Linux-Misc Digest #528, Volume #27                Wed, 4 Apr 01 13:13:02 EDT

Contents:
  Re: Linux games (Michel SIMIAN)
  Re: apache as proxy server (Robert Lynch)
  Re: Linux games (Bart Friederichs)
  Re: File transfer apps (Jeremiah DeWitt Weiner)
  Re: why doesn't fdisk -l work? (Leonard Evens)
  Re: commandline based article download tools (* Tong *)
  Re: Linux games (nordi)
  Re: PLEASE HELP!, MY LINUX have been HACKED~ ("Peter T. Breuer")
  Free Telnet Servers (Ninja IX)
  Re: Free Telnet Servers
  Re: Secure File deletion ("Peter T. Breuer")
  Re: Free Telnet Servers (Grant Edwards)
  Re: Secure File deletion ("Peter T. Breuer")
  Re: Free Telnet Servers
  No network until tcpdump?! (Anton Suchaneck)
  Re: File transfer apps (Frank da Cruz)
  Re: Free Telnet Servers ([EMAIL PROTECTED])
  Re: quark express  (Ninja IX)
  linux-2.4.2 and modules... (Luca Zancan)

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

From: Michel SIMIAN <[EMAIL PROTECTED]>
Subject: Re: Linux games
Date: Wed, 04 Apr 2001 17:18:58 +0200

Cray Drygu wrote:
> 
> [EMAIL PROTECTED] (Bart Friederichs) wrote in
> <[EMAIL PROTECTED]>:
> 
> >Are there any good rocking games on Linux, that are free? Loki has a lot
> >of cool games, but they are not free. Almost everything is totally free
> >on the Linux system, including pretty large projects such as Office
> >suites (not to mention the OS itself), but no games. Yeah, there are
> >some, but I haven't found Quake-like games or some racing games.
> >
> >Anybody has a thought on this?
> 
> Yes...and a pretty good explanation, IMHO.
> 
> Games are, aside from an OS maybe, the most expensive and time-consuming
> software to try and develop.  Note I'm talking about Quake-type games here,
> not the endless minesweeper and solitaire clones.
> 
[.../...]

and someone specialized in OS, as well as multitasking, RPC for
net-games
and probably a senior manager all over the team. Really, a game is
a big project.
 
-- 
Michel SIMIAN

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

From: Robert Lynch <[EMAIL PROTECTED]>
Reply-To: [EMAIL PROTECTED]
Subject: Re: apache as proxy server
Date: Wed, 04 Apr 2001 08:20:27 -0700

Marian Heddesheimer wrote:
> 
> Hi,
> 
> I want to set up a Linux SuSE 7.1 machine with apache server and want
> to use apache as proxy. It works so far for http access but if I try
> to use an ftp-Adress the browser tells me that this is switched off
> for security reasons.
> 
> Can anybody point me to a tutorial or text where I can see how to use
> apache as proxy for ftp connections too? Is this possible with apache
> or do I have to use squid?
> 
> Marian

I don't know how to solve your problems with security and ftp
proxy, but just want to note that there is a bug in the Apache
ftp proxy code.

Specifically, the code is unsophisticated and assumes that
listings are uniform down the page.  The result is that some
sites have a missing letter in some file listings, e.g.

... file specs....  issing_first_letter

When viewed in a browser, for example, this leads to a mysterious
problem that a click on a link results in "no such file" (there
is no file "issing_first_letter", but there IS a file
"missing_first_letter").

FWIW. Bob L.
-- 
Robert Lynch     Berkeley CA USA    [EMAIL PROTECTED]

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

From: Bart Friederichs <[EMAIL PROTECTED]>
Subject: Re: Linux games
Date: Wed, 04 Apr 2001 15:16:57 GMT

Cray Drygu wrote:

> >Anybody has a thought on this?
> 
> Yes...and a pretty good explanation, IMHO.
> 
It is, indeed, a good explanation. Although it's everything I figured
out myself.

> Games are, aside from an OS maybe, the most expensive and time-consuming
> software to try and develop.  Note I'm talking about Quake-type games here,
> not the endless minesweeper and solitaire clones.

Me too.

[.. snip ..]

Actually you're telling it costs a lot of talent, people and time, but
doesn't creating an OS or an office suite also? Maybe the thing here is
that there are not that many artists and all those artists are taken by
iD and the like /or/ the artists aren't active in the open source
community. 

> And *that's* why you don't see quality, free 3D games very often.
Not very often, or not at all? And it's not just 3d games I am looking
for. Ports of old games is good too, think about killer games like Dune
2. One of the main advantages about those games is that they will work
flawlessly on current hardware.

Bart
-- 
=======================================================================
The internet is a too slow way of doing things you'd never do without
it.
                                              Bart Friederichs, 1998
=========================================================================

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

From: Jeremiah DeWitt Weiner <[EMAIL PROTECTED]>
Subject: Re: File transfer apps
Date: 4 Apr 2001 15:35:00 GMT

Vasilis Serghi <[EMAIL PROTECTED]> wrote:
> Is there an application out there, like laplink, that can be used for
> transferring files over parallel or serial cables? Something that is
> transparent to the OS type would be good.

        Depends how you want to do it.  You could just hook up the two
systems with the correct cable and do "cat myfile > /dev/ttyS1" on one
and "cat /dev/ttyS1 > myfile" on the other, but that's pretty ugly.
Alternately, you could set up a getty on one machine's serial port, and
then log in with something like minicom over the serial line, and use
zmodem (or xmodem, or kermit...)  Or you could set up SLIP/PPP (for a 
serial line) or PLIP (for a parallel connection) and then you can use
anything that works on top of TCP/IP.

JDW


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

From: Leonard Evens <[EMAIL PROTECTED]>
Subject: Re: why doesn't fdisk -l work?
Date: Wed, 04 Apr 2001 10:22:27 -0500

Pumpkinhead wrote:
> 
> quick one,
> 
> when trying to check my partition scheme i always use " fdisk -l " but
> on all my rh7.0 machines i don't receive any output at all.  just get
> the prompt back.  any ideas?
> 
> df -h is my alternative but...

You probably need to specify a disk as in
fdisk -l /dev/hda
for the first IDE drive.

-- 

Leonard Evens      [EMAIL PROTECTED]      847-491-5537
Dept. of Mathematics, Northwestern Univ., Evanston, IL 60208

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

From: * Tong * <[EMAIL PROTECTED]>
Crossposted-To: comp.unix.questions
Subject: Re: commandline based article download tools
Date: 04 Apr 2001 12:46:56 -0300

Christopher Albert <[EMAIL PROTECTED]> writes:

> * Tong * wrote:
> > 
> > Hi,
> > 
> > Speaking of commandline based download tools, wget is the best.
> > But there is a problem for it to download articles like:
> > 
> >    http://www.informit.com/newsletter.asp?link=159
> > 
> 
> lynx -source http://www.informit.com/newsletter.asp?link=159 > 159.html

Thanks Chris, but when a problem can't be solved by wget, it won't
be any easy as above lynx solution. 

The problem is to download the *whole article", not only the first
page. and the last paragraph of the full article is many clicks
away... As I said:

,-----
| If you have experienced with any download tools, you will know that
| the above url is very tricky. there's no better way to download the
| whole article (only) without (potentially) download the whole site.
`-----

-- 
Tong (remove underscore(s) to reply)
  *niX Power Tools Project: http://xpt.sourceforge.net/
  http://members.xoom.com/suntong001/
  - All free contribution & collection

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

From: nordi <[EMAIL PROTECTED]>
Subject: Re: Linux games
Date: Wed, 04 Apr 2001 17:53:46 +0200

Bart Friederichs wrote:
> Cray Drygu wrote:
> > >Anybody has a thought on this?
> > Yes...and a pretty good explanation, IMHO.
> It is, indeed, a good explanation. Although it's everything I figured
> out myself.
> 
> > Games are, aside from an OS maybe, the most expensive and time-consuming
> > software to try and develop.  Note I'm talking about Quake-type games here,
> > not the endless minesweeper and solitaire clones.
> 
> Me too.
> [.. snip ..]
> Actually you're telling it costs a lot of talent, people and time, but
> doesn't creating an OS or an office suite also?

I think that the issue is not only whether it takes a lot of time
but if it is done at universities. Programming databases and operating
systems is a great way to understand how they work. For this reason
this is done at universities, because they are important for everyone
that has a lot to do with computers. Understanding the internals of
3D games is not that important, and that's why it is not done at
universities. This way a lot of OS stuff is done at universtities.

Another reason is that for making an OS/office suite you just need
a bunch of skilled programmers. That's it. For developing a game
you need programmers, designers and musicians. So it involves a
much greater variety of people, and at a university it is going
to be virtually impossible to bring them together to a single
project. Talking about variety: for operating systems and the
like there are well established standards that reduce the need
for communication between the developers. Games on the other
hand are only interesting if they are unique. This produces
a great need for communication if all people are supposed to
be developing the same game. And communicating just via email
is not enough in this case.

Or maybe it's just that the game developers were playing too
much with their own game and forgot about the developing part.
Who knows...

nordi

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

From: "Peter T. Breuer" <[EMAIL PROTECTED]>
Subject: Re: PLEASE HELP!, MY LINUX have been HACKED~
Crossposted-To: alt.os.linux,alt.linux,comp.os.linux.help,comp.os.linux
Date: Wed, 04 Apr 2001 16:19:47 GMT

In comp.os.linux.help [EMAIL PROTECTED] wrote:
> In comp.os.linux.misc Leo <[EMAIL PROTECTED]> wrote:
>> One more thing after i read the email, I  checked my log file (
>> /var/log/message ) to see what happened.   Apparently, i have lost ALL the
>> stuff  before date APRIL 3rd ( day of mail send )...   So i couldn't trace

> Looks like somebody have installed some trojan that tryied to
> mail out information about your system.

SOunds like he's been tr0n'ed. Lion or ramen worm. attack through 
bind (< 8.2.2? ) or wuftp <= 2.6.0.

> I suggest you re-install your machine to get rid of the
> trojan and other rootkit that could be in your machine.
> I also suggest some good readings about security and
> firewalling.

Yep.

Peter

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

From: Ninja IX <[EMAIL PROTECTED]>
Subject: Free Telnet Servers
Crossposted-To: comp.os.linux.networking,comp.protocols.tcp-ip,linux.redhat
Date: Wed, 04 Apr 2001 16:32:07 GMT

Hello

Does anyone know any list or just one telnet or rlogin server, allowing 
anonymous users to log in and use basic shell commands

It could be usefull to make tests from outside on a internet-connected box.

Thx


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

From: [EMAIL PROTECTED] ()
Subject: Re: Free Telnet Servers
Date: Wed, 04 Apr 2001 16:36:55 GMT

On Wed, 04 Apr 2001 16:32:07 GMT, Ninja IX <[EMAIL PROTECTED]> wrote:
>Hello
>
>Does anyone know any list or just one telnet or rlogin server, allowing 
>anonymous users to log in and use basic shell commands
>
>It could be usefull to make tests from outside on a internet-connected box.
>

Create an account called anonymous without a password.  Use *any* telnet
server.

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

From: "Peter T. Breuer" <[EMAIL PROTECTED]>
Subject: Re: Secure File deletion
Date: Wed, 04 Apr 2001 16:40:27 GMT

Frank Ranner <[EMAIL PROTECTED]> wrote:
> "Peter T. Breuer" wrote:
>> [Hilarious piffle discarded]
> See your piffle and raise you an strace :-)

> [Early boring stuff snipped]

> close(0)                                = 0
> open("junk.in", O_RDONLY|O_LARGEFILE)   = 0
> close(1)                                = 0
> open("junk.dat", O_WRONLY|O_CREAT|O_TRUNC|O_LARGEFILE, 0666) = 1

Well, O_TRUNC is interesting! Now try it with seek=1. OK, I'll spare
you it:

 open("/dev/zero", O_RDONLY|0x8000)      = 0
 close(1)                                = 0
 open("foo", O_RDWR|O_CREAT|0x8000, 0666) = 1

No O_TRUNC.

> -- this is a result of the command 'dd if=junk.in of=junk.dat' where
> junk.in was 84 bytes
> -- and junk.dat was several k. I'd say that O_TRUNC is significant.

So would I! But it's likely a programming error. Send them a bug report.

> Regards, Frank Ranner


Peter

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

From: [EMAIL PROTECTED] (Grant Edwards)
Subject: Re: Free Telnet Servers
Date: Wed, 04 Apr 2001 16:43:56 GMT

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

>>Does anyone know any list or just one telnet or rlogin server,
>>allowing anonymous users to log in and use basic shell commands
>>
>>It could be usefull to make tests from outside on a
>>internet-connected box.
>
>Create an account called anonymous without a password.  Use
>*any* telnet server.

Sounds to me like a security hole big enough to sail the Queen
Mary thorugh...

-- 
Grant Edwards                   grante             Yow!  I'm protected by
                                  at               a ROLL-ON I rented from
                               visi.com            AVIS...

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

From: "Peter T. Breuer" <[EMAIL PROTECTED]>
Subject: Re: Secure File deletion
Date: Wed, 4 Apr 2001 18:35:23 +0200

[EMAIL PROTECTED] wrote:
> If I do an fopen() in write mode, abstract thought (all I was really
> relying on) doesn't conclusively establish things either way.  

The low-level semantics is not defined in the manpage (though POSIX may
have something more to say), just the algebraic properties ..

> It would make as much sense for a new open() to result in the old
> version of the file being unlinked, with new version as a fresh file,

Well, I know what you're saying: that these low-level semantics would be
indistinguishable in the algebra of file ops. Unfortunately you've
given yourself too small an algebra and the conclusion is incorrect.
Yes, you most certainly can distinguish, the difference is observable:
try doing it twice, in parallel.  According to you that should lower the
df output count from the fielsystem by twice the number of blocks, while
the write is happening.  It doesn't. Both writes go to the same file,
and the df count never changes.

> as it does to attach to the existing one, and then truncate once it's
> closed.  Rather like the two different assumptions of noneuclidean
> algebra.

>> > -> Then, a new set of blocks get written to, by dd.
>> > -> The file is now tied to the new sequence of blocks.
>> 
>> [Hilarious piffle discarded]

> Piffle, perhaps, but I don't think it was ludicrously irrational.

:-). Well, I apologize. Yes, it's rational, in the sense that it's reasoned.
It just happens to be reasoned wrongly, rather like an appealing proof
that reaches a wrong conclusion :-). I can't quite distinguish the
rules of inference that you've applied, or the axiom structure, but
it seems to go awry with the basic premise that open works its magic on
an entirely new inode. It doesn't. It gives you a new descriptor, which
should be a pointer to the same old inode under most circumstances
(even with O_TRUNC). As to what it does with the existing blocks
of that inode, well, the only way I know of for detaching them is to
use O_TRUNC on open.

It's really too bad. I don't know a good way to _shrink_ a file a little
once you have it! You can only get rid of all its blocks with O_TRUNC.

> Thanks for illuminating things by documenting what it _really_ does.
> [Which is mostly honest, though you should detect a whiff of sarcasm
> there, seeing as how there wasn't much documentation...]

Sorry. Someone shoul rewrite the manpages sensibly.

peter

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

From: [EMAIL PROTECTED] ()
Subject: Re: Free Telnet Servers
Date: Wed, 04 Apr 2001 16:50:03 GMT

On Wed, 04 Apr 2001 16:43:56 GMT, Grant Edwards <[EMAIL PROTECTED]> wrote:
>In article 
><[EMAIL PROTECTED]>, 
>[EMAIL PROTECTED] wrote:
>
>>>Does anyone know any list or just one telnet or rlogin server,
>>>allowing anonymous users to log in and use basic shell commands
>>>
>>>It could be usefull to make tests from outside on a
>>>internet-connected box.
>>
>>Create an account called anonymous without a password.  Use
>>*any* telnet server.
>
>Sounds to me like a security hole big enough to sail the Queen
>Mary thorugh...

Yup.  

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

From: Anton Suchaneck <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.networking
Subject: No network until tcpdump?!
Date: Wed, 4 Apr 2001 18:52:36 +0100

Hi guys,

I have a problem:
Sometimes not even pings can reach me until I start tcpdump. I have a 
Xircom PCMCIA Card in a Dell Inspiron 3800. Any ideas?

Bye

Anton

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

From: [EMAIL PROTECTED] (Frank da Cruz)
Crossposted-To: comp.protocols.kermit.misc
Subject: Re: File transfer apps
Date: 4 Apr 2001 16:51:07 GMT

In article <9afev4$pmc$[EMAIL PROTECTED]>,
Jeremiah DeWitt Weiner  <[EMAIL PROTECTED]> wrote:
: Vasilis Serghi <[EMAIL PROTECTED]> wrote:
: > Is there an application out there, like laplink, that can be used for
: > transferring files over parallel or serial cables? Something that is
: > transparent to the OS type would be good.
: 
:       Depends how you want to do it.  You could just hook up the two
: systems with the correct cable and do "cat myfile > /dev/ttyS1" on one
: and "cat /dev/ttyS1 > myfile" on the other, but that's pretty ugly.
:
And risky, definitely not guaranteed to work since there is no error
detection or correction.

: Alternately, you could set up a getty on one machine's serial port, and
: then log in with something like minicom over the serial line, and use
: zmodem (or xmodem, or kermit...)  Or you could set up SLIP/PPP (for a 
: serial line) or PLIP (for a parallel connection) and then you can use
: anything that works on top of TCP/IP.
: 
I'm not aware of anything that can use a parallel cable connection in
UNIX for data transfer, but with serial ports and a null-modem cable,
you can use Kermit:

  http://www.columbia.edu/kermit/ckermit.html

No need for minicom -- it calls Kermit anyway to transfer files, and Kermit
itself already does everything minicom does.  IMPORTANT (no matter what
software you use): the null-modem cable should be fully populated with all
the appropriate crossovers, especially RTS and CTS.

Two scenarios:

 1. On the "far" machine, if you have a getty, then you can make the serial
    port connection to it from the near machine, log in, start Kermit on the
    far machine, and transfer files in the normal way.

 2. If you do not have a getty on the far machine, but it is physically
    accessible to you, you can start Kermit on it and direct it to the port
    to which the null-modem cable is connected, put it in server mode, and
    then go back to the other computer and give client commands to send and
    get files.

Example for scenario 1, starting on the near computer:

  $ kermit
  C-Kermit> set port /dev/ttyS0  <-- Substitute appropriate port
  C-Kermit> set speed 57600      <-- Or other desired speed
  C-Kermit> set flow rts/cts     <-- Use hardware flow control
  C-Kermit> connect
  <press Enter key if you don't see anything>
  login: jdw
  Password: blah 
  $ kermit                       <-- Start Kermit on far computer
  C-Kermit>                      <-- Far Kermit's prompt
  C-Kermit> server               <-- Put the far Kermit in server mode
  <Type Ctrl-\ and then the letter c to escape back to near Kermit>
  C-Kermit>                      <-- Near Kermit's prompt

Now you can give SEND, GET, REMOTE xxx, and BYE commands to the near
Kermit.  Type HELP at the C-Kermit> prompt if you need help.

Example for scenario 2:

On the far computer:

  $ kermit 
  C-Kermit> set port /dev/ttyS2  <-- Substitue appropriate port
  C-Kermit> set speed 57600      <-- Or other desired speed
  C-Kermit> set flow rts/cts     <-- hwfc
  C-Kermit> server               <-- Start server mode

On the near computer:

  $ kermit 
  C-Kermit> set port /dev/ttyS0  <-- Substitue appropriate port
  C-Kermit> set speed 57600      <-- Must be same speed as far computer
  C-Kermit> set flow rts/cts     <-- hwfc
  C-Kermit>

Now you can give SEND, GET, REMOTE xxx, and BYE commands to the near
Kermit.

- Frank

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

From: [EMAIL PROTECTED]
Subject: Re: Free Telnet Servers
Date: 4 Apr 2001 16:55:06 GMT

In comp.protocols.tcp-ip Ninja IX <[EMAIL PROTECTED]> wrote:
> Hello

> Does anyone know any list or just one telnet or rlogin server, allowing 
> anonymous users to log in and use basic shell commands

All you need is already installed.

Read manpages for inetd and bash , the rest is left as an excersize.

> It could be usefull to make tests from outside on a internet-connected box.
Sound dangerous to mee ..
> Thx


-- 
Peter H�kanson         
        IPSec  Sverige      (At the Riverside of Gothenburg, home of Volvo)
           Sorry about my e-mail address, but i'm trying to keep spam out.
           Remove "icke-reklam"and "invalid"  and it works.

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

From: Ninja IX <[EMAIL PROTECTED]>
Subject: Re: quark express 
Date: Wed, 04 Apr 2001 17:00:18 GMT

Sparkzz wrote:

> Hello,

Hi

> 
> Is there any Linux software comparable to Quark????
> 

What is Quark ?

> 
> thanks
> .

nothing

> .
> ....Ken

;)

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

From: Luca Zancan <[EMAIL PROTECTED]>
Subject: linux-2.4.2 and modules...
Date: Wed, 04 Apr 2001 19:04:03 +0200
Reply-To: [EMAIL PROTECTED]

Hi everybody,

I've got a RedHat 6.2 (kernel version 2.2.14-5.0) and I've tried to
install a 2.4.2 kernel (downloaded in its "tar.gz" form directly from
ftp.kernel.org - I would like to try this, instead of the rpm package of
RedHat...).

I've got some problems in creating a RAM disk with mkinitrd...
I must do it because my system boots from a SCSI disk controlled by an
Adaptec2940UW adapter (I've got 2 SCSI disk and A SCSI cd-rom on the
same controller).

mkinitrd tells me that it can't find the "aic7xxx" module (in make
menuconfig I've selected "loadable module support" and all the stuff
needed, including the driver for the Adaptec as a module).
I've noted that the modules stuff is now in
"/lib/modules/2.4.2/kernel/drivers/" (while before was in
"/lib/modules/2.2.14-5.0/")...
Then I think that there is some problem in my "/etc/modules.conf"
file...
I've tried to add "path[scsi]=/lib/modules/2.4.2/kernel/drivers/scsi"
and "path[toplevels]=/lib/modules/2.4.2"
Now "/sbin/mkinitrd    /boot/initrd2-4-2    2.4.2" runs but it says
that:
"All of your loopback devices are in use!"
and, above all, it DOES NOT create the "/boot/initrd2-4-2" file!!!

Can someone tell me what can I try to solve the situation???
Obviously, after modifying /etc/lilo.conf and running /sbin/lilo, I
cannot load the new kernel (it stops shortly after decompressing it) and
I can only mount the old 2.2.14-5.0 one...

Thank you very much,
__________________________________________________

Luca Zancan
Logica S.r.l.
e-mail [EMAIL PROTECTED]
URL http://www.logicaonline.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 by posting to comp.os.linux.misc.

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