Linux-Misc Digest #312, Volume #26               Tue, 14 Nov 00 23:13:02 EST

Contents:
  Re: True GTK+ will eliminate Qt in next few years? ("Evan DiBiase")
  Re: FreeDialup for Linux? (Julian Bordas)
  Re: rexec not working (ljb)
  Re: ethernet speed! (Mike Pumford)
  v4l & xawtv & ibmcam problems ([EMAIL PROTECTED])
  Re: Multithreaded RPC Server Question ([EMAIL PROTECTED])
  Re: Program to convert Unix file format to DOS / Windows format. (Scott Alfter)
  future of the Desktop Wars (Tyler Larson)
  Re: LinkSys betrayed us!  Poor prospects for Linux. (ljp)
  Problems with printing to HP 500C under Debian Linux 2.2 (potato) (Charles Pouliot)
  Re: configuration for mainboard-incorporated video card ([EMAIL PROTECTED])
  cannot modify file by root!!! ("Hello World")
  Re: True GTK+ will eliminate Qt in next few years? (Donovan Rebbechi)
  Re: xmms playing CDs? (John Scudder)
  Is there some kind of "graphical" trafshow? (Dr Aldo Medina)

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

From: "Evan DiBiase" <[EMAIL PROTECTED]>
Subject: Re: True GTK+ will eliminate Qt in next few years?
Date: Tue, 14 Nov 2000 19:23:48 +0500
Crossposted-To: comp.os.linux.advocacy,comp.os.linux.x

In article <[EMAIL PROTECTED]>,
[EMAIL PROTECTED]  wrote:

> In article <[EMAIL PROTECTED]>, Henrik Keiler
> <[EMAIL PROTECTED]> wrote:
>> GTK+ is history. Microsoft is planning a massive lawsuit against GTK+
>> authors and Gnome  fondation starting in Q1/2001. They claim that GTK+
>> violates both US  patents (we don't care about software patents here in
>> europe, but...)  and their intellectual property (even europe cares
>> about that !!). Shipping a product based on GTK+ will be dangerous - if
>> this lawsuit  will be won by microsoft (and GTK+ _is_ violating their
>> patents - that  is the primary problem: M$ has patents covering what
>> GTK+ does... ;-(( )  all vendors who are using GTK+ and GTK+-based
>> products (Gnome) will have  to pay license fees. I won't recomment
>> using GTK+ until these "issues" have been solved -  otherwise you may
>> have to pay many $$ to M$... ;-(

Bah, I'll believe it when I see the links to the patents that GTK
violates. Until then, this reeks of a scare tactic (why someone would
want to scare people away from GTK, I don't know).

-Evan


====== Posted via Newsfeeds.Com, Uncensored Usenet News ======
http://www.newsfeeds.com - The #1 Newsgroup Service in the World!
=======  Over 80,000 Newsgroups = 16 Different Servers! ======

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

From: Julian Bordas <[EMAIL PROTECTED]>
Subject: Re: FreeDialup for Linux?
Date: Wed, 15 Nov 2000 11:44:03 +1100

Dave Nash wrote:
> 
> J wrote:
> 
> > Does anyone know of Free Dialup Service for Linux?
> 
> Are you talking about an ISP?  Most use standard PPP with CHAP or PAP, even
> if they say you need a Windows or a CD.  Ask for the connection details, or
> install on Win and copy the details from the dial-up networking icon.
> 
> Dave

You do get free dial up ISP's that use a windows dial up programme This
doubles as an advert downloader.  I have been unable to connect to this
via wine under linux.


-- 
Julian Bordas

"Windows is yucky. It keeps breaking"
Edward Bordas Age 5.5

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

From: [EMAIL PROTECTED] (ljb)
Subject: Re: rexec not working
Date: 15 Nov 2000 01:52:47 GMT

[EMAIL PROTECTED] wrote:
>Hi,
>    I'm trying to get rexec to work on my RH 7.0 machine.  When I try to rexec
>to a unix box I get an error that says connection refused.  When I try to rexec
>to my own linux box I get the same error.  Any ideas?

"Connection refused" means no process is listening at that port. rexec's
server uses the "exec" service, TCP port 512. Does netstat -ant show any
listening process there? Do you have the appropriate line uncommented from
/etc/inetd.conf (rexecd or in.rexecd)?

By the way, you will probably get replies saying that rexec is a bad
thing, security-wise (it sends clear-text passwords over the net).
Be advised.

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

From: [EMAIL PROTECTED] (Mike Pumford)
Crossposted-To: comp.unix.bsd.freebsd.misc
Subject: Re: ethernet speed!
Date: 15 Nov 2000 00:21:15 -0000

In article <[EMAIL PROTECTED]>,
Antony Mak <[EMAIL PROTECTED]> wrote:
>How to know it is running 10MB or 100MB and Full or half duplex?
>
Here is an example taken from my system:

$ ifconfig -a
dc0: flags=8843<UP,BROADCAST,RUNNING,SIMPLEX,MULTICAST> mtu 1500
        inet 192.168.1.2 netmask 0xffffff00 broadcast 192.168.1.255
        inet6 fe80::2a0:ccff:fe5f:7d17%dc0 prefixlen 64 scopeid 0x1 
        ether 00:a0:cc:5f:7d:17 
        media: autoselect (10baseT/UTP) status: active
        ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
This is the relevent output line which tells me that the card is currently
doing 10MB half duplex ethernet.
        supported media: autoselect 100baseTX <full-duplex> 100baseTX
                 10baseT/UTP <full-duplex> 10baseT/UTP none
        ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
A 100MB card would report one of the 100baseTX media types.

If ifconfig -a says something like media: manual then you will have to
do performance tests to measure the link speed. You may also be able
to examine status lights on the hub to determine the speed.

Mike



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

From: [EMAIL PROTECTED]
Subject: v4l & xawtv & ibmcam problems
Date: Wed, 15 Nov 2000 01:58:10 GMT

I've been struggling with a cheapo camera that was given to me a couple
of days ago.  I dual boot between Win98 and RH 7.  I got the camera up
an running in Windows in a matter of minutes.  Works great.

My Linux installation is another matter...   And I need some help.

I downloaded the xawtv package and built and installed it with no
problems.  Well, one problem- it doesn't work.

/dev/video does exist and does link to /dev/video0.  When I run xawtv
the ibmcam and videodev modules load up.  /proc/bus/usb/devices shows
the camera.  So I think I have the drivers and USB system set up
properly.

When I run 'xawtv -v 2' I get the following output:

This is xawtv-3.23, running on Linux/i686 (2.2.16-22)
visual: id=0x23 class=4 (TrueColor), depth=16
visual: id=0x24 class=4 (TrueColor), depth=16
x11 socket: me=localhost, server=localhost
DGA version 2.0
Xv: 0 adaptors available.
Xv: no usable video port found
Xv: no usable hw scaler port found
x11: color depth: 16 bits, 2 bytes - pixmap: 2 bytes
x11: color masks: red=0x0000f800 green=0x000007e0 blue=0x0000001f
x11: server byte order: little endian
x11: client byte order: little endian
x11: 1024x768, 16 bit/pixel, 2048 byte/scanline, DGA
init: trying: v4l2...
v4l-conf: using X11 display :0
dga: version 2.0
mode: 1024x768, depth=16, bpp=16, bpl=2048, base=0xd6000000
/dev/video: no overlay support
got sigchild
waitpid: No child processes
v4l-conf had some trouble, trying to continue anyway
init: failed: v4l2
init: trying: v4l...
v4l-conf: using X11 display :0
dga: version 2.0
mode: 1024x768, depth=16, bpp=16, bpl=2048, base=0xd6000000
/dev/video: no overlay support


The 'no overlay support' and Xv messages have me a little worried.  Any
advice?  It works once for a while (but the video is very choppy on my
800 MHz Athlon with 16 MB TNT card).  The second time it does not work
and I have to reboot to get the light on the card to shut off.

Please help!
Steve


Sent via Deja.com http://www.deja.com/
Before you buy.

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

From: [EMAIL PROTECTED]
Subject: Re: Multithreaded RPC Server Question
Date: Wed, 15 Nov 2000 02:00:26 GMT

Thanks for your help.  In text below:

In article <8usafp$nb5$[EMAIL PROTECTED]>,
  Arun Taylor <[EMAIL PROTECTED]> wrote:
> In article <8u821c$li8$[EMAIL PROTECTED]>,
>   [EMAIL PROTECTED] wrote:
> > Hello,
> > Is there a way to write a multithreaded server to
> > service SUN RPCs?  I need to write a service that
> > would handle short requests from a few hundreed
> > clients on an MP box and do that as efficiently as
> > possible, and hopefully scale.  I am sure I was
> > not looking hard enough, but I had trouble finding
> > a way of doing that.  I could only find iterative
> > examples.  Thanks for any help.
> >
> >     Greg
> >
>
> There is a pre-compiled multi-threading example (tutorial18) that you
> can download from
>
>         ftp://ftp.accord.com/dist
>
> for Linux(x86) and Solaris(Sparc + x86).  The example can be re-
compiled
> with a download from
>
>         http://www.accord.com/products/doc/IFCC.html
>
> In the above example, a service creates a worker thread everytime a
> request comes in.  Eventually, when a worker thread finishes work, a
> response is returned to the client and the thread exits.

I can overwrite (hack) the stub-generated server code to fork processes
or create threads for handling each call, but this introduces huge
overhead since requests are small.  Thread pool grows and machine
grinds to a halt.  It would be nicer to have a thread pool where
threads would not exit after finishing with a call if there were more
calls to be handled.

> Multi-theading is very useful when you have requests whose compute
times
> vary over a wide range, i.e. from a few milliseconds to tens of
seconds
> or more.  Ideally, the requests that take a short amount of time will
> have high priority and have their responses sent out as soon as
> possible. The requests that take more time would be threaded and have
> their responses sent when they are eventually completed.
>
> If the processing times of the requests are all approximately the
same,
> then I do not think multi-threading is going to make the service
> response times any better.  It will probably make it even slower,
> because of the thread overhead that is involved.  The only exception
to
> this would be if the kernel also supported multi-threading within a
> process.  It this is the case, then the worker thread should be
created
> with concurrency flag set.  This would allow, e.g., a process with two
> threads to run on two processors.

This is exactly what I am shooting at I have servers with 4+ processors
and would like to use as much of the potential throughput as possible.

> Before doing multi-threading, you may want to try and see if a single
> threaded version is acceptable.  If not, then is it possible to have
two
> or more instances of the same service running?  It is much easier to
> scale as well as improve response times by adding more instances of a
> service, either on the same system, or on multiple systems.
>
> A good example that comes to mind is nfs.  If one has many users
> requesting service, one simply increases the number of nfs daemons.
And,
> when that reaches its limit, one buys another system, divides up the
> filesystems and the cycle continues.

As I mentioned, I can do load ballancing, etc. but the purpose of my
current project is to find out the scalability of Sun RPC in the
context of a single process on a multiprocessor system.  If the
dominant paradigm for achieving high scalability in such a situation is
through using multiple daemons, I suppose I have to reconsider the
current test architecture.  The only problem is that it could make
comparisons with other PRC implementations pointless.

Thanks again for your help.

Greg


Sent via Deja.com http://www.deja.com/
Before you buy.

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

From: [EMAIL PROTECTED] (Scott Alfter)
Subject: Re: Program to convert Unix file format to DOS / Windows format.
Date: Wed, 15 Nov 2000 02:39:12 -0000

=====BEGIN PGP SIGNED MESSAGE=====
Hash: SHA1

In article <NakQ5.2390$[EMAIL PROTECTED]>,
Dave Nash  <[EMAIL PROTECTED]> wrote:
>mike wrote:
>
>>     I would like a program to convert Unix format to DOS
>
>If you mean text files, and if you have Windows, open the file in 
>Wordpad/Write.
>
>Alternatively some Unices (eg Solaris) have "dos2unix" or "unix2dos".  My 
>Mandrake 7.1 doesn't have it installed but that doesn't mean you can't get 
>it for Linux.

I usually use something like this:

awk '{printf("%s\r\n",$0)}' in >out

It works, and awk is available everywhere.  Going the other way is even
simpler:

cat in | tr -d "\r" >out

Who needs separate monolithic programs for every little thing when you can
recombine the existing tools to do what you want?  That's the beauty of UN*X
and systems like it (perl and emacs notwithstanding :-) ).

  _/_
 / v \
(IIGS(  Scott Alfter (remove Voyager's hull number for email address)
 \_^_/  http://salfter.dyndns.org
=====BEGIN PGP SIGNATURE=====
Version: GnuPG v1.0.2 (GNU/Linux)
Comment: For info see http://www.gnupg.org

iD8DBQE6Efg6VgTKos01OwkRAs86AKCfb6qgNNc56qwznw54f5fD/43ajgCg6Jap
SLqfwCaZgx+RRCitdoqDOd4=
=6UJn
=====END PGP SIGNATURE=====

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

From: Tyler Larson <[EMAIL PROTECTED]>
Subject: future of the Desktop Wars
Date: 15 Nov 2000 02:59:47 GMT

I have both KDE and Gnome installed on my computer (running RH6.2) and
have come to the conclusion that I like Gnome more for its customizability,
but am still not content with either one.  It seems that KDE has won
as far as stability and ease-of-use is concerned, but Gnome still has won
over more users over KDE where I work (where KDE, Gnome, IceWM and many others
come installed on all the systems, and are equally accessable to every user)
because the people like all the wonderful things they can do to it (Gnome).  
Gnome can certainly be made to suit a user's preferences and wishes much 
easier and to a greater degree; but at the same time, stability and 
consistency are very important as far as productivity is concerned.

With the release of KDE 2, it seems that a lot of the issues I had against
KDE were addressed, and if they add a few more features, I may switch.  The
opinion amoung some of the other SysAdmins here is that eventually Gnome
will will "win" (once it becomes more stable and consistant) because it
caters more to the type of user who runs Linux anyway: the type of user
who needs to be able to change anything and everything he wants to about his
computer setup in order to be content--no restrictions.

What do you think that the future of the Unix/Linux desktop will look like?
Which desktop environment do you see becoming the standard for this platform,
or will there never be such a standard?  I'm not asking which *is* better, 
I'm asking which do you think *will be* better 5 years from now.  Where
do you see the most potential?

-- 
Tyler Larson  |  http://www.tlarson.com  |  [EMAIL PROTECTED]
The only certainty is that nothing is certain.
                -- Pliny the Elder

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

From: ljp <[EMAIL PROTECTED]>
Subject: Re: LinkSys betrayed us!  Poor prospects for Linux.
Crossposted-To: comp.os.linux.hardware,comp.os.linux.networking
Date: Wed, 15 Nov 2000 03:05:57 GMT

Michael Erskine wrote:

> Arctic Storm babbled:
> 
> > LinkSys wants us to *download* the
> > necessary files/drivers, but without the drivers, I can't get on the
> > internet to download them.

hmmm, that seems to be a quandary, eh?

> > 
> > Linux has a long way to go before it can become a common platform, if at
> > all.

Seems perfectly fine to me, I use it everyday.

> 
> > Linux is for hobbiest who have time to tinker with their computers.
> 

Windows is for people that like to reboot just to install software.


> > There are no simple ways to click-and-drag to get things working.


click and drag? whats that?

> > Everything is a struggle;

life is a struggle.


> > you have to learn something new for every petty
> > task.

ahh, such id life, eh?


> >  Imagine if you had to know how the car's engine transferred power via
> > the transmission system before you can drive your car,...
> 

Then we'd probably have a lot less accidents with all the intelligent 
drivers about.

> > Few of us know
> > how a calculator works, and we take it for granted and use it as a
> > fuctional tool.
> 

Calculators work by pushing the little numbers, and then pushing add, or 
subtract, and such. Is that so hard?

> > That's what a computer should be; a functional tool to increase
> > productivity.
> 

hmm, I can do calculations on a computer. I can even from the command line


> > Too much time/effort is required to use Linux.

Too much time is required rebooting Windows.

> > However, Win2K is just as stable, but easy and user-friendly.

I never used it, it's too expensive for me

> > How much is my time worth?
> 
> Probably not much, you don't understand networking, Linux, or computers.

I agree.

> 
> > How much is Win2K?

WAYYY too expensive, troll.


> > Win2K starts to seem pretty attractive,...

If you like wasting your money, thats fine.

> Thanks for your uneducated, lazy, worthless opinion...

ditto...



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

From: Charles Pouliot <[EMAIL PROTECTED]>
Subject: Problems with printing to HP 500C under Debian Linux 2.2 (potato)
Date: Wed, 15 Nov 2000 02:55:21 GMT

 I bought my first printer to use with Linux, a HP 500C because I didn't want
to take the chance of being disappointed my less than optimal support for
newer, more expensive printers. The machine I'm printing from was recently
ugraded to potato (2.2) from an original installation of slink (2.1) I
quickly got the printer to successfully print text.  When I discovered the
need to set up magicfilter to print postscript and other formats, I created
two more entries in my printcap (beyond my basic lp entry, which I left for
printing text): hp500 for black/white printing and hp500c for color printing
(which I'm not using yet because I haven't bought the color cartridge).  It
appears that I'm supposed to use the same magicfilter to print B/W as is used
for an HP500 (black/white printer), which is what I did.  When I print to
that entry, my printer prints a very rough version of my printout.  I tried
both a .gif I had downloaded, and a postscript file generated from HTML by
Netscape.  It seems that the printer prints about an inch, form feeds the
paper, prints anywhere from one millimeter to four inches on subsequent
pages, using up about 5-7 sheets of paper for what should be a single-page
printout.  It mostly resembles what it ought to print out, but has stray
lines and gaps at various locations and symbols and other nonstandard
characters near the left margin. I haven't messed with any settings on either
Debian Linux or the printer, apart from what I mentioned above, though the
printer is second-hand. Can anybody help me with this problem?  Is my
magicfilter incorrect?  Is this peculiar to the particular printer or series
I'm using? Any help would be appreciated. If this is a trite question, please
excuse me.  I looked for information about my printer model and Debian Linux
and found very little - and it seems to me a fine-tuning, more model-specific
problem. Thanks, Charles Pouliot


Sent via Deja.com http://www.deja.com/
Before you buy.

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

From: [EMAIL PROTECTED]
Subject: Re: configuration for mainboard-incorporated video card
Date: Wed, 15 Nov 2000 03:05:06 GMT



friend with a beautiful name:

Great thanks for your patient help.
I have gotten .rpm from intel website by the clue you provided.
Yes, my mainboard is intel 801 and distro is RedHat6.22. Next i will
install the rpm to see weather it works properly.
thanks again.

lxx


Sent via Deja.com http://www.deja.com/
Before you buy.

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

From: "Hello World" <[EMAIL PROTECTED]>
Subject: cannot modify file by root!!!
Date: Wed, 15 Nov 2000 11:43:13 +0800

with permission 0644 and owned by root, i cannot modify or delete the file
with the root account ........ keep saying that permission denied. can
anyone suggest and reason of this and how can i solve this?



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

From: [EMAIL PROTECTED] (Donovan Rebbechi)
Crossposted-To: comp.os.linux.advocacy,comp.os.linux.x,comp.unix.solaris
Subject: Re: True GTK+ will eliminate Qt in next few years?
Date: 15 Nov 2000 03:45:47 GMT

On Tue, 14 Nov 2000 23:51:12 +0100, Matthias Warkus wrote:
>It was the 14 Nov 2000 19:53:48 GMT...
>...and Donovan Rebbechi <[EMAIL PROTECTED]> wrote:
>> For example, ORBit still only supports C, so all 
>> languages besides C are left out of the CORBA framework. 
>
>There are extensions for ORBit to support at least Python and C++ if I
>recall correctly.  

ORBit-C++ requires the bleeding-edge version of glib that "should not be
used to develop stable software". Support for C++ is still in its infancy.

-- 
Donovan

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

From: John Scudder <[EMAIL PROTECTED]>
Subject: Re: xmms playing CDs?
Date: Tue, 14 Nov 2000 22:51:52 -0500

Thanks to all... Got it and am listening to it now.   Now one more
thing..can I edit the names 'CD Audio Track 01', etc. to the actual name
of the tune?

John

> > Can xmms be configured to play CDs?
>
> Yes, you can. Depending on what version you have (I've got 1.2.3), you
> just slot your CD into the drive, select Play File, browse over to your
> CD mount point (or wherever is specified in your settings for the CD
> Audio input plugin) and all the tracks should appear there.
>
> --
> Craig Holyoak
> [EMAIL PROTECTED]
> http://www.uq.net.au/craigh/


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

From: Dr Aldo Medina <[EMAIL PROTECTED]>
Subject: Is there some kind of "graphical" trafshow?
Date: Tue, 14 Nov 2000 22:06:18 -0600

And which is the trafshow's homepage?

-- 
Linux User #98419         -o)    | A man's got to do what a man's got 
http://counter.li.org      /\    | to do. A woman must do what he
ICQ 94335020              _\_v   | can't - Rhonda Bombeck  
Si quieres ayudarme, ponme de    | 
referencia en www.puntosclub.com |

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


** FOR YOUR REFERENCE **

The service address, to which questions about the list itself and requests
to be added to or deleted from it should be directed, is:

    Internet: [EMAIL PROTECTED]

You can send mail to the entire list (and comp.os.linux.misc) via:

    Internet: [EMAIL PROTECTED]

Linux may be obtained via one of these FTP sites:
    ftp.funet.fi                                pub/Linux
    tsx-11.mit.edu                              pub/linux
    sunsite.unc.edu                             pub/Linux

End of Linux-Misc Digest
******************************

Reply via email to