-hcd
AUDIO ehci-hcd
AUDIO snd-hda-intel
AUDIO uhci-hcd
AUDIO uhci-hcd
AUDIO uhci-hcd
AUDIO ehci-hcd
AUDIO ata_piix
AUDIO i2c-i801
AUDIO ata_piix
AUDIO ohci1394
AUDIO firewire-ohci

Hope that helps ...

> We are on the other things. ;)
> 
> Thanks!

I thank you!

Stefan

-- 
Stefan Boresch
Institute for Computational Biological Chemistry
University of Vienna, Waehringerstr. 17       A-1090 Vienna, Austria
Phone: -43-1-427752715                        Fax:   -43-1-427752790


From stefan at mdy.univie.ac.at  Tue Nov 11 04:16:09 2008
From: stefan at mdy.univie.ac.at (Stefan Boresch)
Date: Tue, 11 Nov 2008 13:16:09 +0100
Subject: [cAos] caos-nsa-1.0rc2 again
In-Reply-To: <571f1a060811100932m38be206dt1fa14e518fb7441c@mail.gmail.com>
References: <20081107085302.GH5928@loop.mdy.univie.ac.at>
	<571f1a060811071339k6d400910j47004c0f1e38f1cd@mail.gmail.com>
	<20081110162828.GS5928@loop.mdy.univie.ac.at>
	<571f1a060811100932m38be206dt1fa14e518fb7441c@mail.gmail.com>
Message-ID: <20081111121609.GB5928@loop.mdy.univie.ac.at>

Greg,

On Mon, Nov 10, 2008 at 09:32:13AM -0800, Greg Kurtzer wrote:
> What versions of the intel drivers, kernel, dri, and Xorg does ubuntu use?
> 
> Any ideas on what else might be different?
> 

here is short list of the package versions (Ubuntu 8.04.1 LTS), as far as I can
figure them out. How many specific patches were made I can't say.

Kernel is 2.6.24 based
xorg-server-core 1.4.1
xorg-server 7.3
xserver-org-video-intel 2.2.1
xserver-org-video-vesa  1.3.0
libdrm2   2.3.0
libgl1-mesa-dri 7.0.3

anything else ??

Note that this latest report applies to a machine only supported by
the VESA driver. The new onboard Intel G45 in the "other" machine is a
different story (and this hardware is not supported at all by Ubuntu
8.04). Haven't tested 8.10 yet ...

I don't know what else might be different. Interestingly, I tried the
Ubuntu xorg.conf on the Intel machine (with caos installed); and X
wouldn't start, complaining about insufficient info ... I had
to provide a driver line (vesa or intel) to get it beyond this point.
Why/how this works on Ubuntu I don't know.

Best regards,

Stefan


> Thanks!
> 
> On Mon, Nov 10, 2008 at 8:28 AM, Stefan Boresch <stefan at mdy.univie.ac.at> wrote:
> > On Fri, Nov 07, 2008 at 01:39:09PM -0800, Greg Kurtzer wrote:
> >> I don't know why but your email's always go to my SPAM box. Gmail,... go figure!
> >
> > Strange ... Probably because I still use a local mail server for outbound
> > mail, which is not the MX for incoming for (mdy.)univie.ac.at (i.e., I have
> > no SPF(?) record ...)
> >
> >> > 2) X Server was initially unusable. On this hardware VESA *is* the only
> >> > option, but the server guesses the screen resolution wrong (and on
> >> > a low quality TFT this results in completely blurred fonts).
> >> > X -configure goes wrong as well, had to force 1280x1024 by hand.
> >> > This is not necessarily bad, but somehow Ubuntu 8.04 gets it right
> >> > automatically -- no clue how they do it, since in 8.04 they basically
> >> > also rely on the hardware recognition of the X Server ...
> >>
> >> What video card is it? do you happen to have a copy of their xorg.conf?
> >>
> >
> > I attach the Ubuntu xorg.conf below. As you can see, it's a complete dummy,
> > but it seems to work (slightly) better than caos' no xorg policy. It's
> > also strange that Ubuntu got the screen resolution right, whereas in the
> > caos case It guessed wrong, which led to the "fuzzy" fonts ... So I am
> > not sure that it's going to help you!
> >
> > Best regards,
> >
> > Stefan
> >
> > -----------------------------
> >
> > xorg.conf, following a default Ubuntu 8.04 install. The graphics
> > "card" is an on-board
> >
> > 06:02.0 VGA compatible controller: ATI Technologies Inc ES1000 (rev 02)
> >
> > # xorg.conf (X.Org X Window System server configuration file)
> > #
> > # This file was generated by dexconf, the Debian X Configuration tool, using
> > # values from the debconf database.
> > #
> > # Edit this file with caution, and see the xorg.conf manual page.
> > # (Type "man xorg.conf" at the shell prompt.)
> > #
> > # This file is automatically updated on xserver-xorg package upgrades *only*
> > # if it has not been modified since the last upgrade of the xserver-xorg
> > # package.
> > #
> > # If you have edited this file but would like it to be automatically updated
> > # again, run the following command:
> > #   sudo dpkg-reconfigure -phigh xserver-xorg
> >
> > Section "InputDevice"
> >        Identifier      "Generic Keyboard"
> >        Driver          "kbd"
> >        Option          "XkbRules"      "xorg"
> >        Option          "XkbModel"      "pc105"
> >        Option          "XkbLayout"     "us"
> > EndSection
> >
> > Section "InputDevice"
> >        Identifier      "Configured Mouse"
> >        Driver          "mouse"
> >        Option          "CorePointer"
> > EndSection
> >
> > Section "Device"
> >        Identifier      "Configured Video Device"
> > EndSection
> >
> > Section "Monitor"
> >        Identifier      "Configured Monitor"
> > EndSection
> >
> > Section "Screen"
> >        Identifier      "Default Screen"
> >        Monitor         "Configured Monitor"
> >        Device          "Configured Video Device"
> > EndSection
> >
> > Section "ServerLayout"
> >        Identifier      "Default Layout"
> >        Screen          "Default Screen"
> > EndSection
> >
> > -----------------------------
> >
> > --
> > Stefan Boresch
> > Institute for Computational Biological Chemistry
> > University of Vienna, Waehringerstr. 17       A-1090 Vienna, Austria
> > Phone: -43-1-427752715                        Fax:   -43-1-427752790
> > _______________________________________________
> > cAos mailing list
> > cAos at caoslinux.org
> > http://lists.caosity.org/mailman/listinfo/caos
> >
> 
> 
> 
> -- 
> Greg Kurtzer
> http://www.infiscale.com/
> http://www.runlevelzero.net/
> http://www.perceus.org/
> http://www.caoslinux.org/
> _______________________________________________
> cAos mailing list
> cAos at caoslinux.org
> http://lists.caosity.org/mailman/listinfo/caos
> 

-- 
Stefan Boresch
Institute for Computational Biological Chemistry
University of Vienna, Waehringerstr. 17       A-1090 Vienna, Austria
Phone: -43-1-427752715                        Fax:   -43-1-427752790


From stefan at mdy.univie.ac.at  Tue Nov 11 06:33:13 2008
From: stefan at mdy.univie.ac.at (Stefan Boresch)
Date: Tue, 11 Nov 2008 15:33:13 +0100
Subject: [cAos] perceus vs. caos-nsa kernel [was Re: Automated perceus
	configuration]
In-Reply-To: <571f1a060811101034u1844deabw8fdfcfbde2e7ef51@mail.gmail.com>
References: <20081110164821.GT5928@loop.mdy.univie.ac.at>
	<571f1a060811100930m7d533cbft67dffe2e93c5e7cb@mail.gmail.com>
	<20081110174035.GU5928@loop.mdy.univie.ac.at>
	<571f1a060811101034u1844deabw8fdfcfbde2e7ef51@mail.gmail.com>
Message-ID: <20081111143313.GC5928@loop.mdy.univie.ac.at>

Greg,

On Mon, Nov 10, 2008 at 10:34:54AM -0800, Greg Kurtzer wrote:
> Can you get me the vendor/model ID for that card (lspci;lspci -n) and
> I will check on that.
> 

tried to do some debugging via the perceus initrd ...

It seems to me that there are two issues: (1) /sbin/detect (in the
perceus initrd) doesn't detect the e1000e card. (2) When modifying
/etc/rc.perceus and /sbin/hw_load to load the e1000e kernel module not
matter what, the module gets loaded, but doesn't do anything. The
reason is simple: The driver that comes with 2.6.25-perceus seems to
be too old (v. 0.2.0) By contrast, the driver that comes with
caos-nsa-1.0rc2 is 0.3.3-k2.  I would have assumed that the two
kernels were built from the same source; apparently not so ...

Best regards,

Stefan

-- 
Stefan Boresch
Institute for Computational Biological Chemistry
University of Vienna, Waehringerstr. 17       A-1090 Vienna, Austria
Phone: -43-1-427752715                        Fax:   -43-1-427752790


From gmkurtzer at gmail.com  Tue Nov 11 09:43:47 2008
From: gmkurtzer at gmail.com (Greg Kurtzer)
Date: Tue, 11 Nov 2008 09:43:47 -0800
Subject: [cAos] perceus vs. caos-nsa kernel [was Re: Automated perceus
	configuration]
In-Reply-To: <20081111143313.GC5928@loop.mdy.univie.ac.at>
References: <20081110164821.GT5928@loop.mdy.univie.ac.at>
	<571f1a060811100930m7d533cbft67dffe2e93c5e7cb@mail.gmail.com>
	<20081110174035.GU5928@loop.mdy.univie.ac.at>
	<571f1a060811101034u1844deabw8fdfcfbde2e7ef51@mail.gmail.com>
	<20081111143313.GC5928@loop.mdy.univie.ac.at>
Message-ID: <571f1a060811110943w4461bf81w2e684f3ca05af18b@mail.gmail.com>

The perceus initramfs is using kernel 2.6.25.4 while NSA is using
2.6.26.5. The Perceus 1.5 tree has an updated kernel (2.6.28rc3) that
you can build from source or I can post a kernel/initramfs.img for you
to drop into your 1.4 Perceus tree (/var/lib/perceus/tftp/) which
should do the job.

Let me know. :)

Greg

On Tue, Nov 11, 2008 at 6:33 AM, Stefan Boresch <stefan at mdy.univie.ac.at> wrote:
> Greg,
>
> On Mon, Nov 10, 2008 at 10:34:54AM -0800, Greg Kurtzer wrote:
>> Can you get me the vendor/model ID for that card (lspci;lspci -n) and
>> I will check on that.
>>
>
> tried to do some debugging via the perceus initrd ...
>
> It seems to me that there are two issues: (1) /sbin/detect (in the
> perceus initrd) doesn't detect the e1000e card. (2) When modifying
> /etc/rc.perceus and /sbin/hw_load to load the e1000e kernel module not
> matter what, the module gets loaded, but doesn't do anything. The
> reason is simple: The driver that comes with 2.6.25-perceus seems to
> be too old (v. 0.2.0) By contrast, the driver that comes with
> caos-nsa-1.0rc2 is 0.3.3-k2.  I would have assumed that the two
> kernels were built from the same source; apparently not so ...
>
> Best regards,
>
> Stefan
>
> --
> Stefan Boresch
> Institute for Computational Biological Chemistry
> University of Vienna, Waehringerstr. 17       A-1090 Vienna, Austria
> Phone: -43-1-427752715                        Fax:   -43-1-427752790
> _______________________________________________
> cAos mailing list
> cAos at caoslinux.org
> http://lists.caosity.org/mailman/listinfo/caos
>



-- 
Greg Kurtzer
http://www.infiscale.com/
http://www.runlevelzero.net/
http://www.perceus.org/
http://www.caoslinux.org/


From gmkurtzer at gmail.com  Tue Nov 11 10:04:21 2008
From: gmkurtzer at gmail.com (Greg Kurtzer)
Date: Tue, 11 Nov 2008 10:04:21 -0800
Subject: [cAos] caos-nsa-1.0rc2 again
In-Reply-To: <20081111121609.GB5928@loop.mdy.univie.ac.at>
References: <20081107085302.GH5928@loop.mdy.univie.ac.at>
	<571f1a060811071339k6d400910j47004c0f1e38f1cd@mail.gmail.com>
	<20081110162828.GS5928@loop.mdy.univie.ac.at>
	<571f1a060811100932m38be206dt1fa14e518fb7441c@mail.gmail.com>
	<20081111121609.GB5928@loop.mdy.univie.ac.at>
Message-ID: <571f1a060811111004u337d3549y1df938409b93f15a@mail.gmail.com>

On Tue, Nov 11, 2008 at 4:16 AM, Stefan Boresch <stefan at mdy.univie.ac.at> wrote:

> here is short list of the package versions (Ubuntu 8.04.1 LTS), as far as I can
> figure them out. How many specific patches were made I can't say.
>
> Kernel is 2.6.24 based

We are newer here. ;)

> xorg-server-core 1.4.1

xorg-server 1.4

> xorg-server 7.3

What is contained in this package?

> xserver-org-video-intel 2.2.1

Hrmm, I wonder where they got this. X11R7.3 has 2.1.1, and X11R7.4 has
2.4.2. Do they document a link to the source tarball?

> xserver-org-video-vesa  1.3.0

Same

> libdrm2   2.3.0

Same.

> libgl1-mesa-dri 7.0.3

We have 7.0.2

>
> anything else ??

This is a good start.

>
> Note that this latest report applies to a machine only supported by
> the VESA driver. The new onboard Intel G45 in the "other" machine is a
> different story (and this hardware is not supported at all by Ubuntu
> 8.04). Haven't tested 8.10 yet ...

Ahhhhh, I thought ubuntu worked on the Intel G45. I wonder if Ubuntu
8.10 drives it.

Did Caos NSA work on this system, or have you not tested?

>
> I don't know what else might be different. Interestingly, I tried the
> Ubuntu xorg.conf on the Intel machine (with caos installed); and X
> wouldn't start, complaining about insufficient info ... I had
> to provide a driver line (vesa or intel) to get it beyond this point.
> Why/how this works on Ubuntu I don't know.

Yea, thats kinda weird. I wonder if they have some tweaks/patches to
the xorg-server which is entirely possible.

Thanks for the debugging assistance!

Greg


-- 
Greg Kurtzer
http://www.infiscale.com/
http://www.runlevelzero.net/
http://www.perceus.org/
http://www.caoslinux.org/


From stefan at mdy.univie.ac.at  Tue Nov 11 13:51:10 2008
From: stefan at mdy.univie.ac.at (Stefan Boresch)
Date: Tue, 11 Nov 2008 22:51:10 +0100
Subject: [cAos] perceus vs. caos-nsa kernel [was Re: Automated perceus
	configuration]
In-Reply-To: <571f1a060811110943w4461bf81w2e684f3ca05af18b@mail.gmail.com>
References: <20081110164821.GT5928@loop.mdy.univie.ac.at>
	<571f1a060811100930m7d533cbft67dffe2e93c5e7cb@mail.gmail.com>
	<20081110174035.GU5928@loop.mdy.univie.ac.at>
	<571f1a060811101034u1844deabw8fdfcfbde2e7ef51@mail.gmail.com>
	<20081111143313.GC5928@loop.mdy.univie.ac.at>
	<571f1a060811110943w4461bf81w2e684f3ca05af18b@mail.gmail.com>
Message-ID: <20081111215110.GA23176@loop.mdy.univie.ac.at>

On Tue, Nov 11, 2008 at 09:43:47AM -0800, Greg Kurtzer wrote:
> The perceus initramfs is using kernel 2.6.25.4 while NSA is using
> 2.6.26.5. The Perceus 1.5 tree has an updated kernel (2.6.28rc3) that
> you can build from source or I can post a kernel/initramfs.img for you
> to drop into your 1.4 Perceus tree (/var/lib/perceus/tftp/) which
> should do the job.
> 
> Let me know. :)
> 

This is a tough call. I sort of should get these machines into
production reasonably soon and with reasonable effort. This, in my 
opinion would speak for the initramfs.img from you. On the other hand,
given that these boards are somewhat rough even with a 2.6.27 kernel,
I wonder whether I should not switch (and follow development) to
perceus 1.5 and 2.6.28 (I presume I could also upgrade caos-nsa and the
vnfs itself to 2.6.28 ?) Which of course begs the question: how rough 
is perceus 1.5? (I didn't check, are there already src.rpms to build 
from?)

Best regards,

Stefan

-- 
Stefan Boresch
Institute for Computational Biological Chemistry
University of Vienna, Waehringerstr. 17       A-1090 Vienna, Austria
Phone: -43-1-427752715                        Fax:   -43-1-427752790


From stefan at mdy.univie.ac.at  Tue Nov 11 14:07:18 2008
From: stefan at mdy.univie.ac.at (Stefan Boresch)
Date: Tue, 11 Nov 2008 23:07:18 +0100
Subject: [cAos] caos-nsa-1.0rc2 again
In-Reply-To: <571f1a060811111004u337d3549y1df938409b93f15a@mail.gmail.com>
References: <20081107085302.GH5928@loop.mdy.univie.ac.at>
	<571f1a060811071339k6d400910j47004c0f1e38f1cd@mail.gmail.com>
	<20081110162828.GS5928@loop.mdy.univie.ac.at>
	<571f1a060811100932m38be206dt1fa14e518fb7441c@mail.gmail.com>
	<20081111121609.GB5928@loop.mdy.univie.ac.at>
	<571f1a060811111004u337d3549y1df938409b93f15a@mail.gmail.com>
Message-ID: <20081111220718.GB23176@loop.mdy.univie.ac.at>

Greg,

On Tue, Nov 11, 2008 at 10:04:21AM -0800, Greg Kurtzer wrote:
> On Tue, Nov 11, 2008 at 4:16 AM, Stefan Boresch <stefan at mdy.univie.ac.at> wrote:

[snip]

> > Note that this latest report applies to a machine only supported by
> > the VESA driver. The new onboard Intel G45 in the "other" machine is a
> > different story (and this hardware is not supported at all by Ubuntu
> > 8.04). Haven't tested 8.10 yet ...
> 
> Ahhhhh, I thought ubuntu worked on the Intel G45. I wonder if Ubuntu
> 8.10 drives it.
> 
> Did Caos NSA work on this system, or have you not tested?
> 

OK, I see I confused you. I have been working with caos/perceus on
three systems.

i) the machine where the skge driver wasn't detected/loaded 
automatically. This is an old machine (first generation amd64) with no
"interesting" hardware whatsoever. Once the skge issue was resolved, 
this machine worked flawlessly. (Obviously, all ubuntus work well on 
this machine)

ii) my prospective clusterhead, with the ATI onboard graphics, 
requiring/using VESA. I have several other machines of this type, 
running Ubuntu 8.04 perfectly. Caos worked fine as well, but X initially
was almost unusable and required some handtuning of xorg.conf.

iii) My prospective cluster nodes, with the Intel g45 based boards
(http://www.intel.com/products/desktop/motherboards/DG45ID/DG45ID-overview.htm)

Here I have two issues: the one is the difference in perceus vs Caos-NSA
1.0rc2 kernel, with the former having no network :-(

The other (just for the fun of it) is to get X running on this board 
(after all, they should also make good workstations ..) The 
xf86-xserver-intel-2.4.2 you sent me works fine for 2D, but a working 
combination for 3D acceleration has yet to be found. These boards don't 
work at all under Ubuntu 8.04, since the 2.6.24 kernel doesn't recognize 
the network (and those people that compiled their own network driver are 
stuck with the VESA driver for X). Now, Ubuntu 8.10 *should* work on 
these boards, at least as far as the network is concerned. If you think that 
it would be useful, then I'll try to throw 8.10 on one of these machines 
... Let me know ...

Stefan

PS: I'll check into the package contents you wanted tomorrow!

-- 
Stefan Boresch
Institute for Computational Biological Chemistry
University of Vienna, Waehringerstr. 17       A-1090 Vienna, Austria
Phone: -43-1-427752715                        Fax:   -43-1-427752790


From gmkurtzer at gmail.com  Tue Nov 11 15:14:24 2008
From: gmkurtzer at gmail.com (Greg Kurtzer)
Date: Tue, 11 Nov 2008 15:14:24 -0800
Subject: [cAos] perceus vs. caos-nsa kernel [was Re: Automated perceus
	configuration]
In-Reply-To: <20081111215110.GA23176@loop.mdy.univie.ac.at>
References: <20081110164821.GT5928@loop.mdy.univie.ac.at>
	<571f1a060811100930m7d533cbft67dffe2e93c5e7cb@mail.gmail.com>
	<20081110174035.GU5928@loop.mdy.univie.ac.at>
	<571f1a060811101034u1844deabw8fdfcfbde2e7ef51@mail.gmail.com>
	<20081111143313.GC5928@loop.mdy.univie.ac.at>
	<571f1a060811110943w4461bf81w2e684f3ca05af18b@mail.gmail.com>
	<20081111215110.GA23176@loop.mdy.univie.ac.at>
Message-ID: <571f1a060811111514w26ed8c6el8a936740831d89e0@mail.gmail.com>

I wouldn't recommend to use all of Perceus 1.5 at this time, just the
stage1 kernel and initramfs.img which works fine with 1.4.




On Tue, Nov 11, 2008 at 1:51 PM, Stefan Boresch <stefan at mdy.univie.ac.at> wrote:
> On Tue, Nov 11, 2008 at 09:43:47AM -0800, Greg Kurtzer wrote:
>> The perceus initramfs is using kernel 2.6.25.4 while NSA is using
>> 2.6.26.5. The Perceus 1.5 tree has an updated kernel (2.6.28rc3) that
>> you can build from source or I can post a kernel/initramfs.img for you
>> to drop into your 1.4 Perceus tree (/var/lib/perceus/tftp/) which
>> should do the job.
>>
>> Let me know. :)
>>
>
> This is a tough call. I sort of should get these machines into
> production reasonably soon and with reasonable effort. This, in my
> opinion would speak for the initramfs.img from you. On the other hand,
> given that these boards are somewhat rough even with a 2.6.27 kernel,
> I wonder whether I should not switch (and follow development) to
> perceus 1.5 and 2.6.28 (I presume I could also upgrade caos-nsa and the
> vnfs itself to 2.6.28 ?) Which of course begs the question: how rough
> is perceus 1.5? (I didn't check, are there already src.rpms to build
> from?)
>
> Best regards,
>
> Stefan
>
> --
> Stefan Boresch
> Institute for Computational Biological Chemistry
> University of Vienna, Waehringerstr. 17       A-1090 Vienna, Austria
> Phone: -43-1-427752715                        Fax:   -43-1-427752790
> _______________________________________________
> cAos mailing list
> cAos at caoslinux.org
> http://lists.caosity.org/mailman/listinfo/caos
>



-- 
Greg Kurtzer
http://www.infiscale.com/
http://www.runlevelzero.net/
http://www.perceus.org/
http://www.caoslinux.org/


From gmkurtzer at gmail.com  Tue Nov 11 15:22:42 2008
From: gmkurtzer at gmail.com (Greg Kurtzer)
Date: Tue, 11 Nov 2008 15:22:42 -0800
Subject: [cAos] caos-nsa-1.0rc2 again
In-Reply-To: <20081111220718.GB23176@loop.mdy.univie.ac.at>
References: <20081107085302.GH5928@loop.mdy.univie.ac.at>
	<571f1a060811071339k6d400910j47004c0f1e38f1cd@mail.gmail.com>
	<20081110162828.GS5928@loop.mdy.univie.ac.at>
	<571f1a060811100932m38be206dt1fa14e518fb7441c@mail.gmail.com>
	<20081111121609.GB5928@loop.mdy.univie.ac.at>
	<571f1a060811111004u337d3549y1df938409b93f15a@mail.gmail.com>
	<20081111220718.GB23176@loop.mdy.univie.ac.at>
Message-ID: <571f1a060811111522sd06b592w18e14ba28447ee59@mail.gmail.com>

Got it. That helped! :)

So:

#1) Should be fixed at this point.

#2) The Vesa support works fine with Caos NSA as well, as long as you
specify in the Xorg.conf the resolutions and perhaps modelines? Have
you tried the Sidekick X config module? (sidekick | Base System |
Xorg)

#3) Grab the Perceus stage 1 from
http://www.runlevelzero.net/perceus/perceus-1.4.9d-stage1.tar.gz.
Place the 2 files in there at /var/lib/perceus/tftp/, restart the
Perceus init script (to ensure no caching of these files), and try to
boot a node. This should work until Perceus 1.5 gets released stable.

As far as the Intel X issues, that is where we are waiting. :)



On Tue, Nov 11, 2008 at 2:07 PM, Stefan Boresch <stefan at mdy.univie.ac.at> wrote:
> Greg,
>
> On Tue, Nov 11, 2008 at 10:04:21AM -0800, Greg Kurtzer wrote:
>> On Tue, Nov 11, 2008 at 4:16 AM, Stefan Boresch <stefan at mdy.univie.ac.at> wrote:
>
> [snip]
>
>> > Note that this latest report applies to a machine only supported by
>> > the VESA driver. The new onboard Intel G45 in the "other" machine is a
>> > different story (and this hardware is not supported at all by Ubuntu
>> > 8.04). Haven't tested 8.10 yet ...
>>
>> Ahhhhh, I thought ubuntu worked on the Intel G45. I wonder if Ubuntu
>> 8.10 drives it.
>>
>> Did Caos NSA work on this system, or have you not tested?
>>
>
> OK, I see I confused you. I have been working with caos/perceus on
> three systems.
>
> i) the machine where the skge driver wasn't detected/loaded
> automatically. This is an old machine (first generation amd64) with no
> "interesting" hardware whatsoever. Once the skge issue was resolved,
> this machine worked flawlessly. (Obviously, all ubuntus work well on
> this machine)
>
> ii) my prospective clusterhead, with the ATI onboard graphics,
> requiring/using VESA. I have several other machines of this type,
> running Ubuntu 8.04 perfectly. Caos worked fine as well, but X initially
> was almost unusable and required some handtuning of xorg.conf.
>
> iii) My prospective cluster nodes, with the Intel g45 based boards
> (http://www.intel.com/products/desktop/motherboards/DG45ID/DG45ID-overview.htm)
>
> Here I have two issues: the one is the difference in perceus vs Caos-NSA
> 1.0rc2 kernel, with the former having no network :-(
>
> The other (just for the fun of it) is to get X running on this board
> (after all, they should also make good workstations ..) The
> xf86-xserver-intel-2.4.2 you sent me works fine for 2D, but a working
> combination for 3D acceleration has yet to be found. These boards don't
> work at all under Ubuntu 8.04, since the 2.6.24 kernel doesn't recognize
> the network (and those people that compiled their own network driver are
> stuck with the VESA driver for X). Now, Ubuntu 8.10 *should* work on
> these boards, at least as far as the network is concerned. If you think that
> it would be useful, then I'll try to throw 8.10 on one of these machines
> ... Let me know ...
>
> Stefan
>
> PS: I'll check into the package contents you wanted tomorrow!
>
> --
> Stefan Boresch
> Institute for Computational Biological Chemistry
> University of Vienna, Waehringerstr. 17       A-1090 Vienna, Austria
> Phone: -43-1-427752715                        Fax:   -43-1-427752790
> _______________________________________________
> cAos mailing list
> cAos at caoslinux.org
> http://lists.caosity.org/mailman/listinfo/caos
>



-- 
Greg Kurtzer
http://www.infiscale.com/
http://www.runlevelzero.net/
http://www.perceus.org/
http://www.caoslinux.org/


From stefan at mdy.univie.ac.at  Wed Nov 12 04:02:34 2008
From: stefan at mdy.univie.ac.at (Stefan Boresch)
Date: Wed, 12 Nov 2008 13:02:34 +0100
Subject: [cAos] perceus vs. caos-nsa kernel [was Re: Automated perceus
	configuration]
In-Reply-To: <571f1a060811111514w26ed8c6el8a936740831d89e0@mail.gmail.com>
References: <20081110164821.GT5928@loop.mdy.univie.ac.at>
	<571f1a060811100930m7d533cbft67dffe2e93c5e7cb@mail.gmail.com>
	<20081110174035.GU5928@loop.mdy.univie.ac.at>
	<571f1a060811101034u1844deabw8fdfcfbde2e7ef51@mail.gmail.com>
	<20081111143313.GC5928@loop.mdy.univie.ac.at>
	<571f1a060811110943w4461bf81w2e684f3ca05af18b@mail.gmail.com>
	<20081111215110.GA23176@loop.mdy.univie.ac.at>
	<571f1a060811111514w26ed8c6el8a936740831d89e0@mail.gmail.com>
Message-ID: <20081112120234.GE5928@loop.mdy.univie.ac.at>

Greg,

On Tue, Nov 11, 2008 at 03:14:24PM -0800, Greg Kurtzer wrote:
> I wouldn't recommend to use all of Perceus 1.5 at this time, just the
> stage1 kernel and initramfs.img which works fine with 1.4.
> 

thanks for the stage1 kernel/initramfs; this works like a
charmm and I booted my first node successfully. Thus, I can go
back to finetuning things.

Two minor observations/questions:

(1) Could it be that the openmpi package does not support Fortran?
At least, I can't find the include files / modules ... [not a big
deal, since I will play with different MPIs anyways (or, rather, my
users want them) ..., so it's just one more recompile ...]

(2) As I mentioned, I had to initially kill the vnfs/perceus
initialization (via sidekick) because it took too long ... (the system
had already made entries in /etc/hosts and /private for > 5000
nodes). I then repeated the procedure for 20 nodes (which will be
enough at the moment ...) I noted, however, that all the old entries
were still there. Is this deliberate?

I probably should move to the perceus list, since most of this
is now perceus and not caos-nsa related ...

Thanks,

Stefan

-- 
Stefan Boresch
Institute for Computational Biological Chemistry
University of Vienna, Waehringerstr. 17       A-1090 Vienna, Austria
Phone: -43-1-427752715                        Fax:   -43-1-427752790


From gmkurtzer at gmail.com  Wed Nov 12 06:32:44 2008
From: gmkurtzer at gmail.com (Greg Kurtzer)
Date: Wed, 12 Nov 2008 06:32:44 -0800
Subject: [cAos] perceus vs. caos-nsa kernel [was Re: Automated perceus
	configuration]
In-Reply-To: <20081112120234.GE5928@loop.mdy.univie.ac.at>
References: <20081110164821.GT5928@loop.mdy.univie.ac.at>
	<571f1a060811100930m7d533cbft67dffe2e93c5e7cb@mail.gmail.com>
	<20081110174035.GU5928@loop.mdy.univie.ac.at>
	<571f1a060811101034u1844deabw8fdfcfbde2e7ef51@mail.gmail.com>
	<20081111143313.GC5928@loop.mdy.univie.ac.at>
	<571f1a060811110943w4461bf81w2e684f3ca05af18b@mail.gmail.com>
	<20081111215110.GA23176@loop.mdy.univie.ac.at>
	<571f1a060811111514w26ed8c6el8a936740831d89e0@mail.gmail.com>
	<20081112120234.GE5928@loop.mdy.univie.ac.at>
Message-ID: <571f1a060811120632r630dbfdand1d3ca628c8fc270@mail.gmail.com>

On Wed, Nov 12, 2008 at 4:02 AM, Stefan Boresch <stefan at mdy.univie.ac.at> wrote:
> Greg,
>
> On Tue, Nov 11, 2008 at 03:14:24PM -0800, Greg Kurtzer wrote:
>> I wouldn't recommend to use all of Perceus 1.5 at this time, just the
>> stage1 kernel and initramfs.img which works fine with 1.4.
>>
>
> thanks for the stage1 kernel/initramfs; this works like a
> charmm and I booted my first node successfully. Thus, I can go
> back to finetuning things.
>
> Two minor observations/questions:
>
> (1) Could it be that the openmpi package does not support Fortran?
> At least, I can't find the include files / modules ... [not a big
> deal, since I will play with different MPIs anyways (or, rather, my
> users want them) ..., so it's just one more recompile ...]

The Caos NSA mpif* commands do support fortran:

# mpif90 -show
x86_64-caos-linux-gnu-gfortran -pthread -I/usr/lib64 -L/usr/lib64
-lmpi_f90 -lmpi_f77 -lmpi -lopen-rte -lopen-pal -ldl
-Wl,--export-dynamic -lnsl -lutil -lm -ldl

Are you getting an error?

>
> (2) As I mentioned, I had to initially kill the vnfs/perceus
> initialization (via sidekick) because it took too long ... (the system
> had already made entries in /etc/hosts and /private for > 5000
> nodes). I then repeated the procedure for 20 nodes (which will be
> enough at the moment ...) I noted, however, that all the old entries
> were still there. Is this deliberate?

Sidekick doesn't remove any entries from /etc/hosts, but that is the
only place it should be pre-populating them. Go ahead and remove
extras by hand. :)

>
> I probably should move to the perceus list, since most of this
> is now perceus and not caos-nsa related ...

Sounds good. :)

Greg

-- 
Greg Kurtzer
http://www.infiscale.com/
http://www.runlevelzero.net/
http://www.perceus.org/
http://www.caoslinux.org/


From stefan at mdy.univie.ac.at  Wed Nov 12 06:36:52 2008
From: stefan at mdy.univie.ac.at (Stefan Boresch)
Date: Wed, 12 Nov 2008 15:36:52 +0100
Subject: [cAos] perceus vs. caos-nsa kernel [was Re: Automated perceus
	configuration]
In-Reply-To: <571f1a060811120632r630dbfdand1d3ca628c8fc270@mail.gmail.com>
References: <20081110164821.GT5928@loop.mdy.univie.ac.at>
	<571f1a060811100930m7d533cbft67dffe2e93c5e7cb@mail.gmail.com>
	<20081110174035.GU5928@loop.mdy.univie.ac.at>
	<571f1a060811101034u1844deabw8fdfcfbde2e7ef51@mail.gmail.com>
	<20081111143313.GC5928@loop.mdy.univie.ac.at>
	<571f1a060811110943w4461bf81w2e684f3ca05af18b@mail.gmail.com>
	<20081111215110.GA23176@loop.mdy.univie.ac.at>
	<571f1a060811111514w26ed8c6el8a936740831d89e0@mail.gmail.com>
	<20081112120234.GE5928@loop.mdy.univie.ac.at>
	<571f1a060811120632r630dbfdand1d3ca628c8fc270@mail.gmail.com>
Message-ID: <20081112143652.GF5928@loop.mdy.univie.ac.at>

On Wed, Nov 12, 2008 at 06:32:44AM -0800, Greg Kurtzer wrote:
> On Wed, Nov 12, 2008 at 4:02 AM, Stefan Boresch <stefan at mdy.univie.ac.at> wrote:
> > (1) Could it be that the openmpi package does not support Fortran?
> > At least, I can't find the include files / modules ... [not a big
> > deal, since I will play with different MPIs anyways (or, rather, my
> > users want them) ..., so it's just one more recompile ...]
> 
> The Caos NSA mpif* commands do support fortran:
> 
> # mpif90 -show
> x86_64-caos-linux-gnu-gfortran -pthread -I/usr/lib64 -L/usr/lib64
> -lmpi_f90 -lmpi_f77 -lmpi -lopen-rte -lopen-pal -ldl
> -Wl,--export-dynamic -lnsl -lutil -lm -ldl
> 

OK, I was too quick. Since I saw that there is /usr/include/openmpi,
I looked for mpif.h there, and found only a hierarchy of subdirs 
suggesting cxx only. But mpif.h is simply in /usr/include ...

Sorry,

Stefan

-- 
Stefan Boresch
Institute for Computational Biological Chemistry
University of Vienna, Waehringerstr. 17       A-1090 Vienna, Austria
Phone: -43-1-427752715                        Fax:   -43-1-427752790


From stefan at mdy.univie.ac.at  Wed Nov 12 08:45:42 2008
From: stefan at mdy.univie.ac.at (Stefan Boresch)
Date: Wed, 12 Nov 2008 17:45:42 +0100
Subject: [cAos] Kernels > 2.6.27 ??
Message-ID: <20081112164542.GI5928@loop.mdy.univie.ac.at>

Sorry if I make a pest out of myself ...

If I am not mistaken, the nice perceus provisioning kernel you made
available to me is 2.6.28 based. It seems that in this kernel a known
incompatibility between the Intel DG45ID board and kernels up to
2.6.27, resulting in a 10-15 second delay during boot (e.g.,
https://bugs.launchpad.net/ubuntu/+source/linux/+bug/275351), is
fixed. At least the perceus provisioning kernel doesn't spit out any
warnings and is much quicker to boot than the older 2.6.25
provisioning kernel that later wouldn't recognize the e1000e ...)  So,
I am wondering whether by any chance you have alreay a source rpm for
generic usage in Caos ... If not I'll try the BIOS trick ...

Stefan

PS: Do you want me to check out Ubuntu 8.10 on these machines to see
how far I get?

-- 
Stefan Boresch
Institute for Computational Biological Chemistry
University of Vienna, Waehringerstr. 17       A-1090 Vienna, Austria
Phone: -43-1-427752715                        Fax:   -43-1-427752790


From gmkurtzer at gmail.com  Thu Nov 13 17:49:13 2008
From: gmkurtzer at gmail.com (Greg Kurtzer)
Date: Thu, 13 Nov 2008 17:49:13 -0800
Subject: [cAos] Kernels > 2.6.27 ??
In-Reply-To: <20081112164542.GI5928@loop.mdy.univie.ac.at>
References: <20081112164542.GI5928@loop.mdy.univie.ac.at>
Message-ID: <571f1a060811131749t31d32307y4410462e8c6030d4@mail.gmail.com>

On Wed, Nov 12, 2008 at 8:45 AM, Stefan Boresch <stefan at mdy.univie.ac.at> wrote:
> Sorry if I make a pest out of myself ...
>
> If I am not mistaken, the nice perceus provisioning kernel you made
> available to me is 2.6.28 based. It seems that in this kernel a known
> incompatibility between the Intel DG45ID board and kernels up to
> 2.6.27, resulting in a 10-15 second delay during boot (e.g.,
> https://bugs.launchpad.net/ubuntu/+source/linux/+bug/275351), is
> fixed. At least the perceus provisioning kernel doesn't spit out any
> warnings and is much quicker to boot than the older 2.6.25
> provisioning kernel that later wouldn't recognize the e1000e ...)  So,
> I am wondering whether by any chance you have alreay a source rpm for
> generic usage in Caos ... If not I'll try the BIOS trick ...
>
> Stefan
>
> PS: Do you want me to check out Ubuntu 8.10 on these machines to see
> how far I get?

Sorry for the late response. I was hoping 2.6.28 was going to release
by now and I was going to build a package for testing. Oh well.. ;)

Here is the only newer src.rpm I have available for testing:

   http://www.runlevelzero.net/caos/caos-kernel-2.6.27.4-1.nsa1.src.rpm

Feel free to give that a whirl and let us know.

Thanks!
-- 
Greg Kurtzer
http://www.infiscale.com/
http://www.runlevelzero.net/
http://www.perceus.org/
http://www.caoslinux.org/


From stefan at mdy.univie.ac.at  Fri Nov 14 02:29:47 2008
From: stefan at mdy.univie.ac.at (Stefan Boresch)
Date: Fri, 14 Nov 2008 11:29:47 +0100
Subject: [cAos] Kernels > 2.6.27 ??
In-Reply-To: <571f1a060811131749t31d32307y4410462e8c6030d4@mail.gmail.com>
References: <20081112164542.GI5928@loop.mdy.univie.ac.at>
	<571f1a060811131749t31d32307y4410462e8c6030d4@mail.gmail.com>
Message-ID: <20081114102947.GB9259@loop.mdy.univie.ac.at>

Greg,

On Thu, Nov 13, 2008 at 05:49:13PM -0800, Greg Kurtzer wrote:
> On Wed, Nov 12, 2008 at 8:45 AM, Stefan Boresch <stefan at mdy.univie.ac.at> wrote:
> 
> Sorry for the late response. I was hoping 2.6.28 was going to release
> by now and I was going to build a package for testing. Oh well.. ;)
> 
> Here is the only newer src.rpm I have available for testing:
> 
>    http://www.runlevelzero.net/caos/caos-kernel-2.6.27.4-1.nsa1.src.rpm
> 
> Feel free to give that a whirl and let us know.
> 

thanks a lot, and please don't apologize -- the support I am receiving
here is fantastic!  The 2.6.27 kernel still has the problem (which
isn't really one, it just delays the boot for 10secs unless one turns
off a BIOS setting). In the meantime, I am not even sure that 2.6.28
would be the cure, because I have the feeling that the reduced perceus
provisioning kernel doesn't even touch the problematic subsystem ...

On a more general note, I have my test cluster almost ready, and
it was/is really a pleasure to work with perceus/caos!

Best regards,

Stefan

-- 
Stefan Boresch
Institute for Computational Biological Chemistry
University of Vienna, Waehringerstr. 17       A-1090 Vienna, Austria
Phone: -43-1-427752715                        Fax:   -43-1-427752790


From gmkurtzer at gmail.com  Tue Nov 18 23:30:56 2008
From: gmkurtzer at gmail.com (Greg Kurtzer)
Date: Tue, 18 Nov 2008 23:30:56 -0800
Subject: [cAos] Kernels > 2.6.27 ??
In-Reply-To: <20081114102947.GB9259@loop.mdy.univie.ac.at>
References: <20081112164542.GI5928@loop.mdy.univie.ac.at>
	<571f1a060811131749t31d32307y4410462e8c6030d4@mail.gmail.com>
	<20081114102947.GB9259@loop.mdy.univie.ac.at>
Message-ID: <571f1a060811182330j4bb31435s523d1d2f50bb3ee@mail.gmail.com>

On Fri, Nov 14, 2008 at 2:29 AM, Stefan Boresch <stefan at mdy.univie.ac.at> wrote:
> Greg,
>
> On Thu, Nov 13, 2008 at 05:49:13PM -0800, Greg Kurtzer wrote:
>> On Wed, Nov 12, 2008 at 8:45 AM, Stefan Boresch <stefan at mdy.univie.ac.at> wrote:
>>
>> Sorry for the late response. I was hoping 2.6.28 was going to release
>> by now and I was going to build a package for testing. Oh well.. ;)
>>
>> Here is the only newer src.rpm I have available for testing:
>>
>>    http://www.runlevelzero.net/caos/caos-kernel-2.6.27.4-1.nsa1.src.rpm
>>
>> Feel free to give that a whirl and let us know.
>>
>
> thanks a lot, and please don't apologize -- the support I am receiving
> here is fantastic!  The 2.6.27 kernel still has the problem (which
> isn't really one, it just delays the boot for 10secs unless one turns
> off a BIOS setting). In the meantime, I am not even sure that 2.6.28
> would be the cure, because I have the feeling that the reduced perceus
> provisioning kernel doesn't even touch the problematic subsystem ...

Still waiting on 2.6.28..... ;)

Just incase I don't catch the release announcement right away, feel
free to ping me whenever you see it hit!

>
> On a more general note, I have my test cluster almost ready, and
> it was/is really a pleasure to work with perceus/caos!

That is really great to hear! We should put up a testimonial page on
our sites! hehe

Greg

-- 
Greg Kurtzer
http://www.infiscale.com/
http://www.runlevelzero.net/
http://www.perceus.org/
http://www.caoslinux.org/


From stefan at mdy.univie.ac.at  Wed Nov 19 01:28:49 2008
From: stefan at mdy.univie.ac.at (Stefan Boresch)
Date: Wed, 19 Nov 2008 10:28:49 +0100
Subject: [cAos] mount -n -t tmpfs none /dev/shm
Message-ID: <20081119092849.GL9259@loop.mdy.univie.ac.at>

I am playing around with the Intel MPI library, which insists on the
presence of a configured /dev/shm device. It seems to me that caos
sets it up, albeit without entry in /etc/mtab (mount -n), and this may
be the reason why the Intel installer barfs ...

Is there a reason for the -n option in /etc/rc.sysinit when
mounting /dev/shm? Or am I looking completely in the wrong place ...

Stefan


-- 
Stefan Boresch
Institute for Computational Biological Chemistry
University of Vienna, Waehringerstr. 17       A-1090 Vienna, Austria
Phone: -43-1-427752715                        Fax:   -43-1-427752790


From gmkurtzer at gmail.com  Wed Nov 19 06:51:11 2008
From: gmkurtzer at gmail.com (Greg Kurtzer)
Date: Wed, 19 Nov 2008 06:51:11 -0800
Subject: [cAos] mount -n -t tmpfs none /dev/shm
In-Reply-To: <20081119092849.GL9259@loop.mdy.univie.ac.at>
References: <20081119092849.GL9259@loop.mdy.univie.ac.at>
Message-ID: <571f1a060811190651i45eedaa8lde9611425c160b2e@mail.gmail.com>

Nope, this seems reasonable. Does it work if you remove the -n mount argument?

Thanks!

On Wed, Nov 19, 2008 at 1:28 AM, Stefan Boresch <stefan at mdy.univie.ac.at> wrote:
> I am playing around with the Intel MPI library, which insists on the
> presence of a configured /dev/shm device. It seems to me that caos
> sets it up, albeit without entry in /etc/mtab (mount -n), and this may
> be the reason why the Intel installer barfs ...
>
> Is there a reason for the -n option in /etc/rc.sysinit when
> mounting /dev/shm? Or am I looking completely in the wrong place ...
>
> Stefan
>
>
> --
> Stefan Boresch
> Institute for Computational Biological Chemistry
> University of Vienna, Waehringerstr. 17       A-1090 Vienna, Austria
> Phone: -43-1-427752715                        Fax:   -43-1-427752790
> _______________________________________________
> cAos mailing list
> cAos at caoslinux.org
> http://lists.caosity.org/mailman/listinfo/caos
>



-- 
Greg Kurtzer
http://www.infiscale.com/
http://www.runlevelzero.net/
http://www.perceus.org/
http://www.caoslinux.org/


From stefan at mdy.univie.ac.at  Thu Nov 20 00:30:35 2008
From: stefan at mdy.univie.ac.at (stefan at mdy.univie.ac.at)
Date: Thu, 20 Nov 2008 09:30:35 +0100 (CET)
Subject: [cAos] mount -n -t tmpfs none /dev/shm
In-Reply-To: <571f1a060811190651i45eedaa8lde9611425c160b2e@mail.gmail.com>
References: <20081119092849.GL9259@loop.mdy.univie.ac.at>
	<571f1a060811190651i45eedaa8lde9611425c160b2e@mail.gmail.com>
Message-ID: <36073.62.47.158.156.1227169835.squirrel@www.mdy.univie.ac.at>



> Nope, this seems reasonable. Does it work if you remove the -n mount
> argument?

After removing the -n mount I could install Intel MPI without further
warnings (and /dev/shm shows up in df).  Whether it works (i.e., whether I
can indeed run MPI jobs ;-), I can't tell yet since at the moment I am
somewhat swamped with work, so finetuning this little cluster only gets
10-15 minutes a day :-(

I didn't notice any adverse effects on the system, but then I didn't do
anything serious with it.  BTW, I don't think you should change anything,
but this might qualify as a small FAQ ...

Best regards,

Stefan

>
> Thanks!
>
> On Wed, Nov 19, 2008 at 1:28 AM, Stefan Boresch <stefan at mdy.univie.ac.at>
> wrote:
>> I am playing around with the Intel MPI library, which insists on the
>> presence of a configured /dev/shm device. It seems to me that caos
>> sets it up, albeit without entry in /etc/mtab (mount -n), and this may
>> be the reason why the Intel installer barfs ...
>>
>> Is there a reason for the -n option in /etc/rc.sysinit when
>> mounting /dev/shm? Or am I looking completely in the wrong place ...
>>
>> Stefan
>>
>>
>> --
>> Stefan Boresch
>> Institute for Computational Biological Chemistry
>> University of Vienna, Waehringerstr. 17       A-1090 Vienna, Austria
>> Phone: -43-1-427752715                        Fax:   -43-1-427752790
>> _______________________________________________
>> cAos mailing list
>> cAos at caoslinux.org
>> http://lists.caosity.org/mailman/listinfo/caos
>>
>
>
>
> --
> Greg Kurtzer
> http://www.infiscale.com/
> http://www.runlevelzero.net/
> http://www.perceus.org/
> http://www.caoslinux.org/
> _______________________________________________
> cAos mailing list
> cAos at caoslinux.org
> http://lists.caosity.org/mailman/listinfo/caos
>
>




From gmkurtzer at gmail.com  Thu Nov 20 07:08:57 2008
From: gmkurtzer at gmail.com (Greg Kurtzer)
Date: Thu, 20 Nov 2008 07:08:57 -0800
Subject: [cAos] mount -n -t tmpfs none /dev/shm
In-Reply-To: <36073.62.47.158.156.1227169835.squirrel@www.mdy.univie.ac.at>
References: <20081119092849.GL9259@loop.mdy.univie.ac.at>
	<571f1a060811190651i45eedaa8lde9611425c160b2e@mail.gmail.com>
	<36073.62.47.158.156.1227169835.squirrel@www.mdy.univie.ac.at>
Message-ID: <571f1a060811200708t34f4cf6fq9974cd1c234c3522@mail.gmail.com>

Caos and Perceus are part of an Intel Cluster Ready certification and
various recipes. It should work just fine, but keep us informed just
to be sure!

Good luck!

Greg

On Thu, Nov 20, 2008 at 12:30 AM,  <stefan at mdy.univie.ac.at> wrote:
>
>
>> Nope, this seems reasonable. Does it work if you remove the -n mount
>> argument?
>
> After removing the -n mount I could install Intel MPI without further
> warnings (and /dev/shm shows up in df).  Whether it works (i.e., whether I
> can indeed run MPI jobs ;-), I can't tell yet since at the moment I am
> somewhat swamped with work, so finetuning this little cluster only gets
> 10-15 minutes a day :-(
>
> I didn't notice any adverse effects on the system, but then I didn't do
> anything serious with it.  BTW, I don't think you should change anything,
> but this might qualify as a small FAQ ...
>
> Best regards,
>
> Stefan
>
>>
>> Thanks!
>>
>> On Wed, Nov 19, 2008 at 1:28 AM, Stefan Boresch <stefan at mdy.univie.ac.at>
>> wrote:
>>> I am playing around with the Intel MPI library, which insists on the
>>> presence of a configured /dev/shm device. It seems to me that caos
>>> sets it up, albeit without entry in /etc/mtab (mount -n), and this may
>>> be the reason why the Intel installer barfs ...
>>>
>>> Is there a reason for the -n option in /etc/rc.sysinit when
>>> mounting /dev/shm? Or am I looking completely in the wrong place ...
>>>
>>> Stefan
>>>
>>>
>>> --
>>> Stefan Boresch
>>> Institute for Computational Biological Chemistry
>>> University of Vienna, Waehringerstr. 17       A-1090 Vienna, Austria
>>> Phone: -43-1-427752715                        Fax:   -43-1-427752790
>>> _______________________________________________
>>> cAos mailing list
>>> cAos at caoslinux.org
>>> http://lists.caosity.org/mailman/listinfo/caos
>>>
>>
>>
>>
>> --
>> Greg Kurtzer
>> http://www.infiscale.com/
>> http://www.runlevelzero.net/
>> http://www.perceus.org/
>> http://www.caoslinux.org/
>> _______________________________________________
>> cAos mailing list
>> cAos at caoslinux.org
>> http://lists.caosity.org/mailman/listinfo/caos
>>
>>
>
>
> _______________________________________________
> cAos mailing list
> cAos at caoslinux.org
> http://lists.caosity.org/mailman/listinfo/caos
>



-- 
Greg Kurtzer
http://www.infiscale.com/
http://www.runlevelzero.net/
http://www.perceus.org/
http://www.caoslinux.org/


From gmkurtzer at gmail.com  Tue Nov 25 09:50:44 2008
From: gmkurtzer at gmail.com (Greg Kurtzer)
Date: Tue, 25 Nov 2008 09:50:44 -0800
Subject: [cAos] Caos NSA 1.0 Released!
Message-ID: <571f1a060811250950m4b3521dk1022d902d4c3809a@mail.gmail.com>

Announcing the formal release of Caos Linux NSA 1 for x86 and x86_64!

The Caos team of developers and contributors from Infiscale are proud
to announce the public release of Caos Linux NSA version 1. Caos Linux
is a community-managed and openly-maintained distribution of Linux
focusing on areas where Linux naturally leads and excels: high
performance computation (HPC), servers (especially LAMP and general
Web), and custom appliances (such as file servers and firewalls).

This release identifies the stabilization and validation of the core
operating system, fully tested on some of the world's fastest public
and private systems and architectures.

While Caos Linux NSA signifies a new focus in Linux distributions, it
also represents a return to what got people involved with Linux in the
first place. It is built using known standards and good practices of
the open source community with the end users' input and needs taken
into consideration. Features are never held back or forced in solely
for monetary reasons; technical decisions are always made by technical
people for technical reasons. Caos is clean, simple, solid, and fast
while still maintaining the level of stability and resilience demanded
by enterprise and high-performance computing.

Many distributions no longer focus on the core competencies that
always made Linux so proficient, choosing instead to emulate Windows
on the desktop and commercial Unix vendors on the server at the same
time. In stark contrast to these diametrically opposed goals, the Caos
core operating system has been tuned using very large and
tightly-coupled HPC cluster systems. Careful tweaking and
optimizations have resulted in a very streamlined, lightweight
operating system still suitable for a large spectrum of uses, from
simple desktops to petaflop supercomputers. Future development also
includes a release for embedded systems.



Background:

The Caos project was founded in 2003 with the goal of creating a
global open development structure for a completely
community-maintained, RPM-based distribution of Linux. Shortly after
the project's inception, the Caos developers also began work on
CentOS, a rebuild of the source packages comprising Red Hat Enterprise
Linux, to provide a development platform and transitional mechanism
for Caos, their Community-Assembled Operating System. The team has
attracted some very talented developers over the years and are now
actively growing the project again due to a new resurgence in use,
industry commitment, and great community feedback and support.



Development Team:

At the heart of Caos Linux lies a group of developers dedicated to
high performance. These same developers have already produced some
very well-respected projects: general-purpose Caos Linux 1 and 2,
CentOS, Warewulf, Perceus, and RazorWire. Along with Caos Linux NSA,
the soon-to-be-released Abstractual (cloud management system) and
Equilibria (load balancer) projects now join this wide array of
quality open source solutions.

The Caos development team is currently seeking software developers,
package maintainers, technical writers, evangelists (including trade
show booth staff), and other supporters. Internships are also
available at Infiscale and other organizations currently assisting
with development.



Cooperation with industry:

Currently several large commercial entities are standardizing on and
supporting Caos Linux, and a number of very large high-performance
cluster systems at Top500 are running or evaluating it as well. Caos
Linux NSA is the only freely-downloadable Linux distribution that is
certified as Intel Cluster Ready (ICR), guaranteeing out-of-the-box
compatibility with independent software applications such as Abaqus,
Anima, Fluent, Petrel, and numerous others. Caos is even available
embedded on all Intel Motherboards supporting Rapid Boot.

The Caos developers are committed to working closely with vendors
(both IHV and ISV) and have a history of being on the leading edge of
new/custom features and latest hardware support. For Instance the new
Intel Nahalem CPU's are fully compatible with Caos and offer embedded
capabilities including embedded Perceus on the new Intel motherboards,
so the most high end hardware is supported even before it comes out!
The ability to use the latest and greatest hardware is no longer
reserved solely for those running commercial operating systems.



About Infiscale:

In 2006 the developers of Caos and GravityPark decided to merge and
create a new organization called Infiscale. Far too often, projects
that begin as free and open source software disappear or become
closed/commercial offerings at the first sign of success, leaving the
community scrambling to find alternatives. Infiscale's goal is to keep
open tools open and free projects free, to ensure the survival of the
software many users in the community have grown to rely upon.
Infiscale provides a corporate interface to open source projects when
needed. For instance, vendors hoping to guarantee compatibility with
their products or services may partner with Infiscale to facilitate
this achievement. Infiscale can also provide support and consulting
services for clients requiring assistance or service-level agreements.
The funding provided by these commercial offerings allows Infiscale's
development team to continue contributing great free software to the
community.


-- 
Greg Kurtzer
http://www.infiscale.com/
http://www.runlevelzero.net/
http://www.perceus.org/
http://www.caoslinux.org/


From distro.watch at msa.hinet.net  Tue Nov 25 16:49:09 2008
From: distro.watch at msa.hinet.net (Ladislav Bodnar)
Date: Wed, 26 Nov 2008 08:49:09 +0800
Subject: [cAos] Caos NSA 1.0 Released!
In-Reply-To: <571f1a060811250950m4b3521dk1022d902d4c3809a@mail.gmail.com>
References: <571f1a060811250950m4b3521dk1022d902d4c3809a@mail.gmail.com>
Message-ID: <200811260849.09814.distro.watch@msa.hinet.net>

On Wednesday 26 November 2008, Greg Kurtzer wrote:
> Announcing the formal release of Caos Linux NSA 1 for x86 and x86_64!

What does NSA stand for? I've looked around the CAOS web site briefly, but 
couldn't find the answer.

My second question is - how do you spell the name of your distro? I've seen 
it spelt as cAos, CAOS, Caos - sometimes all three variants on one page or 
one article (sometimes on one line - see the subject title of this email for 
example). So which is correct?

Thanks in advance :-)

Ladislav


From gmkurtzer at gmail.com  Tue Nov 25 16:58:56 2008
From: gmkurtzer at gmail.com (Greg Kurtzer)
Date: Tue, 25 Nov 2008 16:58:56 -0800
Subject: [cAos] Caos NSA 1.0 Released!
In-Reply-To: <200811260849.09814.distro.watch@msa.hinet.net>
References: <571f1a060811250950m4b3521dk1022d902d4c3809a@mail.gmail.com>
	<200811260849.09814.distro.watch@msa.hinet.net>
Message-ID: <571f1a060811251658s40c06d8eseb808ae58959bb23@mail.gmail.com>

NSA stands for "Node Server Appliance" (High Performance compute,
Servers and various appliances). I will have our web person define NSA
on our front page. Thanks for pointing that out.

It used to be "cAos", but as we matured we changed it to "Caos".

Many thanks for a wonderful site, and always showing your support for
us. Please feel free to let me know if you have any further questions!

Greg

On Tue, Nov 25, 2008 at 4:49 PM, Ladislav Bodnar
<distro.watch at msa.hinet.net> wrote:
> On Wednesday 26 November 2008, Greg Kurtzer wrote:
>> Announcing the formal release of Caos Linux NSA 1 for x86 and x86_64!
>
> What does NSA stand for? I've looked around the CAOS web site briefly, but
> couldn't find the answer.
>
> My second question is - how do you spell the name of your distro? I've seen
> it spelt as cAos, CAOS, Caos - sometimes all three variants on one page or
> one article (sometimes on one line - see the subject title of this email for
> example). So which is correct?
>
> Thanks in advance :-)
>
> Ladislav
> _______________________________________________
> cAos mailing list
> cAos at caoslinux.org
> http://lists.caosity.org/mailman/listinfo/caos
>



-- 
Greg Kurtzer
http://www.infiscale.com/
http://www.runlevelzero.net/
http://www.perceus.org/
http://www.caoslinux.org/


From stefan at mdy.univie.ac.at  Tue Nov 25 23:45:47 2008
From: stefan at mdy.univie.ac.at (Stefan Boresch)
Date: Wed, 26 Nov 2008 08:45:47 +0100
Subject: [cAos] Caos NSA 1.0 Released!
In-Reply-To: <571f1a060811250950m4b3521dk1022d902d4c3809a@mail.gmail.com>
References: <571f1a060811250950m4b3521dk1022d902d4c3809a@mail.gmail.com>
Message-ID: <20081126074547.GJ9259@loop.mdy.univie.ac.at>

On Tue, Nov 25, 2008 at 09:50:44AM -0800, Greg Kurtzer wrote:
> Announcing the formal release of Caos Linux NSA 1 for x86 and x86_64!
> 

Congratulations! I am very new user, but the experience has been
enjoyable, not to mention the great support I have received. 
And, I want to second the passage about many distributions
forgetting Linux' core competencies, so Caos is a pleasant change!

Stefan

-- 
Stefan Boresch
Institute for Computational Biological Chemistry
University of Vienna, Waehringerstr. 17       A-1090 Vienna, Austria
Phone: -43-1-427752715                        Fax:   -43-1-427752790


From avrilmaomao at gmail.com  Wed Nov 26 04:50:15 2008
From: avrilmaomao at gmail.com (maomao sun)
Date: Wed, 26 Nov 2008 20:50:15 +0800
Subject: [cAos] Some questions about the partition during the installation
Message-ID: <8ecf88d10811260450s7023f6c4y719854bb05ab9715@mail.gmail.com>

My hardisk  are as follows:

Device Boot      Start         End      Blocks   Id  System
/dev/sda1   *           1        1020     8193118+   7  HPFS/NTFS
/dev/sda2            1021        6880    47070450    f  W95 Ext'd (LBA)
/dev/sda3            6881        8096     9767520   83  Linux
/dev/sda4            8097       10011    15382237+  83  Linux
/dev/sda5            1021        3825    22531131    7  HPFS/NTFS
/dev/sda6            3826        6757    23551258+   b  W95 FAT32
/dev/sda7            6758        6770      104391   83  Linux
/dev/sda8            6771        6880      883543+  82  Linux swap / Solaris

I've already have   windows and Fedora installed and I've data on sda5(ntfs)
and sda6(fat).
Now I want to install Caos Linux NSA to replace fedora.But I want to keep
windows and my data,If I choose the autoinstall mode,can I achieve my goal?
If not,what shall I do?


Best regards.


-- 
MAo oAM
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.infiscale.org/pipermail/caos/attachments/20081126/4beaea6e/attachment.html 

From dmadsen at ikk-nord.de  Wed Nov 26 05:29:21 2008
From: dmadsen at ikk-nord.de (Dethlef Madsen)
Date: Wed, 26 Nov 2008 14:29:21 +0100
Subject: [cAos] Caos NSA 1.0 Released!
References: <571f1a060811250950m4b3521dk1022d902d4c3809a@mail.gmail.com>
Message-ID: <001801c94fca$fdb5d4b0$0564c80a@CZC5110LOJ>

Congratulations !

I like NSA very much. It's stable, fast and very well maintained  So keep on 
the good work.

Thank you !

Dethlef  (private user) 



From stefan at mdy.univie.ac.at  Wed Nov 26 05:57:22 2008
From: stefan at mdy.univie.ac.at (Stefan Boresch)
Date: Wed, 26 Nov 2008 14:57:22 +0100
Subject: [cAos] Some questions about the partition during the
	installation
In-Reply-To: <8ecf88d10811260450s7023f6c4y719854bb05ab9715@mail.gmail.com>
References: <8ecf88d10811260450s7023f6c4y719854bb05ab9715@mail.gmail.com>
Message-ID: <20081126135722.GN9259@loop.mdy.univie.ac.at>

Sorry, no answer, but I am highly interested in the answer because
I have a related problem/question:

On Wed, Nov 26, 2008 at 08:50:15PM +0800, maomao sun wrote:
> My hardisk  are as follows:
> 
[snip]
> I've already have   windows and Fedora installed and I've data on sda5(ntfs)
> and sda6(fat).
> Now I want to install Caos Linux NSA to replace fedora.But I want to keep
> windows and my data,If I choose the autoinstall mode,can I achieve my goal?
> If not,what shall I do?

I also have a dual-boot machine, where nuking the Windows partions is
out of the question. I have two Linux partitions, where I would
like to replace Ubuntu by Caos. These partitions correspond to / and /home, 
but there is *no* /boot (which caos seems to like/require??).

How should I go about this ...? (Could parted "split" / into
/ and /boot? After all, it was perfectly capable of shrinking my Windows
partition(s) and creating additional partitions into the free space.
I could always do this from a live CD (Ubuntu comes to mind ;-) ...)

Best regards,

Stefan

-- 
Stefan Boresch
Institute for Computational Biological Chemistry
University of Vienna, Waehringerstr. 17       A-1090 Vienna, Austria
Phone: -43-1-427752715                        Fax:   -43-1-427752790


From avrilmaomao at gmail.com  Wed Nov 26 06:53:06 2008
From: avrilmaomao at gmail.com (maomao sun)
Date: Wed, 26 Nov 2008 22:53:06 +0800
Subject: [cAos] Some questions about the partition during the
	installation
In-Reply-To: <20081126135722.GN9259@loop.mdy.univie.ac.at>
References: <8ecf88d10811260450s7023f6c4y719854bb05ab9715@mail.gmail.com>
	<20081126135722.GN9259@loop.mdy.univie.ac.at>
Message-ID: <8ecf88d10811260653h12cf9ce1wc5675ea29682cbfa@mail.gmail.com>

We're facing exactly the same problem.[?]
I think the best way is to try.[?]
I'm trying to figure it out by using the virtual box.

On Wed, Nov 26, 2008 at 9:57 PM, Stefan Boresch <stefan at mdy.univie.ac.at>wrote:

> Sorry, no answer, but I am highly interested in the answer because
> I have a related problem/question:
>
> On Wed, Nov 26, 2008 at 08:50:15PM +0800, maomao sun wrote:
> > My hardisk  are as follows:
> >
> [snip]
> > I've already have   windows and Fedora installed and I've data on
> sda5(ntfs)
> > and sda6(fat).
> > Now I want to install Caos Linux NSA to replace fedora.But I want to keep
> > windows and my data,If I choose the autoinstall mode,can I achieve my
> goal?
> > If not,what shall I do?
>
> I also have a dual-boot machine, where nuking the Windows partions is
> out of the question. I have two Linux partitions, where I would
> like to replace Ubuntu by Caos. These partitions correspond to / and /home,
> but there is *no* /boot (which caos seems to like/require??).
>
> How should I go about this ...? (Could parted "split" / into
> / and /boot? After all, it was perfectly capable of shrinking my Windows
> partition(s) and creating additional partitions into the free space.
> I could always do this from a live CD (Ubuntu comes to mind ;-) ...)
>
> Best regards,
>
> Stefan
>
> --
> Stefan Boresch
> Institute for Computational Biological Chemistry
> University of Vienna, Waehringerstr. 17       A-1090 Vienna, Austria
> Phone: -43-1-427752715                        Fax:   -43-1-427752790
> _______________________________________________
> cAos mailing list
> cAos at caoslinux.org
> http://lists.caosity.org/mailman/listinfo/caos
>



-- 
MAo oAM
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.infiscale.org/pipermail/caos/attachments/20081126/46663938/attachment.html 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: image/gif
Size: 541 bytes
Desc: not available
Url : http://lists.infiscale.org/pipermail/caos/attachments/20081126/46663938/attachment.gif 

From glykos at mbg.duth.gr  Wed Nov 26 07:03:43 2008
From: glykos at mbg.duth.gr (Nicholas M Glykos)
Date: Wed, 26 Nov 2008 17:03:43 +0200 (EET)
Subject: [cAos] RealTek network cards woes ...
Message-ID: <Pine.LNX.4.62.0811261655180.29682@aspera.cluster.mbg.gr>


Dear All,

This is definitely not a CAOS-specific problem, so please bear with me:
The set-up is: head-node with 3 interfaces running CAOS NSA 1.0rc2 and 
diskless nodes running caos-nsa-node-1.0rc1-1.stateless.x86_64.vnfs. The 
nodes have two network cards, the first is a RTL8168c (on-board), the 
second is a RTL8169sb. 

The story is: PXE runs smoothly using the first (on-board) card, Perceus 
boots, fails to detect the RTL8168c card (that is, the card that was used 
for TFTP in the first place), uses the second card for DCHP, provisioning 
goes smoothly via the second card, I end-up with nodes with just one 
interface (corresponding to the not-on-board card). Nothing in dmesg about 
the on-board nic. 

The notable thing (at least for someone with as little knowledge as 
myself) is that if I grab a USB CDROM and boot a node with the CAOS NSA 
1.0rc2 CD, both cards are properly detected by the r8169 kernel module 
(which is identical with the one in the VNFS).

My current working hypothesis is that PXE leaves the onboard card in a 
wrong state (whatever this may mean to someone that knows what is talking 
about). I have tried forcing unloading and reloading of the r8169 module, 
but unfortunately to no avail. I have also tried the Realtek-distributed 
kernel module (r8168), also with negative results.

I would be grateful for any ideas or comments,
Thanks,
Nicholas


-- 


             Dr Nicholas M. Glykos, Department of Molecular 
         Biology and Genetics, Democritus University of Thrace,
   University Campus, 68100 Alexandroupolis, Greece, Fax +302551030613
      Tel ++302551030620 (77620),  http://www.mbg.duth.gr/~glykos/




From mej at caoslinux.org  Wed Nov 26 08:09:59 2008
From: mej at caoslinux.org (Michael Jennings)
Date: Wed, 26 Nov 2008 08:09:59 -0800
Subject: [cAos] Some questions about the partition during
	the	installation
In-Reply-To: <20081126135722.GN9259@loop.mdy.univie.ac.at>
References: <8ecf88d10811260450s7023f6c4y719854bb05ab9715@mail.gmail.com>
	<20081126135722.GN9259@loop.mdy.univie.ac.at>
Message-ID: <20081126160959.GI27833@kainx.org>

On Wednesday, 26 November 2008, at 14:57:22 (+0100),
Stefan Boresch wrote:

> Sorry, no answer, but I am highly interested in the answer because
> I have a related problem/question:
> 
> On Wed, Nov 26, 2008 at 08:50:15PM +0800, maomao sun wrote:
> > My hardisk  are as follows:
> > 
> [snip]
> > I've already have   windows and Fedora installed and I've data on sda5(ntfs)
> > and sda6(fat).
> > Now I want to install Caos Linux NSA to replace fedora.But I want to keep
> > windows and my data,If I choose the autoinstall mode,can I achieve my goal?
> > If not,what shall I do?

I highly recommend testing my suggestion in a VM setup before you try
it on a live system, for obvious reasons.

That said, passing the "layout=manual" option to "autoinstall" should
allow you to use fdisk to create your partitions and specify your
fstab layout by hand.  Clearly this is not suitable for inexperienced
users, but it doesn't sound like you guys fit in that category. :)

We hope to improve cinch in the future with more layout options and
better manual partitioning, but for now that's the best answer we
have.  (Insert "patches welcome" comment here.)  ;-)

Michael

-- 
Michael Jennings (a.k.a. KainX)  http://www.kainx.org/  <mej at kainx.org>
Linux Server/Cluster Admin, LBL.gov       Author, Eterm (www.eterm.org)
-----------------------------------------------------------------------
 "This is how the world ends: swallowed in fire, but not in darkness.
  You will live on.  The voice of all our ancestors, the voice of our
  fathers and our mothers to the last generation.  We created the
  world we think you would have wished for us, and now we leave the
  cradle for the last time."                           -- Babylon Five


From stefan at mdy.univie.ac.at  Wed Nov 26 08:17:54 2008
From: stefan at mdy.univie.ac.at (Stefan Boresch)
Date: Wed, 26 Nov 2008 17:17:54 +0100
Subject: [cAos] Some questions about the partition during
	the	installation
In-Reply-To: <20081126160959.GI27833@kainx.org>
References: <8ecf88d10811260450s7023f6c4y719854bb05ab9715@mail.gmail.com>
	<20081126135722.GN9259@loop.mdy.univie.ac.at>
	<20081126160959.GI27833@kainx.org>
Message-ID: <20081126161754.GP9259@loop.mdy.univie.ac.at>

Michael,

thanks for the feedback.

On Wed, Nov 26, 2008 at 08:09:59AM -0800, Michael Jennings wrote:
> On Wednesday, 26 November 2008, at 14:57:22 (+0100),
> Stefan Boresch wrote:
> 
> That said, passing the "layout=manual" option to "autoinstall" should
> allow you to use fdisk to create your partitions and specify your
> fstab layout by hand.  Clearly this is not suitable for inexperienced
> users, but it doesn't sound like you guys fit in that category. :)
> 

I am aware of that option, but the one time I tried it (on a virgin
machine), the system would not boot since I had not made a separate
/boot partition.  (Basically, I went through the fdisk + vi /etc/fstab
routine and created / , /var , /opt and /usr/local, I believe, but ended
up with a nonfunctional grub. There was a brief error message from cinch
flashing by, but the system didn't stop or alert me as to a serious problem.
I only found out when trying to reboot.) In this case it wasn't a problem
since I could easily start from scratch, adding /boot , but in the
case I have in mind that's not an option ...

Best regards,

Stefan

-- 
Stefan Boresch
Institute for Computational Biological Chemistry
University of Vienna, Waehringerstr. 17       A-1090 Vienna, Austria
Phone: -43-1-427752715                        Fax:   -43-1-427752790


From mej at caoslinux.org  Wed Nov 26 08:26:46 2008
From: mej at caoslinux.org (Michael Jennings)
Date: Wed, 26 Nov 2008 08:26:46 -0800
Subject: [cAos] Some questions about the partition during
	the	installation
In-Reply-To: <20081126161754.GP9259@loop.mdy.univie.ac.at>
References: <8ecf88d10811260450s7023f6c4y719854bb05ab9715@mail.gmail.com>
	<20081126135722.GN9259@loop.mdy.univie.ac.at>
	<20081126160959.GI27833@kainx.org>
	<20081126161754.GP9259@loop.mdy.univie.ac.at>
Message-ID: <20081126162646.GJ27833@kainx.org>

On Wednesday, 26 November 2008, at 17:17:54 (+0100),
Stefan Boresch wrote:

> I am aware of that option, but the one time I tried it (on a virgin
> machine), the system would not boot since I had not made a separate
> /boot partition.  (Basically, I went through the fdisk + vi /etc/fstab
> routine and created / , /var , /opt and /usr/local, I believe, but ended
> up with a nonfunctional grub. There was a brief error message from cinch
> flashing by, but the system didn't stop or alert me as to a serious problem.
> I only found out when trying to reboot.) In this case it wasn't a problem
> since I could easily start from scratch, adding /boot , but in the
> case I have in mind that's not an option ...

grub issues are actually a different problem. :)

Did grub give you a prompt, or did it fail altogether?

If you got a prompt, what did you try?

Michael

-- 
Michael Jennings (a.k.a. KainX)  http://www.kainx.org/  <mej at kainx.org>
Linux Server/Cluster Admin, LBL.gov       Author, Eterm (www.eterm.org)
-----------------------------------------------------------------------
 "It takes courage to look inside yourself, and even more courage to
  write it for other people to see."           -- Captain Sisko, "DS9"


From stefan at mdy.univie.ac.at  Wed Nov 26 08:40:19 2008
From: stefan at mdy.univie.ac.at (Stefan Boresch)
Date: Wed, 26 Nov 2008 17:40:19 +0100
Subject: [cAos] Some questions about the partition during
	the	installation
In-Reply-To: <20081126162646.GJ27833@kainx.org>
References: <8ecf88d10811260450s7023f6c4y719854bb05ab9715@mail.gmail.com>
	<20081126135722.GN9259@loop.mdy.univie.ac.at>
	<20081126160959.GI27833@kainx.org>
	<20081126161754.GP9259@loop.mdy.univie.ac.at>
	<20081126162646.GJ27833@kainx.org>
Message-ID: <20081126164019.GR9259@loop.mdy.univie.ac.at>

On Wed, Nov 26, 2008 at 08:26:46AM -0800, Michael Jennings wrote:
> On Wednesday, 26 November 2008, at 17:17:54 (+0100),
> Stefan Boresch wrote:
> 
> > I only found out when trying to reboot.) In this case it wasn't a problem
> > since I could easily start from scratch, adding /boot , but in the
> > case I have in mind that's not an option ...
> 
> grub issues are actually a different problem. :)
> 
> Did grub give you a prompt, or did it fail altogether?
> 
> If you got a prompt, what did you try?
> 

If I remember correctly, grub did not come up (it was
as if it wasn't installed).  I certainly did not get a prompt.
(My original report was
http://lists.caosity.org/pipermail/caos/2008-November/002504.html)
At that point I didn't care too much and redid the procedure with 
/boot ...

So, I take it that you haven't seen this -- I guess I should try to
recreate the problem ... 

Stefan

-- 
Stefan Boresch
Institute for Computational Biological Chemistry
University of Vienna, Waehringerstr. 17       A-1090 Vienna, Austria
Phone: -43-1-427752715                        Fax:   -43-1-427752790


From gmkurtzer at gmail.com  Wed Nov 26 10:08:36 2008
From: gmkurtzer at gmail.com (Greg Kurtzer)
Date: Wed, 26 Nov 2008 10:08:36 -0800
Subject: [cAos] Some questions about the partition during the
	installation
In-Reply-To: <20081126164019.GR9259@loop.mdy.univie.ac.at>
References: <8ecf88d10811260450s7023f6c4y719854bb05ab9715@mail.gmail.com>
	<20081126135722.GN9259@loop.mdy.univie.ac.at>
	<20081126160959.GI27833@kainx.org>
	<20081126161754.GP9259@loop.mdy.univie.ac.at>
	<20081126162646.GJ27833@kainx.org>
	<20081126164019.GR9259@loop.mdy.univie.ac.at>
Message-ID: <571f1a060811261008v7609650ci27d5f9f1922fe6f0@mail.gmail.com>

I will try and recreate this evening as well, but feel free to let us
know if you get any further on debugging it.

Thanks!

Greg

On Wed, Nov 26, 2008 at 8:40 AM, Stefan Boresch <stefan at mdy.univie.ac.at> wrote:
> On Wed, Nov 26, 2008 at 08:26:46AM -0800, Michael Jennings wrote:
>> On Wednesday, 26 November 2008, at 17:17:54 (+0100),
>> Stefan Boresch wrote:
>>
>> > I only found out when trying to reboot.) In this case it wasn't a problem
>> > since I could easily start from scratch, adding /boot , but in the
>> > case I have in mind that's not an option ...
>>
>> grub issues are actually a different problem. :)
>>
>> Did grub give you a prompt, or did it fail altogether?
>>
>> If you got a prompt, what did you try?
>>
>
> If I remember correctly, grub did not come up (it was
> as if it wasn't installed).  I certainly did not get a prompt.
> (My original report was
> http://lists.caosity.org/pipermail/caos/2008-November/002504.html)
> At that point I didn't care too much and redid the procedure with
> /boot ...
>
> So, I take it that you haven't seen this -- I guess I should try to
> recreate the problem ...
>
> Stefan
>
> --
> Stefan Boresch
> Institute for Computational Biological Chemistry
> University of Vienna, Waehringerstr. 17       A-1090 Vienna, Austria
> Phone: -43-1-427752715                        Fax:   -43-1-427752790
> _______________________________________________
> cAos mailing list
> cAos at caoslinux.org
> http://lists.caosity.org/mailman/listinfo/caos
>



-- 
Greg Kurtzer
http://www.infiscale.com/
http://www.runlevelzero.net/
http://www.perceus.org/
http://www.caoslinux.org/


From carlopmart at gmail.com  Wed Nov 26 12:02:01 2008
From: carlopmart at gmail.com (carlopmart)
Date: Wed, 26 Nov 2008 21:02:01 +0100
Subject: [cAos] Product life cycle for cAos-NSA
Message-ID: <492DAB39.5010202@gmail.com>

Hi all,

  First of all, congratulations for this new release: cAos-NSA-1.0. And now my 
question, which is the support cycle for this new release: 3-5 year like cAos-2 
or 18, 16, 6 months like fedora or ubuntu does??

Many thanks.

-- 
CL Martinez
carlopmart {at} gmail {d0t} com


From avrilmaomao at gmail.com  Wed Nov 26 19:14:52 2008
From: avrilmaomao at gmail.com (maomao)
Date: Thu, 27 Nov 2008 11:14:52 +0800
Subject: [cAos] Some questions about the partition during theinstallation
References: <8ecf88d10811260450s7023f6c4y719854bb05ab9715@mail.gmail.com><20081126135722.GN9259@loop.mdy.univie.ac.at><20081126160959.GI27833@kainx.org><20081126161754.GP9259@loop.mdy.univie.ac.at><20081126162646.GJ27833@kainx.org><20081126164019.GR9259@loop.mdy.univie.ac.at>
	<571f1a060811261008v7609650ci27d5f9f1922fe6f0@mail.gmail.com>
Message-ID: <E8806EEFE28F4CD4895660C5434064F5@hichongzi>

I thought you'd already seen Michael Jennings's message.
I 've found that his method was a great way ,at least in my virtual box.
Use the layout=manual option,after the fdisk process,it will prompt to edit 
the fstab so that we can keep our windows partition and install caos on 
certain partitions.

Good Luck!!

--------------------------------------------------
From: "Greg Kurtzer" <gmkurtzer at gmail.com>
Sent: Thursday, November 27, 2008 2:08 AM
To: "Community-Assembled Operating System" <caos at caoslinux.org>
Subject: Re: [cAos] Some questions about the partition during 
theinstallation

> I will try and recreate this evening as well, but feel free to let us
> know if you get any further on debugging it.
>
> Thanks!
>
> Greg
>
> On Wed, Nov 26, 2008 at 8:40 AM, Stefan Boresch <stefan at mdy.univie.ac.at> 
> wrote:
>> On Wed, Nov 26, 2008 at 08:26:46AM -0800, Michael Jennings wrote:
>>> On Wednesday, 26 November 2008, at 17:17:54 (+0100),
>>> Stefan Boresch wrote:
>>>
>>> > I only found out when trying to reboot.) In this case it wasn't a 
>>> > problem
>>> > since I could easily start from scratch, adding /boot , but in the
>>> > case I have in mind that's not an option ...
>>>
>>> grub issues are actually a different problem. :)
>>>
>>> Did grub give you a prompt, or did it fail altogether?
>>>
>>> If you got a prompt, what did you try?
>>>
>>
>> If I remember correctly, grub did not come up (it was
>> as if it wasn't installed).  I certainly did not get a prompt.
>> (My original report was
>> http://lists.caosity.org/pipermail/caos/2008-November/002504.html)
>> At that point I didn't care too much and redid the procedure with
>> /boot ...
>>
>> So, I take it that you haven't seen this -- I guess I should try to
>> recreate the problem ...
>>
>> Stefan
>>
>> --
>> Stefan Boresch
>> Institute for Computational Biological Chemistry
>> University of Vienna, Waehringerstr. 17       A-1090 Vienna, Austria
>> Phone: -43-1-427752715                        Fax:   -43-1-427752790
>> _______________________________________________
>> cAos mailing list
>> cAos at caoslinux.org
>> http://lists.caosity.org/mailman/listinfo/caos
>>
>
>
>
> -- 
> Greg Kurtzer
> http://www.infiscale.com/
> http://www.runlevelzero.net/
> http://www.perceus.org/
> http://www.caoslinux.org/
> _______________________________________________
> cAos mailing list
> cAos at caoslinux.org
> http://lists.caosity.org/mailman/listinfo/caos 



From avrilmaomao at gmail.com  Wed Nov 26 19:16:08 2008
From: avrilmaomao at gmail.com (maomao)
Date: Thu, 27 Nov 2008 11:16:08 +0800
Subject: [cAos] Some questions about the partition duringthe	installation
References: <8ecf88d10811260450s7023f6c4y719854bb05ab9715@mail.gmail.com><20081126135722.GN9259@loop.mdy.univie.ac.at>
	<20081126160959.GI27833@kainx.org>
Message-ID: <1556DE2F11F6477C934783F317BE62E5@hichongzi>

Thanks a lot.

I 've tried on my virtual box.It was great.^_^

--------------------------------------------------
From: "Michael Jennings" <mej at caoslinux.org>
Sent: Thursday, November 27, 2008 12:09 AM
To: "Community-Assembled Operating System" <caos at caoslinux.org>
Subject: Re: [cAos] Some questions about the partition duringtheinstallation> On Wednesday, 26 November 2008, at 14:57:22 (+0100),
> Stefan Boresch wrote:
>
>> Sorry, no answer, but I am highly interested in the answer because
>> I have a related problem/question:
>>
>> On Wed, Nov 26, 2008 at 08:50:15PM +0800, maomao sun wrote:
>> > My hardisk  are as follows:
>> >
>> [snip]
>> > I've already have   windows and Fedora installed and I've data on 
>> > sda5(ntfs)
>> > and sda6(fat).
>> > Now I want to install Caos Linux NSA to replace fedora.But I want to 
>> > keep
>> > windows and my data,If I choose the autoinstall mode,can I achieve my 
>> > goal?
>> > If not,what shall I do?
>
> I highly recommend testing my suggestion in a VM setup before you try
> it on a live system, for obvious reasons.
>
> That said, passing the "layout=manual" option to "autoinstall" should
> allow you to use fdisk to create your partitions and specify your
> fstab layout by hand.  Clearly this is not suitable for inexperienced
> users, but it doesn't sound like you guys fit in that category. :)
>
> We hope to improve cinch in the future with more layout options and
> better manual partitioning, but for now that's the best answer we
> have.  (Insert "patches welcome" comment here.)  ;-)
>
> Michael
>
> -- 
> Michael Jennings (a.k.a. KainX)  http://www.kainx.org/  <mej at kainx.org>
> Linux Server/Cluster Admin, LBL.gov       Author, Eterm (www.eterm.org)
> -----------------------------------------------------------------------
> "This is how the world ends: swallowed in fire, but not in darkness.
>  You will live on.  The voice of all our ancestors, the voice of our
>  fathers and our mothers to the last generation.  We created the
>  world we think you would have wished for us, and now we leave the
>  cradle for the last time."                           -- Babylon Five
> _______________________________________________
> cAos mailing list
> cAos at caoslinux.org
> http://lists.caosity.org/mailman/listinfo/caos 



From gmkurtzer at gmail.com  Wed Nov 26 21:32:48 2008
From: gmkurtzer at gmail.com (Greg Kurtzer)
Date: Wed, 26 Nov 2008 21:32:48 -0800
Subject: [cAos] Product life cycle for cAos-NSA
In-Reply-To: <492DAB39.5010202@gmail.com>
References: <492DAB39.5010202@gmail.com>
Message-ID: <571f1a060811262132m404ae99bl9ba0ca8d70f60dda@mail.gmail.com>

Hello,

We view Caos releases as production solutions thus the life span will
be 3-5 years (pending user and developer support).

That is not to say that the packages will be completely static for the
life time of the system. So for example, it will maintain a static
core that will only get critical bug and security fixes to guarantee a
base line binary compatibility.

The rest of the operating system (where possible) will follow the
upstream maintainers as closely as possible. So for example, while gcc
will be locked at 4.3, openssh will follow the current release as will
other utilities, X applications, etc.. Other packages that are not
core, but heavily used and linked by other apps maybe updated on
periodic dot release upgrades (NSA 1.0.x -> 1.1.x). And all dot
upgrades will be upgradable from every previous release.

Let me know if this isn't clear. Also, this information will be posted
on the website shortly.

Great question BTW!

Greg



On Wed, Nov 26, 2008 at 12:02 PM, carlopmart <carlopmart at gmail.com> wrote:
> Hi all,
>
>  First of all, congratulations for this new release: cAos-NSA-1.0. And now my
> question, which is the support cycle for this new release: 3-5 year like cAos-2
> or 18, 16, 6 months like fedora or ubuntu does??
>
> Many thanks.
>
> --
> CL Martinez
> carlopmart {at} gmail {d0t} com
> _______________________________________________
> cAos mailing list
> cAos at caoslinux.org
> http://lists.caosity.org/mailman/listinfo/caos
>



-- 
Greg Kurtzer
http://www.infiscale.com/
http://www.runlevelzero.net/
http://www.perceus.org/
http://www.caoslinux.org/


From gmkurtzer at gmail.com  Thu Nov 27 00:11:10 2008
From: gmkurtzer at gmail.com (Greg Kurtzer)
Date: Thu, 27 Nov 2008 00:11:10 -0800
Subject: [cAos] RealTek network cards woes ...
In-Reply-To: <Pine.LNX.4.62.0811261655180.29682@aspera.cluster.mbg.gr>
References: <Pine.LNX.4.62.0811261655180.29682@aspera.cluster.mbg.gr>
Message-ID: <571f1a060811270011s6f95eba7p9e572391b7c45476@mail.gmail.com>

Nicholas,

One idea is that the Perceus stage1 just isn't supporting your
hardware properly (it is not the same kernel as a running Caos system
or the installer).

Can you run the command:

   /sbin/detect -ai | grep NETWORK

on that node or a similar node running Caos, the Caos installer or the
Perceus stage 1 shell, and send me the output? From that I can help
determine if the Perceus stage1 properly supports that driver.

An alternative is to test the 1.5 development tree of Perceus. I can
make available for you a kernel and initramfs.img from the 1.5 tree
that you can test.

Warm regards,

Greg

On Wed, Nov 26, 2008 at 7:03 AM, Nicholas M Glykos <glykos at mbg.duth.gr> wrote:
>
> Dear All,
>
> This is definitely not a CAOS-specific problem, so please bear with me:
> The set-up is: head-node with 3 interfaces running CAOS NSA 1.0rc2 and
> diskless nodes running caos-nsa-node-1.0rc1-1.stateless.x86_64.vnfs. The
> nodes have two network cards, the first is a RTL8168c (on-board), the
> second is a RTL8169sb.
>
> The story is: PXE runs smoothly using the first (on-board) card, Perceus
> boots, fails to detect the RTL8168c card (that is, the card that was used
> for TFTP in the first place), uses the second card for DCHP, provisioning
> goes smoothly via the second card, I end-up with nodes with just one
> interface (corresponding to the not-on-board card). Nothing in dmesg about
> the on-board nic.
>
> The notable thing (at least for someone with as little knowledge as
> myself) is that if I grab a USB CDROM and boot a node with the CAOS NSA
> 1.0rc2 CD, both cards are properly detected by the r8169 kernel module
> (which is identical with the one in the VNFS).
>
> My current working hypothesis is that PXE leaves the onboard card in a
> wrong state (whatever this may mean to someone that knows what is talking
> about). I have tried forcing unloading and reloading of the r8169 module,
> but unfortunately to no avail. I have also tried the Realtek-distributed
> kernel module (r8168), also with negative results.
>
> I would be grateful for any ideas or comments,
> Thanks,
> Nicholas
>
>
> --
>
>
>             Dr Nicholas M. Glykos, Department of Molecular
>         Biology and Genetics, Democritus University of Thrace,
>   University Campus, 68100 Alexandroupolis, Greece, Fax +302551030613
>      Tel ++302551030620 (77620),  http://www.mbg.duth.gr/~glykos/
>
>
> _______________________________________________
> cAos mailing list
> cAos at caoslinux.org
> http://lists.caosity.org/mailman/listinfo/caos
>



-- 
Greg Kurtzer
http://www.infiscale.com/
http://www.runlevelzero.net/
http://www.perceus.org/
http://www.caoslinux.org/


From glykos at mbg.duth.gr  Fri Nov 28 02:05:23 2008
From: glykos at mbg.duth.gr (Nicholas M Glykos)
Date: Fri, 28 Nov 2008 12:05:23 +0200 (EET)
Subject: [cAos] [Solved] RealTek network cards woes ...
In-Reply-To: <Pine.LNX.4.62.0811261655180.29682@aspera.cluster.mbg.gr>
References: <Pine.LNX.4.62.0811261655180.29682@aspera.cluster.mbg.gr>
Message-ID: <Pine.LNX.4.62.0811281200220.19596@aspera.cluster.mbg.gr>



Thanks to generous help from Greg Kurtzer who offered me a copy of the 1.5 
development tree of Perceus (perceus-1.4.9d-stage1.tar.gz), the problem I 
reported earlier is now solved and closed.

Thanks again,
Nicholas


-- 


             Dr Nicholas M. Glykos, Department of Molecular 
         Biology and Genetics, Democritus University of Thrace,
   University Campus, 68100 Alexandroupolis, Greece, Fax +302551030613
      Tel ++302551030620 (77620),  http://www.mbg.duth.gr/~glykos/




From eca.ict at gmail.com  Sat Nov 29 07:33:20 2008
From: eca.ict at gmail.com (Egbert Aarts)
Date: Sat, 29 Nov 2008 16:33:20 +0100
Subject: [cAos] MCRYPT
Message-ID: <aeb32ce00811290733h1187125fi4fdaa12cecb5bbdf@mail.gmail.com>

Hello,

My name is Egbert Aarts from the Netherlands and I am totally new to
CAOS Linux. I am presently setting up a webserver.
At this moment I am running into problems with setting up MCRYPT. The
last few day's I did a lot of research and testing but I had no
success to get the problem fixed.

I hope someone can give me some advice or a link to an rpm package
which I can use with the CAOS Linux Distribution.

Thanks in advance for any assistance!


Regards,
Egbert Aarts


From gmkurtzer at gmail.com  Sat Nov 29 12:19:20 2008
From: gmkurtzer at gmail.com (Greg Kurtzer)
Date: Sat, 29 Nov 2008 12:19:20 -0800
Subject: [cAos] MCRYPT
In-Reply-To: <aeb32ce00811290733h1187125fi4fdaa12cecb5bbdf@mail.gmail.com>
References: <aeb32ce00811290733h1187125fi4fdaa12cecb5bbdf@mail.gmail.com>
Message-ID: <571f1a060811291219l778a090era037e87ddda0048d@mail.gmail.com>

We are on it and will let you know as soon as the package is
available. Someone from the development team will let you know when it
is available for testing (probably tomorrow).

Thanks for your patience.

Greg



On Sat, Nov 29, 2008 at 7:33 AM, Egbert Aarts <eca.ict at gmail.com> wrote:
> Hello,
>
> My name is Egbert Aarts from the Netherlands and I am totally new to
> CAOS Linux. I am presently setting up a webserver.
> At this moment I am running into problems with setting up MCRYPT. The
> last few day's I did a lot of research and testing but I had no
> success to get the problem fixed.
>
> I hope someone can give me some advice or a link to an rpm package
> which I can use with the CAOS Linux Distribution.
>
> Thanks in advance for any assistance!
>
>
> Regards,
> Egbert Aarts
> _______________________________________________
> cAos mailing list
> cAos at caoslinux.org
> http://lists.caosity.org/mailman/listinfo/caos
>



-- 
Greg Kurtzer
http://www.infiscale.com/
http://www.runlevelzero.net/
http://www.perceus.org/
http://www.caoslinux.org/


From gmkurtzer at gmail.com  Sat Nov 29 23:08:25 2008
From: gmkurtzer at gmail.com (Greg Kurtzer)
Date: Sat, 29 Nov 2008 23:08:25 -0800
Subject: [cAos] MCRYPT
In-Reply-To: <571f1a060811291219l778a090era037e87ddda0048d@mail.gmail.com>
References: <aeb32ce00811290733h1187125fi4fdaa12cecb5bbdf@mail.gmail.com>
	<571f1a060811291219l778a090era037e87ddda0048d@mail.gmail.com>
Message-ID: <571f1a060811292308o416d4a10p6b4f71327b145513@mail.gmail.com>

Egbert,

OK, we have php-mcrypt integrated in Caos NSA 1 which is currently in
our testing repositories. Please do the following to test:

# sidekick
   Navigate to: Packages -> Repositories -> select "nsa-testing"
# smart upgrade php*
# smart install php-mcrypt
# /etc/init.d/httpd restart

Please test and let us know if that resolves your question and how
Caos NSA is working out for you.

Warm regards,

Greg



On Sat, Nov 29, 2008 at 12:19 PM, Greg Kurtzer <gmkurtzer at gmail.com> wrote:
> We are on it and will let you know as soon as the package is
> available. Someone from the development team will let you know when it
> is available for testing (probably tomorrow).
>
> Thanks for your patience.
>
> Greg
>
>
>
> On Sat, Nov 29, 2008 at 7:33 AM, Egbert Aarts <eca.ict at gmail.com> wrote:
>> Hello,
>>
>> My name is Egbert Aarts from the Netherlands and I am totally new to
>> CAOS Linux. I am presently setting up a webserver.
>> At this moment I am running into problems with setting up MCRYPT. The
>> last few day's I did a lot of research and testing but I had no
>> success to get the problem fixed.
>>
>> I hope someone can give me some advice or a link to an rpm package
>> which I can use with the CAOS Linux Distribution.
>>
>> Thanks in advance for any assistance!
>>
>>
>> Regards,
>> Egbert Aarts
>> _______________________________________________
>> cAos mailing list
>> cAos at caoslinux.org
>> http://lists.caosity.org/mailman/listinfo/caos
>>
>
>
>
> --
> Greg Kurtzer
> http://www.infiscale.com/
> http://www.runlevelzero.net/
> http://www.perceus.org/
> http://www.caoslinux.org/
>



-- 
Greg Kurtzer
http://www.infiscale.com/
http://www.runlevelzero.net/
http://www.perceus.org/
http://www.caoslinux.org/


From eca.ict at gmail.com  Sun Nov 30 01:35:49 2008
From: eca.ict at gmail.com (Egbert Aarts)
Date: Sun, 30 Nov 2008 10:35:49 +0100
Subject: [cAos] MCRYPT
In-Reply-To: <571f1a060811292308o416d4a10p6b4f71327b145513@mail.gmail.com>
References: <aeb32ce00811290733h1187125fi4fdaa12cecb5bbdf@mail.gmail.com>
	<571f1a060811291219l778a090era037e87ddda0048d@mail.gmail.com>
	<571f1a060811292308o416d4a10p6b4f71327b145513@mail.gmail.com>
Message-ID: <aeb32ce00811300135g62f6a147ha986630e0497bed5@mail.gmail.com>

Greg,

This is super that you have so quick a package for MCRYPT available.
I only can install the package later today or early tomorrow morning.
As soon as I know the outcome I will let you know.

Thanks again, this is super service.
Egbert



2008/11/30 Greg Kurtzer <gmkurtzer at gmail.com>:
> Egbert,
>
> OK, we have php-mcrypt integrated in Caos NSA 1 which is currently in
> our testing repositories. Please do the following to test:
>
> # sidekick
>   Navigate to: Packages -> Repositories -> select "nsa-testing"
> # smart upgrade php*
> # smart install php-mcrypt
> # /etc/init.d/httpd restart
>
> Please test and let us know if that resolves your question and how
> Caos NSA is working out for you.
>
> Warm regards,
>
> Greg
>
>
>
> On Sat, Nov 29, 2008 at 12:19 PM, Greg Kurtzer <gmkurtzer at gmail.com> wrote:
>> We are on it and will let you know as soon as the package is
>> available. Someone from the development team will let you know when it
>> is available for testing (probably tomorrow).
>>
>> Thanks for your patience.
>>
>> Greg
>>
>>
>>
>> On Sat, Nov 29, 2008 at 7:33 AM, Egbert Aarts <eca.ict at gmail.com> wrote:
>>> Hello,
>>>
>>> My name is Egbert Aarts from the Netherlands and I am totally new to
>>> CAOS Linux. I am presently setting up a webserver.
>>> At this moment I am running into problems with setting up MCRYPT. The
>>> last few day's I did a lot of research and testing but I had no
>>> success to get the problem fixed.
>>>
>>> I hope someone can give me some advice or a link to an rpm package
>>> which I can use with the CAOS Linux Distribution.
>>>
>>> Thanks in advance for any assistance!
>>>
>>>
>>> Regards,
>>> Egbert Aarts
>>> _______________________________________________
>>> cAos mailing list
>>> cAos at caoslinux.org
>>> http://lists.caosity.org/mailman/listinfo/caos
>>>
>>
>>
>>
>> --
>> Greg Kurtzer
>> http://www.infiscale.com/
>> http://www.runlevelzero.net/
>> http://www.perceus.org/
>> http://www.caoslinux.org/
>>
>
>
>
> --
> Greg Kurtzer
> http://www.infiscale.com/
> http://www.runlevelzero.net/
> http://www.perceus.org/
> http://www.caoslinux.org/
> _______________________________________________
> cAos mailing list
> cAos at caoslinux.org
> http://lists.caosity.org/mailman/listinfo/caos
>


From eca.ict at gmail.com  Sun Nov 30 13:15:08 2008
From: eca.ict at gmail.com (E. Aarts)
Date: Sun, 30 Nov 2008 22:15:08 +0100
Subject: [cAos] MCRYPT
In-Reply-To: <571f1a060811292308o416d4a10p6b4f71327b145513@mail.gmail.com>
References: <aeb32ce00811290733h1187125fi4fdaa12cecb5bbdf@mail.gmail.com>	<571f1a060811291219l778a090era037e87ddda0048d@mail.gmail.com>
	<571f1a060811292308o416d4a10p6b4f71327b145513@mail.gmail.com>
Message-ID: <4933025C.1040002@gmail.com>

Hello Greg,

With your instructions "it was a piece of cake"

The remark in phpMyAdmin that MCRYPT is not available is gone and PHP 
info is showing very nice that MCRYPT is available.

Thanks for the super service!

Regards,
Egbert Aarts



Greg Kurtzer wrote:
> Egbert,
>
> OK, we have php-mcrypt integrated in Caos NSA 1 which is currently in
> our testing repositories. Please do the following to test:
>
> # sidekick
>    Navigate to: Packages -> Repositories -> select "nsa-testing"
> # smart upgrade php*
> # smart install php-mcrypt
> # /etc/init.d/httpd restart
>
> Please test and let us know if that resolves your question and how
> Caos NSA is working out for you.
>
> Warm regards,
>
> Greg
>