Re: [OT]: < 10pt in LaTeX?

[Carel Fellinger]

On Sun, Feb 23, 2003 at 10:31:37PM -0500, Nori Heikkinen wrote:
...
> that's cool ... i was looking for an option i could just pass to
> \documentstyle, i guess, but you can only tell it to be 10, 11, and
> 12pt.  whatever; this is likely the only thing i'll ever do in \tiny
> :)

You could try _extsizes_ (see CTAN):

   *extsizes*  Extends article and report with extra sized fonts. 
   Provides two classes, extarticle and extreport, which allow for
   documents with a base font of size 8--20pt.
   Visit CTAN:macros/latex/contrib/other/extsizes/ Download from
   somewhere nearby (Search) Download from Dante License: lppl 
   Catalogued: 1999/12/07 

Or you could try KOMA script classes, but that probably involves more
work to get it going, anyhow it uses extsizes under the cover.

-- 
groetjes, carel


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Bad Debian (L.A.H.)

[Carel Fellinger]

On Sun, Mar 09, 2003 at 09:53:08PM -0600, Nathan E Norman wrote:
...
> IMO, you can avoid anything printed by Prentice Hall except stuff
> written by W. Richard Stevens.

I think that's ill informed advice.  Some of the best books on informatics
are from them, like:

  A discipline of programming,  by Edgard Dijkstra

  Operating Systems, by Tanenbaum

  Object Oriented Software Construction, by Meyer

Maybe in recent years their catalogue has watered down, I don't know,
but at least they used to be top notch.

-- 
groetjes, carel


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Installing on a FACELESS pc?

[Carel Fellinger]

On Thu, Jan 23, 2003 at 11:10:54AM -0600, will trillich wrote:
...
> headless would be a monitor, keyboard and mouse, with no cpu
> box.

Not quit.  Headless would be without monitor, cpu and memory.
Lack of cpu and memory could be called brainless, but lack of cpu
alone lacks any resemblance with known life forms.


To be serious again, what's really lacking to get headless machines
to fully work is lack of support in most (all?) BIOS'ses to use a
serial line instead of directly attached monitor and keyboard.
If your motherboard is supported the LinuxBIOS might solve that one.

-- 
groetjes, carel


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Installing on a FACELESS pc?

[Carel Fellinger]

On Fri, Jan 24, 2003 at 04:40:05PM +, Pigeon wrote:
> On Fri, Jan 24, 2003 at 03:33:19PM +0100, Carel Fellinger wrote:
...
> > To be serious again, what's really lacking to get headless machines
> > to fully work is lack of support in most (all?) BIOS'ses to use a
> > serial line instead of directly attached monitor and keyboard.
> > If your motherboard is supported the LinuxBIOS might solve that one.
> 
> You could make an adapter with a few resistors and diodes to drive the
> keyboard input of a PC from the parallel or serial port of another
> machine. You'd be typing blind though. You'd have to give an exact
> script to get the Debian installer to the point where the network
> interface is running. OK if nothing goes wrong.

Maybe, just maybe this would allow you to alter BIOS settings, but it
certainly wouldn't make it possible to react to BIOS warnings, like
when your S.M.A.R.T.-enabled harddrive yelled ALARM, or any other
failure your BIOS deals with prior to linux taking over:(

-- 
groetjes, carel


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Initrd diskless boot

[Carel Fellinger]

On Fri, Jan 24, 2003 at 09:55:06PM -0500, Jameson C. Burt wrote:
> I re-emphasize the www.ltsp.org approach. 
...
> I bought a diskless workstation 3 weeks ago from a link on ltsp.org,
> a workstation the size of your outstretched hand yet having 
> audio, USB, parallel, serial, and ethernet ports -- all with 
> a 15 watt power supply and no fans.  

I never could really find the info whether that 15Watt power supply is
part of the bricks or not and whether it needs any fan itself.
Could you enlighten me?

-- 
groetjes, carel


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Initrd diskless boot

[Carel Fellinger]

On Sun, Jan 26, 2003 at 09:52:16PM -0500, Jameson C. Burt wrote:
> On Sun, Jan 26, 2003 at 03:30:32AM +0100, Carel Fellinger wrote:
...
> > I never could really find the info whether that 15Watt power supply is
> > part of the bricks or not and whether it needs any fan itself.
> > Could you enlighten me?

> The power supply is external, the size of a large eraser,
> in the middle of a cord like
...
> They mention 15 watts on their webpage [URL probably wrapped thru email],
> http://www.disklessworkstations.com/cgi-bin/cat/info/jam125.html?Bpdb2RQX;;99

Thanks for the info.


-- 
groetjes, carel


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Cdrecord and audio cd

[Carel Fellinger]

On Wed, Jan 29, 2003 at 08:47:34PM +0530, Sridhar M.A. wrote:
...
> The cd does not spin. CD channel is unmuted. I can play other audio
> cd's. No, I am not attempting to mount an audio cd. All cd players under
> linux gave me the same error.

I missed some of your possed and don't feel like going over them in
the archive, so if it's already covered forget I posted:)

Does it play under Windows, or on a different computer?  If not, could
it be that the disc is a protected disc like Sony likes to make, the
ones that don't play on cd-rom players?

-- 
groetjes, carel


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: find a command i have recently used in bash

[Carel Fellinger]

On Fri, Jan 31, 2003 at 03:09:54PM +0100, Ulf Rompe wrote:
...
> As an example, another useful (for me) addition to the inputrc is
> this one: 
> 
> # Ctrl-Left/Right jumps wordwise on cmd line
> "\e[D": backward-word
> "\e[C": forward-word

I fail to see the Control part here, it just looks like the codes for
right and left.  I would be very interested if it could work with
control/meta, would make it easier for me to remember all the search
commands.
 
> This redefines the commands normally known as M-b / M-f to more
> intuitive key strokes.

I must be missing something:)  Aren't those already bound to word
movement?  Besides, the above inputrc snippet doesn't touch M-b/M-f,
but the cursor keys.

-- 
groetjes, carel


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Root partition stuck in read-only mode.

[Carel Fellinger]

On Fri, Jan 31, 2003 at 10:26:26PM -0500, Lloyd Zusman wrote:
> Shawn Lamson <[EMAIL PROTECTED]> wrote...
...
> > mount -o remount,rw /
> 
> Well, that didn't work, either.  Here's the error:
> 
>   EXT3-fs: Unrecognized mount option 0
>   mount: / not mounted already, or bad option

I think you used the number 0, but it should be the letter o!
Hence the Unrecognized error message. Try again, but this time
make sure you type -o (letter o) instead of -0 (number zero)

-- 
groetjes, carel


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Root partition stuck in read-only mode.

[Carel Fellinger]

On Fri, Jan 31, 2003 at 11:17:12PM -0500, Lloyd Zusman wrote:
> I have some more info about my problem that might be useful.
> 
> I booted up off of disk 1 of my Woody installation CDROM set.
> My root partition is /dev/sda2, so I entered the following at
> the "boot:" prompt ...
> 
>   rescue root=/dev/sda2

hm, SCSI disks...
 
> However, this errored out quickly.  I got a couple screens'
...
>   Kernel panic: VFS: Unable to mount root fs on 08:02
...
> Would this indicate a hardware problem?  Or is there still hope that
> this can be fixed with software?

...more likely your kernel doesn't know about SCSI disks, so software.


-- 
groetjes, carel


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Root partition stuck in read-only mode.

[Carel Fellinger]

On Sat, Feb 01, 2003 at 05:30:36PM -0500, Lloyd Zusman wrote:
> Quoting Carel Fellinger <[EMAIL PROTECTED]>:
> > ...more likely your kernel doesn't know about SCSI disks, so
> > software.
> 
> I've been booting up and running off of SCSI disks day
> in and day out for weeks on gthe same kernel, when I boot
> off the disk itself and not a boot floppy.  This read-only
> problem only started a couple days ago, and I haven't changed
> my kernel in 3 weeks or so.
> 
> So how could my kernel forget about SCSI disks when I
> create a boot floppy?  I did it via mkboot and also via the 
> "yard" suite.  Same results each time.
> 
> Is there something special I have to do to tell my kernel
> that there are SCSI disks at the time I make the boot floppy?

That all depends:)
It's a long time since I made boot floppies (prefer bootable CD's)
but if I'm not mistaken mkboot uses your current kernel.  So if all
the scsi stuff is compiled in all's swell.  But if (part of) the SCSI
stuff is compiled as modules you need to use initrd to get to those
modules in time.


-- 
groetjes, carel


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Root partition stuck in read-only mode.

[Carel Fellinger]

On Mon, Feb 03, 2003 at 08:02:45PM -0500, Lloyd Zusman wrote:
...
> Well, I guess I'll have to go back to the drawing board to
> figure out how to make a usable boot floppy ... or perhaps a
> boot CD ... I really don't care which one I have, as long as
> I have _some_ source for a boot record besides my hard disk.

try Timo's rescue CD (debian based):

   http://rescuecd.sourceforge.net/download.html

If I'm not mistaken his bootimage uses initrd, and does some
autodetection of drives.  If the autodetection fails, it's possible to
get a shell early in the boot process and load the needed scsi
modules.  I've never done that so I don't really know how it works,
but I expect "modprobe driver-for-your-scsi-card" would suffice.


> I guess I'm going to have to gain the knowledge necessary for
> using initrd during the process of building a boot disk/CD,
> since mkboot and the yard suite are currently at the limits
> of my knowledge.
> 
> Without my Linux box working for more than a few minutes
> at a time, I have at best intermittent internet access, so
> could someone point me to a step-by-step cookbook that
> describes how to make a bootable floppy or CD, when initrd is
> needed to load added SCSI modules?  Also, something that
> tells me how to find out which SCSI modules I need, and which
> of these are and aren't compiled into my kernel would also be
> useful.  I have the standard, unmodified 2.4.20-686-smp kernel,
> by the way.

Look in /boot/config-2.4.20-686-smp, it lists all the config params
used to compile that kernel.  Look for scsi-generic and scsi-disk and
the apropriate module for your scsi card.



-- 
groetjes, carel


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: saving mutt email headers

[Carel Fellinger]

On Wed, Feb 12, 2003 at 02:36:43AM +, p wrote:
...
> in "v"iew mode, i try to "s"ave
> the email into a file.  

don't go into view mode, but save from the index instead.

-- 
groetjes, carel


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Windows multiboot (aaargh!)

[Carel Fellinger]

On Thu, Sep 11, 2003 at 12:07:08AM +1200, cr wrote:
> This may seem an odd place to ask this, but I'll bet some of the folks on 
> this list know more about the technicalities of booting Windoze than Windoze 
> users do   ;)

I know next to nothing about Windows and prefer to keep it like that:),
but as I've kids that, like kids like, like to play games, I've been
forced to atleast find out how play this trick.  (example at the end)

>From experience, Win95/98 needs to be on the first drive, needs to be
in a bootable primary partition which needs to be the only/first primary
partition.

Don't despair, GRUB is perfectly capable of _hiding_ specific partitions
in the bootprocess, and many(?) BIOSses allow to swap the order of drives.

So create on the disk of choice enough primary partitons, use GRUBS to
hide all but one, swap the BIOS drive order such that the drive of
choice seems to be the first one, and all is swell.

Well it should be, but experience has learned me otherwise:(
The biggest problem being ghost/phantom drives appearing under windows.

I used to think it was related to windows weared way of determining to
use or not to use lineair addressing mode on large drives, and hence
its inability to obey the partitioning scheme used.  So I was very
carefull to partition disks in such a way that the 1024 cylinder (this
being one of the clues windows uses) felt on a partition boundary.

But nowadays I think that what's really needed is to take care that all
windows partitions have there first sector(s?) cleaned prior to letting
windows format those, as it seems that windows prefers the partitioning
information it finds in that(those) first sector(s) to the real partition
table, and what is worst, there is a bootstrap problem as windows format
prog assumes that info is valid and hence will use it unless it's cleared.


As far as I know the 1024 cylinder thing is still very much relevant
to Win95 as wel as chosing the right partition type, so probably the
safest is to partition carefully.  And be warned, I've lost linux
partitions (on my fathers machine thanks to my sister) using windows
tools to repartition, reformat and reinstall windows:(  --this was
prior to me being very carefull to wipe the first sector, so maybe it
works better now.  Anyway, i've always used linux tools to do the
partitioning, the only way for me to be able to _predict_ the end
result--


...
> I'd like to be able to boot into DOS, Win95 and Win98.
...
> I'm just wondering how practical that is.Can W95 and W98 coexist on the 
> same disk in diferent partitions and still both be bootable?If not, any 

Yes, that's what I've been doing for years.  Though the safest thing
would be to have seperate disks for each windows install and use the
BIOS capability to swap drives, you could do like I do (and pray:).


My disks lookes like:

# two bootable windows partitions on hda
# one bootable windows partition on hdb


My /boot/grub/menu.lst lookes like:

title   Windows from second disc
map (hd0) (hd1)
map (hd1) (hd0)
root(hd1,0)
makeactive
chainloader +1


title   Win95 from first disc second partition
hide(hd0,0)
unhide  (hd0,1)
root(hd0,1)
makeactive
chainloader +1


title   Win98 from first disc first partition
unhide  (hd0,0)
hide(hd0,1)
root(hd0,0)
makeactive
chainloader +1



-- 
groetjes, carel


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Windows multiboot (aaargh!)

[Carel Fellinger]

On Wed, Sep 10, 2003 at 11:28:24PM +0100, Pigeon wrote:
> On Wed, Sep 10, 2003 at 10:30:06PM +0200, Carel Fellinger wrote:
> > But nowadays I think that what's really needed is to take care that all
> > windows partitions have there first sector(s?) cleaned prior to letting
> > windows format those, as it seems that windows prefers the partitioning
,,,
> This is true. This is one reason why it is better to use DOS FDISK to
> create DOS/Windoze partitions - it wipes the first sector when you
> create the partition.

True as this might be, it has the severe disadvantage of being stuck
with DOS FDISK.  Maybe it's just my ignorance on windows things but
somehow DOS FDISK always managed to ruin my carefull layed out
partitioning.  Especially trying to change a partition table
afterwards proved to be troublsome.

-- 
groetjes, carel


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Windows multiboot (aaargh!)

[Carel Fellinger]

On Thu, Sep 11, 2003 at 11:02:53PM +1200, cr wrote:
> On Thursday 11 September 2003 08:30, Carel Fellinger wrote:
...
> > From experience, Win95/98 needs to be on the first drive, needs to be
> > in a bootable primary partition which needs to be the only/first primary
> > partition.
> 
> On the first drive?   I only ask ...

Now that you ask, I'm not sure:)  Could well be that it only needs to
be the first (bootable) primary partition DOS/Windows finds/recognizes.


-- 
groetjes, carel


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Sound capture via fake driver

[Carel Fellinger]

On Mon, Sep 15, 2003 at 08:40:44PM -0400, Carl Fink wrote:
> I remember, sometime last year, using a replacement sound module to capture
> streaming audio to a file.  It worked quite well.

Are you thinking of something like vsound?


-- 
groetjes, carel


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: [Solved] fetchmail rules for logcheck?

[Carel Fellinger]

On Fri, Oct 04, 2002 at 02:08:45PM -0400, Brian P. Flaherty wrote:
> Hello,
> 
> Sorry for the wasted bandwidth, but it had not occurred to me to check

Not wasted at all, atleast not for me:)  I've postponed looking into this
for month now, yet the shere amount of fetchmail blurbs made it harder to
pick up the interesting bits then I liked.

> the workstation and server directories.  Rules for fetchmail are in
> both of those.

So, thanks!

-- 
groetjes, carel


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: USB Mouse problems (Solved)

[Carel Fellinger]

On Sun, Oct 06, 2002 at 03:35:15AM -0400, lameth wrote:
...
> The previous distrobution of linux that I used was Mandrake 8.0. The 
> installs always went fine but then after a month or two of using it I 
> would start seeing messages about non-contiguous data on the hard drive. 

That's just `fsck' periodically run at boot time, to check the disk
for errors.  It's quite normal for a disk to not have all its data
contiguous (imagine all data being contigous and you delete a file,
then most likely you get a hole somewhere and the data is no longer
contigous), but as long as the non-contiguous fraction is small your
system will run just fine.  Fortunately both ext2 and reiserfs are
designed to handle fragmentation very well, in fact they have counter
measurements to `unfragment' all the time, so there hardly ever is a
need to run a defragment program like you'll have to do under Windows.

...snipped an *awfull* lot of no longer relevant thread history
please try to weed out such parts

-- 
groetjes, carel


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: debian emacs policy and configuring mutt

[Carel Fellinger]

On Mon, Oct 07, 2002 at 11:56:59PM +0200, Robert Wilhelm Land wrote:
> On the way trying to convince mutt
> to use emacs -nw instead of vi by
> editing .muttrc as:
> 
> set editor=emacs -nw

There is a space in the command, so you need quotes, try :

  set editor="emacs -nw"


-- 
groetjes, carel


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: tiny mouse howto update - console gpm, X and mouse wheel

[Carel Fellinger]

On Sun, Oct 06, 2002 at 06:03:58PM -0700, Osamu Aoki wrote:
> Hi,
> 
> I am going over ML for the next "Debian Reference" over the mouse
> configuration.
> 
> I found a thread with your names quite interesting.  In order to refresh
> all, I put all participants in TO list.  Excuse me.
> 
> I have 3 questions:
> 
> 1. imps2 issue
> 
> Is "imps2" really needed for woody gpm?  Woody gpm default seems to
> have moved to "autops2".  I thought autops2 will take care both.

IIRC I had problems with autops and some mice I have, be it a Logitech
Mouse Man Wheel, or the simpler wheel version.

> 
> FYI:"gpm -t help|less" command in woody/sarge should provide.
...

In comming versions or right now?
On my Woody machine the order differs, and there is a bonus fuimps2.


> 2. What do you think of Debian Reference description?
> 
> http://www.debian.org/doc/manuals/reference/ 
> Chapter 3.3 Mouse configuration

I had a hard time reading this, and only when I switched from links to
galeon things started to make sense, partly because it now fitted on
one screen but mainly cause I read Dale's suggestion that folded the
two variations showing the one and only difference clearly.  Much more
to my liking:)  Besides, I myself would try to persuede people to use
"repeat_type=raw", so that one unfoldable line would read:

   repeat_type=raw   (or ms3)

And then I would spent a line explaining that value for `type' (in
/etc/gpm.conf) and `Protocol' (in X11's config file) depend on the
brand and model of the mouse in use.


One other thing I noted: just below in that section 3.3, it says:

  Create a symlink /dev/gpmdata --> /dev/mouse to make some

To me it looks as if the symlink is reversed.

 
> 3. Was timothy bauscher"s problem typo?

don't remember


-- 
groetjes, carel


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: tiny mouse howto update - console gpm, X and mouse wheel

[Carel Fellinger]

On Tue, Oct 08, 2002 at 01:51:09AM +0200, Carel Fellinger wrote:
> On Sun, Oct 06, 2002 at 06:03:58PM -0700, Osamu Aoki wrote:
...
> > 2. What do you think of Debian Reference description?
> > 
> > http://www.debian.org/doc/manuals/reference/ 
> > Chapter 3.3 Mouse configuration

O, and why don't you use `/dev/mouse' in X11's config file?
Would make X11's config file independend from gpm.  And the
use of the symlink is easily explained as well:

  In X11's config file use: /dev/mouse,
  in gpm's config file use: /dev/your-real-mouse-device,
  and when you start or stop using gpm, fix where /dev/mouse links to:
iff your using gpm: /dev/mouse --> /dev/gpmdata
iff your NOT using gpm: /dev/mouse --> /dev/your-real-mouse-device

-- 
groetjes, carel


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: tiny mouse howto update - console gpm, X and mouse wheel

[Carel Fellinger]

On Tue, Oct 08, 2002 at 01:39:23AM -0700, Osamu Aoki wrote:
> On Tue, Oct 08, 2002 at 01:51:09AM +0200, Carel Fellinger wrote:
..
> > IIRC I had problems with autops and some mice I have, be it a Logitech
> > Mouse Man Wheel, or the simpler wheel version.
> 
> It is "autops2".  I checked source and this is debian patch.

yep, that's what I ment to type:), anyhow, it doesn't work flawless on my
hardware, whereas imps2 works like a charm.


> > > FYI:"gpm -t help|less" command in woody/sarge should provide.
> > In comming versions or right now?
> > On my Woody machine the order differs, and there is a bonus fuimps2.
> 
> Yes, order is my edit (For making it easy to read...  was my
> intention...)

that's fine with me, just wanted to make sure we were talking the same
version here.


> > > 2. What do you think of Debian Reference description?
> > > 
> > > http://www.debian.org/doc/manuals/reference/ 
> > > Chapter 3.3 Mouse configuration
> > 
> > I had a hard time reading this, and only when I switched from links to
> 
> ???
> How many chars per line do you use?  I use 80char/line and fine.

it's not the chars per line, it's the lines per screen:)

It took me some time to see were the first example ended and the next
started, at first it wasn't even clear that there were two examples!
(the first ended right on the first page, the second started right at
the top of the second page:)  And then it was not obvious just were the
two examples differed (having to page up and down losing context sure
didn't help).  But when I read Dale's proposal it with clear at first
look.  The whole example fitted on a single screen, and as a bonus I
didn't need to compare the two examples line by line to find the one
and only difference, it was very conveniently spelled out for me.

Oh, and thanks for incorporating list knowledge into the debian manual.

-- 
groetjes, carel


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: using locate properly

[Carel Fellinger]

On Tue, Oct 08, 2002 at 03:15:19PM +0200, Robert Wilhelm Land wrote:
...
> rland@MINI:~$ locate [Mm]uttrc
> ... no reaction.
> 
> Rereading the info gave me:
...snipped where you found out about metachars and quoting

> .so I tried:
> 
> rland@MINI:~$ locate '[Mm]uttrc'
> rland@MINI:~$ locate "[Mm]uttrc"

> But:
> rland@MINI:~$ locate &pi0;[Mm]uttrc'

Does &pi0 denote a backtick?
Maybe it's time to RFM of bash about quoting?
...
> What has happened here?

You were so glad you found a solution that you forgot to read the rest:)
Really, if you had only read the next paragraph in locate's man page you
would have seen that patterns containing metachars are matched against
complete path names, so you should have specified a leading *, like:

  $ locate *[Mm]uttrc

(thanks to file globbling pecularities and the fact that in the current
directory no file matched the above pattern I could do without quoting
or escaping meta chars.  So I did, just to save a few keystrokes:)

In this playing with file globbling, quoting and meta chars it's good
to realise who is doing what.  The shell does the file globbling, i.e.
it tries to expand patterns on the commandline into file names to pass
on to the command specified at the start of the commandline.  When a
pattern doesn't match an existing file, the pattern is passed on `as is'.
The locate program on its turn treats its arguments *again* as file
globbling patterns, now to be matched against its file name database.

-- 
groetjes, carel


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Why debian sucks.

[Carel Fellinger]

On Sun, Oct 27, 2002 at 11:41:49PM -0700, Bob Proulx wrote:
...
> > Your problem is that, since you have changed your server,  your
> > unsubscribe message is coming from the wrong address.  You will have to
> > be removed manually.
> 
> If that is true then you should be able to go to this page and paste
> in your subscription address.  It will send a message to you for
> confirmation of the unsubscribe request.  Confirm it and you should be
> unsubscribed.

Unfortunately this doesn't seem to work:(
The otherday I tried to unsubscribe from debian-ipv6 to find out that
I subscribed under a different name, so I tried to unsubscribe with
that other name filled in, I got a conformation messages, but simply
replying to it didn't work. I had to temporarely make exim rewrite
my headers to conform to the old address I used when subscribing.

...
>   Return-Path: <[EMAIL PROTECTED]>
> 
> This will tell you the address that you need to use to unsubscribe.

Yep, though the address is mangled.  So splitoff `bounce-debian-user='
and `@lists.debian.org' and replace the remaining `=' with `@'.

-- 
groetjes, carel


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Why debian sucks.

[Carel Fellinger]

On Mon, Oct 28, 2002 at 07:57:45PM -0800, Paul Johnson wrote:
> On Mon, Oct 28, 2002 at 05:31:54PM +0100, Carel Fellinger wrote:
> > Unfortunately this doesn't seem to work:(
> > The otherday I tried to unsubscribe from debian-ipv6 to find out that
> > I subscribed under a different name, so I tried to unsubscribe with
> > that other name filled in, I got a conformation messages, but simply
> > replying to it didn't work. I had to temporarely make exim rewrite
> > my headers to conform to the old address I used when subscribing.
> 
> Or you can simply just use mutt as your MUA, which allows you to play
> with the From header on your own.

Nah, not while I'm using exim to do (From-)header rewriting:) Maybe
that's the problem, I used the web page to unsubcribe (no exim there)
but my reply to the confirmation request went through exim and header
rewriting.  I'll try again.

-- 
groetjes, carel


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: parted boot CD

[Carel Fellinger]

On Mon, Nov 04, 2002 at 04:19:31PM -0600, Alex Malinovich wrote:
> I'm in a bit of a bind. I need to resize the root partition on one of my
> systems that doesn't have a floppy drive. The only thing that I have
> that I can boot from is a CD-ROM drive. I found a rescue floppy that had
> parted installed and tried using mkisofs to convert it to a CD image,
> but it doesn't boot. I get an error about running out of data while

You could try Tim's Rescue CD, debian based and all.
The only drawback is he choose vim instead of emacs:)

-- 
groetjes, carel


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Newbie networking problems; pppd with Woody.

[Carel Fellinger]

On Fri, Nov 08, 2002 at 07:44:29AM -0500, Barry Mathieu wrote:
...
> I found a solution to this issue by changing the first line of
> resolv.conf file from, "search mindspring.com" to "order hosts,bind".

What versions are you running?  According to the docs on Debian 3.0
there is no such option for resolv.conf.  There is in host.conf
though.  I think it likely that no nameservers are read from your
incorrect resolv.conf at all.

> My resolv.conf file now looks like:
> order hosts,bind
> nameserver 207.69.188.185
> nameserver 207.69.188.186
> 
> I don't get a frustrating error that stops the process of ping or
> fetchmail. For example, a ping response returns no packets on the first
> transmission, but ping continues and by the second transmission returns
> packets. I'm certain the first ping response is returning no packets
> because the ip-up scripts and DNS determination hasn't completed.   

Could it be that somewhere in your ip-up scripts your incorrect
resolv.conf gets replaced by a correct on?


-- 
groetjes, carel


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: power down and power off also?

[Carel Fellinger]

On Sat, Nov 23, 2002 at 03:29:46PM +0530, Sandip P Deshmukh wrote:
> On Sat, Nov 23, 2002 at 12:33:19AM -0800, Osamu Aoki wrote:
...
> > Yep, how about reading my "Debian Reference".
> > 
> >   http://www.debian.org/doc/manuals/reference/
> > 
> > For above question, it is detailed 
> >http://www.debian.org/doc/manuals/reference/ch-install.en.html#s-apm
> 
> sure i will give it a try. but they sound too technical for a newbie
> like me. but sure give it a try

That's fine with us, then you're the one to help improve those docs!
Next time you're stuck, try to read Osamu's very fine reference manual.
If you can't find your way in his reference, come here again, tell us
what your problem is and *also* where you tried to find the anwser and
in his docs.  Then we can start to understand how those reference
manuals actually should have been structured and what language we
should have used and improve on that.

You see Osamu spent a great deal of time and effort in writing that
manual, and did a mighty good job.  Besides it was ment for guys like
you.  So if you're lost, please help improve it even further.

-- 
groetjes, carel


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: /dev/cdrom, /dev/scd0, /dev/sg0 ?

[Carel Fellinger]

On Fri, Nov 29, 2002 at 06:01:33PM +1100, Rob Weir wrote:
...
> A weird thing came up on #debian today, which maybe someone else can
> explain to _me_: a guy was trying to setup CD ripping with ide-scsi
> emulation enabled, and had enormous trouble since cdparanoia claimed it
> couldn't find a generic device.  He only had one SCSI device (real and
> emulated), which was mountable from /dev/scd0, and he had write perms on
> both scd0 and sg0.  In the end, it turned out that he had to have write
> permissions on 'sg1' as well.  Why is that?  I was under the impression
> that sgn was the generic device for scdn, but apparently not...

I'm not much of an expert in anything, but isn't it weird to have one
device both as real and as emulated drive?  And iff you do so, isn't it
then logical that both real and emulated drive get their own special file?

-- 
groetjes, carel


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: which is the best partition table format?

[Carel Fellinger]

On Fri, Nov 29, 2002 at 09:59:59PM -0500, Derrick 'dman' Hudson wrote:
...
> I started looking through the HOWTO.  I'll need some better
> visualization of how it works before I really "get it".  Anyways, I
> found the "linux-lvm" mailing list, so I think I'll try learning some
> stuff there.

Mind to report back here when you've learned enough?
Almost every time my kids install a new game just to ruin
my carefull crafted windows setup I curse the day I gave in
and let them have a windows running game machine, increasing
my desire to expell the last trace of it on my own macnines.

-- 
groetjes, carel


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: ES1869 sound chip driver

[Carel Fellinger]

On Sat, Nov 30, 2002 at 06:54:19AM -0600, Russ Cook wrote:
> Thanks!  Xmms has the same problem - repeating portions of the audio
> multiple times before moving to the next portion.  I checked the system
> log with dmesg as suggested and get the following indicators:
> 
> SB 3.01 detected OK (220)
> ESS chip ES1869 specified
> sb: Interrupt test on IRQ7 failed - Probable IRQ conflict
>  at ox220 irq 7 dma 1,5
> Sound: DMA (output) timed out - IRQ/DRQ config error?

Just seconds agoo I fixed such a problem over here, I had deactivated
th eprinter in the BIOS, but forgot all about /etc/modules.conf, and
indeed over there I still told linux to use IRQ-7 for the printer:)

So look again at dmesg's output, and lookout for an other device
claiming IRQ-7.

Who knows, maybe you're lucky and it's this simple.

-- 
groetjes, carel


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Exim permissions

[Carel Fellinger]

On Sat, Nov 30, 2002 at 11:14:04PM +, Pigeon wrote:
> Re time delay: I've just given it over 15 minutes to see what would
> happen, and it didn't flush its queue, and doing ps ax every so often
> has revealed no trace of exim or any other mail programs running.

Not shure you have a real problem, as exim in the default install
tends to be running just to delivery and then stop, so you would
have a hard time finding a trace of exim with ps.  To be sure that
exim really does his queue check upon ppp connection you could add
this short line to /etc/ppp/ip-up.d/exim:

touch /tmp/exim-did-run

To see the last time exim did run it's ppp trigered queue run do:

ls -l /tmp/exim-did-run

-- 
groetjes, carel


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Sounds of silence--I want volume!

[Carel Fellinger]

On Mon, Dec 09, 2002 at 12:20:38AM +, Chris Owen wrote:
> Haralambos Geortgilakis wrote:
...
> Try running (as root)
> chmod ugo+rwx /dev/cdrom

I think it's ill advice, not worthy of this list --sory for the rant,
it's not personally, it's just that you're not the first to give such
nonsensical advice, it seems that it's even a favourite one lately:(--
, to advice people to mess with those flags where the proper way is to
add users to specific groups.  In this case the cdrom group.  And
whilst the OP is add it, he might as well check he's in the audio group
too.


> If your sound card uses other devices, such as /dev/dsp or /dev/mixer, 
> you may need to do the same to these too.

I knew it:), to access those you need to be in the audio group.

Haralambos, maybe you should check your system again, and undo all those
chmod commands, and instead use the appropriate groups?

-- 
groetjes, carel


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Sounds of silence--I want volume!

[Carel Fellinger]

On Mon, Dec 09, 2002 at 04:13:50AM +0100, Carel Fellinger wrote:
> On Mon, Dec 09, 2002 at 12:20:38AM +, Chris Owen wrote:
...
> I knew it:), to access those you need to be in the audio group.
> 
> Haralambos, maybe you should check your system again, and undo all those

I ment both Hatralambos and Chris here.

-- 
groetjes, carel


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Sounds of silence--I want volume!

[Carel Fellinger]

On Tue, Dec 10, 2002 at 11:29:08AM +1300, Haralambos Geortgilakis wrote:
> 
> Hi All & Carel
> 
> who confesses to ranting.
> 
> Please don't.

I'm lost, don't do what?  confessing or ranting:)

> Your fix isn't.

You mean that adding youself to the group cdrom, changing back the
ownership/access permissions, logging in again, didn't result in sound?

 
> My 'cdrom" is actually a link to my ASUS "dvdrom."

Ah, then you'll have to look at the ownership/access permissions on the
device you're really using. ( /dev/hd[bc] are likely candidates)

 
> What I needed to do, I think, is create this group & then add myself to 
> it too?

These groups should exist, but yes, you should add all who need to
have sound and be able to access the cdrom. e.g.:

   # adduser your-user-name sound
   # adduser your-user-name cdrom


> So, how do I undo the previous cmd, just with a "-"?

argg, top posting, I'm lost again:) You want to get the proper
ownership/access permissions back on your dvd player? e.g.:

  ### let us asume your dvdplayer is on /dev/hdac
  # chown root:cdrom /dev/hdc
  # chmod ug=rw,o= /dev/hdc




-- 
groetjes, carel


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Sounds of silence--I want volume!

[Carel Fellinger]

On Mon, Dec 09, 2002 at 09:51:11PM +, Pigeon wrote:
> On Mon, Dec 09, 2002 at 04:13:50AM +0100, Carel Fellinger wrote:
...
> > I think it's ill advice, not worthy of this list --sory for the rant,
> > it's not personally, it's just that you're not the first to give such
> > nonsensical advice, it seems that it's even a favourite one lately:(--

Oh me and my big mouth, hope nobody calls for arguments:(

> > , to advice people to mess with those flags where the proper way is to
> > add users to specific groups.  In this case the cdrom group.  And
> > whilst the OP is add it, he might as well check he's in the audio group
> > too.
... 
> Why is this advice nonsensical, though? As you say, several people
> have given it recently. Rather fewer people have responded, as you did,
> saying it's a bad idea. Nobody has explained WHY it's a bad idea. What
> harm does it do if the world and his dog can read my CD-ROM?

I wasn't particular concerned with the full access part, but with
bypassing the group protection layer in general.  This layer is used
to make it possible to hand over parts of the hardware to selected
programs to do their low level things without handing over the entire
machine. It's also used to give individual users access to selected
programs and hardware parts.  Thus deviding and restricting access
helps managing security.

One can argue what the relevance is for say a cdrom, but e.g. for
disks it would be dramatic (giving all more or less root access).
Once people start circumventing it for simple things as sound, discs
will follow (IIRC even adviced on this list, though not recently)


-- 
groetjes, carel


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Sounds of silence--I want volume!

[Carel Fellinger]

On Tue, Dec 10, 2002 at 12:07:25AM +, Chris Owen wrote:
> >
> >
> >Carel Fellinger wrote:
...
> >>I think it's ill advice, not worthy of this list --sory for the rant,
> >>it's not personally, it's just that you're not the first to give such
> >>nonsensical advice, it seems that it's even a favourite one lately:(--
...
> Consider me humbly chastised for being "not worthy" of contributing to 

Never ment to imply you not being worthy etc.!  Guess I owe you an
apology then.  Sorry for the tone and such Chris.
 
> the list, I will wait until I get my PHd in Debian Linux before trying 
> to be helpful again.  However, perhaps you can explain to this unworthy 
> soul why my advice was "nonsensical"?  I merely assumed that, like 99.9% 
> of users, Haralambos wants all the users on his machine to be able to 
> access his CD-ROM...

I was reacting to the bypassing of the group machanism in general, not
so much in this particular case. Should have been much clearer on that.

Once people get used to bypassing the group based protection layer they
will do it in other cases too, atleast that's what I fear. (on the other
hand, I've seen people add themselves to the disc group, just to be able
to read scsi cdroms:)

Debian is set up to use the group system and uses it to make the
system more reliable and secure, and I think we as Debian users should
advice each other to go along with it, and not to undermine it.

> I accept that your method is more sophisticated, and even better; but 
> not that mine is nonsensical.  Mine does have the advantage that it 

I agree with you that in this particular case your's is not nonsensical,
I apologies again for wording it like that.  But I do believe it's the
wrong route.

> works, there's no problem with it that I can see, and he won't have to 
> remember this whole rigmarole again if he ever adds a new user then 
> wonders why they can't access the CD (wait, I'm sure I had this problem 

hm, it works, I admitt to that, and me too I don't see no problems in
this case, but to fool with access controll just to compensate for our
bad memory, yack:) I guess something can be done with adduser's config
file to *enhance* our memory instead...nop, let's try the man page
then...hm..costum script..yep.  Just do:

   # echo 'adduser $1 /usr/local/sbin/cdrom'  >>/usr/local/sbin/adduser.local

So there you have it, no need for your trickery:)

-- 
groetjes, carel


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: pon - Modem keeps redialing

[Carel Fellinger]

On Mon, Dec 09, 2002 at 07:34:55PM -0500, Chip Rose wrote:
...
> I've deleted all entries to exim in cron.weekly and cron.daily, the only
> places I found any reference to exim, and it still dials in every five
> minutes.  For now, I'll have to just let it dial away until I find the
> answer, but will mute the modem so I won't have to listen to the noise all
> the time. I'll run pon when completely away from the computer to keep my
> ISP from firing me for the incessant dialups!  Thanks for the response!

Could it be some DNS query?

-- 
groetjes, carel


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: pon - Modem keeps redialing

[Carel Fellinger]

On Mon, Dec 09, 2002 at 07:31:26PM -0600, John Hasler wrote:
> Carel Fellinger writes:
> > Could it be some DNS query?
> 
> Such as a browser tab loaded with to an auto-updating Web site?

Or exim or fetchmail doing some name lookup, or...

-- 
groetjes, carel


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: OT: mutt deletes folder when last message is removed

[Carel Fellinger]

On Tue, Dec 10, 2002 at 09:48:35AM +0100, Holger Rauch wrote:
...
> I've recently switched from Pine to mutt on Debian Woody and am now faced
> with the problem that as soon as I remove the last message from a folder,
> the folder is also deleted.
> 
> Any idea why this could happen and how to avoid it?

whilst in mutt press F1/help once and read to your hearts delight the
very thorough yet compact manual.  It's *all* defined there though it
might take some getting used to to appreciate its thoroughness.

>From that manual:

 6.3.193.  save_empty

  Type: boolean
  Default: yes

  When unset, mailboxes which contain no saved messages will be removed
  when closed (the exception is $spoolfile which is never
  removed).  If set, mailboxes are never removed.

  Note: This only applies to mbox and MMDF folders, Mutt does not delete
  MH and Maildir directories.


-- 
groetjes, carel


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: quick shell questions

[Carel Fellinger]

On Tue, Dec 10, 2002 at 01:00:51PM -0500, Drew Cohan wrote:
> 1.  How do I combine these two (JPG vs jpg):
> 
> for f in /path/to/*.JPG; do mv "$f" `date +%N`.jpg; done
> for f in /path/to/*.jpg; do mv "$f" `date +%N`.jpg; done

that's really simple:

  for f in /path/to/*.[jJ][pP][gG]; do mv "$f" $(date +%N).jpg; done

But, what's the %N format supposed to do? Can't find it in the docs.
Maybe you ment the -uTsec flag instead?  Anyhow, it seems that this
snipped will leave you with a very few files. as most get the same
name:]

> I'm trying to avoid duplicate filenames during my renaming sessions.  

Ah, demanding are we, he:) One way to keep the names uniq is to number
them, this will complicate the script beyond what I like for shell
scripting, so why not switch to the best you can get for (interactive)
scripting and go for Python instead?  But if you insist:

  n=1
  for f in /path/to/*.[jJ][pP][gG]; do
  mv "$f" $(date -uTsec)+$((n++)).jpg
  done


> 2.  What does the "2>/dev/null>&1" mean that I see in a lot of examples
> (eg ls 2>/dev/null/>&1).

I doubt you'll ever find that one, it looks like a contraction of:

  2>/dev/null  this redirects stderr to the bottomless bitbucket.

and

  2>&1 this redirects stderr to stdout, so it merges the two


> I really should take a shell scripting class.  :)

Nah, better go for Python (or Perl if you're into SM:)

-- 
groetjes, carel


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: C-_ on the console

[Carel Fellinger]

On Tue, Aug 12, 2003 at 06:32:38PM -0400, Brian White wrote:
> When I hit C-_ ("undo") in an X window, it works fine.  If I hit it from
> a console window, it does nothing and I have to hit C-- (no shift) for it
> to work.  I tried "M-x describe-key C-_" but it doesn't even seem to notice
> when I try to press C-_.

try

  $ dumpkeys | grep "keycode  12 ="

it should/could look like this:

  keycode  12 = minusunderscore   backslashControl_underscore 
Control_underscore

it means that that key produces - and when shifted _ and when altRed |
and whether Ctrled or shifted+Ctrled you get Ctrl-_

-- 
groetjes, carel


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Fetchmail stuck on bad messages

[Carel Fellinger]

On Mon, Aug 18, 2003 at 12:12:28AM -0500, Michael D Schleif wrote:
> ...Of course, I need something to act as MTA; but, what I am
> unclear about is that which exim contributes as LDA that cannot --
> readily -- be accomplished via procmail.  What are reasons to continue
> to use exim -- or equivalent -- between fetchmail and procmail?

It's a necessity in a multi-drop setup, i.e. getting mail for different
people in the same mailbox at your IPS, you need a MTA to deliver to
all the different users.

I for instance have one mailbox at my IPS with 5 aliases.  Being Dutch
and all, we use those aliases so all members of the family seem to
have there own email address, and I configured fetchmail to get the
mail and deliver depending on header info.  It used to work flawless,
but then the Envelop/Deliverd-TO header got mangled at my IPS and hence
meaningless:(, and now I've to rely on (incomplete) Received headers.


A second reason I prefer the setup I have, is that I can switch to
maildrop, gnus elisp filters or cyrus's sieve (imap site filtering)
without having to change a thing in my fetchmail setup.  If I were to
use procmail directly, I would have to both change my general mail
setup *and* my fetchmail/procmail setup.


-- 
groetjes, carel


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Fetchmail stuck on bad messages

[Carel Fellinger]

On Mon, Aug 18, 2003 at 10:53:24PM -0400, Bijan Soleymani wrote:
> On Tue, Aug 19, 2003 at 03:22:33AM +0200, Carel Fellinger wrote:
...
> > but then the Envelop/Deliverd-TO header got mangled at my IPS and hence
> > meaningless:(, and now I've to rely on (incomplete) Received headers.
> 
> Can't you use procmail to filter on headers. That's the point of
> procmail :)

Well, how does procmail deliver to other users?  Likely he asks
exim to do it, so why not let exim do it in the first place:)

Besides, when you have to rely on the Received headers things get very
complicated, you'll have to look at some of those headers, try other
headers all the same and still find the occasional email you can't
figure out where to sent.  That's all done *and* maintained in
fetchmail, and from what I remember it is not straightforward to do it
in procmail.

Ofcourse, with proper Delivered/Envelope-TO headers the proper procmail
recipe is dead simple.  But your IPS can change them overnight:(


 
> > A second reason I prefer the setup I have, is that I can switch to
> > maildrop, gnus elisp filters or cyrus's sieve (imap site filtering)
...
> Ahh well, that's ok. But the reason I like procmail is that I can use
> gnus, mutt, etc. All at the same time and have the filtering happen
> somewhere else, so I don't have to worry about looking at my mail in one
> program and having the filters in the other mess up.

Ah, but in my setup exim hands over the final delivering to procmail,
so I *do* use procmail for my filtering:)



-- 
groetjes, carel


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: OT: The Hurd

[Carel Fellinger]

On Thu, May 29, 2003 at 02:14:44PM -0500, Kent West wrote:
...
> I've wiped the drive and started over, and started a new, much cleaner 
> document. It can be found at 
> http://faculty.acu.edu/~westk/A_Beginners_Second_Attempt_to_Install_the_Hurd.html

looking at your scribbles, I noticed that one of your current beefs is
that you need to type in a lot to get grub to do its thing.  But that's
not how it's suppposed to be:)  You could do better and install grub on
a flop instead of merely dd-ing it to it, like (on a debian system):

# superformat --superverify  /dev/fd0  hd

# dd  if=/dev/null of=/dev/fd0  bs=512 count=1
# mke2fs  /dev/fd0

# mount /floppy
# mkdir -p /floppy/boot/grub

# cp /usr/lib/grub/i386-pc/stage{1,2} /floppy/boot/grub

# cat >/flop py/boot/grub/menu.lst <

Re: boot-floppies package

[Carel Fellinger]

On Sun, Jun 01, 2003 at 06:06:08AM -0400, Haines Brown wrote:
...
> If you do have a running system, then simply have root visit grub's
> working directory and copy the two stage files over to an unmounted
> floppy:
> 
>  # dd if=stage1 of=/dev/fd0 bs=512 count=1
>  # dd if=stage2 of=/dev/fd0 bs=512 seek=1
> 
> Then mount the floppy and copy splash.xpm.gz to it, create a file
> named grub.conf and a symlink to it from menu.lst. And then set up the
> grub.conf file as per the manual.

I seriously doubt this is sane:)
According to the grub.info (node: Images):

 While Stage 2 cannot generally be embedded in a fixed area as the
 size is so large, Stage 1.5 can be installed into the area right
 after a MBR, or the boot loader area of a ReiserFS or a FFS.

In other words, putting stage2 directly after the bootrecord destroys
any filesytem on the disk, better use stage1.5 or do a true install on
the floppy as I posted recently or as described in the FAQ (node FAQ).

-- 
groetjes, carel


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: boot-floppies package

[Carel Fellinger]

On Sun, Jun 01, 2003 at 11:14:20AM -0400, Haines Brown wrote:
...
> > I seriously doubt this is sane:)
> > According to the grub.info (node: Images):
> > 
> >  While Stage 2 cannot generally be embedded in a fixed area as the
> >  size is so large, Stage 1.5 can be installed into the area right
> >  after a MBR, or the boot loader area of a ReiserFS or a FFS.
> > 
> > In other words, putting stage2 directly after the bootrecord destroys
> > any filesytem on the disk, better use stage1.5 or do a true install on
> > the floppy as I posted recently or as described in the FAQ (node FAQ).
> 
> Ouch! I hope I didn't give bum advice :-(. I understand your point
> about the MBR, but if the grub files (besides the two files above, I
> also copied over the 1_5 fs drivers for about 200 Kb altogether) are
> on diskette, and I don't see how the warning would apply. I've been
> merrily using the floppy to boot both RedHat (ext3) and Debian (ext3
> and reiserfs).

As stated before, there is really limited space on a floppy in between
the MBR and the actual file system.  (Oh, and I'm not even sure there
is enough room there on a floppy to stack stage1.5)  Depending on the
size of stage2 you're very likely to have overwritten part of the
filesystem.  Whether you'll ever notice this depends on your usage,
but to me it seems a disaster waiting to happen.  Whether the grub
files are somewhere else on that disk really doesn't matter, you've
ruined the filesystem.

The ruined file system might lead to you overwriting part of stage2
whenever you put something on the floppy, like changing menu.lst,
or putting the other grub files on disk.

Ofcourse if you first put those files on the flop, change menu.lst,
and only then use dd to put stage1 and stage2 in place, you might
be lucky and overwrite only some of those superfluous grub files.


-- 
groetjes, carel


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: boot-floppies package

[Carel Fellinger]

On Sun, Jun 01, 2003 at 04:50:58PM -0400, Haines Brown wrote:
> Carel,
> 
> Now I'm really getting worried, and trying to recall were I picked up
> the suggestion to create a grub floppy (it wasn't my own idea). I was
> from Linux Journal, but I can't seem to pin down the issue.

I'm not sure I understand your fear, there is nothing wrong with a grub
boot floppy, on the contrary, it's very very handy, has saved my day on
many occasions, so I hardly leave the house without one.

But, when you "dd" stage2 right after stage1, that is right after the
MBR, then you're very likely to be overwriting the FAT tables
(assuming we're dealing with DOS formatted floppies here).  Ofcourse
it all depends on the way the floppy is formatted, check it with
"minfo" and look for "reserved" and "hidden" fields (normally 1 sector
is reserved for the MBR and 0 for hidden sectors).

I don't know the details of ext2fs, but I assume there too not a lot
of space is waisted;)


> I also find a reference to doing so in the clone-HOWTO, section
> 5.3. Also
> http://www.mathematik.uni-marburg.de/sys/os/linux/install/mkgrubfd.html 

After some puzzled looks, I conclude that they err.  At one time they
were describing how to create a bootflop for a RedHat system. At that
time they were using a setup where stage2 was to be found on hd0.
Then it's alright to remove stage* from the floppy after the setup as
those files weren't used to begin with.

Later they changed to a full floppy based setup much like the way I
suggested in an earlier post this week.  Now stage2 on the flop is
used, so removing it is *bad* advice.

[stage1 is always copied to the MBR, so it doesn't matter where it
 was during setup]

> here's another venture:
> http://www.uruk.org/orig-grub/install.html
> I find that the Linux Journal article was from the May issue:
> http://www.linuxjournal.com/article.php?sid=4622
>  
> These discussions propose two or three somewhat different approaches
> to creating the floppy, and I see that I didn't use the Linux Journal
> method after all. But these methods would not see to have any bearing
> on your concern about overwriting the floppy's filesystem.

Well, they don't mention it, but they do take precorsions not to fool
things up:)  The linux journal does like I posted, which takes care
not to overwrite the filesystem, it merely puts stage1 into the MBR
and alters the MBR to point to stage2 on the floppy.  www.uruk.org on
the otherhand explicitly warns:

  NOTE: This will destroy any data currently on the floppy.

So, yes they use dd for stage2, but they know that doing so will
destroy any filesystem currently on the floppy.  There is no way
you can safely add a menu.lst file on that flop and use it.

 
> If stage2 does intrude on the file system, what exactly do you mean by
> the "file system?" The diskette's partition table? The 1_5 fs drivers?

the FAT tables and the root directory to begin with (unless there were
enough sectors 'reserved' or 'hidden').
  

> > The ruined file system might lead to you overwriting part of stage2
> > whenever you put something on the floppy, like changing menu.lst, or
> > putting the other grub files on disk.
> 
> At least you suggest the floppy is in danger of failing. But beyond
> ending up with a non-functioning floppy, are you suggesting it might
> affect the file system on the hard disk? If so, how?

I can't remember suggesting such a thing!
 

> > Ofcourse if you first put those files on the flop, change menu.lst,
> > and only then use dd to put stage1 and stage2 in place, you might be
> > lucky and overwrite only some of those superfluous grub files.
> 
> I'm not sure what surperfluous files you refer to. I run multiple file
> systems and so need some of the drivers, but, of course not all. But

stage1 and stage1.5* are superfluous.  stage1 is needed during setup,
is copied to the MBR and updated to know where stage2 resides.
stage1.5* can only be put to use on harddisk as there is some spare
room between the MBR and the actual file system, during setup it is
copied into that spare room and stage1 is linked to it instead of to
stage2.  This is very usefull as stage1.5 is able to find stage2 on a
file system without having to resort to having to list the sectors it
lives on, so one can e.g. upgrade stage2 without having to run setup
again.  Maybe you should try to read more on it in the fine manual?


> since one can never know, you are implying that a grub floppy is a BAD
> IDEA to begin with? 

What on earth have I done to make you think such evil thoughts:)


-- 
groetjes, carel


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: OT: The Hurd

[Carel Fellinger]

On Wed, Jun 11, 2003 at 09:41:25AM -0500, Kent West wrote:
> Carel Fellinger wrote:
... snipped all but the strangly important dd line
> >   # dd  if=/dev/null of=/dev/fd0  bs=512 count=1
...
> Thanks for the response, Carel; I didn't mean to leave you hanging for 
> so long without an acknowlegement, but I just haven't had time to get 
> back to playing with the Hurd/grub. Looks like your post will help a lot 
> when I get back to it though; thanks!

your welcome:) Oh, and when you do make a grub flop like this, be sure
not to skip the above "dd ... count=1" line when you opt for having a
non fat filesystem on that flop.  It took me quit some time to find
out that that was wat kept mount and grub from recognising the ext2 fs
I put on it.

-- 
groetjes, carel


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

init=/bin/bash not working with initrd?

[Carel Fellinger]

Hai,

when tonight I tried to boot into my system bypassing init I found to
my surprise that "init=/bin/bash" doesn't work anymore.  Atleast not
with kernel-image.2.4.20 and initrd.  It still works with 2.2 kernels
without initrd.  Googling and searching the deb archives didn't give
me any hint why it stopped working, but I suspect it has to do with
initrd.  Anyone in the know wanting to share his/her insights?


-- 
groetjes, carel


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: finding network activity

[Carel Fellinger]

On Sun, Dec 15, 2002 at 12:45:48PM -0700, John Schmidt wrote:
> Hi,

Hi to you too.  Let me warn you first, me head is not working properly
so what I say may make no sense, but...

> I have demand dialing turned on and would like to determine why ppp is 
> starting up at 15 minute intervals.  I am looking at syslog and not 
> seeing anything that would initiate outside traffic.  I also have 
> tcpdump -i any turned on and not seeing anything either.  I removed 

...not sure, but the other month when I had similar problems I noticed
that "tcpdump -i" was of no use as the interface wasn't up yet so
tcpdump didn't monitor the ppp line, or the line was up and ready
allright but the culprit of bringing it up had vanished before I could
type tcpdump:( IIRC I switched off demand dailing temporarily, brought
up ppp, started tcpdump and waited for 15 minutes, and bingo: DNS:)


-- 
groetjes, carel


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: /etc/dhcpc/dhcpcd-eth0.exe

[Carel Fellinger]

On Wed, Dec 18, 2002 at 01:58:32PM -0800, Expert User wrote:
> When the dhcpclient used to be dhcpcd, there was a way to run some
> script after the dhcp has run by putting a script in /etc/dhcpc/dhcpcd-eth0.exe.
> 
> Now that I have dhclient, how do I achieve the same result?
> 
> I learned from the man page that there is /etc/dhclient-exit-hooks
> script I can create, but how to associate it with a perticular
> interface?

check $reason and $interface within your /etc/dhclient-exit-hooks script?


-- 
groetjes, carel


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: ntpdate tip -- hmm? [NEW TOPIC]

[Carel Fellinger]

On Sat, Dec 28, 2002 at 11:28:26AM -0700, Bob Proulx wrote:
...
> However, the pointer to rtfm is probably too much for most newbies.
> It would be great if ntpdate and ntp-simple both offered to do the
> very simple configuration of using the existing DNS servers as NTP
> servers and having the debconf set up the configuration files
> accordingly.  I think I will wishlist that.

I doubt that's a good idea:(

I tested a few ISP's here in Holland, and none of them had their DNS
servers double as a NTP server.  And yes, atleast one of them is a
unix based ISP, so there goes your argument that unix based IPS's
would do it like you claim:)


-- 
groetjes, carel


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Removed from lists due to email errors.

[Carel Fellinger]

On Sun, Jan 12, 2003 at 12:37:35PM -0800, Xavian-Anderson Macpherson wrote:
> "Your mail address, [EMAIL PROTECTED], has been removed
> from the following mailing lists, because it generated an
> exessive number of bounced mails:"

bounced messages are normally not generated by mail clients like Kmail
and mozilla, but by mail servers like exim or -most likely- the one at
your IPS.  One reason for your IPS's mail server to bounce messages is
that your mailbox overthere is full.

 
> Someone else mention the same thing to me before, but I have no idea 
> what is causing it or how to fix it.  I have no idea what I may have 
> done to cause this.  All I know is that Kmail wound up duplicating 
> emails in my mail box.  So I went back to using mozilla.  I didn;t know 
> that it was causing the same problem remotely.  But I guess that is 
> what's happening.  So how do I fix this?  Please help!!

more likely, but all depending on your particular setup, there is a
hickup in the process getting your email from your IPS to your system.
I once had a message that gave rise to an error during retrieval, my
fetchmail process barked off leaving all just retrieved mail on the
server.  So the next run I got the same mail (duplicates all over the
place:) and the same error until I deleted the offending mail in my
IPS's mailbox by hand.


-- 
groetjes, carel


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Rack mounted backend ideas

[Carel Fellinger]

On Mon, Jan 13, 2003 at 01:59:15PM -0500, Mark L. Kahnt wrote:
..
> The current desktop box would become a combination firewall and X
> terminal, but it all raises one key question: with the second rack
> computer being meant to be rebooted regularly as I switch between
> various o/s, how reliable is it to control the boot process/selection
> from a separate machine? I know that Lilo and Grub do have the "serial"

Not sure where your doubts stem from, but the only reason to have a
screen and keyboard attached to the machine it self is to fiddle with
the BIOS.  Linux doesn't care, log messages can be directed to the
serial port.  Grub and lilo are capable to read from the serial port to
get there instructions, heck, they're even capable during boot to let
you decide to switch to console or serial.

And ofcourse running programs on those servers with there output directed
to your X-terminal is just a matter of setting it up that way:)

> option for sending the info to another machine (I'm hoping that means
> that Minicom would handle it fine.) Any heads-up of use, given that it

yep, minicom is fine

> is one thing to point clients to the HOWTOs when setting things up, but
> I haven't regularly run a system this way - only installed and
> troubleshot them?

Here me English is lacking, atleast I can't parse the above.

Do or don't you have experience with using a headless server?

Do you have clients in the sence of costumers that you want
to use those rack mounted systems?


-- 
groetjes, carel


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Rack mounted backend ideas

[Carel Fellinger]

On Mon, Jan 13, 2003 at 11:18:58AM -0800, nate wrote:
> Mark L. Kahnt said:
> 
> 
> > a separate machine? I know that Lilo and Grub do have the "serial" option
> > for sending the info to another machine (I'm hoping that means that
> > Minicom would handle it fine.) Any heads-up of use, given that it is one
> 
> I haven't tried GRUB, but with LILO the serial options doesn't let me

GRUB is fine, I use it and it nevers fails:)

> interact with lilo, it just spits the information out to the serial port.

This I find hard to believe, did you try to send a break?  According to
the lilo docs one needs to send a break instead of pressing shift to get
lilo's attention.

-- 
groetjes, carel


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Rack mounted backend ideas

[Carel Fellinger]

On Mon, Jan 13, 2003 at 04:46:14PM -0500, Mark L. Kahnt wrote:
> On Mon, 2003-01-13 at 14:48, Carel Fellinger wrote:
...
> > Here me English is lacking, atleast I can't parse the above.
...
> Yeah, I did say something weird there :)
> 
...sniped the explanation

Okee, I see.  In my experience most linux related things go swell with
a headless box.  Bootup selections are easily handled over the serial
line, (root) login over the serial line is almost like the console but
for the fact that I haven't managed yet to get minitel to deal with
colors nor to pass on Home, Page-up, Ctrl-Alt-Del, etc keys.  Not that
big of a problem for me, as I very often log in onto such system with
ssh over the ethernet from a normal console or from an xterm.

The one thing that's really troublesome is things the BIOS tries to
tell, like the other day I had a harddisk start to fail, and the BIOS
tolled me so thanks to S.M.A.R.T.  But on a headless system such
messages would have been invisible, and depending on your BIOS you
have a change that the BIOS will keep waiting for you to press a
key:(, preventing it to complete the booting.  There is some S.M.A.R.T.
aware software under linux, but I haven't tried it out yet.

So yes it works. Have it hooked up onto your local network to easily
access the servers using ssh (and X-forwarding), use a serial line for
boot control and root access, enable S.M.A.R.T. under linux, and prey
the linux S.M.A.R.T. software detects harddisk failures before the
BIOS does.


-- 
groetjes, carel


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Switching smart relays automatically in Exim...

[Carel Fellinger]

On Sat, Jan 18, 2003 at 07:26:00AM -0500, Darryl L. Pierce wrote:
...
> My question, is there a way to configure Exim to use my ISP's SMTP 
> server and my SMTP server at work without having to manually change 
> settings when I go from one place to the other?

There are many ways to achieve this, most with their pecularities:)
But assuming you're not trying to send a million emails, the router
option "queryprogram" might be of use to you.  You'll have to build
your own queryprogram to check which mail server to use, e.g. by
checking IP nrs.

-- 
groetjes, carel


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

security leak in ppp.log file

[Carel Fellinger]

Sorry for reposting this question, but somehow my posts to the newsgroup
never make it to the list and many posting of others seam to miss too:(,
so please reply by email too. thanks


Hai,

for starters: the longer I have Debian GNU/Linux the happier I get,
  and the more often I find what I need in the docs,
  but not this one:(

Recently my IPS changed from chat to PAP, and now I find my password
in the ppp.log file. I know that file has restricted access rights,
but I prefer to have no passwords lying around, besides Debian ships
with an X configuration where the console messages are routed to an
xterm on the XDM login screen, so there they are for everyone to read!
Previously I used chat scripts only (no PAP authentication once logged
into my IPS), and in there you have flags (\q) to suppress listing of
the password in the ppp.log files. How to achief this with PAP or CHAP?

---

I got this one reply of Eric G. Miller suggesting to use pppconfig and
have the password and userid in the pap-secrets file. Unfortunately that
was no cure, as that was exactly what I already had done:(.

the ppp.log file reads something like:

Jun 27 13:30:21 vvs pppd[16672]: Using interface ppp0
Jun 27 13:30:21 vvs pppd[16672]: Connect: ppp0 <--> /dev/ttyS1
...
Jun 27 13:30:43 vvs chat[16672]: send (\d)
Jun 27 13:30:44 vvs pppd[16671]: Serial connection established.
Jun 27 13:30:45 vvs pppd[16671]: Using interface ppp0
Jun 27 13:30:45 vvs pppd[16671]: Connect: ppp0 <--> /dev/ttyS1
Jun 27 13:30:45 vvs pppd[16671]: sent [LCP ConfReq id=0x2  ]
Jun 27 13:30:46 vvs pppd[16671]: rcvd [LCP ConfAck id=0x2  ]
Jun 27 13:30:46 vvs pppd[16671]: rcvd [LCP ConfReq id=0x28]
Jun 27 13:30:46 vvs pppd[16671]: sent [LCP ConfAck id=0x28]
Jun 27 13:30:46 vvs pppd[16671]: sent [LCP EchoReq id=0x0 magic=0x1b9a3fac]
Jun 27 13:30:46 vvs pppd[16671]: sent [PAP AuthReq id=0x1 user="my-user-id" 
password="my-password"]
Jun 27 13:30:49 vvs pppd[16671]: sent [PAP AuthReq id=0x2 user="my-user-id" 
password="my-password"]

and there they are, so what did I do wrong? how to prevent this?
just ran pppconfig, selected PAP protocol and kept all the default awnsers.
by the way, I'm still running (mainly) ham (Debian 2.0)

-- 
groetjes, carel


-- 
groetjes, carel

security leak in ppp.log file

[Carel Fellinger]

Hai,

thanks for all the (email-) replies, it's fixed now.

As one told me it is indeed a flaw in older versions of pppd,
dealt with in later versions with an extra flag 'hide-password'.
To circumvent it with older versions you simply have to edit
/etc/ppp/peers/provider and comment the 'debug' option out.

-- 
groetjes, carel

Apt-cdrom and Ham cd's

[Carel Fellinger]

tired of multi-cd/dselect's inability to handle debian cd's, I downloaded
alt-0.3.3. But elas, apt-cdrom doesn't find the package files on the cd's.
what magic want do i have to use to direct apt-cdrom to the proper place
so that it can find the package files? And related, were can I find a
description of how debian cd's are structured. I could find no note on
where the package files should live, nor on the .disk subdir.

thanks for any pointers.

I'm running a partly upgraded HAM, and try to use non standard ham-cd's from
Datom/Schlittermann.
-- 
groetjes, carel

Re: E-mail client for POP3?

[Carel Fellinger]

Kent West <[EMAIL PROTECTED]> wrote:
...
> Which boils down to this: fetching from the server is not a real
> good solution for me. Is there a text-based email client that
> uses IMAP rather than POP? Alternatively, if fetchmail had an

you could consider using Pine. It's IMAP based, though not completely
free if i remember correctly.

-- 
groetjes, carel

Re: tofrodos

[Carel Fellinger]

[EMAIL PROTECTED] wrote:
> does anyone know where i can find the util "tofrodos".  it converts text
> files between Dos and Unix Formats.. todos - fromdos.

never heard of tofrodos, but in HAMM you get:

zgrep todos /var/lib/dpkg/Contents-Debian2-hamm.gz
/usr/bin/todos  utils/sysutils
/usr/man/man1/todos.1.gzutils/sysutils


-- 
groetjes, carel

Re: dselect/apt/cdrom

[Carel Fellinger]

Martin Fluch <[EMAIL PROTECTED]> wrote:
> You should install the dpkg-multicd package and than choose the multi_cd
> access methode in dselect.

or you could do as I was advised, get apt-0.3.12, compile it your self and
follow the apt-cdrom man page. works swell on debian-hamm. solved all the
problems i had with i(multi_cd) not installing things that i told it too
with dselect. YMMV

-- 
groetjes, carel

Re: LILO won't load win95, but no error...??

[Carel Fellinger]

Martin Waller <[EMAIL PROTECTED]> wrote:
> Oops - you're quite right.

> fdisk /MBR from a dos boot disk...?

>>From: Tom Pfeifer <[EMAIL PROTECTED]>
>>>This is a guess based on your description of what happens. It almost
>>sounds like you may have, at one point, had this line in your lilo.conf
>>file...
>>
>>boot=/dev/hda1

if the above is the case an easier way is to uninstall lilo as follows:

lilo -u /dev/hda1

this will restore the saved boot sector from /boot/boot. of the device
specified with the -u option.

-- 
groetjes, carel

Re: Naming convention invalid (local) domain

[Carel Fellinger]

On Tue, Sep 24, 2002 at 01:16:52AM +0200, Robert Ian Smit wrote:
...
> I am using dyndns, and was wondering if I could use my subdomain
> from that service as a domain. I only use dyndns to be able to log

yes ofcourse (I take it you mean .dyndns.org).  You know,
a valid domainname doesn't imply there is a server out there with that
name, well hell, there even doesn't have to be a machine at all.
The important thing is that you don't use someone else's domain!
Otherwise all complains about your net behaviour would get directed
to the wrong guy:)


> I could ofcourse also make up a domain name. For instance robian.inv
> or something like that. Would that make more sense than just @robian
> in the MID? I want to be sure that should my domainname ever "get out",
> (i.e. mail headers) I don't cause problems or confuse people.

The official sanctionized way of doing it is using *your* real
domainname.  If you don't have a domainname, or don't want to use it,
add ".invalid" to whatever hostname you happen to use indicating to
the world out there it's a fake domain.  By some RFC it's spelled
".invalid" instead of your ".inv", better stick to that.

If it's just for mail you can get away by letting exim rewrite all
headers of outgoing mail.  Take care that you include atleast a valid
return adress, and often a (seamingly) valid Envelope From adress is
needed too, but all others may be fake ones preferrrably ending in
".invalid".


-- 
groetjes, carel


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Naming convention invalid (local) domain

[Carel Fellinger]

On Tue, Sep 24, 2002 at 09:01:54AM +0200, Robert Ian Smit wrote:
...
> I think that would be the address rewriting rule that by default
> looks in /etc/email-addresses. I never liked it very much to have
> user stuff like that in /etc, but at the time it was the only way I

That's because it's here were user and system administrator meet:) In
the old days the mailer software would accept what ever to user put in
the From header.  But too many peole misbehaved and mailed using names
of other people.  So the Sender header came to be, and flags to tell
the UMA and to TMA to add it when needed.  The default value for those
flags is often 'add', but you as a system administrator can change this
at will, sometimes even for specific users.  But be warned, very often
this flag is used to (dis)allow other things too, like in exim the
'trusted-users' flag allows special flags to pass in the call to exim.

> knew how to have the "smarthost" at my isp accept my relay requests.
> At the time, I didn't know that the MID from my local exim would
> show up in the headers which potentially makes some people think
> less of me (i.e. borked-message-headers highlighting in mutt)

MID is Message ID?
I used to tackle this by applying a sed filter to the mail just prior
to handing it over to my uucp setup.  Nowadays I don't think it's that
important anymore, many IPS's adding their own ID's in the Received
headers and all.

> Eventually I want to be independent of their mail infrastructure, but
> I don't feel opening my MTA to the outside world is a good idea for
> now.

If you just stick to sending mail out and leave the receiving wows to
your IPS I don't think you're really opening up your MTA to the
outside world.  Just aslong as exim isn't listening on an outside port
you should be safe I guess (but I'm no mail and certainly no security
guru).

-- 
groetjes, carel


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Woody IPv6 problem finally, really, solved

[Carel Fellinger]

On Thu, Sep 26, 2002 at 11:14:27AM +0200, Jan Willem Stumpel wrote:
...
> With the help of the IPv6 developer list I finally found the real 
> solution: Woody versions of some Debian packages, like telnet, 
> and, it seems, exim, *require* IPv6 addresses for your local 
> machines (including localhost!) in /etc/hosts, or they will call 
> your ISP to ask for them. And of course your ISP's nameservers 
> don't have them either. Specifying local lookup ("files") in 
> /etc/nsswitch.conf does not help.

Thanks for clearing this up, but...
 
> So if your /etc/hosts is, for instance:
,,,
> 192.168.1.3 venus.my.homevenus
> 
> Then at the end of the file you have to add
...
> :::192.168.1.3 venus.my.homevenus

...now there is one name with to addresses, how will other programs react
to that. e.g. dnsmasq?

-- 
groetjes, carel


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Email (was Setting up Exim was Setting up Sendmail)

[Carel Fellinger]

[ sorry for sending this offlist too, but as you have email problems I'm not
  sure list trafic will reach you.  If it does, pray tell and I wil adjust ]

On Tue, Oct 01, 2002 at 04:47:48PM -0700, Michael Olds wrote:
...
> I have Exim set up. I believe it is set up properly as I used the defaults
> mostly. I can send mail from my Linux box, but I cannot receive mail, either
> internally or from the outside. The log indicates that mail is being

so it isn't set up properly:)  For starters, we need some more info.
You say you've used the defaults during exim install/setup.  Did you
choose the "Internet site" or "Satellite" or...

The next question is, how much do you need/want your machine to do:)
I normally prefer to have an ISP take care of all delivery and receiving
problems, but others really want to run their own full blown mail server.


> received and is being delivered no errors. I am using the Kmail e-mail
> client, which I have uninstalled and re-installed a couple of times just to
> check. If there is a better client I am listening. I am sure this is

The email client just hands it over to a email server, and depending
on your configuration of your email client that could well be your
IPS's mail server.  To shortcircuit such debugging fun, you could try
to use "mail" (install mailutils or mailx if you're missing this one)
like this:

   # to check the mail gets delivered open a log window and enter your passwd
   $ xterm -e su - root "tail -f /var/log/exim/*log" &

   # send a mail from and to your normal non root login
   $ echo just testing | mail -s test $USER

   # look at the log window, this should look like:

==> /var/log/exim/paniclog <==

==> /var/log/exim/rejectlog <==

==> /var/log/exim/mainlog <==
2002-10-02 02:20:10 17wXFK-0005dU-00 <= [EMAIL PROTECTED] U=carel P=local S=319
2002-10-02 02:20:11 17wXFK-0005dU-00 => [EMAIL PROTECTED] <[EMAIL PROTECTED]> 
R=smarthost T=remote_smtp H=schaduw.felnet [192.168.1.1]
2002-10-02 02:20:11 17wXFK-0005dU-00 Completed


> something silly, like permissions, but I do not know where to look next for
> the solution. As far as I can tell the relevant directories and files are
> user me and group mail, not sure if it is correct or safe, but also with
> read, write, and execute for both.

I'm not sure what files you're talking here, but I sure hope your not
refering to e.g. files in /etc?
 

-- 
groetjes, carel


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Email (was Setting up Exim was Setting up Sendmail)

[Carel Fellinger]

On Tue, Oct 01, 2002 at 06:02:00PM -0700, Michael Olds wrote:
...
> You say you've used the defaults during exim install/setup.  Did you
> choose the "Internet site" or "Satellite" or...
> 
> So "Internet" or what was the #1 choice.

yep, that's "Internet site".  The proper general choice for what you want.

> 
> The next question is, how much do you need/want your machine to do:)
> I normally prefer to have an ISP take care of all delivery and receiving
> problems, but others really want to run their own full blown mail server.
> 
> In general what I am aiming to do is to get to the point where I have a self
...snipped

Okee so you want to have a full blown mail server, relaying mail for
your own domain (that Windows machine), delivering locally and sending
and receiving from the net.  That's a lot to check out, so let's break
it up and start of with checking local delivery on the linux box...


> 
...
> ==> /var/log/exim/mainlog <==
> 2002-10-02 02:20:10 17wXFK-0005dU-00 <= [EMAIL PROTECTED] U=carel P=local
> S=319
> 2002-10-02 02:20:11 17wXFK-0005dU-00 => [EMAIL PROTECTED]
> <[EMAIL PROTECTED]> R=smarthost T=remote_smtp H=schaduw.felnet
> [192.168.1.1]
> 2002-10-02 02:20:11 17wXFK-0005dU-00 Completed
> 
> Little confusion here, but it comes down to the fact that the log reads as
> does your last entry here.

...so the local delivery test seems to work, i.e. exim is happy.
Though I'm not sure how versed you are in reading the logs, (i'm
not:), and maybe you're overlooking something here (the logs should be
quit different, for one there shouldn't be a R=smarthost entry), so
could you share the relevant part of the log?


> I am talking about when I send mail from the Windows machine to the Linux
> box, the linux box is recording such entries in /var/log/exim/mainlog and
> similar entries for mail being sent from root to userme. I am just not able
> to get them to show up in my email client. Again, I can send from Linux to
> the Windows box no problem.
> 
> 
> 
> > something silly, like permissions, but I do not know where to look next
> for
> > the solution. As far as I can tell the relevant directories and files are
> > user me and group mail, not sure if it is correct or safe, but also with
> > read, write, and execute for both.
> 
> I'm not sure what files you're talking here, but I sure hope your not
> refering to e.g. files in /etc?
> 
> in /etc/exim
> I have permissions for /etc/exim/ user root rwe; group mail re (is it Debian
> or Krusader that gives permissions in terms of "show entries" "write
> entries" "enter" why not use the numbers, or the usual rwx?)

most likely it's Krusader

> for /etc/exim/exim.conf user root rw; group root r
> for /home/userme/mail user me rwe group mail rwe

why group mail?  Is it your system mailbox?

> for /var/mail/userme user root rwe group mail rwe set Gid others re sticky

user root? Gid? e bit?  Are You using maildirs?
I think even for maildirs the owner of /var/mail/userme should be userme

And talking about defaults, I doubt debian's exim setup defaults to maildir.
And I doubt it's default is to use ~user/mail as system mailbox. 

So I'm wondering, how did you manage to get those, e.g. what email server
was installed prior to exim, did you change things yourself?

-- 
groetjes, carel


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Email (was Setting up Exim was Setting up Sendmail)

[Carel Fellinger]

On Wed, Oct 02, 2002 at 05:05:03AM +0200, Carel Fellinger wrote:
...
> user root? Gid? e bit?  Are You using maildirs?
> I think even for maildirs the owner of /var/mail/userme should be userme

Moreover when the owner is root, it's highly unlikely that a mer userme will
be able to read it.  So change it to owner userme and try again.
 
> And talking about defaults, I doubt debian's exim setup defaults to maildir.
> And I doubt it's default is to use ~user/mail as system mailbox. 
> 
> So I'm wondering, how did you manage to get those, e.g. what email server
> was installed prior to exim, did you change things yourself?

Oh, and could you share the output of:

  $ ls -l /var/mail/userme ~userme/mail

both prior and after you send userme some mail?
If the outpur differs, mail is delivered.

-- 
groetjes, carel


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Email (was Setting up Exim was Setting up Sendmail)

[Carel Fellinger]

On Tue, Oct 01, 2002 at 09:45:37PM -0500, Shyamal Prasad wrote:
...
> One other thing: I noticed the word 'smarthost' in the log file
> entries you posted on this. Does you use a smarthost or not?

nah, that was an example log snipped from *me*, not from Michael

-- 
groetjes, carel


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Email (was Setting up Exim was Setting up Sendmail)

[Carel Fellinger]

On Tue, Oct 01, 2002 at 08:35:37PM -0700, Michael Olds wrote:
> Carel:
> 
> OK! I ran ls -l... with these results:
> 
> mail  0   datedrafts
>   0   inbox
>   840 sentmail
>   1613trash
> 
> I looked in the trash directory and there was nothing there.
> 
> But this looks like it is giving me what is in /home/userme, not what is in
> /var/mail/userme

I've to guess here, cause I can't see the real output, nor the exact
command you gave, but it looks like you just listed ~userme/Mail.  I
guess some more, you're doing your mail on a windows machine and have
a hard time copying output from the linux box into the email:) I sure
hope you've two machines running and don't need to reboot while doing
email just to run some checks.

So let's make the checks simpler and less output copious:)

1) On the linux box login as userme.

2) To check if it's a normal mailbox or a directory or whatever, use:

  $ file /var/mail/userme

   On my machine this looks like:

  $ file /var/mail/userme
  /var/mail/carel: ISO-8859 mail text, with very long lines

   so it's in standard mbox format.  If it were in maildir format, it
   would have been a directory and the output would likely look like:

  $ file /var/mail/userme
  /var/mail/carel: setgid directory

3) To see whether new mail arives, use:

 $ du /var/mail/userme
 $ echo just testing | mail -s test userme
 $ du /var/mail/userme

   On my machine this looks like:

 $ du /var/mail/carel 
 132/var/mail/carel
 $ echo just testing | mail -s test carel
 $ du /var/mail/carel 
 136/var/mail/carel

4) To look at the mail itself we need to know the format of the mail spool.
   in case it's in mbox format (see test 2), use:

 $ tail /var/mail/userme

   on my machie this looks like:

 $ tail /var/mail/cf 
 id 17wiLZ-yt-00
 for <[EMAIL PROTECTED]>; Wed, 02 Oct 2002 14:11:21 +0200
 To: [EMAIL PROTECTED]
 Subject: test
 Message-Id: <[EMAIL PROTECTED]>
 From: Carel Fellinger <[EMAIL PROTECTED]>
 Date: Wed, 02 Oct 2002 14:11:21 +0200

 just testing


If the above 4 steps yield similar results, then exim is definitely
working as it should, and you can move on to fixing your email client
setup.

-- 
groetjes, carel


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Email (was Setting up Exim was Setting up Sendmail)

[Carel Fellinger]

On Wed, Oct 02, 2002 at 06:09:07AM -0700, Michael Olds wrote:
> Good morning Carel,
> 
> Live on the East Coast?

nah, I'm living in the Netherlands (Europe), but I've strange sleeping habits:)

> Thank you again for your careful attention to this! I have followed your
> instructions and the output would indicate that Exim was running ok. (??) My
> output next to yours below:

All in all it looks like exim is working fine, but...
 
> 1) On the linux box login as userme. X
> 
> 2) To check if it's a normal mailbox or a directory or whatever, use:
> 
>   $ file /var/mail/userme  X
> 
>On my machine this looks like:
> 
>   $ file /var/mail/userme
>   /var/mail/carel: ISO-8859 mail text, with very long lines
> 
>   On mine it was similar:
>   /var/mail/userme ISO-8859 mail text but nothing else

Okee, so it's a standard mailbox.
To finish this part of let's double check on ownership and protections. It
should be like this:

owner: userme
group: mail
protection: -rw-rw
   user:  read, write
   group: read, write
   others:


> 3) To see whether new mail arives, use: X
> 
>  $ du /var/mail/userme
>  $ echo just testing | mail -s test userme
>  $ du /var/mail/userme
> 

What bothers me is that "du /var/mail/userme" shows the file hasn't
grown...some thinking later, ofcourse, it's a large file by now, so
du shows the size in kilobytes, and such a small test message isn't
going to change the size that much, so goes by unnoticed:(

If you want to be dead sure, you could rerun step 3, but instead of
"du /var/mail/userme" now switch to "du -b /var/mail/userme".  And for
added fun instead of "just testing" use some other odd string, like
"last call for alcohol" so as to make it simpler to see with step 4
that it's really the newly sent message that's in your mail spool?

So presuming the above few extra checks are okee, lets move on to step
6, and try to use a simple email client to look at your emails.

6) To look at the mail use (still loged in as userme):

   $ mail

   to quit this program press the q-key followed by the enter-key.

   On my machine this gives a screen full of output looking like:

  Mail version 8.1.2 01/15/2001.  Type ? for help.
  "/var/mail/carel": 26 messages 26 new
  >N  1 [EMAIL PROTECTED] Tue Oct 01 22:02  120/8086  animus 2002...
   N  2 [EMAIL PROTECTED] Tue Oct 01 23:02   56/3295  animus 2002...
 
   N 19 [EMAIL PROTECTED] Wed Oct 02 13:02   27/1225  animus 2002...
   N 20 [EMAIL PROTECTED] Wed Oct 02 13:11  205/12000 Anacron job...
  & q
  Held 26 messages in /var/mail/carel


If this works, then the next thing to do is to choose an email client and
we will help configuring it.

-- 
groetjes, carel


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Email (Solved)...next...POP3 setup

[Carel Fellinger]

On Wed, Oct 02, 2002 at 11:49:31AM -0700, Michael Olds wrote:
> Carel,
> 
> OK, I got it.
...

great.
 
> ...now...on to POP3 configuration...I am using qpopper. I see the
> qpopper.conf in /etc/ but the package installed with no configuration dialog
> and the instructions say to configure using ./configure which I assume is
> from the install directory, only there is no install directory...
> 
> ...or...any better suggestions for a POP3 server?

I would advice to use an imap server on your linux box.  This makes it
easy to read your linux mail both from your Windows as wel as from
your linux machine without the trouble of having to remember were you
read and stored that one mail you're after.

As I promissed my father to help him switch to linux I've been
examining imap daemons lately.  The one that performs best with both
Mozilla Mail Client and Outlook Express in my limited tests has been
cyrus-imapd.  The disadvantage of this package is that it uses its own
mailbox format.

-- 
groetjes, carel


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Headers in this ML

[Carel Fellinger]

On Thu, Oct 03, 2002 at 01:34:09PM +0100, Richard Kimber wrote:
> I've just received a message from the list with no reference in the
> headers to debian.  Since I sort my mail on the basis of headers, the
> message went into the wrong box.  Has there been some change in how the
> list is administered?

Are you using standard mbox format?  Could it be that it's just the end
of another email that just happened to have a proper "From " header in
its *body* ?

-- 
groetjes, carel


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: debootstrap

[Carel Fellinger]

On Sun, Dec 14, 2003 at 12:56:04AM -0500, Emma Jane Hogbin wrote:
> Hi,
> 
> I'm in the process of trying to convert a leased rack space from RedHat to
> Debian. I'm working my way through:
>   http://www.debian.org/releases/stable/i386/ch-preparing.en.html
>   http://trilldev.sourceforge.net/files/remotedeb.html
> 
> I'd like to use the RPM that is referred to in both of the documents I'm
> reading. It says it's supposed to be here, but I get 403 forbidden. Any
> ideas?
>   http://people.debian.org/~blade/install/debootstrap

make it yourself:),  I think it's merely "alien --to-rpm" on the deb-file.
for your convenience I've put such an alienised rpm on the web:

http://www.xs4all.nl/~cfelling/debootstrap-0.2.18-2.i386.rpm

be aware, it's the one from sarge, and I've never ever used rpm-nised
debs, so it might not work at all.

-- 
groetjes, carel


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Web browsers for Linux (was: Re: Netscape Bus Error)

[Carel Fellinger]

Hai,

On Wed, Aug 16, 2000 at 05:19:29PM -0300, Rogerio Brito wrote:
> On Aug 16 2000, André Dahlqvist wrote:

> > quiet a lot of people who seam to like using Netscape to handle
> > their mail, and I think it's nice to give those people that option.
...
>   BTW, I also notice how much people use Netscape to handle
>   their mail and when I install Linux for my friends I install
>   it also, for the following convenience: you don't need an MTA
>   in your machine for the (conceptually) simple tasks of
>   receiving and sending e-mails -- it incorporates both a POP3
>   and a SMTP client in a single program.
> 
>   That is the reason why I don't install mutt for other people
>   (that might not know how to fix the problems when they
>   happen).  But *if* I knew of other e-mailers with the same
>   functionality already packaged for Debian, I would consider
>   them.

You could use mutt's recently build in support for POP and IMAP servers
(or you could use fetchmail:) and use ssmtp just to send the mail (seems
a simple program to install). But I don't see how you can do without local
mail on a linux system, local services need to be able to send reports if
things go wrong.

Re: Is the 3COM 3C509B PCI (PCI PCI PCI __not__ ISA) supported?

[Carel Fellinger]

On Wed, Aug 30, 2000 at 11:15:04AM -0700, Chris Majewski wrote:
> I bought a 3c509b PCI thinking this would be the easiest and most
> reliable thing to get working.. bullshit. 
> /proc/pci says:
> 
>   Bus  0, device  11, function  0:
> Ethernet controller: 3Com 3C905B 100bTX (rev 48).

So it isn't a 3c509B afterall, is it:)

...
> but compiling the 3c509 driver into the kernel (2.2.17) does nothing,

and so you need the 3c59x driver instead.

-- 
groetjes, carel

Re: Netscape 4.75: Can't open plain text files

[Carel Fellinger]

On Fri, Sep 01, 2000 at 03:31:50PM +0200, Frank Mehnert wrote:
Hi,

> with Netscape 4.75 I installed from potato-proposed-updates I can't open
> plain text files anymore. If I install 4.75 by the tar.gz archive,
> everything works find. If I use the debian package, after loading the
> text file (status bar bottom left finished), nothing happens.

The same here, if you run netscape from a xterm you will notice an error
message like: "Netscape: Ignoring unsupported format code in mailcap file: %{"
Checking the /etc/mailcap file you'll see entries for text which use
the %{..} feature of mailcap. Somehow netscape has troubles with it.

Could you try with the tarball one if it gives the same error and whether
it uses the mailcap file at all?

-- 
groetjes, carel

Re: parsing existing mbox

[Carel Fellinger]

On Sat, Sep 02, 2000 at 11:23:17AM -0500, Will Trillich wrote:
...
> you've got a mbox with hundreds of messages
> in it, and you've updated your .procmailrc or
> exim.conf with new filtering rules, and you
> wanna run them on your mbox to split it up
> into pieces, perhaps forwarding certain messages
> on to other folks, or deleting those matching
> pattern X.

you could try the script down below, but beware I'm not a guru
and myself I have troubles using formail on the debian-user-digest,
it chops of some attachments it seems (or is that an error of the
digest at debian.org?, never checked it).

You can do without the locking if you do
   # mv mbox mbox.unsorted
   # formail -s procmail  my failing memory tells me it involved an odd
> option along the lines of "-0f0" or "-f1"...

The big question is ofcourse for which program:)

++
# cat sortmail
#!/bin/bash

unset noclobber

ORGMAIL=/var/spool/mail/$LOGNAME #adjust to need
export MAILDIR=~/Mail

if cd $HOME &&
   test -s $ORGMAIL &&
   lockfile -r0 -l3600 .newmail.lock 2>/dev/null
then
   trap "rm -f .newmail.lock" 1 2 3 15
   umask 077
   lockfile -l3600 -ml
   cat $ORGMAIL >>.newmail && cat /dev/null >$ORGMAIL
   lockfile -mu
   formail -s procmail <.newmail && rm -f .newmail
   rm -f .newmail.lock
fi
exit 0
++

-- 
groetjes, carel

nfs and firewall

[Carel Fellinger]

Hai,

I'm trying to secure my system, I ran pmfirewall and some tests.
It seems that rpc.mountd still listens on port 1024 even on the
outgoing ethernet.

I am trying hard to read up to this subject, but in the time being
I would feel much better if I were able to shut off *all* services
from this machine to the hostile internet. So if some kind soul
could shed some light onto this, I would be much obliged:)

My setup is a firewall and several local machines on a local net,
the firewall doing masquerading and firewalling. For ease of upgrading
I want the firewall to be able to mount a debian mirror on another
local machine. In the end I also think of letting the firewall machine
act as a local mail and news server (is that deemed secure?).

-- 
groetjes, carel

Re: nfs and firewall

[Carel Fellinger]

On Sat, Sep 02, 2000 at 08:23:08PM -0500, Phil Brutsche wrote:
> A long time ago, in a galaxy far, far way, someone said...
> 
> > Hai,
> > 
> > I'm trying to secure my system, I ran pmfirewall and some tests.
> > It seems that rpc.mountd still listens on port 1024 even on the
> > outgoing ethernet.
...
> I would remove the nfs-server (or nfs-kernel-server, whichever you have
> installed) package.  You don't need that package to connect to an NFS
> server; only if you're going to *be* the NFS server do you need it.

Okee, removed it.

> > local machine. In the end I also think of letting the firewall machine
> > act as a local mail and news server (is that deemed secure?).
> 
> It can be a bad thing: I call having "too many" services on one system
> "too many eggs in one basket".  I've seen situations in the past where an
> exploit in one piece of software will expose the entire system to the
> attacker, and let him/her gain access to all that computer offers.

agreed, but...
I only want to run it as a local service, not as a service to the net.
The reason being that my firewall is the only machine on 24/7, so it
seems the logical place to provide *local-only* services to my localnet.
But being new to this securing thingy I don't know whether such a setup
would compromise security, neither do I know how to disable internet
access to those services, and how rigidly that can be done. I've a lot
of reading to do:)

-- 
groetjes, carel

Re: nfs and firewall

[Carel Fellinger]

On Sun, Sep 03, 2000 at 06:01:09AM +0200, Sebastian Ritter wrote:
> On Sat, 2 Sep 2000, Carel Fellinger wrote:
> 
> > Hai,
> > 
> > I'm trying to secure my system, I ran pmfirewall and some tests.
> > It seems that rpc.mountd still listens on port 1024 even on the
> > outgoing ethernet.
'''
> You can find a lot of informations on how to set up Firewalls in the
> IPCHAINS-HOWTO. You can find that document under http://www.linuxdoc.org/.

I know, I'm reading it. But it takes time to fully understand it:(

> Using the firewall as a mail and news server is extremely dangerous. The
> best firewall would be a dedicated machine which ONLY acts as a
> firewall and does nothing more. I think any company that's a little bit
> nervous about security should afford that.

I'm not a company:), and I never intent to provide internet services.
Those services are for the localnet only! I want them to get denied on
the external (internet) ethernet. I don't know yet whether that still
compromises security (I've a lot of reading to do:), so for the time
being I would appreciate a verdict from a more experienced person.
Do you think that even in the above situation local only mail/news
services are a bad thing? And is that because once you get cracked
the cracker has access to your local news and mail spool?

> It seems to me that you are very new to IP security. I'd strongly advise
> you to buy external support or read lots of related books, e. g. "Building
> Internet Firewalls 2nd Edition" by O'Reilly to gain the basic 
> skills. Otherwise it's very likely that you'll get cracked. ;-)

I've no money to spent on this, so I will have to read and read and read...
It's just that in the mean time i would prefer to have a safe machine:)
I understood from reading sofar that as long as you don't expose any service
to the outside world you are safe, don't know for sure yet though.

-- 
groetjes, carel

digest from bounce-debian-user-digest

[Carel Fellinger]

Hai,

since I think sunday I get my debian-user digest from

  [EMAIL PROTECTED]

I've temporarilly done stupid things with my proc/fetchmail setup,
but corrected it. Do I need to do something to get the digest of
the 'bounce'?  Or has it nothing to do with my setup and is this
again a mysterious change in digest policy?

-- 
groetjes, carel

Re: Netscape mime-types

[Carel Fellinger]

On Mon, Sep 04, 2000 at 02:42:22PM -0300, Marcio Rosa da Silva wrote:
> Hi all!
> 
> Last week I made an apt-get update; apt-get upgrade and my Netscape
> changed from 4.73 to 4.75.

upgrade again:) Had the same problems, but the version of today solved it.

-- 
groetjes, carel

Re: digest from bounce-debian-user-digest

[Carel Fellinger]

On Tue, Sep 05, 2000 at 11:39:07AM -0400, Chris Gray wrote:
> On Mon, Sep 04, 2000 at 03:09:48PM +0200, Carel Fellinger wrote:
> > 
> > since I think sunday I get my debian-user digest from
> > 
> >   [EMAIL PROTECTED]
.. 
> This has been around for quite a while now on debian-user -- months at
> least.  It was announced at the time.  It is supposed to be a long-term

thanks for the info. I only noticed it this weekend, as I now switched
to having fetchmail grab my mail in 'multidrop' mode. So I had to figure
out how to get to the "From " envelop header, never saw the real "From "
before:)

However, I never saw an announcement either, can't find
anything on "bounce" in the archives too, so where are such things
announced? in some privat mailinglist:)  I guess, I would feel better
if I could read some official statement on this somewhere.
I'll check the policy guides.

-- 
groetjes, carel

Re: Netscape mime-types

[Carel Fellinger]

On Mon, Sep 04, 2000 at 07:14:08PM -0700, Dale L . Morris wrote:
> I just tryed installing 4.75 and had the same problem. I installed it

we are talking about 4.75-2, yes?
I think that's what fixed it for me.

-- 
groetjes, carel

Python License Headaches

[Carel Fellinger]

Hai,

Lately the Python language suffered from a change in the license that is
deemed incompatible with the GPL by RMS himself!

Quoting Guido van Rossum in news:comp.lang.python:

> There's one bit of sad news: according to Richard Stallman, this
> version is no more compatible with the GPL than version 1.6 that was
> released this morning by CNRI, because of a technicality concerning
> the choice of law provision in the CNRI license.  Because 2.0b1 has to
> be considered a derivative work of 1.6, this technicality in the CNRI
> license applies to 2.0 too (and to any other derivative works of 1.6).
> CNRI is still trying to work this out with Stallman, so I hope that we
> will be able to release future versions of Python under a
> GPL-compatible license.

In other postings one is adviced to vent concers in this matter directly
to CNRI and FSF, so as to urge them to resolve the matter.

Now that some debian projects use python, I think it would make sence to
'vent our concerns' to them as the debian project. I'm a mere user:), so
I don't qualify to mail them on behalve of the debian project. I'm posting
here hoping that some debian developer will take this upon him.

-- 
groetjes, carel

Netscape and spurious net access

[Carel Fellinger]

Hai,

thanks to xnetload I noticed that netscape has a lot of net activity
on startup and on exit. Even when used to view a local html file
(netscape localfile.html) a lot of net activity is going on. I wonder
what he is doing. Anyone in the know?

-- 
groetjes, carel

PGP and Mutt

[Carel Fellinger]

Hai,

I admit upfront, I haven't read all the docs, but I would be chilled
if someone could point me to the relevant part of the docs, or maybe
even spoils the fun of having to read it all and explains:)

I've upgraded from hamm to potato, and decided to try pgp due to all
the debian-user mails that are signed. I've told mutt to use pgp5 but
now I get the following error message once I open a signed mail:

   [-- PGP output follows (current time: Sat Sep  9 03:17:03 2000) --]
   Signature by unknown keyid: 0x55F2B9B0
   This signature applies to another message
   Opening file "/dev/null" type text.
   [-- End of PGP output --]

what's wrong here?

-- 
groetjes, carel

Re: Netscape and spurious net access

[Carel Fellinger]

On Fri, Sep 08, 2000 at 01:56:29PM -0700, Nate Amsden wrote:
> Carel Fellinger wrote:
> > thanks to xnetload I noticed that netscape has a lot of net activity
> > on startup and on exit. Even when used to view a local html file
...
> i suggest loading something like iptraf or netwatch to see whats going
> on. in many versions of netscape, whenever it loads it tries to resolve
> a bunch of different netscape domain names to see if your DNS is
> working, this is the only thing i can think of that it's doin.

and right you were, it's mere dns lookups.

-- 
groetjes, carel

Re: PGP and Mutt

[Carel Fellinger]

Brian and others,

thanks for the replys, I guess I can't solve all without reading:)

On Fri, Sep 08, 2000 at 07:09:34PM -0700, brian moore wrote:
> On Sat, Sep 09, 2000 at 03:21:58AM +0200, Carel Fellinger wrote:
> > now I get the following error message once I open a signed mail:
> > 
> >[-- PGP output follows (current time: Sat Sep  9 03:17:03 2000) --]
> >Signature by unknown keyid: 0x55F2B9B0
> 
> You need to get the key for that key id.  GNUPG has an option to
> autofetch unknown keys.
> 
> >This signature applies to another message
> 
> That's normal: it means the signature is detached (as in multi-part).

I see. This was the one that worried me the most and ment the least.

Now trying to fill my keyrings...

-- 
groetjes, carel

Re: debian-user-digest down?

[Carel Fellinger]

On Sun, Sep 10, 2000 at 10:19:27AM -0400, Kent Pirkle wrote:
> I've not received the debian-user-digest or debian-devl-digest for
> the last couple of days. Is anyone else having this problem?

Same here:(, so I finally switched to the non-digest

-- 
groetjes, carel

Re: Using RSA authentication with SSH - mini HOWTO

[Carel Fellinger]

On Mon, Sep 11, 2000 at 11:29:14AM -0700, kmself@ix.netcom.com wrote:
> On Mon, Sep 11, 2000 at 01:02:18AM -0400, S . Salman Ahmed ([EMAIL 
> PROTECTED]) wrote:

clear description snipt but the one typeo...
> 
> Troubleshooting:  If Something Breaks
> -
...
> o Your public key cannot be writeable by anyone but you, but needs
>   group and world write permissions.
^
read I presume:)

-- 
groetjes, carel

Re: OFFTOPIC: Reading data from a parallel port

[Carel Fellinger]

On Fri, Sep 15, 2000 at 11:25:45AM -0400, [EMAIL PROTECTED] wrote:
...
> My problem is, I need to be able to read the data being sent to and from the
> device to reverse engineer it. Therefore, I need someway to log the data that
> goes across the parallel port. Thus, I'm looking for a windows app that will
> allow me to do this. Nothing special, just a way to log data going across the
> parallel port.
> 
> I know Linux has some packages to do log parallel port data, but, the software
> runs under Windows...not Linux, so, I'll need a windows app. Also, I looked 
> into
> the Diamond Rio mp3 player utility, and they claimed to have used a custom VxD
> (virtual device driver?). This doesn't mean much to me.

This being really offtopic I guess I might as well detour you a little more:)

One painfully slow alternative might be to hook this Windows machine's
parallel port to a linux running one, get the first step of the protocol.
Implement it on the linux machine, and try to talk to the mp3 player from
the linux machine with this first step of the protocol. Get the mp3 player's
respons. Hook up the windows machine again, start the conversation, and use
the mp3's responce of the previous step, get the respons from windos, etc.
etc etc etc.

Might be easier if you have more machines hanging around,
but atleast you need 2. Doupt this is what you need:)

-- 
groetjes, carel

Re: mutt macro syntax

[Carel Fellinger]

On Tue, Sep 19, 2000 at 12:22:20PM +0200, Egbert Bouwman wrote:
...
> I tried:
> 
>macro pager - delete-thread exit 
> 
> but that doesn't work.
> Does somebody know the right syntax ? 
> The manual is not particularly helpful.

Indeed the manual is a little sparse on the subject:), but taking into
account that mutt is so wonderfull, I think you may find the following
to work:

   macro pager - "delete-thread; exit"

let me try...no good...reading some in the fine manual...aha
macro expects keysequences, no functions. try again:

   macro pager - ':exec delete-thread; exec exit'\n

yep this works. Mind the trailing \n, needed to persuede mutt to doit.

-- 
groetjes, carel

System.map-2.2.17-ide and kernel incompatibility

[Carel Fellinger]

Hai,

the other day I noticed that running ps gives the following warning:

   $ ps
   {module_list} {module_list_R__ver_module_list}
   Warning: /boot/System.map-2.2.17-ide does not match kernel data.
   PID TTY  TIME CMD
   ...

I'm running kernel 2.2.17-ide. When I switch back to 2.2.17 sic all
is swell. (I'm trying to compile the kernel here, but things have
changed since hamm, and I can't get it to work yet)


-- 
groetjes, carel

Re: System.map-2.2.17-ide and kernel incompatibility

[Carel Fellinger]

On Sat, Sep 23, 2000 at 02:44:47PM +0200, Carel Fellinger wrote:
> Hai,
> 
> the other day I noticed that running ps gives the following warning:
> 
>$ ps
>{module_list} {module_list_R__ver_module_list}
>Warning: /boot/System.map-2.2.17-ide does not match kernel data.
>PID TTY  TIME CMD
>...
> 
> I'm running kernel 2.2.17-ide. When I switch back to 2.2.17 sic all
> is swell. (I'm trying to compile the kernel here, but things have
> changed since hamm, and I can't get it to work yet)

Thanks to advice in an other threat I managed to compile the kernel
with fakeroot, and indeed this one, with the ide patches is okee.

So is this a bug worth filing?

-- 
groetjes, carel