devil-linux-develop — developer discussions

Re: [Devil-linux-develop] [fmII] BusyBox 1.0.0-pre5 released (Default branch)

[Diego T. <dt...@co...>]

On Tue, Dec 23, 2003 at 01:55:53AM -0800, no...@fr... wrote:

> This email is to inform you about the release of version '1.0.0-pre5' of
> 'BusyBox' through freshmeat.net. All URLs and other useful information can

well... here it is, the long awaited pre5 release :)) why we don't wait to the final 1.0.0 ?


-- 
-- gnupg keyfingerprint -- 48AF 5BF9 8F54 2966 64CC  2327 7CD0 DD91 B09D 5799
-- Use of a keyboard or mouse may be linked to serious injuries or disorders.  
Diego Torres - dtorres at anthalia dot org - Madrid / España

Re: [Devil-linux-develop] [PATCH] fix subdir handling of jail script (update)

[Friedrich L. <fl...@fl...>]

Heiko Zuerker wrote on 23.12.2003 15:33 MET:
> Friedrich Lobenstock wrote:
> 
>> Friedrich Lobenstock wrote on 23.12.2003 04:58 MET:
>>
>>> Attached you'll find a patch to the jail script. It fixes the case
>>> when one has a subdirectory eg. /etc/named/sec/ in /etc/named/.
>>> It also preserves the ownership of the directory created.
>>>
>>> Please test and if it does not give you troubles please apply.
>>
>>
>>
>> Attached you will find the patch against the latest version of the
>> jail script.
>>
> I'm confused now, did my changes not work?
> Was there still trouble with the subdirectories?

I used a version before you fixed that. I did not burn the CD so
I can blame someone else for that :-)

BUT my patch should do it in a more general way even handling
links in subdirs. Yours does not ;-)

-- 
MfG / Regards
Friedrich Lobenstock
____________________________________________________________________
Friedrich Lobenstock                       Linux Services Lobenstock
URL: http://www.lsl.at/                         Email: fl...@fl...
____________________________________________________________________

Re: [Devil-linux-develop] Boot bug on latest 1.1.

[Emmanuel E. <ee...@fr...>]

hi,

Heiko Zuerker a =E9crit:
> Tim Tait wrote:
>=20
>> Bruce Smith wrote:
>>
>>> I just downloaded devil-linux-1.1.1-2003-12-21-i586-SMP.tar.bz2 (sinc=
e I
>>> didn't have a recent compile of my own at work), and installed it on =
a
>>> new box.
>>>
>>> It booted up, ask me to copy a blank config to media (I said "y"), it
>>> booted, I setup the box, ran save-config, and rebooted.  It came back
>>> up, and I noticed that I didn't have a root password, which was set. =
I
>>> also noticed my network wasn't setup, and NONE of my changes were the=
re.
>>> After repeating the process a couple times, I finally figured out tha=
t
>>> the boot process was creating and reading my config on a local hard
>>> drive (ext3 /dev/hda2), and save-config was saving to my USB stick on
>>> /dev/sda1.
>>>
>>> - BS
>>> =20
>>>
>> I confused, why doesn't it write back to the same place it found it on=
=20
>> boot? AIR, the location is saved into a environment variable. Can't=20
>> that be preserved somehow and used in save-config?
>=20
>=20
> The reason is easy: initrd is already updated to use devfs and=20
> save-config is old (out-dated).
> I still have to copy the specific part over to the save-config script.
>=20
> We could store the config location somewhere in the system and re-use i=
t=20
> for save-config, but we would loose the flexibility of switching config=
=20
> devices while the system is up.
> Any more thoughts/opinions about this ?

Yes, add a parameter in /etc/sysconfig/config for the save-config device=20
to use ...
>=20
> Heiko
>=20
>=20
>=20
>=20
> -------------------------------------------------------
> This SF.net email is sponsored by: IBM Linux Tutorials.
> Become an expert in LINUX or just sharpen your skills.  Sign up for IBM=
's
> Free Linux Tutorials.  Learn everything from the bash shell to sys admi=
n.
> Click now! http://ads.osdn.com/?ad_id=3D1278&alloc_id=3D3371&op=3Dclick
> _______________________________________________
> Devil-linux-develop mailing list
> Dev...@li...
> https://lists.sourceforge.net/lists/listinfo/devil-linux-develop
>=20
>=20

Re: [Devil-linux-develop] [PATCH] fix subdir handling of jail script (update)

[Heiko Z. <he...@zu...>]

Friedrich Lobenstock wrote:

> Friedrich Lobenstock wrote on 23.12.2003 04:58 MET:
>
>> Attached you'll find a patch to the jail script. It fixes the case
>> when one has a subdirectory eg. /etc/named/sec/ in /etc/named/.
>> It also preserves the ownership of the directory created.
>>
>> Please test and if it does not give you troubles please apply.
>
>
> Attached you will find the patch against the latest version of the
> jail script.
>
I'm confused now, did my changes not work?
Was there still trouble with the subdirectories?

Heiko

Re: [Devil-linux-develop] chroot to lfs build system failed - "kernel too old"?

[Heiko Z. <he...@zu...>]

Tim Tait wrote:

>
> I just finished setting up Woody 3r02 and it won't let me execute the 
> chroot to the lfssystem., it tells me "fatal: Kernel is too old". I 
> have 2.4.18 on the system... whats too old? The woody kernel? The lfs 
> system?

It can't really have anything todo with the chroot, except when the 
chroot support in your kernel is messed up.
But I don't expect that this is the case.

Heiko

Re: [Devil-linux-develop] Boot bug on latest 1.1.

[Heiko Z. <he...@zu...>]

Tim Tait wrote:

> Bruce Smith wrote:
>
>> I just downloaded devil-linux-1.1.1-2003-12-21-i586-SMP.tar.bz2 (since I
>> didn't have a recent compile of my own at work), and installed it on a
>> new box.
>>
>> It booted up, ask me to copy a blank config to media (I said "y"), it
>> booted, I setup the box, ran save-config, and rebooted.  It came back
>> up, and I noticed that I didn't have a root password, which was set. I
>> also noticed my network wasn't setup, and NONE of my changes were there.
>> After repeating the process a couple times, I finally figured out that
>> the boot process was creating and reading my config on a local hard
>> drive (ext3 /dev/hda2), and save-config was saving to my USB stick on
>> /dev/sda1.
>>
>> - BS
>>  
>>
> I confused, why doesn't it write back to the same place it found it on 
> boot? AIR, the location is saved into a environment variable. Can't 
> that be preserved somehow and used in save-config?

The reason is easy: initrd is already updated to use devfs and 
save-config is old (out-dated).
I still have to copy the specific part over to the save-config script.

We could store the config location somewhere in the system and re-use it 
for save-config, but we would loose the flexibility of switching config 
devices while the system is up.
Any more thoughts/opinions about this ?

Heiko

Re: [Devil-linux-develop] Handling patch files?

[Heiko Z. <he...@zu...>]

Martin Glazer wrote:

>>Here are the magic lines (you find it in the file prepare):
>>             if [ ! -f devil-linux-kernel-patches.done ]; then
>>                         touch devil-linux-kernel-patches.done
>>                         for PATCH in $(ls $WORKDIR/kernel-patches/* 2>
>>/dev/null )
>>                         do
>>                                     echo applying patch: $PATCH
>>                                     patch -p1 -i $PATCH || exit 1
>>                         done
>>             fi
>>
>>When the patches need to be installed in a specific order, just rename them
>>with a number in front:
>>01_first_patch
>>02_second_patch
>>and so on
>>
>>    
>>
>
>The problem I am having is not so much the order of the patches, but rather 
>that some of them have to be patched with p0 and some with p1. I suppose I 
>could always redo the exception patch file, but I'de rather not mess with it 
>as they are not mine.
>
>Is there any other elegant way of fixing a p0/p1 patch file?
>  
>
Put them in 2 different folders:
yourpatchset/p0
yourpatchset/p1

But that only works when the order doesn't matter, but you said this is 
your case anyway.

Heiko

Re: [Devil-linux-develop] Handling patch files?

[Heiko Z. <he...@zu...>]

Friedrich Lobenstock wrote:

> Martin Glazer wrote on 23.12.2003 08:37 MET:
>
>> On December 16, 2003 11:25 am, hzu...@ra... wrote:
>>
>>> Here are the magic lines (you find it in the file prepare):
>>>             if [ ! -f devil-linux-kernel-patches.done ]; then
>>>                         touch devil-linux-kernel-patches.done
>>>                         for PATCH in $(ls $WORKDIR/kernel-patches/* 2>
>>> /dev/null )
>>>                         do
>>>                                     echo applying patch: $PATCH
>>>                                     patch -p1 -i $PATCH || exit 1
>>>                         done
>>>             fi
>>>
>>> When the patches need to be installed in a specific order, just 
>>> rename them
>>> with a number in front:
>>> 01_first_patch
>>> 02_second_patch
>>> and so on
>>>
>>
>>
>> The problem I am having is not so much the order of the patches, but 
>> rather that some of them have to be patched with p0 and some with p1. 
>> I suppose I could always redo the exception patch file, but I'de 
>> rather not mess with it as they are not mine.
>>
>> Is there any other elegant way of fixing a p0/p1 patch file?
>
>
> Create subdirs p0, p1 and so one where you put the patches and then
> you know which patches should be applied which way. Just a thought.
>
I should first check if somebody already posted an answer.... ;-)

Heiko

Re: [Devil-linux-develop] Bug? - copy_man

[Heiko Z. <he...@zu...>]

Martin Glazer wrote:

>On December 17, 2003 06:40 am, hzu...@ra... wrote:
>  
>
>>You're right. Good job !
>>Can you send us a patch?
>>    
>>
>
>Didn't get a chance to look at this again until today, and in the latest 
>build, I didn't have the same problem, although this function hasn't changed.
>
>I checked a bit further and it looks like these directories should have been 
>created already in the create_cd_dirs script, called during prepare. As this 
>was the case before the latest udate, I don't know why it wasn't working the 
>first time round?
>
>Maybe just something to keep a lookout for in future?
>  
>
I actually forgott about this problem. ;-)
I  quickly  added (just now )  the "mkdir" to the script, just in case....

Heiko

Re: [Devil-linux-develop] chroot to lfs build system failed - "kernel too old"?

[Bruce S. <bw...@ar...>]

> I just finished setting up Woody 3r02 and it won't let me execute the 
> chroot to the lfssystem., it tells me "fatal: Kernel is too old". I have 
> 2.4.18 on the system... whats too old? The woody kernel? The lfs system?

I had the same problem after switching from Fedora to SuSE 9.0.

The problem went away when I installed a fresh copy of the lfssystem.
Try that (but save your build directory so you don't have to download
everything again).

 - BS

Re: [Devil-linux-develop] Handling patch files?

[Friedrich L. <fl...@fl...>]

Martin Glazer wrote on 23.12.2003 08:37 MET:
> On December 16, 2003 11:25 am, hzu...@ra... wrote:
>
>>Here are the magic lines (you find it in the file prepare):
>>             if [ ! -f devil-linux-kernel-patches.done ]; then
>>                         touch devil-linux-kernel-patches.done
>>                         for PATCH in $(ls $WORKDIR/kernel-patches/* 2>
>>/dev/null )
>>                         do
>>                                     echo applying patch: $PATCH
>>                                     patch -p1 -i $PATCH || exit 1
>>                         done
>>             fi
>>
>>When the patches need to be installed in a specific order, just rename them
>>with a number in front:
>>01_first_patch
>>02_second_patch
>>and so on
>>
> 
> 
> The problem I am having is not so much the order of the patches, but rather 
> that some of them have to be patched with p0 and some with p1. I suppose I 
> could always redo the exception patch file, but I'de rather not mess with it 
> as they are not mine.
> 
> Is there any other elegant way of fixing a p0/p1 patch file?

Create subdirs p0, p1 and so one where you put the patches and then
you know which patches should be applied which way. Just a thought.

-- 
MfG / Regards
Friedrich Lobenstock
____________________________________________________________________
Friedrich Lobenstock                       Linux Services Lobenstock
URL: http://www.lsl.at/                         Email: fl...@fl...
____________________________________________________________________

[Devil-linux-develop] [fmII] BusyBox 1.0.0-pre5 released (Default branch)

[<no...@fr...>]

This email is to inform you about the release of version '1.0.0-pre5' of
'BusyBox' through freshmeat.net. All URLs and other useful information can
be found at

    http://freshmeat.net/projects/busybox/

The changes in this release are as follows:
This release includes a number of random bugfixes. 

Project description:
BusyBox combines tiny versions of many common UNIX utilities into a
single small executable. It provides minimalist replacements for most
of the utilities you usually find in GNU fileutils, shellutils, etc.
The utilities in BusyBox generally have fewer options than their
full-featured GNU cousins; however, the options that are included
provide the expected functionality and behave very much like their
GNU counterparts. BusyBox provides a fairly complete POSIX
environment for any small or embedded system. 

Trove categories:
[Development Status  ] 5 - Production/Stable
[Environment         ] Console (Text Based)
[Intended Audience   ] End Users/Desktop
[License             ] OSI Approved :: GNU General Public License (GPL)
[Operating System    ] POSIX :: Linux
[Programming Language] C
[Topic               ] Software Development :: Embedded Systems, System ::
Boot :: Init 

If you would like to cancel subscription to releases of this project,
login to freshmeat.net and choose 'home' from the personal menubar at the
top of the page. You'll be presented with a list of projects you're
subscribed to in the right column, which you may cancel by highlighting
the project in question and clicking the 'delete' button.

Sincerely,
freshmeat.net 

Re: [Devil-linux-develop] Handling patch files?

[Martin G. <sou...@gl...>]

On December 16, 2003 11:25 am, hzu...@ra... wrote:
> Hey,
>
> I would suggest you do it similar to wait I do with the kernel patches,
> since this is better when you have to apply multiple patches.
> I have one tarball kernel-patches.tar.bz2 which I keep up to date.
>

That will make things easier...

> Here are the magic lines (you find it in the file prepare):
>              if [ ! -f devil-linux-kernel-patches.done ]; then
>                          touch devil-linux-kernel-patches.done
>                          for PATCH in $(ls $WORKDIR/kernel-patches/* 2>
> /dev/null )
>                          do
>                                      echo applying patch: $PATCH
>                                      patch -p1 -i $PATCH || exit 1
>                          done
>              fi
>
> When the patches need to be installed in a specific order, just rename them
> with a number in front:
> 01_first_patch
> 02_second_patch
> and so on
>

The problem I am having is not so much the order of the patches, but rather 
that some of them have to be patched with p0 and some with p1. I suppose I 
could always redo the exception patch file, but I'de rather not mess with it 
as they are not mine.

Is there any other elegant way of fixing a p0/p1 patch file?

Martin

Re: [Devil-linux-develop] Bug? - copy_man

[Martin G. <sou...@gl...>]

On December 17, 2003 06:40 am, hzu...@ra... wrote:
> You're right. Good job !
> Can you send us a patch?

Didn't get a chance to look at this again until today, and in the latest 
build, I didn't have the same problem, although this function hasn't changed.

I checked a bit further and it looks like these directories should have been 
created already in the create_cd_dirs script, called during prepare. As this 
was the case before the latest udate, I don't know why it wasn't working the 
first time round?

Maybe just something to keep a lookout for in future?

Martin

> On 12/17/2003 04:24:13 AM Martin Glazer wrote:
> >Hey all
> >
> >Just came across something strange with respect to the man pages
> >created on
> >the cdtree - in /usr/share/man
> >
> >My /usr/share/man/man5 is not a directory, but rather an actual file.
> >Happens
> >to be the syslog-ng.conf man page.
> >
> >I think there is a bug in the copy_man function in build/functions
> >
> >copy_man ()
> >{
> >BD=$1
> >[ -z "$BD" ] && BD=$(pwd)
> >for I in 1 2 3 4 5 6 7 8 9
> >do
> >cp -dp $BD/*.$I $CDDIR/usr/share/man/man$I 2> /dev/null
> >cp -dp $BD/*.$I.gz $CDDIR/usr/share/man/man$I 2> /dev/null
> >cp -dp $BD/*.$I.bz2 $CDDIR/usr/share/man/man$I 2> /dev/null
> >done
> >return 0
> >}
> >
> >Correct me if I'm wrong, but if the manX directory does not yet exist
> >when
> >this script is run, then instead of copying the file into the manX
> >directory,
> >it will actually copy the file into a file called manX and hence the
> >problem
> >I am seeing.
> >
> >This causes a few problems later in the build/install process, but
> >nothing
> >serious enough to warrent an exit error.
> >
> >If I am correct, then I would suggest one workaround would be to create
> >the
> >actual man directories before doing the cp - ie add the following line
> >after
> >the do
> >mkdir -p $CDDIR/usr/share/man/man$I 2>/dev/null
> >
> >thoughts?

[Devil-linux-develop] chroot to lfs build system failed - "kernel too old"?

[Tim T. <t....@co...>]

I just finished setting up Woody 3r02 and it won't let me execute the 
chroot to the lfssystem., it tells me "fatal: Kernel is too old". I have 
2.4.18 on the system... whats too old? The woody kernel? The lfs system?

Tim

Re: [Devil-linux-develop] Boot bug on latest 1.1.

[Tim T. <t....@co...>]

Bruce Smith wrote:

>I just downloaded devil-linux-1.1.1-2003-12-21-i586-SMP.tar.bz2 (since I
>didn't have a recent compile of my own at work), and installed it on a
>new box.
>
>It booted up, ask me to copy a blank config to media (I said "y"), it
>booted, I setup the box, ran save-config, and rebooted.  It came back
>up, and I noticed that I didn't have a root password, which was set. I
>also noticed my network wasn't setup, and NONE of my changes were there.
>After repeating the process a couple times, I finally figured out that
>the boot process was creating and reading my config on a local hard
>drive (ext3 /dev/hda2), and save-config was saving to my USB stick on
>/dev/sda1.
>
> - BS
>  
>
I confused, why doesn't it write back to the same place it found it on 
boot? AIR, the location is saved into a environment variable. Can't that 
be preserved somehow and used in save-config?

Tim

Re: [Devil-linux-develop] [PATCH] fix subdir handling of jail script (update)

[Friedrich L. <fl...@fl...>]

Friedrich Lobenstock wrote on 23.12.2003 04:58 MET:
> Attached you'll find a patch to the jail script. It fixes the case
> when one has a subdirectory eg. /etc/named/sec/ in /etc/named/.
> It also preserves the ownership of the directory created.
> 
> Please test and if it does not give you troubles please apply.

Attached you will find the patch against the latest version of the
jail script.

-- 
MfG / Regards
Friedrich Lobenstock
____________________________________________________________________
Friedrich Lobenstock                       Linux Services Lobenstock
URL: http://www.lsl.at/                         Email: fl...@fl...
____________________________________________________________________

[Devil-linux-develop] [PATCH] fix subdir handling of jail script

[Friedrich L. <fl...@fl...>]

Hi!

Attached you'll find a patch to the jail script. It fixes the case
when one has a subdirectory eg. /etc/named/sec/ in /etc/named/.
It also preserves the ownership of the directory created.

Please test and if it does not give you troubles please apply.

-- 
MfG / Regards
Friedrich Lobenstock
____________________________________________________________________
Friedrich Lobenstock                       Linux Services Lobenstock
URL: http://www.lsl.at/                         Email: fl...@fl...
____________________________________________________________________

[Devil-linux-develop] [ devil-linux-Bugs-864025 ] snort init interface support

[SourceForge.net <no...@so...>]

Bugs item #864025, was opened at 2003-12-21 12:26
Message generated for change (Settings changed) made by smiley73
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=410643&aid=864025&group_id=34096

Category: Configuration / Scripts
>Group: v1.1
>Status: Closed
>Resolution: Fixed
Priority: 5
Submitted By: Heiko Zuerker (smiley73)
>Assigned to: Heiko Zuerker (smiley73)
Summary: snort init interface support

Initial Comment:
currently the snort init script doesn't specify on
which interface it should run.
add functionality to specify multiple interfaces. only
one interface per snort daemon can be specified, so
i.e. /etc/snort/snort-<interface>.conf as config file
would be a good default.

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

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=410643&aid=864025&group_id=34096

Re: [Devil-linux-develop] nice

[Bruce S. <bw...@ar...>]

> > >as always, we could made that a configuration option!
> > > 
> > It is probably a little bit slower, but I don't think that you will 
> > "feel" it.
> > It simply tells the system that other processes have a higher priority 
> > and so your system is more usable during a compile.
> 
> yes, i know. i actually was asking for the nice "value" (0..19 :)

It's not worth the hassle, unless you specifically set priorities on
other processes you run.

 - BS

Re: [Devil-linux-develop] nice

[Diego T. <dt...@co...>]

On Mon, Dec 22, 2003 at 08:22:55PM -0500, Heiko Zuerker wrote:
> >as always, we could made that a configuration option!
> > 
> It is probably a little bit slower, but I don't think that you will 
> "feel" it.
> It simply tells the system that other processes have a higher priority 
> and so your system is more usable during a compile.


yes, i know. i actually was asking for the nice "value" (0..19 :)

-- 
-- gnupg keyfingerprint -- 48AF 5BF9 8F54 2966 64CC  2327 7CD0 DD91 B09D 5799
-- Use of a keyboard or mouse may be linked to serious injuries or disorders.  
Diego Torres - dtorres at anthalia dot org - Madrid / España

Re: [Devil-linux-develop] nice

[Heiko Z. <he...@zu...>]

Diego Torres wrote:

>On Mon, Dec 22, 2003 at 04:13:49PM -0500, Bruce Smith wrote:
>  
>
>>>I was thinking if we should run all scripts in the build environment 
>>>with "nice".
>>>Any objections ?
>>>      
>>>
>>Sounds nice to me.  :-)
>>    
>>
>
>will it go slower or faster? :)
>
>as always, we could made that a configuration option!
>  
>
It is probably a little bit slower, but I don't think that you will 
"feel" it.
It simply tells the system that other processes have a higher priority 
and so your system is more usable during a compile.

Heiko

Re: [Devil-linux-develop] nice

[Diego T. <dt...@co...>]

On Mon, Dec 22, 2003 at 04:13:49PM -0500, Bruce Smith wrote:
> > I was thinking if we should run all scripts in the build environment 
> > with "nice".
> > Any objections ?
> 
> Sounds nice to me.  :-)

will it go slower or faster? :)

as always, we could made that a configuration option!
-- 
-- gnupg keyfingerprint -- 48AF 5BF9 8F54 2966 64CC  2327 7CD0 DD91 B09D 5799
-- Use of a keyboard or mouse may be linked to serious injuries or disorders.  
Diego Torres - dtorres at anthalia dot org - Madrid / España

Re: [Devil-linux-develop] [ devil-linux-Feature Requests-864429 ] add coda filesystem

[Diego T. <dt...@co...>]

On Mon, Dec 22, 2003 at 02:14:46PM -0500, Heiko Zuerker wrote:

sorry for the other mail (the one with the bad address :=) it was my
spam-fighting account, and i forgot to change it back to the "usual" :)

> >and nfs-client/server on userland?
> > 
> >
> Do you know what NFS stands for? No File Security  ;-)
> 
> Since it's such an unsecure protocol, I hesitated so far about including 
> it.
> But as always I'm open for a discussion about it.

one day i needed a network filesystem, and nfs came to my hand :) of course it isn't
secure (i use it with ssh over the network) but without doubt is more standard
for talking with other unix boxes that samba.

good point: it has a userland implementation. if someone breaks it, he/she won't
trash the full box, only a daemon.

bad: we already have coda :) (and i've never used it :)

so: i don't know, i could use coda on my fileserver instead of nfs :) i didn't
want to star a flame, only ask about opinions :)))

-- 
-- gnupg keyfingerprint -- 48AF 5BF9 8F54 2966 64CC  2327 7CD0 DD91 B09D 5799
-- Use of a keyboard or mouse may be linked to serious injuries or disorders.  
Diego Torres - dtorres at anthalia dot org - Madrid / España

Re: [Devil-linux-develop] Boot bug on latest 1.1.

[Heiko Z. <he...@zu...>]

Bruce Smith wrote:

>>>It booted up, ask me to copy a blank config to media (I said "y"), it
>>>booted, I setup the box, ran save-config, and rebooted.  It came back
>>>up, and I noticed that I didn't have a root password, which was set. I
>>>also noticed my network wasn't setup, and NONE of my changes were there.
>>>After repeating the process a couple times, I finally figured out that
>>>the boot process was creating and reading my config on a local hard
>>>drive (ext3 /dev/hda2), and save-config was saving to my USB stick on
>>>/dev/sda1.
>>> 
>>>
>>>      
>>>
>>I would say this is not a bug.
>>
>>The script checks the root directories of all "harddisks" and cdroms for 
>>etc.tar.bz2.
>>save-config is not yet modified to use the devfs, so the search doesn't 
>>include IDE "harddisks" yet.
>>
>>"harddisks" means any devices which are or emulate harddisks (usb 
>>sticks, cf cards)
>>    
>>
>
>Shouldn't it search USB (scsi) before IDE?
>  
>
Yeah why not.
You can change it if you want.

Heiko