devil-linux-develop — developer discussions

Re: [Devil-linux-develop] ddt didnt compile

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

On Sat, Jan 31, 2004 at 10:20:57AM -0500, Heiko Zuerker wrote:
> 
> Yes, the bug is already reported on SF.
> 
> It looks to me, that ddt isn't maintained anymore. I wanted to wait 
> until we get closer to the next major DL release and nuke it, in case 
> there's no update available.

it's a pitty, but since ddt servers where hackedm nothing has been the same.

i vote for nuking it.

-- 
-- 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 options

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

Roland Pabel wrote on 01.02.2004 23:47 MET:
>>It looks like, the anonymous CVS is even older...
>>You patch doesn't apply, I added you the current build-iso file, please
>>create the patch against this one.
> 
> done,

Can this possibly be added to the "custom-cd" script so everybody
can have this option without need to recompile the whole thing?

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

Re: [Devil-linux-develop] boot options

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

> It looks like, the anonymous CVS is even older...
> You patch doesn't apply, I added you the current build-iso file, please
> create the patch against this one.
done,
Roland
-- 
ICQ UIN 49339118	Linux Counter #88774
GPG-Key 1024D/59C6AFA6 2003-02-07 Roland Pabel <ro...@pa...>

Re: [Devil-linux-develop] boot options

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

Friedrich Lobenstock wrote:
> Roland Pabel wrote on 01.02.2004 22:11 MET:
> 
>> Checked out CVS minutes ago, but the ramdisk part was still not fixed 
>> (cvs problems?!)
> 
> 
> The public CVS is as it seems currently at least some days behind the
> project CVS server.
> 
> See http://sourceforge.net/docman/display_doc.php?docid=2352&group_id=1
> 

It looks like, the anonymous CVS is even older...
You patch doesn't apply, I added you the current build-iso file, please 
create the patch against this one.

Heiko

Re: [Devil-linux-develop] boot options

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

Roland Pabel wrote on 01.02.2004 22:11 MET:
> Checked out CVS minutes ago, but the ramdisk part was still not fixed (cvs 
> problems?!)

The public CVS is as it seems currently at least some days behind the
project CVS server.

See http://sourceforge.net/docman/display_doc.php?docid=2352&group_id=1

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

Re: [Devil-linux-develop] boot options

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

> Why don't you just make it a listsel ?
> 0 text mode 80x25
> 1 640x480 VESA
> 2 800x600 VESA
> and so on.
>
> In the build-iso file, then just "cut" it.
>
> This "should" work (of course no promise ;-) )
Ok, here is the patch. works for me ( no promise ;-) )
Checked out CVS minutes ago, but the ramdisk part was still not fixed (cvs 
problems?!)
According to linux/Documentation/ramdisk.txt "ramdisk=<size>" is obsolete, so 
I changed it to "ramdisk_size=<size>"
Roland
-- 
ICQ UIN 49339118	Linux Counter #88774
GPG-Key 1024D/59C6AFA6 2003-02-07 Roland Pabel <ro...@pa...>

Re: [Devil-linux-develop] Re: [Devil-linux-commit] build/scripts/config config_busybox,1.16,1.17

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

Friedrich Lobenstock wrote:
> Hi Heiko!
> 
> I think your busybox-1.00-pre6 problems must have had something
> do to with your following changes:

Yes that could be, but I just leave it there for now. ;-)

Heiko

[Devil-linux-develop] Re: [Devil-linux-commit] build/scripts/config config_busybox,1.16,1.17

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

Hi Heiko!

I think your busybox-1.00-pre6 problems must have had something
do to with your following changes:

smi...@pr... wrote on 01.02.2004 20:35 MET:
> Index: config_busybox
> ===================================================================
> RCS file: /cvsroot/devil-linux/build/scripts/config/config_busybox,v
> retrieving revision 1.16
> retrieving revision 1.17
> diff -u -d -r1.16 -r1.17
> --- config_busybox	24 Jan 2004 17:56:01 -0000	1.16
> +++ config_busybox	1 Feb 2004 19:35:10 -0000	1.17
> @@ -48,15 +48,21 @@
>  CONFIG_TAR=y
>  CONFIG_FEATURE_TAR_CREATE=y
>  CONFIG_FEATURE_TAR_BZIP2=y
> -CONFIG_FEATURE_TAR_EXCLUDE=y
> +# CONFIG_FEATURE_TAR_FROM is not set
>  CONFIG_FEATURE_TAR_GZIP=y
> +# CONFIG_FEATURE_TAR_COMPRESS is not set
>  CONFIG_FEATURE_TAR_OLDGNU_COMPATABILITY=y
>  CONFIG_FEATURE_TAR_GNU_EXTENSIONS=y
> -CONFIG_FEATURE_UNARCHIVE_TAPE=y
> +# CONFIG_FEATURE_TAR_LONG_OPTIONS is not set
>  CONFIG_UNCOMPRESS=y
>  CONFIG_UNZIP=y
>  
>  #
> +# Common options for cpio and tar
> +#
> +CONFIG_FEATURE_UNARCHIVE_TAPE=y
> +
> +#
>  # Coreutils
>  #
>  CONFIG_BASENAME=y
> @@ -87,6 +93,7 @@
>  CONFIG_FEATURE_FANCY_HEAD=y
>  # CONFIG_HOSTID is not set
>  CONFIG_ID=y
> +# CONFIG_INSTALL is not set
>  CONFIG_LENGTH=y
>  CONFIG_LN=y
>  CONFIG_LOGNAME=y
> @@ -109,12 +116,11 @@
>  CONFIG_REALPATH=y
>  CONFIG_RM=y
>  CONFIG_RMDIR=y
> +# CONFIG_SEQ is not set
>  # CONFIG_SHA1SUM is not set
>  CONFIG_SLEEP=y
>  CONFIG_FEATURE_FANCY_SLEEP=y
>  CONFIG_SORT=y
> -CONFIG_FEATURE_SORT_REVERSE=y
> -CONFIG_FEATURE_SORT_UNIQUE=y
>  CONFIG_STTY=y
>  CONFIG_SYNC=y
>  CONFIG_TAIL=y
> @@ -156,13 +162,17 @@
>  CONFIG_FEATURE_HUMAN_READABLE=y
>  
>  #
> +# Common options for md5sum, sha1sum
> +#
> +# CONFIG_FEATURE_MD5_SHA1_SUM_CHECK is not set
> +
> +#
>  # Console Utilities
>  #
>  # CONFIG_CHVT is not set
>  CONFIG_CLEAR=y
>  # CONFIG_DEALLOCVT is not set
>  # CONFIG_DUMPKMAP is not set
> -# CONFIG_LOADACM is not set
>  # CONFIG_LOADFONT is not set
>  # CONFIG_LOADKMAP is not set
>  # CONFIG_OPENVT is not set
> @@ -173,6 +183,7 @@
>  # Debian Utilities
>  #
>  # CONFIG_MKTEMP is not set
> +# CONFIG_PIPE_PROGRESS is not set
>  CONFIG_READLINK=y
>  # CONFIG_RUN_PARTS is not set
>  # CONFIG_START_STOP_DAEMON is not set
> @@ -185,7 +196,6 @@
>  CONFIG_FEATURE_AWK_MATH=y
>  CONFIG_PATCH=y
>  CONFIG_SED=y
> -CONFIG_FEATURE_SED_EMBEDED_NEWLINE=y

Could it have been the above?

>  # CONFIG_VI is not set
>  
>  #
> @@ -203,13 +213,17 @@
>  CONFIG_FEATURE_GREP_FGREP_ALIAS=y
>  CONFIG_FEATURE_GREP_CONTEXT=y
>  CONFIG_XARGS=y
> +# CONFIG_FEATURE_XARGS_SUPPORT_CONFIRMATION is not set
> +# CONFIG_FEATURE_XARGS_SUPPORT_QUOTES is not set
> +# CONFIG_FEATURE_XARGS_SUPPORT_TERMOPT is not set
> +# CONFIG_FEATURE_XARGS_SUPPORT_ZERO_TERM is not set
>  
>  #
>  # Init Utilities
>  #
>  CONFIG_INIT=y
>  # CONFIG_FEATURE_USE_INITTAB is not set
> -# CONFIG_FEATURE_INITRD is not set
> +CONFIG_FEATURE_INITRD=y
>  # CONFIG_FEATURE_INIT_COREDUMPS is not set
>  # CONFIG_FEATURE_EXTRA_QUIET is not set
>  CONFIG_HALT=y
> @@ -239,10 +253,12 @@
>  # CONFIG_CROND is not set
>  # CONFIG_CRONTAB is not set
>  # CONFIG_DC is not set
> +# CONFIG_DEVFSD is not set

Or this?

>  # CONFIG_LAST is not set
>  # CONFIG_HDPARM is not set
>  # CONFIG_MAKEDEVS is not set
>  # CONFIG_MT is not set
> +# CONFIG_RX is not set
>  # CONFIG_STRINGS is not set
>  # CONFIG_TIME is not set
>  # CONFIG_WATCHDOG is not set
> @@ -251,12 +267,9 @@
>  # Linux Module Utilities
>  #
>  CONFIG_INSMOD=y
> -# CONFIG_FEATURE_OLD_MODULE_INTERFACE is not set
> -CONFIG_FEATURE_NEW_MODULE_INTERFACE=y
> -
> -#
> -#   Support new (post 2.1) Linux kernels (Forced enabled)
> -#
> +# CONFIG_FEATURE_2_2_MODULES is not set
> +CONFIG_FEATURE_2_4_MODULES=y
> +CONFIG_FEATURE_2_6_MODULES=y
>  # CONFIG_FEATURE_INSMOD_VERSION_CHECKING is not set
>  # CONFIG_FEATURE_INSMOD_KSYMOOPS_SYMBOLS is not set
>  # CONFIG_FEATURE_INSMOD_LOADINKMEM is not set
> @@ -315,6 +328,7 @@
>  # CONFIG_RENICE is not set
>  # CONFIG_TOP is not set
>  # CONFIG_UPTIME is not set
> +# CONFIG_SYSCTL is not set
>  
>  #
>  # Another Bourne-like Shell
> @@ -343,20 +357,21 @@
>  #
>  # Bourne Shell Options
>  #
> +# CONFIG_FEATURE_SH_EXTRA_QUIET is not set
> +CONFIG_FEATURE_SH_STANDALONE_SHELL=y
> +# CONFIG_FEATURE_SH_APPLETS_ALWAYS_WIN is not set
>  CONFIG_FEATURE_COMMAND_EDITING=y
> +CONFIG_FEATURE_COMMAND_HISTORY=15
>  CONFIG_FEATURE_COMMAND_SAVEHISTORY=y
>  CONFIG_FEATURE_COMMAND_TAB_COMPLETION=y
>  CONFIG_FEATURE_COMMAND_USERNAME_COMPLETION=y
> -CONFIG_FEATURE_COMMAND_HISTORY=15
> -CONFIG_FEATURE_SH_STANDALONE_SHELL=y
> -# CONFIG_FEATURE_SH_APPLETS_ALWAYS_WIN is not set
>  # CONFIG_FEATURE_SH_FANCY_PROMPT is not set
> -# CONFIG_FEATURE_SH_EXTRA_QUIET is not set
>  
>  #
>  # System Logging Utilities
>  #
>  CONFIG_SYSLOGD=y
> +# CONFIG_FEATURE_ROTATE_LOGFILE is not set
>  # CONFIG_FEATURE_REMOTE_LOG is not set
>  # CONFIG_FEATURE_IPC_SYSLOG is not set
>  CONFIG_KLOGD=y

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

Re: [Devil-linux-develop] busybox

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

Heiko Zuerker wrote:
> Friedrich Lobenstock wrote:
> 
>> Heiko Zuerker wrote on 01.02.2004 18:19 MET:
>>
>>> Friedrich Lobenstock wrote:
>>>
>>>> Heiko Zuerker wrote on 01.02.2004 18:02 MET:
>>>>
>>>>> I just tested pre6 and it's still not working properly.
>>>>
>>>>
>>>>
>>>> Any details?
>>>
>>>
>>>
>>> Once the system is fully booted up, just execute a "mount".
>>> With the versions newer then pre3, you'll see that initrd and all 
>>> sub-mounts didn't get umounted.
>>
>>
>>
>> Can you give me more details, kernel, output of mount, and so one,
>> so I can report this on the busybox mailinglist.
>>
> 
> Oh wait a minute, I know why I don't get an error message.
> 
> I'll take another look at it, once I'm finished with lunch.

Great.... now I can't reproduce the problem.

Whatever, I update to pre6.

Heiko

Re: [Devil-linux-develop] busybox

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

Friedrich Lobenstock wrote:
> Heiko Zuerker wrote on 01.02.2004 18:19 MET:
> 
>> Friedrich Lobenstock wrote:
>>
>>> Heiko Zuerker wrote on 01.02.2004 18:02 MET:
>>>
>>>> I just tested pre6 and it's still not working properly.
>>>
>>>
>>> Any details?
>>
>>
>> Once the system is fully booted up, just execute a "mount".
>> With the versions newer then pre3, you'll see that initrd and all 
>> sub-mounts didn't get umounted.
> 
> 
> Can you give me more details, kernel, output of mount, and so one,
> so I can report this on the busybox mailinglist.
> 

Oh wait a minute, I know why I don't get an error message.

I'll take another look at it, once I'm finished with lunch.

cu
   Heiko

Re: [Devil-linux-develop] busybox

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

Friedrich Lobenstock wrote:
> Heiko Zuerker wrote on 01.02.2004 18:19 MET:
> 
>> Friedrich Lobenstock wrote:
>>
>>> Heiko Zuerker wrote on 01.02.2004 18:02 MET:
>>>
>>>> I just tested pre6 and it's still not working properly.
>>>
>>>
>>> Any details?
>>
>>
>> Once the system is fully booted up, just execute a "mount".
>> With the versions newer then pre3, you'll see that initrd and all 
>> sub-mounts didn't get umounted.
> 
> 
> Can you give me more details, kernel, output of mount, and so one,
> so I can report this on the busybox mailinglist.
> 

The problem is, it's hard to see anything.

The only effect I see is that it's not unmounted, no error message, nothing.

Heiko

Re: [Devil-linux-develop] python build

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

Oliver Jehle wrote:
> i've compiled it successfully with kernel 2.6....
> and installed it..
> 
> can you try it with the 3.3 compiler and new binutils ??.. after
> experiments with the gcc compiler build, i've installed it also to
> recompile all the packages.  as i can remember, the original version from
> the lfssystem was 3.2.3 or so...

I'm using the latest gcc and binutils, they're included in the build system.

Don't forgett that the "original" gcc 3.2.3 is only used until we 
compile our own 3.3.2, which happens very early in the build.

Heiko

Re: [Devil-linux-develop] busybox

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

Heiko Zuerker wrote on 01.02.2004 18:19 MET:
> Friedrich Lobenstock wrote:
> 
>> Heiko Zuerker wrote on 01.02.2004 18:02 MET:
>>
>>> I just tested pre6 and it's still not working properly.
>>
>> Any details?
> 
> Once the system is fully booted up, just execute a "mount".
> With the versions newer then pre3, you'll see that initrd and all 
> sub-mounts didn't get umounted.

Can you give me more details, kernel, output of mount, and so one,
so I can report this on the busybox mailinglist.

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

Re: [Devil-linux-develop] busybox

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

Friedrich Lobenstock wrote:
> Heiko Zuerker wrote on 01.02.2004 18:02 MET:
> 
>> Friedrich Lobenstock wrote:
>>
>>> Heiko Zuerker wrote on 22.01.2004 02:48 MET:
>>>
>>>> Howdy folks,
>>>>
>>>> I had to revert  back to busybox 1.0pre3 (again), since it still has 
>>>> mount problems. I even tried the current CVS version, but it still 
>>>> the same.
>>>>
>>>> This means Initrd module loading with Kernel 2.6 won't work anymore.
>>>
>>>
>>>
>>>
>>> Try pre-5. They seem to see pre-releases not as versions that mature
>>> but one where heavy work goes one. So I expect busybox to take a while
>>> to be mature again.
>>
>>
>>
>> I just tested pre6 and it's still not working properly.
> 
> 
> Any details?

Once the system is fully booted up, just execute a "mount".
With the versions newer then pre3, you'll see that initrd and all 
sub-mounts didn't get umounted.

Heiko

Re: [Devil-linux-develop] busybox

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

Heiko Zuerker wrote on 01.02.2004 18:02 MET:
> Friedrich Lobenstock wrote:
> 
>> Heiko Zuerker wrote on 22.01.2004 02:48 MET:
>>
>>> Howdy folks,
>>>
>>> I had to revert  back to busybox 1.0pre3 (again), since it still has 
>>> mount problems. I even tried the current CVS version, but it still 
>>> the same.
>>>
>>> This means Initrd module loading with Kernel 2.6 won't work anymore.
>>
>>
>>
>> Try pre-5. They seem to see pre-releases not as versions that mature
>> but one where heavy work goes one. So I expect busybox to take a while
>> to be mature again.
> 
> 
> I just tested pre6 and it's still not working properly.

Any details?

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

[Devil-linux-develop] python build

[Oliver J. <oli...@mo...>]

i've compiled it successfully with kernel 2.6....
and installed it..

can you try it with the 3.3 compiler and new binutils ??.. after
experiments with the gcc compiler build, i've installed it also to
recompile all the packages.  as i can remember, the original version from
the lfssystem was 3.2.3 or so...

Re: [Devil-linux-develop] busybox

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

Friedrich Lobenstock wrote:
> Heiko Zuerker wrote on 22.01.2004 02:48 MET:
> 
>> Howdy folks,
>>
>> I had to revert  back to busybox 1.0pre3 (again), since it still has 
>> mount problems. I even tried the current CVS version, but it still the 
>> same.
>>
>> This means Initrd module loading with Kernel 2.6 won't work anymore.
> 
> 
> Try pre-5. They seem to see pre-releases not as versions that mature
> but one where heavy work goes one. So I expect busybox to take a while
> to be mature again.

I just tested pre6 and it's still not working properly.

Heiko

Re: [Devil-linux-develop] upgrade-config

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

Heiko Zuerker wrote on 01.02.2004 16:50 MET:
> Friedrich Lobenstock wrote:
> 
>> Bruce Smith wrote on 01.02.2004 03:23 MET:
>>
>>> The grep example is the cleanest bash code, but what about performance?
>>> The upgrade-config script is already a little slow, and I'm more
>>> concerned about speed than how it looks.
>>>
>>> (my guess is still grep, but I'd like hear other thoughts)
>>
>> I think you could try the following (filelist.txt becomes selectlist):
>>
>> # build candidate list (a list of all available files)
>> ....... >> /tmp/candidates
>>
>> # calculate intersection of candidate list with select list
>> TO_SELECT=$(sort /tmp/candidates selectlist | uniq -d)
>>
>> for SELECT in $TO_SELECT; do
>>     # cycle through all the files that should be selected per default
>>         ...set variables...
>> done
>>
>> "uniq -d" gives you those lines of both files that are equal,
>> so you will get exactly a list of file that exist in the current
>> config which should be selected per default.
> 
> 
> I blame it on the early time, but I'm confused now....
> 
> Anyhow, you also need to check the filedate, you can exclude all the 
> 1980 ones. This should further reduce possible errors.

Ok, to clarify this, I thought this is clear but if you only add
the files with date > 1980 to the /tmp/candidates files the result
should be what you want.

PS: I don't have to tell you that you should run mktemp to create
a temporary file (see sample code in "man 1 mktemp") instead of
using /tmp/candidates which was just used to be descriptive.

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

Re: [Devil-linux-develop] upgrade-config

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

Friedrich Lobenstock wrote:
> Bruce Smith wrote on 01.02.2004 03:23 MET:
> 
>>>>>> Shouldn't we move firewall.rules to /etc/sysconfig/network?
>>>>>> /etc/sysconfig will propably be always be selected for upgrade.
>>>>>
>>>>>
>>>> All the other init scripts are in init.d.
>>>> I guess it could go either place...   :-)
>>>
>>>
>>> Was just a thought to make it more intuitive to see which files
>>> will propably need migration/copying from version to version.
>>
>>
>> I don't know which place is better.
>> It's easier to leave it alone.  :-)
> 
> 
> Ok.
> 
> 
>> What's the most efficient way to check each file against a large list of
>> filenames in a script?
>>
>> I'm thinking a bunch of lines like this isn't the best.
>> [ $i = file1 ] && ..."
>> [ $i = file2 ] && ..."
>> [ $i = file3 ] && ..."
>> ....
>>
>> Or a big long case statement?  case $i in
>>   file1 | file2 | file3 | ... )
>>
>> Or putting all the files to check in a text file, and doing a:
>> "[ $(grep -xc $i filelist.txt) -eq 1 ] && ..."
> 
> 
> grep -qx $i filelist.txt &&
> 
> or
> 
> if grep -qx $i filelist.txt; then
>  ....
> fi
> 
>>
>> Or something else?
>>
>> The grep example is the cleanest bash code, but what about performance?
>> The upgrade-config script is already a little slow, and I'm more
>> concerned about speed than how it looks.
>>
>> (my guess is still grep, but I'd like hear other thoughts)
> 
> 
> I think you could try the following (filelist.txt becomes selectlist):
> 
> # build candidate list (a list of all available files)
> ....... >> /tmp/candidates
> 
> # calculate intersection of candidate list with select list
> TO_SELECT=$(sort /tmp/candidates selectlist | uniq -d)
> 
> for SELECT in $TO_SELECT; do
>     # cycle through all the files that should be selected per default
>         ...set variables...
> done
> 
> "uniq -d" gives you those lines of both files that are equal,
> so you will get exactly a list of file that exist in the current
> config which should be selected per default.

I blame it on the early time, but I'm confused now....

Anyhow, you also need to check the filedate, you can exclude all the 
1980 ones. This should further reduce possible errors.

Heiko

Re: [Devil-linux-develop] default gateway changes in 1.1.x

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

DONE

Heiko

Friedrich Lobenstock wrote:
> Hi Heiko!
> 
> Around Dec 30th you did the following changes:
> 
> http://cvs.sourceforge.net/viewcvs.py/devil-linux/build/config/etc/init.d/network?r1=1.22&r2=1.23&diff_format=u 
> 
> http://cvs.sourceforge.net/viewcvs.py/devil-linux/build/config/etc/init.d/network?r1=1.22&r2=1.23&diff_format=u 
> 
> @@ -327,15 +327,13 @@
>          test -n "$ONLY_INTERFACE" && exit
> 
>          #
> -        # If the /etc/sysconfig/network file contains a GATEWAY 
> variable, set
> -        # the default gateway and the interface through which the default
> -        # gateway can be reached.
> +        # If the /etc/sysconfig/network file contains a GATEWAY variable,
> +        # warn the user that we don't support it anymore
>          #
>          if [ -n "$DEFAULTGATEWAY" -a -n "$DEFAULTGATEWAY_IF" ]; then
> -            echo -n "Setting up routing for $DEFAULTGATEWAY_IF interface"
> -            route add default gateway $DEFAULTGATEWAY \
> -                    metric 1 dev $DEFAULTGATEWAY_IF
> -            evaluate_retval
> +            echo -n "Setting of default gateway via 
> /etc/sysconfig/config is not supported anymore"
> +            print_status failure
> +            echo "please set it in the interface configuration file"
>          fi
> 
>          # enable IP Routing
> 
> 
> http://cvs.sourceforge.net/viewcvs.py/devil-linux/build/config/etc/sysconfig/config?r1=1.19&r2=1.20&diff_format=u 
> 
> --- devil-linux/build/config/etc/sysconfig/config    2003/10/09 
> 00:14:08    1.19
> +++ devil-linux/build/config/etc/sysconfig/config    2003/12/30 
> 18:43:56    1.20
> @@ -12,8 +12,6 @@
> 
>  HOSTNAME=Devil
>  DOMAIN=devil-linux.org
> -#DEFAULTGATEWAY=10.90.1.1
> -#DEFAULTGATEWAY_IF=eth0
> 
>  # enable routing -> yes/no
>  START_ROUTING=yes
> 
> Could you please add the corresponding examples on how to configure
> the default gateway to the /etc/sysconfig/nic/*.sample files? Thx.
> 

Re: [Devil-linux-develop] [ devil-linux-Bugs-888279 ] alsa-driver fails under kernel 2.6

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

Bruce Smith wrote:
>>Summary: alsa-driver fails under kernel 2.6
> 
> 
> ALSA is included in the stock 2.6 kernel. 
> I don't think you need the separate driver, only the lib & utils.

Good point, I changed it.

Heiko

[Devil-linux-develop] [ devil-linux-Bugs-888279 ] alsa-driver fails under kernel 2.6

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

Bugs item #888279, was opened at 2004-01-31 14:59
Message generated for change (Comment added) made by smiley73
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=410643&aid=888279&group_id=34096

Category: Package
>Group: v1.1
>Status: Closed
>Resolution: Fixed
Priority: 5
Submitted By: Heiko Zuerker (smiley73)
>Assigned to: Heiko Zuerker (smiley73)
Summary: alsa-driver fails under kernel 2.6

Initial Comment:
alsa doesn't compile under kernel 2.6
tried with latest version 1.0.2

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

>Comment By: Heiko Zuerker (smiley73)
Date: 2004-02-01 10:04

Message:
Logged In: YES 
user_id=112133

we don't need to compile our own driver, it's included in
kernel 2.6 (thx Bruce)

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

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=410643&aid=888279&group_id=34096

Re: [Devil-linux-develop] upgrade-config

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

Bruce Smith wrote on 01.02.2004 03:23 MET:
>>>>>Shouldn't we move firewall.rules to /etc/sysconfig/network?
>>>>>/etc/sysconfig will propably be always be selected for upgrade.
>>>>
>>>All the other init scripts are in init.d.
>>>I guess it could go either place...   :-)
>>
>>Was just a thought to make it more intuitive to see which files
>>will propably need migration/copying from version to version.
> 
> I don't know which place is better.
> It's easier to leave it alone.  :-)

Ok.


> What's the most efficient way to check each file against a large list of
> filenames in a script?
> 
> I'm thinking a bunch of lines like this isn't the best.
> [ $i = file1 ] && ..."
> [ $i = file2 ] && ..."
> [ $i = file3 ] && ..."
> ....
> 
> Or a big long case statement?  
> case $i in
>   file1 | file2 | file3 | ... )
> 
> Or putting all the files to check in a text file, and doing a:
> "[ $(grep -xc $i filelist.txt) -eq 1 ] && ..."

grep -qx $i filelist.txt &&

or

if grep -qx $i filelist.txt; then
  ....
fi

> 
> Or something else?
> 
> The grep example is the cleanest bash code, but what about performance?
> The upgrade-config script is already a little slow, and I'm more
> concerned about speed than how it looks.
> 
> (my guess is still grep, but I'd like hear other thoughts)

I think you could try the following (filelist.txt becomes selectlist):

# build candidate list (a list of all available files)
....... >> /tmp/candidates

# calculate intersection of candidate list with select list
TO_SELECT=$(sort /tmp/candidates selectlist | uniq -d)

for SELECT in $TO_SELECT; do
	# cycle through all the files that should be selected per default
         ...set variables...
done

"uniq -d" gives you those lines of both files that are equal,
so you will get exactly a list of file that exist in the current
config which should be selected per default.

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

[Devil-linux-develop] RESEND ddt didnt compile

[Oliver J. <oli...@mo...>]

looks like a problem between libgcrypt and ddt... 


Making all in packets
make[2]: Entering directory `/build/tmp/ddt-0.5.9/framework/packets'
g++ -DPACKAGE_NAME=\"\" -DPACKAGE_TARNAME=\"\" -DPACKAGE_VERSION=\"\"
-DPACKAGE_STRING=\"\" -DPACKAGE_BUGREPORT=\"\" -DPACKAGE=\"ddt\"
-DVERSION=\"0.5.9\" -DDDT_CLIENT=1 -DCONF_DIR=\"/etc\"
-DPID_DIR=\"/var/run\" -DLOG_DIR=\"/dev/null\" -DVAR_DIR=\"/var/lib\"
-DSTDC_HEADERS=1 -DHAVE_SYS_TYPES_H=1 -DHAVE_SYS_STAT_H=1
-DHAVE_STDLIB_H=1 -DHAVE_STRING_H=1 -DHAVE_MEMORY_H=1 -DHAVE_STRINGS_H=1
-DHAVE_INTTYPES_H=1 -DHAVE_STDINT_H=1 -DHAVE_UNISTD_H=1 -DSTDC_HEADERS=1
-DHAVE_UNISTD_H=1 -DHAVE_DAEMON=1 -DHAVE_LIBDL=1  -I. -I. 
-I../include   -I/usr/include  -I/usr/include -I/usr/include   -g -O2 -c
DdtCipher.cc
In file included from DdtCipher.cc:12:
DdtCipher.h:61: warning: `GCRY_CIPHER_HD' is deprecated (declared at
   /usr/include/gcrypt.h:580)
DdtCipher.cc: In constructor
`DdtCipher::DdtCipher(DdtCipher::CipherAlgos,
   DdtCipher::CipherModes)':
DdtCipher.cc:27: error: cannot convert `gcry_cipher_algos' to `
   gcry_cipher_handle**' for argument `1' to `gcry_error_t
   gcry_cipher_open(gcry_cipher_handle**, int, int, unsigned int)'
make[2]: *** [DdtCipher.o] Error 1
make[2]: Leaving directory `/build/tmp/ddt-0.5.9/framework/packets'
make[1]: *** [all-recursive] Error 1
make[1]: Leaving directory `/build/tmp/ddt-0.5.9/framework'
make: *** [all-recursive] Error 1

Re: [Devil-linux-develop] upgrade-config

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

> >>>Shouldn't we move firewall.rules to /etc/sysconfig/network?
> >>>/etc/sysconfig will propably be always be selected for upgrade.
> >>
> >>That's a good thought, I would agree to this.
> >>Of course upgrade-config needs to take care of this.
> > 
> > It is a script.  An init script.  
> 
> It's the "config file" of /etc/init.d/firewall, isn't it? :-)

It's that too!  :-)

> > All the other init scripts are in init.d.
> > I guess it could go either place...   :-)
> 
> Was just a thought to make it more intuitive to see which files
> will propably need migration/copying from version to version.

I don't know which place is better.
It's easier to leave it alone.  :-)

> > As far as default files to be selected in update-config, I think we
> > should make a table of files (and maybe directories) that are initially
> > selected, and leave everything else turned off.
> > 
> > I REALLY don't want to make a mistake and select anything by default
> > that shouldn't be.  A novice may not know what a file is, and ass-u-me
> > that since we defaulted it to be copied that it should be.
> 
> Absolutely right.

What's the most efficient way to check each file against a large list of
filenames in a script?

I'm thinking a bunch of lines like this isn't the best.
[ $i = file1 ] && ..."
[ $i = file2 ] && ..."
[ $i = file3 ] && ..."
....

Or a big long case statement?  
case $i in
  file1 | file2 | file3 | ... )

Or putting all the files to check in a text file, and doing a:
"[ $(grep -xc $i filelist.txt) -eq 1 ] && ..."

Or something else?

The grep example is the cleanest bash code, but what about performance?
The upgrade-config script is already a little slow, and I'm more
concerned about speed than how it looks.

(my guess is still grep, but I'd like hear other thoughts)

 - BS