devil-linux-develop — developer discussions

Re: [Devil-linux-develop] Kernel abort.

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

> > mrproper, new LFS.
> > Difference: I switched my config from i686 to i585 and turned off SMP.
> > And I got the following:
> 
> > /data/build/tmp/linux-2.4.25/include/net/pkt_sched.h:292:2: #error PSCHED_CLOCK_SOURCE=PSCHED_CPU is not supported on this arch.
> 
> You have to run 'make menuconfig' to change CPU or set
> CONFIG_PSCHED_CLOCK_SOURCE=PSCHED_JIFFIES in
> the devil .config file.

I'm pretty sure that PSCHED_JIFFIES was set in the .config file.
It worked a couple days ago on a i686-SMP compile.  All I did was
switch menuconfig to a i586-UP.

The error message kind of implies that PSCHED_JIFFIES isn't supported by
i586?  I just set it to PSCHED_GETTIMEOFDAY and restarted the compile.
We'll see if that works, in a few hours...

 - BS

Re: [Devil-linux-develop] Kernel abort.

[Jaroslav K. <pe...@pe...>]

On Mon, 5 Apr 2004, Bruce Smith wrote:

> mrproper, new LFS.
> Difference: I switched my config from i686 to i585 and turned off SMP.
> And I got the following:

> /data/build/tmp/linux-2.4.25/include/net/pkt_sched.h:292:2: #error PSCHED_CLOCK_SOURCE=PSCHED_CPU is not supported on this arch.

You have to run 'make menuconfig' to change CPU or set
CONFIG_PSCHED_CLOCK_SOURCE=PSCHED_JIFFIES in
the devil .config file.

-----
Jaroslav Kysela <pe...@pe...>

Re: [Devil-linux-develop] Kernel abort.

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

Same problem as last time?
Is CONFIG_PSCHED_CLOCK_SOURCE defined in build/.config?

Could be a bug in the menuconfig stuff.

Heiko

> mrproper, new LFS.
> Difference: I switched my config from i686 to i585 and turned off SMP.
> And I got the following:
> --------------------------------------------------------------------------------
>
> executing linux with option build (in /data/build/tmp/linux-2.4.25
> /data/build/tmp/linux-2.6.3)
>
> # tail of the log file:
>
> gcc -D__KERNEL__ -I/data/build/tmp/linux-2.4.25/include -Wall
> -Wstrict-prototypes -Wno-trigraphs -O2 -fno-strict-aliasing -fno-common
> -fomit-frame-pointer  -pipe -mpreferred-stack-boundary=2 -march=i586
> -nostdinc -iwithprefix include -DKBUILD_BASENAME=dev  -c -o dev.o dev.c
> In file included from dev.c:96:
> /data/build/tmp/linux-2.4.25/include/net/pkt_sched.h:292:2: #error
> PSCHED_CLOCK_SOURCE=PSCHED_CPU is not supported on this arch.
> make[4]: *** [dev.o] Error 1
> make[4]: Leaving directory `/data/build/tmp/linux-2.4.25/net/core'
> make[3]: *** [first_rule] Error 2
> make[3]: Leaving directory `/data/build/tmp/linux-2.4.25/net/core'
> make[2]: *** [_subdir_core] Error 2
> make[2]: Leaving directory `/data/build/tmp/linux-2.4.25/net'
> make[1]: *** [_dir_net] Error 2
> make[1]: Leaving directory `/data/build/tmp/linux-2.4.25'
>
>
>
>
> -------------------------------------------------------
> This SF.Net email is sponsored by: IBM Linux Tutorials
> Free Linux tutorial presented by Daniel Robbins, President and CEO of
> GenToo technologies. Learn everything from fundamentals to system
> administration.http://ads.osdn.com/?ad_id=1470&alloc_id=3638&op=click
> _______________________________________________
> Devil-linux-develop mailing list
> Dev...@li...
> https://lists.sourceforge.net/lists/listinfo/devil-linux-develop
>
>


-- 
Regards
  Heiko Zuerker
  http://www.devil-linux.org

[Devil-linux-develop] Kernel abort.

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

mrproper, new LFS.
Difference: I switched my config from i686 to i585 and turned off SMP.
And I got the following:
--------------------------------------------------------------------------------

executing linux with option build (in /data/build/tmp/linux-2.4.25 /data/build/tmp/linux-2.6.3)

# tail of the log file:

gcc -D__KERNEL__ -I/data/build/tmp/linux-2.4.25/include -Wall -Wstrict-prototypes -Wno-trigraphs -O2 -fno-strict-aliasing -fno-common -fomit-frame-pointer  -pipe -mpreferred-stack-boundary=2 -march=i586   -nostdinc -iwithprefix include -DKBUILD_BASENAME=dev  -c -o dev.o dev.c
In file included from dev.c:96:
/data/build/tmp/linux-2.4.25/include/net/pkt_sched.h:292:2: #error PSCHED_CLOCK_SOURCE=PSCHED_CPU is not supported on this arch.
make[4]: *** [dev.o] Error 1
make[4]: Leaving directory `/data/build/tmp/linux-2.4.25/net/core'
make[3]: *** [first_rule] Error 2
make[3]: Leaving directory `/data/build/tmp/linux-2.4.25/net/core'
make[2]: *** [_subdir_core] Error 2
make[2]: Leaving directory `/data/build/tmp/linux-2.4.25/net'
make[1]: *** [_dir_net] Error 2
make[1]: Leaving directory `/data/build/tmp/linux-2.4.25'

[Devil-linux-develop] [documentation] suggestion on how to start the build system

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

Hi!

I would like to suggest that we document the following command on how to 
start the DL build system:

# alias devil
alias devil='cp /etc/{resolv.conf,hosts} ~/devil-linux/lfssystem/etc/; 
chroot /root/devil-linux/lfssystem /usr/bin/env -i HOME=/root TERM=$TERM 
/bin/bash --login'

Because we copy now important network settings to the build system we 
really only need the chroot command to be available in the host system. 
Everything else can then be done from the chrooted build system.

PS: In the documentation I would merge section 1.5 with 2.

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

Re: [Devil-linux-develop] small patch

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

Roland Pabel wrote:
> Hi,
> here is a small patch which polishes up apcupsd a  bit and fixes save-config 
> when an empty medium is inserted (/cdrom/config was changed to /config)
> I'm almost done with mailman, quite a beast. I'm not yet sure if it is 
> possible to chroot mailman easily, would need apache inside and config be 
> copied out each time something is changed...

It's added.

Heiko

[Devil-linux-develop] small patch

[Roland P. <pa...@ta...>]

Hi,
here is a small patch which polishes up apcupsd a  bit and fixes save-config 
when an empty medium is inserted (/cdrom/config was changed to /config)
I'm almost done with mailman, quite a beast. I'm not yet sure if it is 
possible to chroot mailman easily, would need apache inside and config be 
copied out each time something is changed...
Roland
-- 
ICQ UIN 49339118	Linux Counter #88774
GPG-Key 1024D/59C6AFA6 2003-02-07 Roland Pabel <ro...@pa...>

[Devil-linux-develop] [ devil-linux-Feature Requests-873874 ] add sudo

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

Feature Requests item #873874, was opened at 2004-01-09 11:33
Message generated for change (Settings changed) made by blubdog
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=410646&aid=873874&group_id=34096

Category: None
Group: None
>Status: Closed
Priority: 5
Submitted By: Bruce Smith (blubdog)
>Assigned to: Bruce Smith (blubdog)
Summary: add sudo

Initial Comment:
sudo provides very flexible administrative delegation
to a chain of administrators.

http://www.courtesan.com/sudo/sudo.html


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

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=410646&aid=873874&group_id=34096

Re: [Devil-linux-develop] thoughs about additional options to save-config (followup)

[<hzu...@ra...>]

On 03/26/2004 07:09:59 AM Arnaud Gomes-do-Vale wrote:
>Heiko Zuerker <he...@zu...> writes:
>
>>> Background: This way one could run
>>>   ssh root@firewall "save-config -dump" > etc.tar.bz2
>>> on a remote host.
>>
>> Hmmmm, like that. Of course one would have to have root login enabled,
>> which I usually don't allow.
>
>That's pretty much what I have been doing for some time with a
>modified version of the save-config script. Now if we could just use
>tftp to load the config... :-)

Is there really interest loading the config over the network?

I was thinking about that feature for a while, but this is only interesting
for LANs, not for DMZ.
I was also more thinking about using wget from a https server, or some
other way to ensure a secure communication.

Heiko

Re: [Devil-linux-develop] thoughs about additional options to save-config (followup)

[Arnaud Gomes-do-V. <Arn...@ir...>]

Heiko Zuerker <he...@zu...> writes:

>> Background: This way one could run
>>   ssh root@firewall "save-config -dump" > etc.tar.bz2
>> on a remote host.
>
> Hmmmm, like that. Of course one would have to have root login enabled,
> which I usually don't allow.

That's pretty much what I have been doing for some time with a
modified version of the save-config script. Now if we could just use
tftp to load the config... :-)

-- 
Arnaud

Re: [Devil-linux-develop] DL and the eagle-usb driver

[Philippe Midol-M. <pmi...@so...>]

Friedrich Lobenstock wrote:
> Philippe Midol-Monnet wrote on 25.03.2004 17:07 MET:
> 
>>
>> The new attached script will solve the problem.
>>
>> I make it very quickly, and I have just enough time to test the "make 
>> dist"
>>
> 
> If you find time please check if you can run configure with an option to 
> tell it where to find and install the dsp_code*.bin files.
> 
That was what I have planned to do this weekend. I can't do that from my 
office (the proxy is totally paranoiac). ;-)

-- 
Philippe Midol-Monnet
pmi...@so...

Re: [Devil-linux-develop] thoughs about additional options to save-config (followup)

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

Heiko Zuerker wrote on 26.03.2004 02:04 MET:
> Friedrich Lobenstock wrote:
> 
>> Hi!
>>
>> Friedrich Lobenstock wrote on 26.03.2004 01:49 MET:
>>
>>>
>>> *The following applies only if one uses eg. a USB stick to save the 
>>> config*
>>>
>>
>> Of course there's something else that came to my mind right after 
>> hitting the send button. So here it goes:
>>
>> 5) option "-really-quit" .... save config with no output at all
> 
> -really-quiet

Typo, of course.

> Do we really need this?

See 6), any text output will mess it up there.


>> 6) option "-dump":   ... implies -really-quit
>>
>> When one runs "save-config -dump" the config will not be saved to the 
>> configuration media but dumped to stdout.
>>
>> Background: This way one could run
>>   ssh root@firewall "save-config -dump" > etc.tar.bz2
>> on a remote host.
> 
> 
> Hmmmm, like that. Of course one would have to have root login enabled, 
> which I usually don't allow.

You got a point.

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

Re: [Devil-linux-develop] thoughs about additional options to save-config

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

Heiko Zuerker wrote on 26.03.2004 02:02 MET:
> Friedrich Lobenstock wrote:
> 
>> 4) /etc/sysconfig/save-config with the following content:
>>
>> #
>> # This config option, only applicable when used with an USB configuration
>> # media, allows one to keep more than one copy of the configuration. This
>> # might be of help when the configuration media goes bad so there is an
>> # increased chance to recover the configuration from one of the 
>> additional
>> # ones.
>> #
>> NR_COPIES="3"
> 
> 
> Nice feature, we should keep the default to 0, which means no backup.

Excatelly.

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

Re: [Devil-linux-develop] thoughs about additional options to save-config (followup)

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

Friedrich Lobenstock wrote:
> Hi!
> 
> Friedrich Lobenstock wrote on 26.03.2004 01:49 MET:
> 
>>
>> *The following applies only if one uses eg. a USB stick to save the 
>> config*
>>
> 
> Of course there's something else that came to my mind right after 
> hitting the send button. So here it goes:
> 
> 5) option "-really-quit" .... save config with no output at all
-really-quiet

Do we really need this?

> 6) option "-dump":   ... implies -really-quit
> 
> When one runs "save-config -dump" the config will not be saved to the 
> configuration media but dumped to stdout.
> 
> Background: This way one could run
>   ssh root@firewall "save-config -dump" > etc.tar.bz2
> on a remote host.

Hmmmm, like that. Of course one would have to have root login enabled, 
which I usually don't allow.

Heiko

Re: [Devil-linux-develop] thoughs about additional options to save-config

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

Friedrich Lobenstock wrote:
> Hi!
> 
> *The following applies only if one uses eg. a USB stick to save the config*
> 
> 1) "-h" .... print some help text and all possible command line options

Good point

> 2) "-b":
> 
> When one runs "save-config -b" the old config will be backed upt to 
> etc-backup.tar.bz2

Good idea.

> 3) option "-t timeout":
> 
> When one runs "save-config -t 10" the config will be saved as 
> etc-tryout.tar.bz2 instead of etc.tar.bz2 and the timeout value 10 will 
> be written to a file named "try-out".
> 
> At the next boot the system will detect both files from above, will load 
> the etc-tryout.tar.bz2 and continue starting the system. Before the 
> first prompt is displayed a "shutdown -r XX &", where XX is the value 
> given via -t to save-config.
> 
> Background: This way one could try out invasive changes to the config 
> even from remote without the risk of loosing the old config in case the 
> new config is crap.

Should be easy to implement, why not.

> 4) /etc/sysconfig/save-config with the following content:
> 
> #
> # This config option, only applicable when used with an USB configuration
> # media, allows one to keep more than one copy of the configuration. This
> # might be of help when the configuration media goes bad so there is an
> # increased chance to recover the configuration from one of the additional
> # ones.
> #
> NR_COPIES="3"

Nice feature, we should keep the default to 0, which means no backup.


> So that's it for now. What do you think?

Go ahead and implement it.

Heiko

[Devil-linux-develop] thoughs about additional options to save-config (followup)

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

Hi!

Friedrich Lobenstock wrote on 26.03.2004 01:49 MET:
> 
> *The following applies only if one uses eg. a USB stick to save the config*
> 

Of course there's something else that came to my mind right after hitting 
the send button. So here it goes:

5) option "-really-quit" .... save config with no output at all

6) option "-dump":   ... implies -really-quit

When one runs "save-config -dump" the config will not be saved to the 
configuration media but dumped to stdout.

Background: This way one could run
   ssh root@firewall "save-config -dump" > etc.tar.bz2
on a remote host.

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

[Devil-linux-develop] [ devil-linux-Feature Requests-923584 ] add

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

add IP Tables network magic SysRq function
Mime-Version: 1.0
Content-Type: text/plain; charset="ISO-8859-1"

Feature Requests item #923584, was opened at 2004-03-25 19:55
Message generated for change (Tracker Item Submitted) made by Item Submitter
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=410646&aid=923584&group_id=34096

Category: Packages
Group: v1.1.x
Status: Open
Priority: 5
Submitted By: Heiko Zuerker (smiley73)
Assigned to: Nobody/Anonymous (nobody)
Summary: add 	 
	
add IP Tables network magic SysRq function

Initial Comment:
ipt_sysrq is a new iptables target that allows you to
do the same as the magic sysrq key on a keyboard does,
but over the network.

http://terminus.sk/~marek/prog/ipt_sysrq.shtml

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

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=410646&aid=923584&group_id=34096

[Devil-linux-develop] thoughs about additional options to save-config

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

Hi!

*The following applies only if one uses eg. a USB stick to save the config*

1) "-h" .... print some help text and all possible command line options


2) "-b":

When one runs "save-config -b" the old config will be backed upt to 
etc-backup.tar.bz2


3) option "-t timeout":

When one runs "save-config -t 10" the config will be saved as 
etc-tryout.tar.bz2 instead of etc.tar.bz2 and the timeout value 10 will be 
written to a file named "try-out".

At the next boot the system will detect both files from above, will load 
the etc-tryout.tar.bz2 and continue starting the system. Before the first 
prompt is displayed a "shutdown -r XX &", where XX is the value given via 
-t to save-config.

Background: This way one could try out invasive changes to the config even 
from remote without the risk of loosing the old config in case the new 
config is crap.


4) /etc/sysconfig/save-config with the following content:

#
# This config option, only applicable when used with an USB configuration
# media, allows one to keep more than one copy of the configuration. This
# might be of help when the configuration media goes bad so there is an
# increased chance to recover the configuration from one of the additional
# ones.
#
NR_COPIES="3"


So that's it for now. What do you think?

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

Re: [Devil-linux-develop] Abort.

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

> > > >/data/build/tmp/linux-2.4.25/include/net/pkt_sched.h:214:25: #if with
> > > >no expression
> > > 
> > > Since it's the packet scheduler, I guess the problem is the new sed command
> > > from Jaroslav
> > 
> > Is variable CONFIG_PSCHED_CLOCK_SOURCE defined in your /build/.config 
> > file? 
> 
> Nope.
> 
> > You have to rerun 'make menuconfig' to create it.
> 
> OK, I'll try again.  Thanks!

And it works now.  Thanks again!

 - BS

Re: [Devil-linux-develop] DL and the eagle-usb driver

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

Philippe Midol-Monnet wrote on 25.03.2004 17:07 MET:
> 
> The new attached script will solve the problem.
> 
> I make it very quickly, and I have just enough time to test the "make dist"
>

If you find time please check if you can run configure with an option to 
tell it where to find and install the dsp_code*.bin files.

Otherwise you might want to contact the original author so he adds an 
option to configure and the Makefile.

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

Re: [Devil-linux-develop] DL and the eagle-usb driver

[Philippe Midol-M. <pmi...@so...>]

hzu...@ra... wrote:
> On 03/22/2004 02:06:51 AM Philippe Midol-Monnet wrote:
> 
>>Heiko Zuerker wrote:
>>
>>
>>
>>>You forgott to add the eagle-usb.help file, I took care of this.
>>>
>>
>>OK
>>
>>
>>>There are 2 huge files in /etc/eagle-usb/dsp_code*.bin , can we move
>>>them to /usr/share/eagle-usb and just symlink them ?
>>>
>>
>>I will build the 1.1.x version and test this modification (it will take
>>some time ;-)
> 
> 
> This change get's quite important now, since the etc.tar.bz2 doesn't fit on
> a floppy disk anymore.
> 

The new attached script will solve the problem.

I make it very quickly, and I have just enough time to test the "make dist"


-- 
Philippe Midol-Monnet
pmi...@so...

Re: [Devil-linux-develop] DL and the eagle-usb driver

[<hzu...@ra...>]

On 03/22/2004 02:06:51 AM Philippe Midol-Monnet wrote:
>Heiko Zuerker wrote:
>
>
>>
>> You forgott to add the eagle-usb.help file, I took care of this.
>>
>OK
>
>> There are 2 huge files in /etc/eagle-usb/dsp_code*.bin , can we move
>> them to /usr/share/eagle-usb and just symlink them ?
>>
>I will build the 1.1.x version and test this modification (it will take
>some time ;-)

This change get's quite important now, since the etc.tar.bz2 doesn't fit on
a floppy disk anymore.

Heiko

Re: [Devil-linux-develop] Re: Feedback about "IP Tables network magic SysRq function"

[<hzu...@ra...>]

On 03/25/2004 04:30:24 AM Marek Zelem wrote:
>On Wed, 24 Mar 2004, Heiko Zuerker wrote:
>
>> Marek Zelem wrote:
>> >  Hello
>> >
>> >   Thank you for your suggestions.
>> >
>> >   I make severel enhancements. Version 0.3 is now available. There
>are no
>> > longer prime numbers needed and no password is compiled into module.
>> > I used SHA1 hash algorithm as you suggest. And finaly I improve
>security
>> > by adding timestamps. Everithing is described at
>> > http://terminus.sk/~marek/prog/ipt_sysrq.shtml and in README file.
>>
>> Sounds good to me. :-))
>>
>> Will different timezones cause problems? If you not already do so, I
>> would suggest that you always use UTC.
>
>I used standard time(2) system call, so timestamps are in UTC.

Cool.

>> Anybody have any more concerns about including this tool?
>
>I only know that is included at linuxlinks.com.

We're actually discussing about including it into Devil-Linux.

Heiko

[Devil-linux-develop] Re: Feedback about "IP Tables network magic SysRq function"

[Marek Z. <ma...@te...>]

On Wed, 24 Mar 2004, Heiko Zuerker wrote:

> Marek Zelem wrote:
> > 	Hello
> >
> >   Thank you for your suggestions.
> >
> >   I make severel enhancements. Version 0.3 is now available. There are no
> > longer prime numbers needed and no password is compiled into module.
> > I used SHA1 hash algorithm as you suggest. And finaly I improve security
> > by adding timestamps. Everithing is described at
> > http://terminus.sk/~marek/prog/ipt_sysrq.shtml and in README file.
>
> Sounds good to me. :-))
>
> Will different timezones cause problems? If you not already do so, I
> would suggest that you always use UTC.

I used standard time(2) system call, so timestamps are in UTC.

>
> Anybody have any more concerns about including this tool?

I only know that is included at linuxlinks.com.

		Marek

[Devil-linux-develop] Re: Feedback about "IP Tables network magic SysRq function"

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

Marek Zelem wrote:
> 	Hello
> 
>   Thank you for your suggestions.
> 
>   I make severel enhancements. Version 0.3 is now available. There are no
> longer prime numbers needed and no password is compiled into module.
> I used SHA1 hash algorithm as you suggest. And finaly I improve security
> by adding timestamps. Everithing is described at
> http://terminus.sk/~marek/prog/ipt_sysrq.shtml and in README file.

Sounds good to me. :-))

Will different timezones cause problems? If you not already do so, I 
would suggest that you always use UTC.

Anybody have any more concerns about including this tool?

Heiko