Alsa not working after reboot

[cothrige]

Having been a fairly long-time Slackware user, since 8.1, and being
somewhat unhappy with some compatibility problems with Gnome apps and
that distro, I decided to see if Debian would work for me.  I don't
have high-speed internet access and so used a work computer which I
have access to periodically to download and burn the first two discs
of Testing.  This seemed a solid place to start and I was confident
that two discs would get a working system up.

I was not surprised to find that basically I was right, though
surprised at some of what was left off of these two discs, eg. alsa, X
headers and libraries, emacs, etc.  But, overall, things went well and
after tinkering about with the installation I soon got a working
system with the basics in place.

After downloading and installing the alsa-1.0.8 packages I ran
alsaconf and had the sound up and running.  However, I discovered
later that if I rebooted the sound never reloaded and I had to run
alsaconf again.  This had happened to me some time back with Slack 9.1
and I managed after some work to get it going, but I cannot really
recall how.  I fear that with Debian I am unsure of where to start.
The modules.conf file is a bit of a shocker for me as it is nothing
like that of Slackware, and has a great deal more going on.  I cannot
really tell if it has what I need in it at all.  The
/etc/modprobe.d/sound file, which I am not familiar with at all, is
there and has a reference to my card, es1688.  Other than that I
haven't a clue what it does and cannot be sure what it should have in
it.

I have noticed that at the end of the boot process a brief reference
to alsa is seen, and something along the lines of "no modules loaded."
And in fact nothing is.  An lsmod shows nothing related to snd at all,
and there is no mixer to adjust.  So my situation is not a muted card
or pcm but in fact a total absence of sound.  Only running alsaconf
again with each boot will make sound usable.  I have tried a number of
google searches only to find numerous muted cards and loaded modules
without sound, but can't seem to locate any with no modules being
loaded at all.  Is there a step or action after alsaconf which will
make this permament and reload the modules with each boot?  Am I
forgetting or overlooking something stupidly obvious?

patrick


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

Re: Alsa not working after reboot

[cothrige]

* Colin ([EMAIL PROTECTED]) wrote:
> 
> alsaconf will load the modules that you need but it won't add them to your
> /etc/modules file to have them load after a reboot.  Do "lsmod" before and
> after alsaconf and put the modules that were added by alsaconf in 
> /etc/modules.

Yes, that worked.  After Mr. Hood's post above, thanks by the way, I
saw this suggested in the README.Debian file.  Actually there was
quite a bit more about hotplug, but this caught my eye as being
straightforward and so I did it first.  Although I only added one
module, snd-es1688, and apparently the rest was loaded automatically.
Now I have quite a few snd modules listed in lsmod, and I rebooted
with nothing  in the /etc/modules file but that one above.  But,
working is what counts.

Many thanks to all,

patrick


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

Re: Alsa not working after reboot

[cothrige]

* Colin ([EMAIL PROTECTED]) wrote:
> 
> I'd suggest putting in snd-pcm-oss and snd-mixer-oss if it isn't there 
> already.

Thanks for the heads up.  I will do that.

patrick


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

Setterm failure

[cothrige]

I recently have begun using the console much more on a slower machine
and have noticed that setterm does not seem to work.  I had set it in
rc.local with "setterm -blank 15" and "setterm -powerdown 45", and
while the screen will blank, it will never powerdown.  I also tried
doing it manually, just to see, by issuing the command "setterm
-powersave powerdown" but this also failed.  I have tried with the
default Debian kernel and one I have compiled, with framebuffer and
without, but nothing has worked.  So far I just can't seem to get the
monitor to powerdown.  Does anybody have any ideas about what I may be
doing wrong?

Thanks,

Patrick


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

Re: OT: sponge burning!

[cothrige]

* Ron Johnson ([EMAIL PROTECTED]) wrote:
> 
> Anyway, voluntary abortion is legal in most countries, including the
> USA, so voluntary abortion does not meet the statutory definition of
> "murder".
>

I am obviously inserting myself into another person's conversation,
and for that I apologize, but having read this I feel compelled to
comment.  According to this distinction above there were no Jews
murdered in Germany during the Nazi period.  I can't think this would
be a very viable definition for 'murder.'

Just my two cents,

Patrick


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

Re: OT: sponge burning!

[cothrige]

* Ron Johnson ([EMAIL PROTECTED]) wrote:
> Ask a pro-lifer whether she'd euthanize a pregnant cat.
> 
> Then you'll know when she *really* thinks life begins.
>

Well, I don't think I could euthanize a pregnant cat.  Do you think
many people could?  Seems a bit cold for me.

Patrick


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

Re: OT: sponge burning!

[cothrige]

* Ron Johnson ([EMAIL PROTECTED]) wrote:
> I was just having this discussion the other day.  Legal killing is
> *not* murder.

I would certainly say that lawful killing is not murder, but might not
agree with the specific choice of legal.  Legal would imply the laws
of the land, whereas lawful would be more broad.  What is legal in
this country, i.e. certain killing, may not be lawful if one accepts
a higher law.  Many, myself included, would believe that many killings
which are in fact completely legal are still in opposition to the
Divine law.  That is why abortion is murder, even though it is legal.

That is my take on it.

Patrick


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

Re: OT: sponge burning!

[cothrige]

* Paul Johnson ([EMAIL PROTECTED]) wrote:
> Roberto C. Sanchez wrote:
> 
> > On Mon, Feb 26, 2007 at 11:38:59AM -0800, Paul Johnson wrote:
> >> 
> >> Name five American national-level politicians in your lifetime that had
> >> moral fortitude.  You can't get that far in US politics without being a
> >> scumbag.
> >> 
> > Reagan
> 
> Absolute liar. ...He makes Barry Goldwater look good.
>

I can't really argue much about Reagan or his legacy, i.e. W, but I am
surprised at the comparison here to Goldwater.  Most of the
neoconservative movement, beginning somewhat with Reagan personally
and very much with the immediate post-Reagan period, has been nothing
more than outrageously expensive fascism, read 'law and order'
Republicanism.  But, Goldwater always seemed much more legitimately
conservative to me, with what appeared to be a very sincere interest in
true small government.  Is there something I have missed about him?

Patrick


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

Re: OT: sponge burning!

[cothrige]

* Paul Johnson ([EMAIL PROTECTED]) wrote:
>  
> He was a full-fledged supporter of McCarthyism, never saw a war he didn't
> like, was very much anti-labor-rights, and verbally abused Eisenhower for
> having a moral compass on those issues.  His only saving grace is he turned
> libertarian in retirement and became rather socially progressive after his
> political career ended.  The fact the revelation came after his retirement
> makes it about as empty and meaningless as Strom Thurmond apologizing for
> being a racist bigot for the first 99 years of his life.
> 

I must state, at the outset, that I am no Goldwater expert, but there
are phrases here which cause me to be dubious of this approach.
Statements like "anti-labor-rights" immediately gives me pause, as
such are so often trotted out as labels without substance.  An example
was the charge that Bob Dole was "anti-elderly" because he didn't
support Medicare.  The fact that he may have felt that this very act
was not pro-elderly does not enter the picture.  And I have seen many
called "anti-labor" when in fact they were merely anti-socialist, a
position very associated with many "labor rights" movements.

I am moderately familiar with the tensions between Goldwater and
Eisenhower, and I am just not too sure it could be called "verbal abuse."
Goldwater was definitely critical of the Eisenhower White House,
comparing it to the New Deal IFIRC, but abusive?  That, again, sounds
like a rather loaded term which may say more of the critic of
Goldwater than it does of Goldwater himself.

I have little doubt that if I went over every detail of the man's positions I
would have much to criticize, but this above just doesn't seem
terribly convincing.  And I certainly wouldn't put him in the current camp
of anti-conservatives which your previous post seemed to imply, though
I may have read into that.

Patrick


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

Re: REALLY OT: News Flash

[cothrige]

* [EMAIL PROTECTED] ([EMAIL PROTECTED]) wrote:
> 
> Just where did Jesus claim to be God?  I've heard lost of people say 
> that Jesus was God, but I haven't seen the place where Jesus himself 
> claims it.
> 
> -- hendrik

I think the clearest place is in the Gospel According to St. John:

The Jews therefore said to him: Thou art not yet fifty years
old, and hast thou seen Abraham?  Jesus said to them: Amen,
amen I say to you, before Abraham was made, I am.  They took
up stones therefore to cast at him. But Jesus hid himself, and
went out of the temple.

The key phrase is "before Abraham was made, I am."  This is eveb
stronger than things like "I and the Father are one." In the Old
Testament, in the burning bush incident, God actually introduces
himself as "I Am" and makes this his personal name.  Therefore, by so
obviously calling himself by this name we see that Jesus is directly
declaring himself not only to be divine in a general sense, but more
precisely is saying that he *is* the very God who spoke to Moses.

Patrick


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

Re: Capture of video stream..

[cothrige]

"Mumia W.." <[EMAIL PROTECTED]> writes:

> On 08/06/2007 01:53 PM, ISHWAR RATTAN wrote:
>>
>> I want to capture streamed video from a youtube
>> link...
[snip]
>
> Maybe the '-dumpstream' option to mplayer can capture such streams.
>
> If you're using the Firefox web browser, the "Fast Video Download"
> extension can make downloading Youtube videos easy.
>

There is a website www.keepvid.com which will make youtube videos
downloadable as well.  This can then be converted to divx or such using
mencoder.  You just have to watch out for some odd framerates which can
throw off the resulting playback.

Patrick


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

Strange video troubles (DPMS & nv driver)

[cothrige]

I have been having a couple of odd developments lately which I think may
be traceable to my Samsung SyncMaster 940BW 19" LCD monitor.  It started
when I noticed I was not getting consistent reactions to my DPMS
settings.  In my xorg.conf I have the following:

Section "ServerFlags"
Option "BlankTime" "10"
Option "StandbyTime" "40"
Option "SuspendTime" "50"
Option "OffTime" "60"
EndSection

...

Section "Monitor"
Identifier "940BW"
HorizSync   30.0 - 81.0
VertRefresh 56.0 - 75.0
Option "DPMS"
EndSection

I would expect that after 10 minutes idle my screen would blank, and
then after 40 it would enter standby mode, and so on.  However,
sometimes it just never does any blanking or sleeping at all.  Other
times it will blank and then never sleep.  Of course, often it does both
just as it should.  I have tried using xscreensaver, which I don't
usually run, with its power management settings, but it didn't seem to
affect anything.  There is just no telling really what may or may not
happen in this area.

In response I have tried to find something in the logs regarding actual
calls to the monitor relevant to DPMS, but have found nothing.  Is there
a way to see when the computer sends a signal to the monitor to blank or
sleep and see how the message was received or responded to?

In trying to solve this problem I thought I would make sure it was not
video driver related and tried the nv driver.  I usually use the nvidia
driver.  However, I could not get nv to work.  Everything was a bit
fuzzy, and my fonts were all jaggy and hard to read, with the vertical
lines being very faint.  Also the screen was too tall for the monitor by
about thirty pixels, and was way off to the side with about two inches
of black down the right.  This could not be corrected with the
horizontal placement setting, and the best I could get was about a half
inch of black on the side, with the other side cut off.  Obviously it
was completely unusable.

I can think of no reason why the nv driver would fail this badly, and
why DPMS is so hit and miss.  Perhaps it is as simple as a bad monitor,
or maybe it is two unrelated coincidences.  I would like to make sure
before replacing it but just can't find much information to make a
judgement by.  Has anyone heard of something like this with 'nv' or have
any suggestions about tracking down DPMS and what is going on there?

Many thanks,

Patrick


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

Re: Strange video troubles (DPMS & nv driver)

[cothrige]

Ralph Katz <[EMAIL PROTECTED]> writes:
>
> wrt dpms:
> Until you get a more knowledgable reply, man xset (especially, the 's'
> option) and man xscreensaver-command (-time) may help.

Many thanks.  I will read up on these right now.

> I use the nv driver, but have a crt.

I have gotten to thinking.  Does nv take any flags like nvidia does?
Could there be a setting I could manipulate when loading it which could
help with LCD monitors, or the like?  I must admit complete ignorance on
this issue, as when I used nv consistently I had a very simple setup
with a CRT as well.  I never needed to do anything but load the driver
for it to work.

> Sometimes, mplayer or another player disables the screen saver and
> forgets to re-enable it, which is annoying.  That accounts for failure
> to engage dpms actions for me for those cases.

I had not thought of this.  I have noticed that at times mplayer will
disable dpms and never reset it, apparently.  In most cases though it
only seems to turn off dpms during play, but perhaps if it dies for some
reason...?  I have also noticed that avidemux2 kills all dpms and does
not restart it.  I very, very rarely use that though, and not nearly as
often as the problem has appeared.  But, it very well could be some
other application I am not thinking of which I use just often enough to
be causing a problem.  Perhaps next time I notice the monitor refusing
to sleep I will try 'xset q' and just see if dpms has been disabled.  I
can't remember if I did that before, but it may be informative.

>
> fwiw, I have no Section "ServerFlags" in my xorg.conf on this etch box.
>  There are no log entries indicating screen saver or dpms actions or
> settings.

But, do you have dpms on?  If so how does it know when to set standby
and so on?  Or do you use xscreensaver or the like?  Do you think
removing the ServerFlags section and then using only xscreensaver may
have an impact?  I may try that.

>
> Good luck!
>
> Ralph Katz

Many thanks Ralph.

Patrick


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

Re: Strange video troubles (DPMS & nv driver)

[cothrige]

Ralph Katz <[EMAIL PROTECTED]> writes:

> On 08/09/2007 09:12 PM, cothrige wrote:
>
>> I have gotten to thinking.  Does nv take any flags like nvidia does?
>> Could there be a setting I could manipulate when loading it which could
>> help with LCD monitors, or the like?  I must admit complete ignorance on
>> this issue, as when I used nv consistently I had a very simple setup
>> with a CRT as well.  I never needed to do anything but load the driver
>> for it to work.
>
> Maybe dpkg-reconfigure xorg will help?  iirc, there was an LCD question
> asked.

Unfortunately, this did not change anything in my setup.  At least not
in regards to using 'nv.'

> I don't know anything about ServerFlags.
>
> DPMS is on:  ~$ grep DPMS /etc/X11/xorg.conf
> Option  "DPMS"
>
> My dpms settings are handled by xscreensaver ($ xscreensaver-demo).

Cool.  I am trying this right now to see if it helps.  Maybe there is
some glitch in how DPMS is operating in my system and so perhaps
xscreensaver will have a more direct method which will work better.

> Also, I sometimes call this from a script in ~/bin when not wanting to
> wait for xscreensaver to kick in:
> xset -display :0.0 dpms force  suspend

I did the same thing with my setup, but lately even that has not worked
right.  For instance, I could open a terminal and type 'xset dpms force
off' and the monitor would sleep.  However, I have always mapped Super-x
in Fluxbox to that command and would hit that to sleep the monitor, but
lately it will only make the display blink.  I really cannot quite
figure out what is going on in this regard.  The good news is that after
switching to xscreensaver at least that did seem to work again, if only
so far.

> Hope this helps,

Immensely.  Much appreciated.

> Ralph

Patrick


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

CD player with CDDB

[cothrige]

I generally use a tiling WM (xmonad, ion, stump, etc.) and listen to
music with mpd.  With that setup I can keep a small text monitor in the
status bar, or other such place, for seeing just what is currently
running.  Every once in a while though I will actually play a cd and I
would like to have such a monitor for that time too, if possible.  I
have gone through the tools in apt and installed all the CLI players
that seemed to be listed but none will return any actual data from
CDDB.  Cdcd seems to imply it can, but it doesn't actually work from
here, instead complaining that it cannot parse the data. (I installed it
and libcdaudio from source just to see but I still got the same error,
btw.)

Does anyone by any chance know of a player, or app, which can return
this information for me?  Since I use a tiling WM gui apps can be
tricky, and I am hoping to find something I can use a command line to
get data from, and google just doesn't seem to return anything I can
use.  Many thanks for any suggestions.

Patrick


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

/dev/cdrom & /dev/dvd symlinks and udev

[cothrige]

I have two dvd drives, hdc is a standard drive and hdd is a cd/dvd
writer.  The problem is that all the cdrom symlinks always point at
/dev/hdd which is not my primary drive.  I like cdrw and dvdrw as they
are, but would like to have cdrom and dvd to point correctly to hdc.  In
trying to do this I have followed a number of online howtos but nothing
has any effect.

I have tried using 10-local.rules, local.rules, z99.rules without luck.
I have also tried a number of entries in these files, including:

SUBSYSTEMS=="ide", KERNELS=="1.0", SYMLINK+="cdrom", ENV{GENERATED}="1"

and:

KERNEL=="hdc", SUBSYSTEM=="block", SYMLINK+="cdrom"

and a couple of others I cannot even recall right now.  I used udevinfo
to get the above data, and udevtest would return that the symlinks would
be created, but they never were.  They continue to point to /dev/hdd
apparently regardless of what I do.  I have restarted udev with the
init.d script and have rebooted, all without any change.

What exactly am I doing wrong here?  Is there something I am
overlooking?  Is there perhaps another rules file which overrules this
one and resets or refuses to allow a reset of the symlinks?

Thanks for any help.

Patrick


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

Re: /dev/cdrom & /dev/dvd symlinks and udev

[cothrige]

Wackojacko <[EMAIL PROTECTED]> writes:

>
> See /etc/udev/rules.d/z25_persistent-cd.rules this is auto generated
> by the script in /etc/udev/.  I think you can change these around if
> you set the $GENERATED variable to 0 (from comments in file).
>
> HTH
>
> Wackojacko
>

Many thanks for the tip, and I will go and look there.  I will admit
that I was a bit nervous about poking around too much in the present
files as so many howtos and so on made a big deal about making your
own.  

Strange thing is that I finally got so fed up with this that just
decided to delete the symlinks and recreate them for what I needed.  I
figured that at least until the next boot I would have /dev/cdrom where
I needed it and cd players and so on would work.  What shocked me was
that they were still that way after reboot.  So now I am wondering if
udev is even setting this stuff?  I had made a rule in 010-local.rules
for my mp3 player, a USB mass storage device, and that has worked fine
from the moment I first restarted udev, even though nothing was
affecting the dvd and cdrom symlinks.  So, I knew that the rules were
being implemented, but the ones for cdrom and dvd were not doing what
they should.  It is all a bit confusing to me.

Thanks again,

Patrick


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

Re: capture real audio stream

[cothrige]

Lorenzo Bettini <[EMAIL PROTECTED]> writes:

> Hi
>
> I'd like to capture some audio stream from a web radio.  I know about
> streamripper, but it does not work for real audio.  And in particular,
> if possible, I'd like to record what is being played, without knowing
> the address of the real audio file (which I actually don't know: bbc
> radio just starts the real player plugin in the web page).
>
> any clue please?
>
> thanks in advance
>   Lorenzo
>
> -- 

This is pretty easy if you have either mplayer or audacity installed.
If you happen to be playing from BBC 7 it is particularly easy, and all
you have to do is right click on the listen link and save the ram file
to your drive.  Then you just read it as it is plain text, and inside
you will find something like:

rtsp://rmv8.bbc.net.uk/bbc7/0930_sun.ra?BBC-UID=34f67dd2ff9daa4baac9cd5c20f0b659244a9c75a07060437bfa8149d4698951&SSO2-UID=

All you need is the stuff up to and including ".ra" which we can use in
an mplayer command as follows:

mplayer rtsp://rmv8.bbc.net.uk/bbc7/0930_sun.ra -vo null -vc null -ao
pcm:file=0930_sun.wav

That will then stream the content live and record it directly to the
hard drive as a wav file.

If you are using some other bbc site, such as BBC 4, then you just left
click on the listen button, which will pop up a new window with a
built-in player.  On the left side is a link for "Play in standalone
Real player" or some such.  Right click on this and save the resulting
ram or rm file and continue just as above.  The BBC radio sites other
than BBC 7 just have a built in litle player window of some kind
abstracting away the ram file, but it is still available in that link.

BTW, if all of that fails for some reason then you can always open
audacity (before the audio player to make sure your sound card is
available for recording), adjust your input to volume and set the volume
levels.  Then you can open the browser and start the player, and begin
the recording in audacity.  This will record the stream as a wav as
well, though if your signal drops out you will have to clean up any
silent patches in the file.  The mplayer way is much, much better as it
will automatically correct any failures in the feed and so the resulting
file will have no blank patches in it, so I really would recommend that
way first.

Hope this helps,

Patrick


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

Re: capture real audio stream

[cothrige]

Lorenzo Bettini <[EMAIL PROTECTED]> writes:

>
> I'm trying with this one
>
> http://www.bbc.co.uk/radio2/musicclub/events_andrewlloydwebber.shtml
>
> and can't find such links, not even with Page Info (as suggested by
> Joachim); could it be they changed something in their site?

Well, neither could I.  I suppose that the Beeb has decided to really
hide the actual content and this time did a bang up job as far as I can
tell.  They seem to have gotten rid of the standalone player link on
their listen again player (though oddly BBC4 still has it and it is also
on the BBC Radio 2 listen live page) and that definitely puts a kink in
things.  I tried viewing the source for each frame and that didn't show
anything useful, at least not that I could recognize. :-) I even looked
via ps for the running realplay in hopes that it would reveal some clue,
but got nothing useful.

In the end I was only able to succeed by accident.  I tried the link in
epiphany and it gave me a popup complaining that totem could not open
the content, and in so doing told me the full file name.  That turned 
out to be rtsp://rmv8.bbc.net.uk/radio2/fridaymusic.ra?BBC-UID=a42\
6cbda207670378b86528c714caba3684a2b4fc040b101329203b36444a5b5&SSO2-UID='
I doubt that this could be reliably reproduced, though perhaps you may
have a broken install of epiphany too, and that could turn out to be a
good thing. 

I do hope this is not the way that the BBC Radio people will be doing
things everywhere, as it is certainly a great deal less convenient.

Patrick


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

Re: capture real audio stream

[cothrige]

Lorenzo Bettini <[EMAIL PROTECTED]> writes:
>
> well that somehow makes me feel better, 'cause I was starting to feel
> stupid since I wasn't able to find no such links ;-)
>

If you go to the home page for BBC radio 2 and click on Listen at the
top right, you will see the "Listen using stand-alone Real Player" link
on the left hand side.  That is the link I had expected on the player
used for Listen again, and it does seem odd that it exists on one player
and not the other.  Maybe BBC has a policy about listen again since it
is effectively "play on demand" where the live player requires you
stream in real time just when it happens to be on?  There is also a
disclaimer of sorts on the BBC Radio 2 pages saying that the listen
again player is having some difficulties.  Maybe this is one?  I am
inclined to think that is not the case though, and likely it is
intentional and has to do with the perceived difference of listen again
and live playing.

> I'm trying the link you provided with the command
>
> mplayer
> "rtsp://rmv8.bbc.net.uk/radio2/fridaymusic.ra?BBC-UID=a426cbda207670378b86528c714caba3684a2b4fc040b101329203b36444a5b5&SSO2-UID='"
> -vo null -vc null -ao pcm:file=/mnt/appo/musica/webber.wav
>
> and it seems to work fine! :-)

Glad to hear.  It has always served me fine.  BTW, the only reason I
have not been prone to using dumpstream as I know others do is that for
some reason a while back I would get a mysterious failure that way and
mplayer would quit midstream.  I really don't know why, but using '-ao
pcm..." has not had that problem as often, though I have run into it
from time to time.

>
> by the way, isn't there a way to record directly in mp3 I suppose?

I suppose it may be possible, but I have never tried.  I always clean up
the resulting file anyway.  The files that mplayer records this way are
almost always quite a bit longer than the actual item being recorded,
and so I like to trim off the extra stuff on the ends, and I use either
audacity or sox to do that.  Wavs mean not having to uncompress and
recompress the stream which keeps it from degrading.  I also like to
normalize it and while it can be done with mp3 it doesn't work on most
players and so I do that to the wav and then compress it to ogg or mp3.

>> I do hope this is not the way that the BBC Radio people will be doing
>> things everywhere, as it is certainly a great deal less convenient.
>
> Later I'll try with epiphany too... but why are you saying that it is
> due to a broken install of epiphany?

Opera and Firefox (Iceweasel) just play the file in a hidden and
seamless way, which is obviously what should happen.  That makes it hard
to know what is being played but it does seem the expected behaviour.
Epiphany, on the other hand, fails and so alerts me to that and tells me
the full name of the file in so doing.  Good for us, but since the file
doesn't play it means that the plugins for realplay are not present and
working and neither is totem (which is what epiphany is trying to open
the file with.)  So, something in there does seem broken right now,
though that is actually a good thing.  It makes me wonder if I moved
realplay would that make the other browsers alert me similarly?  Just a
thought if I were desperate.

Patrick


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

Cannot change default browser

[cothrige]

I have been using kopete, as gaim wasn't behaving quite to my needs,
and found that when it prompts me to open an inbox on one of my
accounts it always opens epiphany.  Of course, the first thing I did
was try to change whatever kopete thought was the right browser, but I
could find nothing at all in the config options which indicated any
selections of browsers.

I don't have a full KDE installed, and so there is no control center
type of thing I can run to change that.  But I switched to gnome long
enough to open the control center there and set firefox as default,
but it did not work.  I opened firefox and had it check for whether it
was the default browser, and then made is such, but this also had no
effect.

After reading the man page and some online helps, I tried
'update-alternatives --set x-www-browser /usr/bin/firefox' but I still
get epiphany.  When I run 'update-alternatives --list x-www-browser'
this is what I get: 
/usr/bin/epiphany 
/usr/bin/firefox 
/usr/bin/opera

If I run 'update-alternatives --config x-www-browser' I get:

There are 3 alternatives which provide `x-www-browser'.

  SelectionAlternative
---
 +1/usr/bin/epiphany
* 2/usr/bin/firefox
  3/usr/bin/opera

Press enter to keep the default[*], or type selection number: 2
Using `/usr/bin/firefox' to provide `x-www-browser'.

I have checked display and it says x-www-browser is set to manual, so
I thought the + would not mean anything, but it still defaults
everytime to epiphany.

Finally, I uninstalled epiphany, causing untold problems in the Gnome
world, and this caused Opera to open.  I actually like opera for some
things, but not this, and so don't really want to have to uninstall
it.  So, now I am back to square one.

Just how do I convince this app not to use any browser but firefox?

Many thanks for any help,

Patrick


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

Re: Cannot change default browser

[cothrige]

* Alan Ianson ([EMAIL PROTECTED]) wrote:
> 
> Try running update-alternatives --all. There is also a x-gnome-browser (or 
> somesuch) that may need an adjustment.
> 

I found x-www-browser, www-browser and gnome-www-browser.  I set them
all just in case.  I find the same result for gnome-www-browser as the
other, and even though I set it it has no effect.  So far, no matter
what I do I get epiphany, or opera if I uninstall epiphany.  BTW, I
also confirmed that the symlink in /etc/alternatives points to
firefox, which seems to indicate that the system is not even accessing
these links.  

Patrick


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

Re: Cannot change default browser

[cothrige]

* Kelly Clowers ([EMAIL PROTECTED]) wrote:
> 
> You could just install kcontrol; it looks like it only depends on a couple
> of things that kopete doesn't.

I tried this, but it is not working.  It installed okay, and opens,
but the left hand pane, where one would expect to see the menus and
choices, is blank.

> Otherwise, try opening ~/.kde/share/config/kdeglobals
> There should be an entry like this:
> [General]
> BrowserApplication=!/usr/bin/epiphany
> 
> Change that to firefox and hopefully you will be ready to go.
> I have not tried this, and I am not responsible if your monitor explodes.

I opened this file, but there was no such entry in it as
BrowserApplication or anything labelled [General].  So, I added them
just in case that may work.  However, I still always get epiphany.
Though, on the up side, my monitor did not explode.

Perhaps this is hard encoded into Kopete these days.  It clearly
doesn't use anything in the system to decide this.

Many thanks for the suggestions,

Patrick


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

Re: Etch: Kcontrol not working: it is empty

[cothrige]

* Mirto Silvio Busico ([EMAIL PROTECTED]) wrote:
> Hi all,
> I've just setup a new machine wigth Etch weekly build of 9 Nov 2006.
> 
> When I start Kcontrol it starts but shows an empty selction panel on the
> left; so it is unusable.
> 
> Anyone is experiencing this bheaviour?

Yep, same here.  But, I figured it was because I didn't install the
entire kde system, and to be honest I wasn't going to install the 338
packages it would take to find out.  Like you it starts fine, but
presents a totally blank left panel.

Here is what apt says I have:
kcontrol:
  Installed: 4:3.5.5a.dfsg.1-1
  Candidate: 4:3.5.5a.dfsg.1-1
  Version table:
 *** 4:3.5.5a.dfsg.1-1 0
500 http://ftp.us.debian.org etch/main Packages
100 /var/lib/dpkg/status

Patrick


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

Re: Where, Doom3 For Linux Download

[cothrige]

Dave Thayer <[EMAIL PROTECTED]> writes:

> On Mon, Sep 10, 2007 at 05:09:27PM -0400, Orestes Leal wrote:
>> Hi Folks, I need a 'direct' download (with suport for resume) for download
>> de Linux version of doom3, I have the original 3 CDs so the only thing
>> that I need it´s the .run? or whatever for linux, any help will be
>> thankfully appreciated,
>
> Have a look at the official "DOOM III GNU/Linux FAQ"
>  It contains the link
> you're looking for as well as a bittorrent tracker link. Of course,
> you'll want to read the other information there too...
>

I had a heck of a time finding a working link a while back for quake2,
and in the end had to go with the loki installer from
http://www.liflg.org/.  I really don't know what makes it a different
way than others, or if it does at all, but the link was there, it
worked, and admittedly the install was easy.  I think they have doom3
there so it may be worth looking into as well.

Patrick

LPR and CUPS

[cothrige]

I have an HP Deskjet printer which I have set up using CUPS.  It has
always functioned fine, and I can use lp to print without any problems,
and openoffice, browsers and such work just fine.  However, I use Emacs
and would like to be able to print from within that app, and it uses
lpr, which I have never installed.  I had wondered if that would cause
problems or have conflicts, but I was able to install it with aptitude
seemingly without error.  But, even though everything seemed okay, when
I tried using lpr for printing it did not work.

Initially it would simply retort with "lpr: lp: unknown printer."  I
"seemed" to fix that by setting the variable PRINTER to my printer.
However, that just made it fail silently.  When I looked at the status
of the printer with "lpc status" it said that the jobs were queued, but
that it was waiting for my computer to "come up."  I had no idea what
this was about, as lpd was running and lpr seemed to be taking jobs.

So, what I am wondering is if it is the CUPS way I have been doing
things that has caused the problems, and if so can that be worked
around?  It would seem that CUPS cannot easily be uninstalled on my
system as so much depends on it.  Could my problem be that lprng is not
installed?  This also cannot be fixed easily as it conflicts with CUPS.
If nothing else, maybe somebody knows how to convince emacs to print
without using lpr?

Many thanks for any possible suggestions which anyone can make.

Patrick


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

Re: LPR and CUPS

[cothrige]

Andrew Sackville-West <[EMAIL PROTECTED]> writes:

>
> cupsys-bsd package provide /usr/bin/lpr. you should not need to
> install an additional package to get lpr functionality. 
>

Well, there you go.  I did not have that particular package installed,
rather having only cupsys and so on.  This fixed it right away, and
without any further configuration than what I already had.  At least not
for getting it to print.

Thank you very much for that helpful suggestion.

Patrick



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

Re: LPR and CUPS

[cothrige]

Andrew Sackville-West <[EMAIL PROTECTED]> writes:

> On Thu, Sep 20, 2007 at 06:35:15PM -0500, cothrige wrote:
>> Andrew Sackville-West <[EMAIL PROTECTED]> writes:
>
> that's better than the typical "cups sucks" flamewar we get... cue in
> 3..2..1..
>
> A

I have to admit that I am pretty much entirely ignorant concerning the
whole thing.  I did read some posts a while back discussing the evil of
cups, but I think having absolutely no experience with the other
approaches I couldn't really follow the reasoning.  I have little doubt
that cups is less than a wonderful tool (so many of these things are and
I still find myself fuming in hate regarding alsa from time to time) but
since up to this time I have been able to use it successfully I have had
no reason to try anything else.

Patrick


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

Re: Debian may lose a user

[cothrige]

Mike McCarty <[EMAIL PROTECTED]> writes:

> John Hasler wrote:
>>
>> I don't see that you provided any useful information.
>
> Then it wasn't directed at you.
>
> There are those here who have expressed a desire for Linux
> to be a viable alternative to Windows for more users. It was
> directed at those people.

After walking in from a day of my kids' soccer matches I noticed this
thread and feel I really must post a comment.  I may be very late to
this, and if so I apologize, but it just seems to me that there is
likely a miscommunication going on here.  I can't help but think that
most people on this list would actually very much welcome the criticism
which seems to be on offer regarding this situation, but surely
everybody would also like to know what they did wrong, so to speak.  I
for one get at least a hundred or two messages on this list per day,
probably more, and certainly wouldn't know from the above references
which posts or questions were those being talked about here.  Really,
with everything being discussed here how could everybody notice and
remember every thread?

I also think everyone here has a desire for Linux to be a viable
alternative to Windows (I happen to believe it already is one since I in
fact use it instead of that OS) and so would certainly assume your
comments were directed at them.  The people on this list are here quite
obviously because they believe in what is done here, and that seems to
most often be offer help to others.  But, honestly, regardless of how
well-intentioned the posts, just how can anyone improve with such
esoteric comments?

I believe that is what the poster above was saying, and most likely you
understood him as being defensive or confrontational.  I really don't
think the people here are trying to be in any way argumentative, but
rather are asking for more info, even if you aren't going to be here to
hear the answers.  It is just a natural desire to see what mistakes were
made, specifically, so that they are not made again.  I personally think
it is a wonderful sentiment to fill people in on a failure, but if
nobody can know how they failed, then really what good is the
information?

At least that is how I would percieve this situation.

Patrick


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

Re: efficiency of windows managers

[cothrige]

"Javier Vasquez" <[EMAIL PROTECTED]> writes:

>
> Don't know about windowMaker, but you might try:
>
> fluxbox
> icewm
> pekwm
> fvwm2
>
> You might find some pretty light, and some besides offering lots of
> fun and good looking features...  I use fluxbox and a machine with
> 512M main, and 64M ati-rage is performing pretty well...

I know my choices may seem rather extreme by some, but if one is really
seeking a lightweight wm I would suggest adding xmonad (with dzen2) to
the above list.  I have tried many small tools in this area, including
ratpoison, ion3, and stumpwm, and xmonad is easily the fastest and has
the smallest memory footprint I have found yet.  As I have no experience
with haskell compiling it was something of a learning experience, but it
really wasn't too bad and has been more than worth it.  It does just
about everything I ever liked in ion3 (the next smallest IMHO), and all
without Tuomo! :-)

Patrick



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

Re: efficiency of windows managers

[cothrige]

"Pál Csányi" <[EMAIL PROTECTED]> writes:

> What about stumpwm?
>
> http://www.nongnu.org/stumpwm/
>
> Use this window manager somebody on Debian Etch?

Yes, I have it installed using etch, and like it very much.  Though it
was kind of hard to get going. You may have already set it up, but just
in case you have not, of in case somebody else is trying, I will briefly
mention what I found worked.  It may help somebody out there.

Firstly, I never got the Debian version working at all.  I am sure I
goofed it up, but I just couldn't find a way to make it start.  The docs
didn't seem to offer anything useful at all either.  In the end I used
the CVS version of stumpwm.  It has a real README which actually has
decent instructions.  By following them I had a working executable in
just a couple of minutes.  However, sbcl from Debian also does not work
right, and under any kind of load the entire system becomes uselessly
unresponsive.  I removed the SBCL installed via apt, and recompiled it
from source.  Then I installed clx using asdf "(require 'asdf) (require
'asdf-install) (asdf-install:install 'clx)" and recompiled stump.  This
worked fine, though the final product tends not to be very lightweight.

I was also able to get stump with clisp working, though in the end it
was less than reliable.  The up-side though is that it is much, much
lighter than sbcl.  Again, I could not get the clisp I installed via apt
to work, and had to compile from source.  It failed again and again to
work with stump though, until I tried CVS on that too.  I also had to
use new-clx instead of mit-clx.  Unfortunately, even after getting it to
work it crashes X fairly regularly (anytime certain dialog boxes are
used, such as print or downloads) and so is rather less than useful.
Really too bad as I was much impressed with its generally faster,
lighter feel than stump with sbcl.

But, I did get sbcl going okay, and with it stump seems solid and
usable.  It is fast overall, and certainly has the nice feel of
ratpoison, with the bonus of a mode-line.  And even though it is not
terribly lightweight considering the sparsity of the thing I still think
it is an excellent choice for a tiling WM.

However, as much as I liked it, xmonad (http://xmonad.org/) was even
more pleasing than everything else.  It also took a little work to get
going, though nothing like stump, but it was worth it in the end.  It is
unbelievably light and fast and even seems smaller than ratpoison.  And
dzen2, a separate app, as a status-bar is much more full-featured than
stump's, and seems about as good as what ion has.  Additionally, it
handles transient windows much more realistically than does stump or
ratpoison.  With it you can also grab the window and resize it making
that window a floating one rather than tiling, all without the hoops you
seemingly had to jump through with ion3 to do the same thing.  Without a
doubt, xmonad with dzen has been the most satisfying and simple approach
I have yet tried in the window manager market.

Patrick

Re: efficiency of windows managers

[cothrige]

"Pál Csányi" <[EMAIL PROTECTED]> writes:

> Hello Patrick!

Howdy,

> No, I have not set it up.  :(
> Thank you for help! :)

Glad to be able to do so.

> The debian stumpwm package dependency are: sbcl cl-clx-sbcl.

Yeah, and I actually met those, and it may have "worked" except I could
never get it started as I just didn't know how.  It may be because I am
not a lisper, but it just seemed very esoteric.  There were no direct
instructions of what to do to get things running.  For instance, what
command do you use and where?  I just never did get it running and the
/usr/share/doc... stuff didn't really help me at all.

Also, finding out later that the Debian SBCL has a problem and causes
the entire system to virtually stop running if you do any CPU intensive
tasks I have very strong doubts it would be a good idea to run stump
that way in the first place.  Much better I think to go ahead and
compile sbcl yourself.

> Would you please give for us the step by step advices how to install
> from CVS stumpwm?

First I installed SBCL via apt.  This is because you have to have a lisp
to compile sbcl (odd, huh?) and that one works fine.  I am sure clisp or
cmucl would work too, but I already had sbcl because I was trying to get
stump working, and so I used it. Of course, YMMV.  I then grabbed the
source to sbcl from their website, untarred it and changed to that
directory.  In there I configured it (not sure exactly how as it was a
while ago) to install in my home directory rather than in /usr/local.
This way it would not interfere, even temporarily with my system sbcl
which I was using to compile it.  It also makes uninstalling as easy as
deleting the sbcl dirs in my home folder.  Apps like sbcl are hard to
uninstall and so this works well for me.  Just export SBCL_HOME in your
.bashrc with something like 
"export SBCL_HOME=/home/cothrige/lisp/lib/sbcl" and then `source ~/.bashrc'.

I then started sbcl and typed the following one at a time at the prompt,
to install clx:

* (require 'asdf) 
* (require 'asdf-install) 
* (asdf-install:install 'clx)

This will download and install everything for you.  BTW, I installed it
system-wide, but I don't think it matters if you use your home directory
for sbcl as I did.  If you use /usr/local or some such it will obviously
matter.  Then I closed SBCL and grabbed the stump source this way:

cvs -d:pserver:[EMAIL PROTECTED]:/cvsroot/stumpwm login
cvs -z3 -d:pserver:[EMAIL PROTECTED]:/cvsroot/stumpwm co stumpwm

(BTW, press return for the password on the first one.)

This will create a stumwpm directory.  Cd there and do the following:

autoconf
./configure
make

That will do it.  It defaults to sbcl, and if you have installed sbcl to
a place in your path you will have no trouble at all.  There is no make
install or the like as you run it from this directory.  What you do is
move the stumpwm folder to wherever you like and then just invoke that
in your .xinitrc.  If I left it in my home directory for instance, it
would be like this:

exec ~/stumpwm/stumpwm

There is a sample .stumpwmrc file in the stumpwm directory, but it has
nothing on the mode-line in it.  If you are interested in that you will
need something for it.  It isn't too bad to get configured once it is up
and running, though I found all of it to be somewhat vague.  If you
think you would like it I will gladly post up a copy of my stumpwmrc
file and you can look at that to see if you see anything which may help
you.

> Just in any case, if you have stumpwm installed from CVS on Debian,
> how can you uninstall (purge) it? (For case if the Debian stumpwm
> package come usable.)

Well, the nice thing is since it doesn't install at all you just delete
the entire stumpwm directory and that really does purge it.  The way I
did things nothing left my home folder at all, which made cleanup much
easier.  And the way I kept messing it up with trial and error that
approach came in handy, believe me.

I hope this helps you get it running.  Once I did I really liked it,
though I didn't end up staying with full time.  Maybe I just liked the
challenge of making it work. ;-)

Good luck on it.

Patrick

Re: efficiency of windows managers

[cothrige]

"Pál Csányi" <[EMAIL PROTECTED]> writes:

> Have you read /usr/share/doc/stumpwm/README.Debian [3]?  It contains
> information about how to start StumpWM.

I tried this too, and I think it takes a great deal for granted.  There
is just too much assumed about what you know for this stuff to work,
IMHO.

> If you want to start StumpWM from Emacs (which is not my advice),
> fire-up SLIME and then give only the two last commands:
>
>   * (asdf:oos 'asdf:load-op 'stumpwm)
>   * (stumpwm:stumpwm)

Yes, I tried this, but it didn't work for me.  And swank has its own
troubles, as far as newbies are concerned.  I did get it working kind
of, but it was never too solid, and tends to cause more problems than it
would ever be worth to me.  Since I am a lisp newb I just doubt that
there is much I can do in the way of hacking stump on the fly. ;-)

> Howewer, because I'm CL newbie, I can't to start stumpwm on Debian Etch.

I feel your pain brother.  Believe me, the thing to do is to just use
CVS stump, compile it to run with sbcl and then load it via .xinitrc.
That way it just runs like any other WM, and is loaded in the same way.
Way better method if you ask me.

Patrick

OT: Amazon DRM-free MP3 Downloads

[cothrige]

Has anyone used this yet?  I bought a few tunes, just to try it out, and
it went quite smoothly.  Just being able to use it with Iceweasel (no
user-agent spoofing either) on Debian without any complaint or glitch
was really nice.  And the files sounded great (256K bitrate) and there
was actually a surprisingly not awful selection.  I would have thought
there would have been more hooplah out there about this, but so far
haven't noticed much commentary online.  Maybe I should finally start
reading blogs?

Patrick


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

Re: OT: Amazon DRM-free MP3 Downloads

[cothrige]

Andrew Sackville-West <[EMAIL PROTECTED]> writes:

> On Wed, Oct 03, 2007 at 09:12:14AM -0500, cothrige wrote:
>
> http://yro.slashdot.org/article.pl?sid=07/09/25/1951219
> http://slashdot.org/articles/07/09/26/1748213.shtml
> http://apple.slashdot.org/article.pl?sid=07/09/30/0148246&from=rss
>
> I know. I'm sick. please help me.
>
> ;-)
>

Ah, you see, I don't follow the right sites anymore.  Slashdot used to
hold some interest for me, but it just seems to have drifted into a
different style these days.  Of course, maybe I am the one that
drifted. :-)

> My understanding is that it works just fine with linux for individual
> tracks, but not for the full album which requires their special
> downloader. but that's hearsay from the above articles.

Downloading a single track definitely worked okay on my end, though I
didn't try an album, yet.  Overall, less impulse usually to do that
anyway.  I usually look at these mp3 downloads when I only like one or
two songs on an album, otherwise I like all the packaging and artwork
that comes with a standard disc.  That is just me though.

Thanks for the heads up on the above articles.

Patrick


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

Finding installed dependencies?

[cothrige]

I have a feeling this is a dumb question.  It seems like something that
should be relatively easy, but searching through man pages and google
has not helped me so far.  Say I have a package, 'pkgx-1.0,' installed.
Is there a way that I can list other installed packages which have that
first one, i.e. 'pkgx-1.0', listed as a dependency?  I was trying to
clean up some junk on my computer, and just would like a nice and easy
way to find out what I have which is using one or another package.

Thanks in advance,

Patrick


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

Re: Finding installed dependencies?

[cothrige]

Kevin Mark <[EMAIL PROTECTED]> writes:

> On Fri, Oct 05, 2007 at 01:55:05PM -0500, cothrige wrote:
>
> 'debfoster' is a neat program to find cruft and remove it.
> If pkgx is install and has dependencies a, b and c, then 'debfoster'
> will show you this and ask you if you want to delete all of them. Read
> the man pages and docs before you use it as its job is to remove stuff.
> you can also just do 'apt-cache show pkgx' to see its dependencies
>

Okay, I have heard of that in passing.  I will definitely look into it,
and the apt-cache command sounds like it may be just what I am looking
for. 

Many thanks,

Patrick


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

Re: Finding installed dependencies?

[cothrige]

Ralph Katz <[EMAIL PROTECTED]> writes:

> On 10/05/2007 02:59 PM, Kevin Mark wrote:
>
> And apt-cache rdepends .  $ man apt-cache
> Look at deborphan as well.
>
> Regards,
> Ralph

Awesome.  Now I have several tools to look into.  Isn't that just the
way it is, so often when it seems hard to do something, you ask for help
and end up with several ways to do it.  That is one reason why I am glad
I switched to Debian.

Thanks very much for all the help guys.

Patrick


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

Re: Which browser is better, firefox?

[cothrige]

"Douglas A. Tutty" <[EMAIL PROTECTED]> writes:

> On Thu, Nov 08, 2007 at 09:52:43PM +0800, Kenlen wrote:
>> recommend w3m... I think better than lynx. . 
>> 
>
> I tried w3m but I couldn't get it to work.  IIRC its just that its a
> different key commands so its a new learning curve.  Since I didn't see
> what it would do better I didn't bother trying to learn it and
> uninstalled w3m instead.

I love text browsers, and generally prefer them if something else isn't
absolutely required.  Having tried all I can I have consistently found
elinks to be the best performer of the lot.  It has the most flexible
configurations, with all the best features of the others combined into
one (with the exception of javascript and https I think).  It has
features such as tabs, a status bar (with a clock), 256 color support,
tables, and so on.  It can even set the background color to that of the
original page which can be quite cool in many instances (if you're not
reading slashdot).  If one is really doing a lot of text browsing elinks
is very much worth the look.

Patrick



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

Re: What am I missing without mutt?

[cothrige]

Daniel Burrows <[EMAIL PROTECTED]> writes:

>   Some cons that annoy me frequently:
>
> * Stack-based interface: mutt's interface is organized around
>   doing a task which may have sub-tasks, and once you're in a
>   sub-task you can't get back to the main one without "quitting".
>   So you can't refer back to a mailbox while composing a message,
>   unless you add the message onto the postponed list and pull it
>   back off when you're ready.
>
>   Running several instances of mutt in different xterms can
>   alleviate this somewhat, at the cost of possibly de-syncing
>   the instances if you aren't careful.

I remember having a similar experience to this when using Mutt.  The way
I handled was via emacs, my chosen editor.  I would open my mbox in
another buffer, and then search out the message I wanted to see.  I
could then switch from one buffer to another, and of course kill and
yank between them, which was mostly what I had wanted.  Of course, this
was a bit ugly as the file was sometimes quite long and had to be
searched through a bit for what I wanted.  Eventually I switched to Gnus
which made this issue irrelevant, since it uses multiple buffers for
messages already.  But I did, and do, miss the Mutt config though, as my
.gnus just scares the crap out of me.

Patrick


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

Re: this list is on google groups

[cothrige]

steve <[EMAIL PROTECTED]> writes:

> its got nothing to with any announcement by debian list.  its web
> crawlers, they touch everything, scary huh. mailing lists are archived
> all over the net. search for your email address on any search engine
> you'll see quite surprising results if you post regularly to any mailing
> list.

Actually, I think the truth is even simpler than this.  Google Groups is
a usenet archive and posting service.  Debian-user is also a usenet
group, lists.debian.user, and all the posts here I think are
automatically crossposted to that newsgroup, and vice-versa.  I have
used that method to browse through a time or two, though I prefer the
mailing list way of doing things a bit.  But, because this is a
newsgroup and a list it has automatically been archived at google
groups.  Or, at least I would think that is what is going on.

Patrick


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

Aptitude stuck on sun-java5-bin

[cothrige]

I have managed to get myself in a bit of a bind regarding packages.
Aptitude is reporting an error regarding an apparent half-finished
upgrade to sun-java5-bin, and this has caused all things to stop.
Unfortunately I have not found a way to resolve or work around this.
Here is what has been coming back from any aptitude commands:


Preparing to replace sun-java5-bin 1.5.0-14-2 (using 
.../sun-java5-bin_1.5.0-14-2_i386.deb) ...
sun-dlj-v1-1 license has already been accepted
Unpacking replacement sun-java5-bin ...
/usr/share/icons/sun-java5.png is not a directory
dpkg: warning - old post-removal script returned error exit status 2
dpkg - trying script from the new package instead ...
/usr/share/icons/sun-java5.png is not a directory
dpkg: error processing 
/var/cache/apt/archives/sun-java5-bin_1.5.0-14-2_i386.deb (--unpack):
 subprocess new post-removal script returned error exit status 2
sun-dlj-v1-1 license has already been accepted
/usr/share/icons/sun-java5.png is not a directory
dpkg: error while cleaning up:
 subprocess post-removal script returned error exit status 2
Errors were encountered while processing:
 /var/cache/apt/archives/sun-java5-bin_1.5.0-14-2_i386.deb
E: Sub-process /usr/bin/dpkg returned an error code (1)

I don't know why it would expect that sun-java5.png would be a
directory, but this has apparently led to some strange confusions.  I
have removed the sun-java5-bin debfile from cache just in case something
may change with a new download, but it still resulted in the same
complaints.  I have also tried uninstalling the package, but it
insists it must install it before removing it, and so that won't work
either.  I am not quite sure how to proceed here, and I am sure I am
overlooking the obvious response simply because I am still a bit of a
newb with apt.

Any ideas of what I may be able to do to get this resolved?  Many thanks
in advance,

Patrick


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

Re: Aptitude stuck on sun-java5-bin

[cothrige]

"Adrian Levi" <[EMAIL PROTECTED]> writes:

>
> you could try running,
>
> #dpkg --configure -a
>
> But likely I suspect it will fail.

Yes, it appears to have changed little.  Still failing.

> Another option is possibly to pre-empt it and go ahead and delete:
> /usr/share/icons/sun-java5.png

Actually, there is no such file.  I assume it is trying to install one,
but I can't be sure.

At some point during this process things changed, and now I am not
getting the same complaint as before.  Rather, all it says now is:

dpkg: error processing sun-java5-bin (--remove):
 Package is in a very bad inconsistent state - you should
 reinstall it before attempting a removal.
Errors were encountered while processing:
 sun-java5-bin
E: Sub-process /usr/bin/dpkg returned an error code (1)

I think my machine is becoming condescending about this. :-) All I can
think is Duh, it is in a "very bad inconsistent state."  But, I cannot
reinstall as the same thing pops up then.  Right now all that I get is
the above complaint, regardless of what I do with apt.  Very confusing I
must say.

Patrick


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

Re: Changed device for camera..

[cothrige]

Charlie <[EMAIL PROTECTED]> writes:

> Thanks for that Ron, but the camera mounts again on /dev/sda1 and when I want 
> to write the udev rules for it they won't write but choke on the attributes 
> of:
>
> * ATTRS{serial} command not found
> * ATTRS{model} command not found
> * ATTRS{modalias} command not found

Why so many ATTRS?  I have an mp3 player and I used only ATTRS{product}
as that is specific enough to only apply to that one device.  The main
problem I ran into back when I was trying to get udev working was
accidentally mixing ATTRS from different blocks of udevinfo, or putting
my ATTRS under a SUBSYSTEM they didn't come from.  Be careful to look at
the output for udevinfo carefully, and whatever ATTR or ATTRS you use
take them from the same block, and when you use a SUBSYSTEM or KERNEL
node be equally careful that the ATTRS that follow it in your rule
actually come from the same device in udevinfo.  This kind of thing
seems to completely stop the rule from working.

What I did was to run `udevinfo -a -p $(udevinfo -q path -n /dev/sda)` 
and then I looked for the most specific ATTR I could find.
For mine that was ATTRS{product}=="Sansa e260R", which was directly
under SUBSYSTEMS=="usb".  Above all of this, in another block, was
SUBSYSTEM=="block", which I also used in the first position. 
So, in /etc/udev/rules.d/10-local.rules I put this: 
SUBSYSTEM=="block", SUBSYSTEMS=="usb", ATTRS{product}=="Sansa e260R", 
SYMLINK+="sansa" 

In this case I used two separate devices to define the rule, the "block"
and "usb" were from separate device sections of udevinfo.  However, the
attribute came from the same device as SUBSYSTEMS=="usb" and so it
worked okay.  From what I can tell this is crucial.  

Something else to consider may be whether you are trying to change the
/dev/sda1 link.  I didn't try that in the end, and instead just used
SYMLINK+= to create a new link in addition to the ones already made by
udev, such as sda1 or sdb1.  You can use the new link as easily as any
other and it is less trouble to get working, in my experience, than is
trying to actually override the rest of what udev wants to do.

Have you considered, though, using /dev/disk/by-uuid instead?  As far as I
can tell that link never changes and is very specific to the device
itself.  For instance, on my system when I plug in my Sansa above there
is created in /dev/disk/by-uuid a symlink 3874-E1C0 which points to
/dev/sda1.  In my fstab I just have UUID=3874-E1C0 instead of
/dev/sda1 and it mounts up just fine.  And no mucking about with udev at
all.

Good luck,

Patrick


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

Re: Changed device for camera..

[cothrige]

Charlie <[EMAIL PROTECTED]> writes:

> Thanks for that very extensive explanation Patrick. It is much appreciated 
> and 
> though I was going to leave the whole thing as is till another time. I am 
> prompted by the time you have taken to explain this to write more.

You're very welcome.  I only hope that it helps you out.

> I don't have /etc/udev/rules.d/10-local.rules But do 
> have /etc/udev/rules.d/udev.rules

Yes, you will have to create that file if you decide to go the udev
route.  I wouldn't use any other already existing rules files, should
you be tempted to do that, just because it may mess up something in your
system udev setup.  I also think the above 10-local.rules loads earlier
than those files, and this is what you want.  

> I did:
>
> # udevinfo -a -p $(udevinfo -q path -n /dev/sda1)
>
> I get two entries for the Olympus camera:
>
[snip udevinfo]
>
> Then did:
>
> KERNEL=="sda1", SUBSYSTEMS=="usb", {serial}=="X11017714", SYMLINK+="cam"

Okay, I think I can see what is causing you trouble here.  Rather than
{serial}==... I believe you should have ATTRS{serial}==...  I am
certainly no expert, and it has been a while since I messed about with
any udev rules as I am trying to cut back on my drinking ;-), but I
do think the entire label is necessary.  That seems backed up by your
result of "command not found" for the {serial}... input.  I believe you
could add the ATTRS and find it will then work.

> As far as uuid is concerned, I get an id, but not a uuid, so I tried to enter 
> the appropriate id into /etc/fstab but it didn't take there either. Not the 
> id. There was no uuid i could discover, but forget what command I used to 
> find the id.

If you type `ls -l /dev/disk/by-uuid` in a terminal you should get back
a list of all your disks uuid symlinks.  If you plugged in your camera
already, one of them should be something like CAMERA-UUID -> ../../sda1.
This is what you are looking for, and you put this symlink in your
fstab.  Just use the format of UUID=CAMERA-UUID in place of /dev/sda1.
For example, mine is "UUID=3874-E1C0 /mnt/sansa vfat user,noauto 0 0"
rather than "/dev/sda1 /mnt/sansa ..."  In the end, much less messy than
udev rules I think.

Hope you get it working.

Patrick


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

Re: Changed device for camera..

[cothrige]

Ron Johnson <[EMAIL PROTECTED]> writes:

>
> /sbin/blkid will also tell you what a device's UUID is.

Very cool.  I was not aware of this one.

Patrick


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

Exim4, ISPs and headers

[cothrige]

I recently took a look at the headers of some of my mail I sent to my
yahoo account from my ISP mail account, and saw some odd stuff which
made me wonder if perhaps my setup is goofed up.  I am curious if
perhaps something like this may cause my mail to be seen as spam or the
like, and thought I would ask here just to be sure.

I have set up exim4 with "mail send by smarthost; received via SMTP or
fetchmail" and told it to hide my local address and use bellsouth.net,
my ISP, instead.  However, in looking at the mail I have sent to my
yahoo mailbox I noticed that there are a couple of "Received:" headers,
and in a couple of these rather than bellsouth.net is my local machine
address and domain.  Because my from address is the ISP and these are
different I fear they will make the mail look like spam or suspicious
somehow.

Should I be concerned about these headers?  If so, how can I address
changing this behaviour.  Should exim4 be hiding the local address even
in these rather than just the from?

Thanks for any help.

Patrick


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

Re: Exim4, ISPs and headers

[cothrige]

Andrei Popescu <[EMAIL PROTECTED]> writes:

> Have a look at my headers, should be the same (I am using a similar 
> setup with postfix). Didn't have any troubles so far (though I am 
> subscribed to the whitelist for Debian lists).

Cool.  I guess I picked the wrong week to clean out my mailboxes and so
I didn't have too many messages lying about to look at the headers.  The
few there were seemed unlike mine, or were from gmail or the like.  But,
I do see that yours is quite similar.

> The only way I know of not getting those Received: headers would be to 
> let your MUA talk directly to the smarthost, but I don't think this is 
> necessary (I'm guessing a lot of people on this list uses similar 
> setups).

Good to know.  I won't worry too much about it then, unless I start
finding my mails dropped by other people.  And I think I will start
glancing at the headers now and then on this list just to see how common
these kinds of things are.

Many thanks for the information Andrei.

Patrick


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

Re: Idea of a Debian Mascot [Was: FW: Bits from the DPL: FTP assistants, marketing team, init scripts, elections]

[cothrige]

Andrew Sackville-West <[EMAIL PROTECTED]> writes:

> I have to say, there is nothing friendly about that damn gnu. And that
> statement is in no way a reflection on my views of GNU or it's
> projects. Just that if we're gonna have a mascot, I lean more towards
> the penguin/snarky-devil side of the debate than the smelly ruminant
> side. 

I think most open source software tends to look like something from a
group of high school kids, and one of the reasons is the whole mascot
idea.  That daemon in the tennis shoes, the puffer fish, Tux and that
aforementioned hippy gnu all end up just looking ameteurish to me.  The
new FreeBSD logo, on the other hand, is very impressive most of all
because it has no mascot type look to it at all, and instead aims for a
professional and finished appearance.  The combination of the logo and
the color scheme for their packaging is just tremendous, and sets them
entirely apart from the rest of the OSS pack really.

As for Debian, I think the swirl is an incredible graphic which is much
better than what FreeBSD started out with.  Even what they have now is
certainly not any better, but only better presented.  What Debian needs
more than anything is some color or finish to go along with the swirl
which would make the project look more mature and professional.  Mascots
generally have the opposite effect.  I am worried that one day I am
going to surf over to debian.org to be met with a big goofy looking
cartoon wallaby.

Patrick


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

Re: Idea of a Debian Mascot [Was: FW: Bits from the DPL: FTP assistants, marketing team, init scripts, elections]

[cothrige]

Hal Vaughan <[EMAIL PROTECTED]> writes:

> There is a revamped image with a new logo and an easy to use GUI 
> installer one can use if Debian is an issue.  It's a fairly large 
> project.  Here's a link to it:
>
> http://ubuntu.com.
>
> (I notice, when checking that home page that it seems like it's been 
> taken over by Dell.)

Man, you aren't kidding there.  For a second I thought the site had
redirected me to dell.com.

...
> If you want a friendly GUI and a nifty and easy install, go for another 
> distro.  There's a reason there's so many distros out there and it's 
> too much to ask for one distro to try to hit more than one or two 
> focused markets.

I can agree with this to a point, I also can't say I am too excited
about any possible reimaging campaigns either.  I certainly am not
particularly interested in mascots.  However, it seems to me that you
may be overstating things a bit.  Why are friendly GUIs and stability
mutually exclusive?  Isn't that a bit like saying a well built truck
can't be attractive?  It may be true that they usually aren't, but I
just am not too sure that things actually have to be that way.

And I don't think an image revamp, dangerous as it may be in the end,
has to mean a change in these other issues.  Consider that FreeBSD did
actually do this very thing in updating their logo and such, and are
still stable and continue to be stupidly impossible to install.  I
actually thought I was pretty old school after several years of using
Slackware partly because I loved the installer, but I was simply not
ready for the evil of FreeBSD.  I think what they have isn't so much an
installer as it is a test to make sure you really want to run their OS
enough to satisfy them.  Well, they win. ;-)

Patrick


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

Re: Help with man page display in Etch

[cothrige]

Thierry Chatelet <[EMAIL PROTECTED]> writes:

> On Wednesday 27 February 2008 15:47, Michelle Konzack wrote:
>> Woops!  This manpage seems to be written in UTF8
>>
>> Do you have tried to us man under an UTF8 locale like en_US.UTF8?
>>
>> Thanks, Greetings and nice Day
>> Michelle Konzack
>
>
> I am in enUS_UTF8 and I have the same “ and so on as Paul. I thought it 
> was because this man was not in UTF8. So next: how do you know the charactere 
> encoding of a man page?
> Thierry

Could your terminal be one of those which doesn't support utf8 encoding,
like mrxvt?  If so you could just set LC_CTYPE or LANG to en_US and that
might clear it up.  Works for me anyway.

Patrick


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

Re: microsoft vs opensource

[cothrige]

Miles Fidelman <[EMAIL PROTECTED]> writes:
> Tzafrir Cohen wrote:
>> I know about copyrights laws. I know about patent laws. I know about
>> trademarks. I know about property laws. I know of contract laws. I
>> am not aware of any  intelectual property laws.
>>   
> "intellectual property law" = patent law + copyright law + trademark
> law + some related fields
>
> These days you see a lot of law firms describe themselves as
> "intellectual property practices"

I do think the term "intellectual property" is commonly used to describe
a large array of legal concepts, but it isn't, AFAIK, a legal term
applying to specific laws.  There is copyright law, contract law, patent
law, trademark law, and so on, but they are not one and the same and
they do not operate together in some manner.  Some have also made some
rather convincing arguments, IMO, that these legal concepts above are
really too disparate to be clumped together with such an umbrella term
at all.  So, for these reasons, I think it is entirely honest and
correct to say that there are no "intellectual property" laws, and also
to describe the use of that term as "confusing" as the earlier post did.

Patrick 


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

Re: microsoft vs opensource

[cothrige]

"Telaman Consultancies" <[EMAIL PROTECTED]> writes:

...
> The ever persistent issue of software patents for example, when
> dealing with programming - which is essentially a language - and
> shouldn't, as a means of transmission of knowledge, even have
> copyright attached to it.

Doesn't this prove my point a bit?  Here you say in reference to
software patents that a language shouldn't be copyrighted.  You appear
to be using patent and copyright interchangeably, but that doesn't work.
And language is specifically what copyright is meant to protect, but not
patents.  Stephen King's latest book is copyrighted by him, but not
patented.  An invention, likewise, can be patented, but not copyrighted.
The term "intellectual property" may be acceptable in some cases, but in
the case above I think it is clear why some would call it confusing.

Patrick


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

Re: microsoft vs opensource

[cothrige]

Miles Fidelman <[EMAIL PROTECTED]> writes:

> There's a difference between:
>
> - what makes sense (for some definition of "makes sense")
>
> - what's "right" (for some definition "right")
>
> - the legal and regulatory  issues involved (there's a lot of dispute
> about whether software should be patentable, copyrightable, both,
> neither)

I don't disagree with any of this.  But, neither do I disagree with
those who point out that "intellectual property" can be a confusing
term.  Yes, people use it, including lawmakers and lawyers.  But, when
one gets down to the facts more specific issues will always be in
discussion, whether it be copyright, trademark, patent or whatever.

> If you want to engage in masturbatory conversation, you can pick the
> terminology you like.  If you want to understand what's going on,
> write software licenses, and/or influence policy - then you have to
> understand and use the terminology the lawyers, politicians, and
> lobbyists use.

I agree.  If you design and build the perfect mousetrap then you should
file for a patent.  Not doing so, and then later trying to sue others
who build their own because you thought you had a copyright may not work
out too well.  Likewise, deciding to reprint Stephen King's Carrie in
full because a patent runs out in 20 years is more than likely going to
be a really big whoops.

Patrick


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

Re: convert video to audio

[cothrige]

KS <[EMAIL PROTECTED]> writes:

> laura eznarriaga wrote:
>> 
>> *
>> *convert video to adio*
>> 
>> how do you convert a dvd or mp4 to adio
>> is their a program for this ?_
>> *
>> 
>
> Try searching for ffmpeg. It can demux the audio and video for you.

One could also try mplayer's dumpaudio, or pcm output.  I use these
quite regularly for extracting the audio from video files.

Patrick


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

Re: Recording TV with Debian

[cothrige]

[EMAIL PROTECTED] writes:

> On Wed, 02 Jul 2008 20:19:12 -0300
>
> With mencode I used the following command (using PAL in my country):
>
> mencoder tv:// -tv
> driver=v4l2:input=1:norm=pal:width=640:height=480:device=/dev/video0:
> freq=775.25:adevice=/dev/dsp1:forceaudio:audiorate=32000 buffersize=64
> -ovc lavc -lavcopts vcodec=mpeg4:vbitrate=400:keyint=30 -oac mp3lame
> -lameopts br=32:cbr:mode=3 -ffourcc divx -o "test.avi"
>
> I tried to change the options, but no matter what I did, the result was
> poor.
>
> With transcode I used this command:
>
> transcode -x v4l2=resync_margin=1:resync_interval=250,v4l2 -g 640x480
> -i /dev/video0 -p /dev/dsp -e 32000,16,2 -N 0x1 -J
> resample,levels,smartyuv,pv -w 4000 -y ffmpeg -F mpeg4 -o test.avi
>
> Which results in some larger files, but with excelent quality.
>
> Any advice on how to improve would be greatly appreciated, especially
> about the file sizes that transcode produces.

I have not used transcode in a long time, but it seems to me at first
glance that the above command has a video bitrate of 4000 whereas the
mencoder is 400.  If the mencoder command is as you have typed it here
such a low bitrate could be at least part of the trouble.

Patrick


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

Re: changing xterm colors

[cothrige]

Jan Willem Stumpel <[EMAIL PROTECTED]> writes:

> Jamie Griffin wrote:
>> Jan Willem Stumpel wrote:
>>> You can edit a text file called .Xresources in your home
>>> directory (or create it if it does not exist).
>>>
>>> Put the following lines in the file:
>>>
>>> xterm*VT100*foreground:  green
>>> xterm*VT100*background:  black
>>> xterm*VT100*cursorColor: red
>
>> I've tried that and it hasn't worked. Not sure what to try next
>> - does anyone have any other ideas?
>
> This definitely *should* work. If it does not work something weird
> is the matter. Are you sure your terminal program is xterm? For
> instance, do you see the "VT Fonts" menu if you do a
> control-rightclick in the terminal window?
>
> Regards, Jan

I missed some of this thread, so this may not be applicable in the above
case for some reason.  But, assuming that the desire is to change the
values for any xterm, I have used one of the following formats in
.Xresources or .Xdefaults, and one or the other has always worked:

xterm*background: black

or:

XTerm.background: black

Maybe this will help the OP.

Patrick


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

Re: xterm fonts

[cothrige]

* Deboo ^ ([EMAIL PROTECTED]) wrote:
> I installed a minimal X sysetm with fluxbox and xterm and two other
> terminal emulators: eterm and mrxvt. But all three of them give very
> small fonts. WIth xterm, I was able to get a reasonable font with the
> HUGE option in the right-click menu but I need bigger fonts. How do
> you exactly specify bigger fonts with xterm? I am aware it's possible
> with -f or -fn option but what comes after that? Say I wan a 10x20
> font? it doesn't work just by typing -f 10x20

You could try putting something like this in your .Xdefaults:

xterm*font: 10x20

Every xterm opened will then use that font.  That is how I have set my
fonts, but I don't switch them around or anything, so it may not work
so well for you.

> Also how to use the dektop icons that DSL is able to use with fluxbox?
> I already installed fbdesk but it seems of no use at all or is it
> buggy?

Perhaps you would like idesk.  It used to be very good, with support for
translucent icons and so on.  Haven't used it lately though, so it may
not be so nice these days.

Patrick


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

Re: xterm fonts

[cothrige]

* Deboo ^ ([EMAIL PROTECTED]) wrote:
> >
> >You could try putting something like this in your .Xdefaults:
> >
> >xterm*font: 10x20
> >
> >Every xterm opened will then use that font.  That is how I have set my
> >fonts, but I don't switch them around or anything, so it may not work
> >so well for you.
> 
> Thanks. This works if I already open a new xterm from within an xterm
> or any terminal emulator but not from the fluxbox right click XShells
> --> xterm.

Interesting.  Perhaps it is not opening an actual xterm, but rather a
different emulator?  Maybe it is set up to open your default terminal
emulator, which may not be xterm.  Is
/etc/alternatives/x-terminal-emulator pointing to xterm?  If not,
changing that might fix your problem.

Patrick


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

Really slow xterm

[cothrige]

I have installed xterm via apt, running etch, and have noticed that it
scrolls really slowly.  I compared it to rxvt by running `time ls` in
/usr/bin with rxvt taking 0.572s and xterm running at 4.633s.  Earlier
it was even worse taking over 10 seconds.  It is a very big difference
in usage, and if I am compiling something it can have very significant
impact.

Has anybody else had this experience?  Is there anything I can change
which may help this?

Thanks in advance,

Patrick


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

Question about Mutt

[cothrige]

There has always been a thing about Mutt which has bugged me, and
lately it has nagged at me even more than usual.  Perhaps somebody
on this list will know of a way to change this.

In the index screen if I mark a message to be deleted it will cease to
be accessible in any way, and if I scroll to it the cursor will skip
it and move to the next message which is not selected for deletion.  I
have always assumed this to be the expected behaviour, and normally it
is not really an issue at all.  However, if I am in a mailbox for
something rather high volume, such as the directory for this list, I
often will delete many messages.  If I delete a hundred messages and
decide that I wish to undelete one in the middle I may not even be
able to read the subject and use that to undelete.  It is also a pain
to have the cursor jump a hundred messages to the top if I move up one
message too far.

So, what I am wondering is if there is some setting which would make
messages marked for deletion available like a normal message, at least
as far as the cursor goes.  On my screen they are red, which is enough
for me to know what will be deleted.  I went through my muttrc line by
line and just didn't see anything, and so hope that maybe it is
possible and somebody here has figured out how to do it.

Many thanks in advance,

Patrick


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

Re: Question about Mutt

[cothrige]

* Ken Irving ([EMAIL PROTECTED]) wrote:
>  
> Just hold the shift key and use j or k (i.e., J or K) to go up or down.
> Without shift deleted messages are skipped, but not so with the shift key.
> 
> Ken

Many thanks for that.  I obviously overlooked that somewhere along the
line.  It is surely a little thing, but I am very glad to know how to
work around it.

Patrick



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

Strange dhcp difficulty with 2.6.20+ kernels

[cothrige]

Not too long ago I was going to upgrade my kernel, I am currently
running 2.6.18.1, and downloaded the latest 2.6.20 tarball.  I went
through the usual steps.  I configured it carefully and then
compiled it via make-kpkg.  Lastly I installed it with dpkg.  When I
rebooted everything seemed okay, but then I found that I had no
connection on eth0.  This is a dhl modem connected via a linksys
router, and is set up with dhcp3-client which has generally worked
fine.

Running ifconfig showed eth0 to be configured, and it looked just as
it usually does, at least as far as I could tell.  But, I could ping
nothing outside of my own box.  I also could not connect to my router
or the modem itself.  If I try pinging anything I can see the lights
on the front of the router flashing, as they should, but nothing is
coming through.

At this point I have tried to compile a 2.6.20-1, 2.6.20-11 and
2.6.21-1, and all have ended up the same.  Actually, the 2.6.21 also
gave me a kernel panic, but I am gathering this is a bug from the
sis900 module that my nic uses.  The mystery for me is that I just
don't know where to start looking for the trouble.  Dmesg looks just
as it always does when setting up eth0, and if I run ifdown eth0 and
ifup eth0 the response looks very normal too.  My resolv.conf is
also the same as always.  I just have no connection and no indication
as to why.

Where might I start looking for the trouble?  Is there a way to get
some more data with a hint of just what is not working as it should?
As of right now I just don't know where to look.

Many thanks,

Patrick


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

Re: Really slow xterm

[cothrige]

* Michelle Konzack ([EMAIL PROTECTED]) wrote:
> Am 2007-05-16 15:59:10, schrieb cothrige:
> > I have installed xterm via apt, running etch, and have noticed that it
> > scrolls really slowly.  I compared it to rxvt by running `time ls` in
> > /usr/bin with rxvt taking 0.572s and xterm running at 4.633s.  Earlier
> > it was even worse taking over 10 seconds.  It is a very big difference
> 
> It seems, there is somethong wrong with WOUR XTerm.

Actually, it has been worse at times than even above.  Yesterday, I
ran the same test and it took 20 seconds.  I thought that was pretty
awful.

> Maybe YOUR "rxvt" (Reduced X VirtualTerminal) load only
> "dash" and xterm the "bash" (Blown Advanced SHell)?

Nope, both are bash.
 
> :-)
> 
> Thanks, Greetings and nice Day
> Michelle Konzack
> Systemadministrator
> Tamay Dogan Network
> Debian GNU/Linux Consultant
> 

Thanks,

Patrick


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

Re: Really slow xterm

[cothrige]

* Thomas Dickey ([EMAIL PROTECTED]) wrote:
> cothrige <[EMAIL PROTECTED]> wrote:
> > Actually, it has been worse at times than even above.  Yesterday, I
> > ran the same test and it took 20 seconds.  I thought that was pretty
> > awful.
> 
> But at the same time rxvt would be running slower (due to your system load)
> since it's based on the load presented.

No, actually rxvt still ran as usual, and mrxvt a bit quicker still.
I would naturally expect both to be a bit quicker than xterm, but I
don't recall it ever being so obvious as it seems now.  While rxvt
seems pretty normal, regardless of the load, xterm always scrolls the
text so slowly that it is very, very obvious.  I used to use
xterm quite often, and it just never looked and acted this way.  It
actually reminds me of using a framebuffer console in how it is
working right now.

> On the other hand, running xterm remotely, I've measured rxvt running 5 times
> slower than xterm.
> 
> (though why someone realistically would do "ls -l /usr/bin" hasn't been said)

Well, the situation is that when I am using xterm the screen is
drawing and outputting very slowly.  Very visibly so.  I ran 'time ls
/usr/bin' simply as a test for the speed of the terminal emulators.
Since that directory was large I figured that I would get the best
most informative results by doing that.

> 
> -- 
> Thomas E. Dickey
> http://invisible-island.net
> ftp://invisible-island.net
> 

Patrick



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

Re: Really slow xterm

[cothrige]

* Thomas Dickey ([EMAIL PROTECTED]) wrote:
> 
> scrolling is one place to measure, but there are several (font choice
> is another, e.g., Xft is notoriously slow).  Older versions of rxvt
> would stop refreshing while they were getting input, and then paint
> the final screen (usually faster, but with some drawbacks).

I did give some thought to fonts, but was unsure of just how to be
thorough.  So, what I did was try various fonts other than the default
that I use, i.e. Terminus.  I tried fixed, 7x14, 10x20 and a couple of
others that seemed obvious.  None of this had any noticeable impact,
btw.  Of course, I am not sure about things like Xft, and how to
bypass that.  My fonts are basically the default setup for Debian
Etch, as I have never had the wherewithal to start messing with that
can of worms.  I can remember some real hairpullers trying to get
decent fonts in Slackware which I would not want to relive. ;-)

> 
> -- 
> Thomas E. Dickey
> http://invisible-island.net
> ftp://invisible-island.net
> 

Many thanks,

Patrick



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

Re: xterm fonts

[cothrige]

* Deboo ^ ([EMAIL PROTECTED]) wrote:
> 
> One more question is: how to get a font that is larger than 12x24? I
> remember having seen a nice SUN console font that was quite big. I
> want something similar. Which packages will install extra and bigger
> fonts for xterm or any other terminal for that matter, to use?
 
I use terminus, which is really very nice on the eyes, and can be
very large.  I run pxlsz (as reported in xfontsel) of 24, and it is
shown as available up to 32.  Actually, quite humongous really.   And
terminus is far and away the easiest on the eyes console font I have
ever used.

What I did was install xfonts-terminus via apt, and then opened the
/usr/share/fonts/X11/misc/fonts.alias file and added this to it:

ter-24 "-xos4-terminus-medium-r-normal-*-24-240-72-72-c-120-iso8859-1"

Then I restarted X and that gave me ter-24 as a font, and I could
avoid using that long string in scripts and configs, i.e. .Xdefaults.
Instead I could add "xterm.font: ter-24" to that file and then I have
a good xterm font working.


> Regards,
> Deboo

Patrick



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

Re: I am ANGRY with Debian.

[cothrige]

* Stephan Seitz ([EMAIL PROTECTED]) wrote:
> On Thu, May 31, 2007 at 08:01:20AM -0400, Marty wrote:
> >license, which *is* considered a free license.  In my opinion, all the 
> >analogies fall short because documentation is not software, regardless 
> >of Debian's dogmatic claims to the contrary.
> 
> If you mean with documentation some files you have on your computer, then 
> they are of course software. They may not be programs but they are 
> software.
> 
> Shade and sweet water!
> 
>   Stephan

I have never thought of this that way before, but it sounds like a
very sound approach.  As a person very supportive of the ideas of RMS
and the FSF I had some initial trouble coming to grips with why Debian
was taking this rather odd looking position.  But, after having read
some commentary and such online it seems to me now like the only thing
to do really.

People often seem to resent what looks like a personal political idea
getting in the way of the system.  In this case it is suggested that
Debian is being petty and fighting over trivial political stuff.
However, it seems to me that it is the other way around.  I think, and
I trust I will be corrected if I am wrong, RMS is trying to treat
documentation and manuals for free software like either a book or a
political manifesto.  I could understand his ideas about much of this
if he were discussing a print or online publication or article, but
not the documentation for free software.

It is argued that these 'invariant' sections must exist since without
them a person could present their own ideas as if they were those of
another, or vice versa.  In a book, or a political tract, this
matters, but really how does it apply to presenting the technicals on
using a piece of software?  And isn't that the same kind of thinking
which some use to attack free programs?  Why should specific comments,
cover sheets and so on have to be maintained in a work of this nature?
It really makes no sense, and comes down to a personal insistence on
the part of RMS that specific political commentary and the like be
kept a part of the work for ever.

I really do like Stallman's ideas, and those of the FSF.  I think they
have done immense good for software and the world in general.  It is
too easy to forget that systems like Debian are here very much because
of the "petty" bickering RMS.  As I see the likes of Ubuntu come up
and the inevitable suggestions about the silly positions of these
"fanatics" which just keeps us from having fancy things like Windows,
I can't help but think people are forgetting what things are all
about.  You should go home with the date that brought you, and we have
so much great software at least in great part thanks to the work of
people like RMS.  Where would free software be without fanatics
protecting free software?  However, sometimes people don't always get
it right, and this time it seems to me that Stallman is looking at
this wrong, and Debian's position just makes more sense overall.

Patrick



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

Re: I am ANGRY with Debian.

[cothrige]

* Nic James Ferrier ([EMAIL PROTECTED]) wrote:
> cothrige <[EMAIL PROTECTED]> writes:
> 
> > People often seem to resent what looks like a personal political idea
> > getting in the way of the system.  In this case it is suggested that
> > Debian is being petty and fighting over trivial political stuff.
> 
> I am NOT trying to debate the decision to regard GFDL as
> non-free. Although I think the decision is wrong that is NOT what I am
> complaining about.

And likewise, I have no intention of labelling you in particular in
any way.  I really meant my comments in a more general way,
considering the many posts and comments I have read online regarding
this issue.

> However, then taking that decision as an action point to remove all
> the documentation without putting it in elsewhere is simply not good
> enough. 

This has not troubled me much, but I can see your point.  However, I
would say that this really seems to bear less on the decision and more
on the implementation.

> A purely political decision has been taken that substantially changed
> *my* user experience (and that of others I'm sure) without warning and
> people think that's ok?

I think here you have hit a nail on the head.  So often, in these
post-Ubuntu days, we hear about this user experience, and how such
focus on things like openness of drivers, or freedom, is causing us to
lose out to the other guys, i.e. Mac and Windows.  I just don't buy
it.  The user experience is as good as it is *because* people have
insisted on this freedom.  The demand by many that things should be
free has given everyone else the opportunities all down the line to
make sure things could improve.  It is inevitable that as long as the
'politics' of free software remain paramount the user experience will
always improve.  However, the opposite could never be said.

Why are people so quick to throw away the very things that made what
they have possible?  I see it all the time here in America.  We had
freedom of speech, thought and association, and because of it built a
great nation.  Now that we have that great nation the first thing we
all seem to want to get rid of are those ridiculous freedoms of
speech, thought and association.  I just find it very strange.  We
should dance with the date that brought us, and in this case that
means insisting that software be and remain free.

And, another thing I think is funny is that so many talk about Debian
making political decisions (I don't know if that is what you mean
above, but I am inferring it from the context of the rest of the post
and thread) in a case where the real decision was by the authors of
the Emacs docs.  Why in the world does the documentation for a piece
of free software need invariant sections?  There are no invariant
sections in the program itself, so why the docs?  That is the 'purely
political decision' which has caused the problems, and not that of
Debian, at least as I see it.

Patrick


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

Re: I am ANGRY with Debian.

[cothrige]

* Manoj Srivastava ([EMAIL PROTECTED]) wrote:
> On Thu, 31 May 2007 19:59:27 +0100, Nic James Ferrier
> <[EMAIL PROTECTED]> said:  
> 
> > My name is Nic Ferrier. I am really ANGRY at Debian.  After 10 years
> > of being a dedicated Debian user I have reached the point at which I
> > am so angry with what is being done that I want to stop using it. I
> > will start to look for viable alternatives to Debian.
> 
> Good luck with your software choices elsewhere.  I suggest you
>  look at the GNU web site to see what OS they recommend; it may be more
>  to your liking, since you consider the GFDL licensed software to be
>  free.

One could go with gNewSense, http://www.gnewsense.org/, which is what
Richard Stallman uses.  It appears to be Ubuntu stripped of anything
that the maintainers would consider non-free.  I am assuming that
would not mean Emacs docs, as RMS endorses it, but would probably be
things like nvidia drivers I suppose.

Patrick


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

Strange reaction to keyboard at first login

[cothrige]

I have a strange thing happening when I try to login immediately after
boot.  It only happens once, and then everything is fine, but it still
bugs me.  After booting, which seems completely normal, the first time
I touch the keyboard this pops up on screen:

"input: AT Translated Set 2 keyboard as /class/input/input4"

This always causes the initial login attempt to fail, and so I need to
do it again.  Little thing, but I can't help thinking it shouldn't do
it, and there must be something I can do to stop it.  I tried doing a
google search but this seems to come up in other people's normal boot
process, rather than waiting for a first input from the keyboard.  So,
I can't seem to find any hint about how to stop it from happening
then, or to happen earlier when perhaps it should.

Is there a setting somewhere which will anticipate this problem?
Have I configured something incorrectly somewhere?  Not being a usb
keyboard or anything I didn't think it needed any configuration, but
maybe there is something I can change to alter the current behaviour.

Many thanks in advance,

Patrick


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

Re: I am ANGRY with Debian.

[cothrige]

* Wesley J. Landaker ([EMAIL PROTECTED]) wrote:
> 
> Well since gNewSense is a derivative of Ubuntu which derives from Debian, by 
> default the emacs docs would be out as well, unless they add them back in 
> themselves. Looking at gNewSense, I got the impression that they only 
> removed things and didn't actually modify packages to add things back in.

I was aware that gNewSense was derived from Ubuntu, but I suppose I
had not given much thought to how Ubuntu themselves handle these
things.  Do they just redistribute these things straight from Debian?
I suppose I simply assumed that they would include the documentation
for Emacs, either by restoring it to the Debian package or by
repackaging it entirely.  It makes me a bit curious thinking about it.

Patrick


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

Re: I am ANGRY with Debian.

[cothrige]

* Romain Francoise ([EMAIL PROTECTED]) wrote:
> 
> It's very simple: Debian is no longer upstream for emacs-snapshot in
> Ubuntu, the packages are taken from my personal repository.
> 
> See https://launchpad.net/ubuntu/+source/emacs-snapshot/1:20070529-1
> 

Thanks for that info.  Does that then mean that the Ubuntu Emacs does
include the docs that Debian removes?

Patrick


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

Re: Strange reaction to keyboard at first login

[cothrige]

* Chris Bannister ([EMAIL PROTECTED]) wrote:
> On Sun, Jun 03, 2007 at 09:21:36AM -0500, cothrige wrote:
> 
> Is this a stock Debian kernel or a self compiled one?
> Is this from the CLI or from a GUI? (i.e x or no x)

Sorry, I should have mentioned that I am running a 2.6.20.12 kernel
that I compiled myself.  And it is a problem that has only existed on
this kernel, so far.

> It seems like the kernel module for the keyboard isn't being loaded at
> boot but on demand - and even then, not loading cleanly for some reason.

My keyboard is a very standard PS/2.  Which module needs to load for
that?  I don't remember any odd configurations for the keyboard when I
compiled, but I will go back and look through the config again just to
see if I am overlooking something.

> Is there anything slightly sus in the syslog boot messages?

Well, not that I can see.  But, I will admit that I am likely
overlooking something.  The complaint, "input: AT Translated Set 2
keyboard as /class/input/input4", is in /var/log/messages.  Earlier in
the same file there are more of the same, but instead of
/class/input/input4 they say /class/input/input0.  I don't remember if
that ever came up on screen, but I only remember the input4 quite
recently, and I think these older ones predate my current kernel and
the problem I am dealing with right now.

Patrick


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

Re: Strange reaction to keyboard at first login

[cothrige]

* Chris Bannister ([EMAIL PROTECTED]) wrote:
> On Mon, Jun 04, 2007 at 12:13:22PM -0500, cothrige wrote:
> > Well, not that I can see.  But, I will admit that I am likely
> > overlooking something.  The complaint, "input: AT Translated Set 2
> > keyboard as /class/input/input4", is in /var/log/messages.  Earlier in
> > the same file there are more of the same, but instead of
> > /class/input/input4 they say /class/input/input0.  I don't remember if
> > that ever came up on screen, but I only remember the input4 quite
> > recently, and I think these older ones predate my current kernel and
> > the problem I am dealing with right now.
> 
> So, have you found the problem?
> 

Nope.  Sad to say I have not.  I have looked into the possible
modules, and I could only find one which seems a candidate for my
problem, ATKBD, and I have this compiled in.  I have yet to find
anything else which may give a clue as to what exactly is going wrong
and so I am just not quite sure what I can tweak right now.  I have
considered recompiling the kernel and using the above as a module and
seeing what happens then, but since I have never used that as a module
before I am just not very confident it will help.

In the meantime I have simply developed a habit of tapping the enter
key twice after booting, and this clears the trouble.  I don't like
that, and I really do think things should work, but at least
everything seems to operate until I can get actually find out what is
going wrong.

Patrick


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

Re: screenshot

[cothrige]

* Frank McCormick ([EMAIL PROTECTED]) wrote:
> 
> Anybody know of relatively simple software to take screenshots of the
> dexktop or of windows of applications. Gnome-utils contains a screenshot
> maker but its a 6 meg download and am 18 meg installation!!

I believe the Gimp has a pretty straightforward tool for this.  And
there is import that comes with ImageMagick.  It is a commandline app
and works very well, in my experience.  The online documentation has
always seemed very lacking, but there are very comprehensive sets
somewhere on the internet, but I cannot recall where exactly.  I
usually have to do a google search everytime I need to figure out a
flag.

Patrick


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

Trouble with cron

[cothrige]

I am a Debian newb, coming from Slackware, and am trying to set up a
cronjob as a user.  In other systems all I did was crontab -e and then
added the job.  After if I ran crontab -l it was listed right there
ready to go.  However, this is not working in Debian.  Cron is
installed and running, but when I enter the command and save it
crontab -l lists nothing at all, and nothing runs.  Is there some step
I am overlooking?  I have never had to do otherwise before, and have
tried reading online for Debian specific tips, but so far everything
comes back the same as I would expect.  But, still, no cronjob.

Any suggestions are greatly appreciated.

Patrick


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

Re: Trouble with cron

[cothrige]

* Johannes Wiedersich ([EMAIL PROTECTED]) wrote:
> 
> Which editor are you using? Please do a
> ll /etc/alternatives/editor
 
Oddly it says nano as well, but my EDITOR and VISUAL variables are
both set to emacs, which is that which I use generally.  And that is
what was invoked with crontab -e both times, and not nano.  BTW, I
found that just as with Slackware, the EDITOR variable does not seem
to change the crontab editor, but VISUAL does.  I found it out as in
trying to intentionally use vi instead I had to change VISUAL to make
it come up.  Nothing online seemed to indicate this is so but the
Debian helps and such all mention EDITOR.  Have you had this
experience?

> since I configured nano as the default editor. I suspect that you might 
> unknowingly run an unfamiliar editor and  might unknowingly quit without 
> saving.

Okay, I think I figured it out.  In typing this email I was going to
yank in the output of crontab -l as it is odd.  At first I was running
very small test tasks just to see if it was doing something.  Nothing
useful was produced by crontab -l. But when I tried some more
complicated echoes and such I found that my prompt was being printed
over the output and then a new prompt put in right after.  I then
realized one mistake I was making, though I doubted it was the main
problem, and that was no newline at the end of the line.  I added this
and it would seem it may have been the problem all along.  I am going
to continue to test this and see for sure, but as of now all I can say
is be careful to include a newline at the end of the last line in the
crontab.  I never realised this was essential.  Probably always had
one without thinking about it before.

> 
> You can change it to your favourite editor (provided it is installed) by 
> running
> 
> update-alternatives --config editor
> 

This a cool feature, but what about the nano thing above.  I don't get
nano for anything, and didn't even know it was installed.  Is the
above configuration really even doing anything?  How can I find out?

> In contrast to other distributions, in debian one can configure 
> different alternatives for common tasks.
> 
> On my sytem, eg., there are 6 alternatives for an editor and
> 3 alternatives for different versions of ghostscript.

I would be interested in knowing how you set that kind of thing up.
Can you direct me to a website with some info?
 
Many thanks for the help Johannes.

Patrick


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

Re: Trouble with cron

[cothrige]

* michael ([EMAIL PROTECTED]) wrote:
[snip...]
> > 
> > Okay, I think I figured it out.  In typing this email I was going to
> > yank in the output of crontab -l as it is odd.  
> 
> hang on, I thought you said it had *nothing* before

Yes, that was my mistake as it looked like nothing.  It was just a
mangled prompt, but that was because it was printing a very short line
from crontab -l without the newline, and then another prompt.  For
some reason I still don't understand, overtop of this on the same line
it would print  yet another prompt, giving something like this:

cothrige (celephais:~) $ crontab -l
cothrige (celephais:~) $ cothrige (celephais:~) $ 

Though apparently there is an output to crontab -l in that, it is
completely covered by the first prompt, and that is because it is so
short.  When I used a longer command it then showed partly and that
tipped me off that it was in fact finding a command, but due to the
missing newline it was mangling the output, and apparently not able to
execute the command.  Now it seems to be working fine.

Thanks again for the help.

Patrick

Re: Trouble with cron

[cothrige]

* Johannes Wiedersich ([EMAIL PROTECTED]) wrote:
> 
> As said by michael in another reply, the alternatives are overriden by 
> EDITOR or VISUAL.
 
Yes, that would make sense, but I had unset these in trying to get vi
to work with crontab.  I guess I had assumed that vi would be the
standard editor and thought that commenting out the variables in my
.bashrc and then sourcing that would bring up vi or vim.  But, emacs
still came up.  In looking back it would seem that nano should have.

It reminds me of gdm.  I don't usually use that kind of thing but
since it was default in Debian I thought what the heck.  So, I
switched to fluxbox to see what happened and it starts up conky.  I
like conky, and do use it, but I didn't tell it to do that.  However,
the desktop was black and so I thought I would set a wallpaper, and to
do that I created a .Xsession file and added fbsetbg to that and
restarted the X server, but nothing.  I copied it to .Xclients and did
it again.  Nothing.  I copied it to .xsession and then .xclients.
Nope.  I checked that they were all executable.  No good.  So, I added
a script to /etc/X11/Xsession.d to set a wallpaper, and still no good.
So, I made the gdm script in /etc/init.d unexecutable and created a
.xinitrc to run with startx.

Admittedly, I am no expert with gdm, or any login manager.  But, I do
recall using it once in Gentoo and I just created the .Xsession file
and that did everything, including the window manager.  Now, nothing.
And the same with EDITOR.  I have to export it explicitly or I get
emacs everytime.  Could Debian just be so good it knows what I want
without having to tell it?  So far I have conky and emacs, both
unrequested.  Man, you people are good! 

> nano is just another editor - not as powerful as emacs, but sufficient 
> for me editing my crontab etc.

Isn't it also called pico?  I recall using that once way back when and
it was okay as I recall.  But Jed has always been my low resource quick
and dirty editor.  At least for when emacs is cumbersome.

> Just install different packages, the corresponding entries should 
> 'automagically' appear, when you run the corresponding 
> update-alternatives. See 'man update-alternatives'.
> 
> There is also a short summary on
> http://www.debian.org/doc/manuals/reference/ch-package.en.html#s-alternatives
> which is always good to have.

I may try this out and see what it does.  Perhaps if I change it to vi
or something and then unset my editor again it will kick it into
action.  Be interesting to see.

> You could install debian's reference to
> /usr/share/doc/Debian/reference/
> by running
> aptitude install debian-reference-en

This sounds like a great suggestion and I will definitely do it.  I
like plenty of documentation at my fingertips.

Continued thanks for all the tips and information.

Patrick


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

Two questions about boot scripts

[cothrige]

I don't particularly like gdm and so I made the script in init.d
nonexecutable.  I feel, though, this was likely the wrong way.  What
is the right way to do it?  The whole Debian bootscript system is
somewhat intimidating to a Slackware user, and so I am hesitant to go
in there willy-nilly.

Also, where would I be best advised to add something to start esd, as
in '/usr/bin/esd -nobeeps'?  Is there a config file which will be seen
and read during boot?

Many thanks for any help,

Patrick


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

Re: Two questions about boot scripts

[cothrige]

* derek ([EMAIL PROTECTED]) wrote:
> Hello
> create the script in /etc/init.d/
> lets say its named myscript
> make it executable
> then do /usr/sbin/update-rc.d myscript defaults
> that will create a sym link in all the run levels
> Derek

That sounds straightforward enough.  Many thanks.

Patrick


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

Re: Two questions about boot scripts

[cothrige]

* Dekxter ([EMAIL PROTECTED]) wrote:
> GDM is disabled by running:
> # update-rc.d -f gdm remove

Thanks a lot.  I can't believe how much trouble I was having just
trying to figure that little item out.

Patrick


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

Re: Trouble with cron

[cothrige]

* Serban Udrea ([EMAIL PROTECTED]) wrote:
> Hello,

Howdy,
 
> I am on Slackware 10.1 box right now and the man page for crontab states 
> that crontab -e will use  /usr/bin/vi or VISUAL. It does not mention 
> EDITOR. On this box if VISUAL is not set, vi is run. If I set VISUAL to 
> something else I get that.

That may be where I got the notion that vi would default.  I can just
remember that in the past vi, or Elvis in Slack I think, was always
what I got when I edited my crontab.  I suppose I had assumed I would
in Debian as well.

> Maybe strace can help to find out what crontab actually does to get emacs 
> started. And maybe first check that VISUAL is indeed unset.

I may just try strace.  Maybe it isn't a big deal, but I am a little
curious about why it is behaving oddly.

Thanks for the advice.

Patrick


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

Re: Two questions about boot scripts

[cothrige]

* Mumia W.. ([EMAIL PROTECTED]) wrote:
>
> Install the Debian Reference (debian-reference-en) and read §§ 2.4.2 and 
> 2.4.3 and 8.1.4 from /usr/share/doc/Debian/reference/reference.en.txt.gz .
 
I am hoping to sit down this weekend and read through the reference.
It looks to be a rather complete coverage of things Debian, and I am
quite hopeful it will iron out much I don't know.  I have been
wondering about specifics for maintenance, such as updates and so on,
and I figure that will be well covered in this and other texts.  I
like the apparent complete approach to documentation in Debian, though
so far things have been a little fragmented it seems.  I am confident
given time everything I need to know will be covered.
 
> You can create a new script such as /etc/init.d/local.sh and put all of 
> your initialization commands in it. Then you would use update-rc.d to 
> tell Debian what runlevels to start and stop that script's daemons and 
> settings, e.g.
> 

I found /etc/rc.local and it seems to be for the purpose I was
looking.  Surprising to see it there, as I thought all the init
scripts were in /etc/rc... or /etc/init.d.

> # # That was a Control-D. We're now out of "cat."
> # update-rc.d local.sh start 2 3 4 5 . stop 0 1 6 .
> 
> Read "man update-rc.d"

Very helpful little tool.  Though I am still wondering why Debian has 
runlevels 2-5 with no apparent difference between them.  I have to
quit assuming that one is for console logins.

Many thanks again for all this help.

Patrick

System maintenance

[cothrige]

In reading online it seems that the standard practice to apply
security patches would be to run 'apt-get update' and then 'apt-get
upgrade'.  I am curious if this really is the best way and if so, how
often should it be done?

I use Fluxbox, and quickly switched from the default Gnome when I
first installed.  But, before doing that, I noticed that there was
monitor of some sort which popped up in the notification area
announcing available updates.  How reliable is this tool, and can it
be used from outside of Gnome?

Just trying to get a grip on system maintenance and hoping to hear
some ideas from those here.  Thanks in advance,

Patrick


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

Re: System maintenance

[cothrige]

* Roberto C. Sanchez ([EMAIL PROTECTED]) wrote:
> Moving this back on list so every gets the benefit.
> 

Very sorry.  I hit the r instead of L.  I tend to do that when I am
not thinking, and that is too often.

> With Etch there will be more updates.  I also recommend against having
> them automatically installed since they may be more disruptive.  The
> main problem is that the freeze is impending.  This means that there is
> a flurry of activity as maintainers try and get updated version of their
> packages into Sid in time for them to migrate to Etch.  Your call.

I looked at the man page, and it is somewhat meager.  I looked online
and most info was equally so, and included mostly recommendations that
a line for MAILON="always" be included.  And the config file is a bit
hairy.  So, I am curious now about what this really does behind the
scenes.

For instance, what does this do differently than apt-get update and
apt-get upgrade (or dist-upgrade)?  When you say you manually install
them does that mean you run dpkg on each package in turn, or that you
merely run apt-get upgrade yourself?  I guess being rather new to the
Debian ways I am still seeing in a fog on some of these things.

Many thanks again for the help.

Patrick


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

Re: System maintenance

[cothrige]

* Kevin Mark ([EMAIL PROTECTED]) wrote:
> Hi Patrick,

Hello Kevin,

> the best option if you want stability and (little or) no breakage is to
> run 'stable'. This is what Debian releases. Although there is now
> security support for testing also.

I had actually intended initially to install Stable, but had troubles
with the installer.  In the end it turned out to be an unreliable cd
drive that was causing my troubles, but I didn't know that at the
time, and so I ended up using etch later.  I figured that as a
Slackware guy I would end up wishing I had found a way to install
stable, but as this is all kind of an experiment perhaps it will work
out okay.

I also kind of assumed that like the last three or four times I tried
Debian I would get very frustrated and go back, but I have surprised
myself in how much I have actually enjoyed it this time round.  I
think something changed in the installer and I actually managed a
better install than I had before, though maybe I just did a better job
of reading.  But, in any case, I am learning a great deal about Debian
and have certainly decided to keep the system and so my decision of
testing vs stable has taken on a greater significance than I had
thought it would when I first started to consider it.

> >=20
> 'apt-get update' gets the latest information about what packages have
> been updated but does not install them. 'apt-get upgrade' installs (but
> does not remove) the new packages while 'apt-get dist-upgrade' can both
> install and remove packages. This is a paraphrase, so read 'man apt-get'
> for more complete info.=20

I have been debating just which I should go with, upgrade or
dist-upgrade.  If I had stable I would almost certainly go with
upgrade, but I can't help but think that since I am certainly going to
get more than security patches then perhaps dist-upgrade will apply
better.  I have noticed that apparently the little notification tool
in Gnome uses dist-upgrade, so I would assume that at some level
somebody thinks of that as default, though just whom and why I
couldn't say.
=20
> Using stable is the main benefit to Debian. If you do not use stable
> (like me, as I use unstable), then all the above does not apply.  You
> have to update/dist-upgrade as often as you can to keep up with bugs and
> security issues and the chance of installing something that may break
> your system is not as near to zero as it is with stable. This does not
> mean that Debian unstable is horrible and broken, it just takes more
> effort to keep working, secure and up-to-date which is why I use
> apt-listbugs and apt-changes!

I am not familiar with apt-listbugs and apt-changes.  What are those?

One thing that I am now curious about is the setup of sources.list and
the release of the next stable.  For instance, my sources.list was
initially setup with entries for 'etch' such as "deb
http://ftp.ndlug.nd.edu/mirrors/debian/ etch main contrib non-free" I
have seen online some people have 'testing' where mine says etch.
Would having etch mean that when the next release of stable comes out,
which I am gathering will be etch, I will then be running stable?  Or
is there some other entry somewhere that makes what I am using remain
'testing'?  The reason I ask is that come that day I am not sure if I
will want to continue to follow testing or possibly have stable, but
it seems that it may be a good time to consider just drifting up to
stable.  In the meantime I cannot imagine how a person would switch
=66rom testing to stable without a full reinstall.  Is that right?

Many thanks

Patrick


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

Re: System maintenance

[cothrige]

* Steve Kemp ([EMAIL PROTECTED]) wrote:
> On Mon, Oct 16, 2006 at 08:36:20AM -0500, cothrige wrote:
> 
> > I am not familiar with apt-listbugs and apt-changes.  What are those?
>  
>http://www.debian-administration.org/articles/44
> 
>They show you outstanding bugs/changes which will be applied when
>   you upgrade.

Very cool.  Good to know about that.  But, I found that my system
doesn't have any idea about anything like apt-listbugs.  I followed
the link in the article and found that there were packages listed for
stable and unstable.  How does one normally proceed from there?  Will
there typically be a testing package later?

> > One thing that I am now curious about is the setup of sources.list and
> > the release of the next stable.  For instance, my sources.list was
> > initially setup with entries for 'etch' such as "deb
> > http://ftp.ndlug.nd.edu/mirrors/debian/ etch main contrib non-free" I
> > have seen online some people have 'testing' where mine says etch.
> > Would having etch mean that when the next release of stable comes out,
> > which I am gathering will be etch, I will then be running stable? 
> 
>   Exactly.

Very nice.  That should make either choice fairly bump free, I would
think.

Thanks,

Patrick


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

Re: System maintenance

[cothrige]

* [EMAIL PROTECTED] ([EMAIL PROTECTED]) wrote:
> 
> If you want a little more control over the details, with a usable 
> text-based user interface, use 'aptitude'.
> after it's started in a text console (very useful if your X is broken)
> the command 'u' updates its package lists, 'U' then does the same as
> apt-get upgrade, except that it just decides which things to upgrade 
> without doing it, 'g' tell sit to go ahead and do it -- except again, it 
> pauses once for you to vies the entire list of proposed changes.  You 
> can then edit the list if you wish, and do another 'g' to get it to 
> actually make the changes.

I have been wanting to try aptitude (so far I have used apt-get) but
it has intimidated me in the past.  But, it really does sound decent
and so I think I will bite the bullet and do it.  After all, this has
all been driven by the desire to learn how to use Debian.  And I
really like the sound of its ability to remove unused dependencies
later.

Many thanks for the info on this.

Patrick


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

Re: System maintenance

[cothrige]

* Andrei Popescu ([EMAIL PROTECTED]) wrote:
> On Mon, Oct 16, 2006 at 08:36:20AM -0500, cothrige wrote:
> 
> Scenario1: You install stable (now sarge) and the entry in your sources.list
> is 'stable'. When etch will be released the next dist-upgrade will upgrade
> your whole system to etch *without* reinstalling. You should read the release
> notes first though, especially for production machines.

Unfortunately I would have to reinstall, and now that I have things
moving along nicely I would hate to go through that.  Plus, since
things are not really messed up yet I suppose I have no reason to go
to stable at this point.  And didn't I read somewhere that December is
the date for the next release?  If so, I see even less reason to hurry.

> Scenario2: You installed testing but your sources.list has 'etch'. If you 
> don't
> change this you will allways stay with etch, when it becomes stable and
> eventualy old-stable. Again, no reinstall needed. 

I think this is likely to be what I do.  As I see it, I can always
upgrade to testing again, at least much more easily than trying to
move down to stable.  My plan therefore is to remain with etch until I
find that I just can't live without the newest things, and since I
have spent the last four years or so with Slack, I have a feeling that
will take a little while.

> If you want to stay with
> testing you could change your sources.list to point to testing. Some people
> on this list say testing gets very broken just after the release, so you
> might want to stay with etch until things stabilize in testing and only
> then do the dist-upgrade.

I am very thankful for this heads up.  Fortunately, I was likely to
remain with etch anyway, at least for a while.  The bleeding edge is
not my natural terrain anway.

Thanks Andrei.

Patrick


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

Re: ls sort order again

[cothrige]

* Ben Breslauer ([EMAIL PROTECTED]) wrote:
> T wrote:
> >Hi
> >
> >I am using Debian testing, I read that the ls is able to sort
> >alphabetically, but mixes uppercase and lowercase together i.e. 'Pearl'
> >comes before 'pearl' but after 'otter'.
> >
> >otter
> >Pearl
> >pearl
> >
> >I want that behavior. How can I do that, instead of the traditional order?
> 
> I'm using testing, and it's doing this for me by default.
> 
> --Ben

I'm also using testing, and mine also does this by default.  And I
don't know if I really like it.  Maybe too long with asciibetical, but
it looks funny to me this way.

Patrick


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

Starting iptables

[cothrige]

I was wondering about the best way to start iptables with each boot in
Debian and so I did some googling.  I found a Debian Wiki and it gave
instructions concerning update-rc.d, but this requires a script for
iptables in init.d and this does not exist.  At least not in my
system.  Can anyone give me a quick idea of the "correct" way to start
iptables at boot?  I suppose I could just add something to rc.local,
but I was sort of assuming there was a more appropriate way.

Many thanks,

Patrick


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

Re: Starting iptables

[cothrige]

* Andrew Sackville-West ([EMAIL PROTECTED]) wrote:
> 
> isn't iptables part of the kernel and therefor up by default when the
> kernel starts executing? 
> 
> A

Yes, iptables as far as I know is part of the kernel, but the rules
must be loaded.  In Slackware I would create a script and put it in
rc.d to be loaded.  I suppose that I could do something similar with
Debian, but would like to make sure that there is not some more
correct way to handle it first.

Patrick


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

Re: Starting iptables

[cothrige]

* Roberto C. Sanchez ([EMAIL PROTECTED]) wrote:
> > 
> Use shorewall.
> 
> Regards,
> 
> -Roberto


I was just looking at a howto on this.  I have never used any of these
tools before as I already had a firewall script which worked.  But,
maybe now is as good a time as any to learn how this works.

Patrick


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

Re: Starting iptables

[cothrige]

* John Hasler ([EMAIL PROTECTED]) wrote:
> Patrick writes:
> > I suppose that I could do something similar with Debian, but would like
> > to make sure that there is not some more correct way to handle it first.
> 
> No more correct but more sensible would be to install one of the several
> packages that do exactly what you want.  I like ipmasq.

In the process of googling for info on this I did find a page at 
http://www.debian-administration.org/articles/445 which had a fairly
simple approach.  There it recommends putting a script in
/etc/network/if-up.d to run iptables-restore.  This is similar to the
Gentoo way, except that was somewhat automated with a script in init.d
which, oddly, Debian is lacking.  Assuming this works it does have
simplicity on its side though, and that is always a good thing.

Patrick


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

Re: Starting iptables

[cothrige]

* H.S. ([EMAIL PROTECTED]) wrote:
> Well, my custom firewall script does take start, stop and restart
> arguments and so I could call it using the rc method. However, I have
> thus far used it by calling it with a pre-up line in the stanza for my eth0:
> pre-up /etc/myfirewall/firewall.sh restart

You added that line to /etc/network/interfaces, right?  Does it matter
just where you put it in the script?

Patrick


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

Re: Starting iptables

[cothrige]

* Kevin Mark ([EMAIL PROTECTED]) wrote:
> > 
> Hi Patrick,
> most folks just run 'shorewall'! And you can add more rules if you need
> to.
> =Kev

This does seem to be the consensus here.  However, as I have never
used this tool it is a bit intimidating.  And the documentation is so
vast it may be a bit of an overkill for my very simple purposes.  You
see, I have only one NIC which is connected to a Linksys router, which
in turn is connected to the modem.  My modem does its own firewalling,
but I cannot bring myself to rely entirely on it, and always set up my
own as well.  But, because I have only one NIC I can never quite
figure out what to do with loc in the zones, which in the
documentation and such is always eth1, which I don't have.  Should I
not have a loc zone?  Or do I just have eth0 for both net and loc?

Patrick



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

Re: Starting iptables

[cothrige]

* Andrew Sackville-West ([EMAIL PROTECTED]) wrote:
> 
> read this
> 
> http://www.shorewall.net/standalone.htm
> 
> A

Well, there you go.  I was completely on the wrong side of the docs.
Thanks for this shortcut.  It seems pretty straightforward too.  Took
me about five minutes to follow it and get things going.  The well
commented config files, especially intefaces, made things a little
better too.

Many thanks,

Patrick


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

Re: Starting iptables

[cothrige]

* [EMAIL PROTECTED] ([EMAIL PROTECTED]) wrote:
> 
> As I see it, you have two choices.  If you just want something that
> should do what you want and don't want to have to set anything up, just
> install ipmasq.  It determines what the untrusted network is by where
> the default route or gateway points; its automatic.  If you want the
> tightest firewall with only the ports you want open, then go with
> shorewall.  

Interesting what you say about ipmasq.  How automatic is it?  I would
have assumed that it had more to do with making your machine a
gateway, which mine isn't, than firewalling itself.  I am assuming
that it does both?  

> The documentation is vast; its like a book.  You wouldn't buy a big book
> on network security and open it to the middle and expect to know what
> was going on.  Start at the beginning and just read it through.  Trust
> your brain to synthesize and develop a plan for your situation.

I know what you mean there.  I think it turned out to be something
like 550 pages, give or take.  And I actually was reading it from the
beginning, but you can imagine what a task that is just to set up a
couple of rules.  And I was beginning to think that it was not set up
to handle a situation as simple as mine.  Of course, I was wrong.

But, this all begs the question of what Shorewall is really trying to
do.  I would think that the point of these firewall tools would be to
get around the rather difficult process of figuring out iptables.
However, shorewall seems to simply replace the very archaic and tricky
iptables commands and structure with its own equally difficult
version.  Why is that exactly?  Couldn't somebody with that kind of
need simply take the same time and learn the very thing that Shorewall
is manipulating, i.e. iptables?

Patrick


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