Linux-Misc Digest #979, Volume #18               Thu, 11 Feb 99 00:13:08 EST

Contents:
  Re: one thing that sux about Linux.... (Frank Carney)
  Re: PPP Dialers ("Michael M. Tung")
  Re: KDE is a Memory Hog. ("David A. Frantz")
  Need good vt420 emulator that can talk over serial port (Gregory Propf)
  Re: Advice for Microsoft-haters (Julian T. J. Midgley)
  Re: Umount won't unmount /usr (Rick Walker)
  Re: VFAT floppy for fstab ([EMAIL PROTECTED])
  Re: glibc2.1 and egcs1.1.1 problem (Ludger Solbach)
  Re: Antivirus (Richard Steiner)
  [Off-topic][Q] How to get swapping times ? (Andrei A. Dergatchev)
  Re: Advice for Microsoft-haters (Julian T. J. Midgley)
  Re: Advice for Microsoft-haters (Julian T. J. Midgley)
  (newbie) shadow help (Greg Evans)
  Re: screwed up fstab (Dan Harris)
  Re: Linux InstallFest -- DC -- 20 Feb 99 (Tom Ballinger)
  Re: hacked login (Chad M. Townsend)
  Circumventing my ISP (Kevin Currie)
  Re: kernel too big? (Kevin Martin)
  Re: How to make a ISA PNP modem work in Linux? (Keon-woo Hong)
  APM/DPMS harddisk spindown? (Thomas Frese)
  Re: Simple but sweet. a.out is not my friend (Ed Finch)

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

From: Frank Carney <[EMAIL PROTECTED]>
Crossposted-To: linux.redhat.misc
Subject: Re: one thing that sux about Linux....
Date: Tue, 09 Feb 1999 20:33:13 -0700

>Here we go again. Look people, I know that Winblows is a piece of crap
- M$
>still hasn't pried the DOS kernel out of the thing. I've been in IT
support
>since Windows 3.0, so I know the architecture inside. As such, I hate
I agree, but I am not sure which should be pried out DOS or Windows.

>Windows with all the same venom and vitriol as the rest of you . But,
it is
>a necessary evil (for now).
It is only necessary as long as this is believed.  As people in college
use Linux instead of M$ the "educated" people will eventually get into
positions (managment) and choose to use Linux there too.  It will just
be a matter of time.

>You expect a CEO to learn e-mail on Linux? Right. His time costs about
$200
>an hour - too valuable to spend learning a new OS, when Windows is
ready and
>able to handle at least this simple chore ( for a while anyway, until
it
>chews the FAT table on the disk and explodes like a roman candle) and
he
Now I do agree this is true that a learning curve exists.  Time is a
great consideration no matter what your position may be.  Now if you
want to sell Linux to a CEO you might convince him that it will in the
end same him time by not having system crashes as you have illustrated.
However, this will no longer be an issue as more people in influential
positions start to use Linux and M$ is phased out.

>mostly knows how to use it. I wish everyone in here would drop the
"holier
>than thou" attitude. You all seem to either have a superiority complex
or
>want to divide the world into technology haves and have nots.
Now this is an interesting way to lump all those involved as having a
"holier than thou" attitude.  First of it assumes that everyone in here
is using Linux and only Linux (and nothing but the Linux : ).  Then it
assumes that everyone in here thinks they are better than those who
would use M$.  Both of these assumptions are false.  Now this seperating
people into halves and have nots has already been done by M$.  If you
try to "compete" you are either bought out or forced to leave the market
by "lower quality" mass distributed products.

Now I do enjoy this type of discussion and I would like to further talk
(if it stays civil) with people about it.  However, there is nothing
wrong with the occasional "M$ sucks" or "Linux sucks".  A lot of times
it helps to let out frustration.  Now if it were to be too excesive then
we might have to open a discussion group just for this maybe
"comp.os.linux.sucks" or "comp.os.win.sucks".   Have a good one.



--
To e-mail me please click here mailto:[EMAIL PROTECTED] or remove
remove "NOSPAM" from the reply-to address.



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

From: "Michael M. Tung" <[EMAIL PROTECTED]>
Subject: Re: PPP Dialers
Date: Mon, 08 Feb 1999 20:00:06 +0100

[EMAIL PROTECTED] wrote:

> HI people ... I am currently using Redhat 5.1
>
> and I was wondering if there are any other PPP dialers on the net like Kppp ?
>
> and .. anyone using Afterstep ??
>
> Alvin
>
> &#137;
>

Hi Alvin,

try this one (the X-ISP Homepage):

http://users.hol.gr/~dbouras/

It's a quite nifty tool to facilitate handling ppp/chat. You need to have
X11/XForms installed to compile/run the program.

Have fun,
                             Mike




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

From: "David A. Frantz" <[EMAIL PROTECTED]>
Subject: Re: KDE is a Memory Hog.
Date: Wed, 10 Feb 1999 22:12:57 -0500

Hi all;

Not to add fuel to the fire but the last time I install KDE memory usage
shoot way way up.    We are talkiing about using more that half of a 128 meg
machine here.     KDE was slow, sluggish, and not very inspiring.    Now
maybe the code has been cleaned up and its resource hogging problems
reduced, but I'm not likely to switch back unitl I see the evidence that
this beast is under control.

Dave

[EMAIL PROTECTED] wrote in message <79pf34$589$[EMAIL PROTECTED]>...
>In article <[EMAIL PROTECTED]>,
>  Frank Hale <[EMAIL PROTECTED]> wrote:
>>  but for me I will keep using WindowMaker as it is
>> > MUCH faster.
>> >
>>
>> Couldn't agree more. Window Maker kicks the crap out of KDE in terms of
>> speed.
>>
>> KDE has become just as bloated as any MS product out there. I mean take
>> a look at the ftp site. You have to download over 10 megs of junk to get
>> any use out of it. I will stick with WindowMaker, much smaller and about
>> 10x as fast.
>
>You are comparing apples and planes.
>
>In those 10 Mb you get a file manager, a web browser, a terminal emulator
>(actually 2), find tool, panel/app/launcher/taskbar, pager, window manager,
>and half a dozen other programs, as well as general use widget libraries,
>session management tools, configuration management, and several other
>things I can't recall right now.
>
>When you download windowmaker you get a window manager, no wonder the
package
>is smaller!
>
>Then, maybe windowmaker is faster than kwm (just a tiny piece of KDE).
>But latest windowmaker works as a drop-in kwm replacement, so, how
>is KDE slower, if you can use windowmaker as part of it?
>
>Your comparison doesn't make a lot of sense.
>
>--
>Roberto Alsina (KDE developer, MFCH)
>
>-----------== Posted via Deja News, The Discussion Network ==----------
>http://www.dejanews.com/       Search, Read, Discuss, or Start Your Own



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

From: Gregory Propf <[EMAIL PROTECTED]>
Subject: Need good vt420 emulator that can talk over serial port
Date: Thu, 11 Feb 1999 03:21:33 GMT

I am saddled with a Vax at work and I connect to it through minicom
under Linux.  This is using a null modem link, not a phone line.  I
would like to know if anyone knows a good vt420 emulator or perhaps a
way to attach an xterm to a serial port (since xterm supports vt420
emulation through the termcap entries).  Even a better comm program than
minicom might help.


-- 

"I wanted plutonium, not Beanie Babies..." 
          - Sadaam Hussein, in a letter to Santa Claus.

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

From: [EMAIL PROTECTED] (Julian T. J. Midgley)
Crossposted-To: alt.linux,alt.os.linux,comp.os.linux.advocacy
Subject: Re: Advice for Microsoft-haters
Date: 10 Feb 1999 04:20:29 -0000

In article <[EMAIL PROTECTED]>, Larry <[EMAIL PROTECTED]> wrote:
>On 29 Jan 1999 15:41:03 GMT, Michael C. Vergallen 
><[EMAIL PROTECTED]> wrote:

>
>You people have no idea what technology is. Wait till you see the 
>next offering the military comes out with in smart weapons, not 
>to mention the next "black aircraft" project.
>
>The latest is the f-22. The most advanced aircraft in the world.
>
>This stuff you're talking about is ancient compaired to the technologies the 
>government and NASA  have been developing over the past decade.
> 
>Anyone who thinks the U.S. isn't the most technologically advanced
>country in the world is living in a vacuum.

That's not to say it's not without it's fair share of idiots- like the
one who decided it would be a good idea to use NT to run a ships
command and control system (*laughs*).


-- 
Julian T J Midgley      |                    [EMAIL PROTECTED]
Trinity Hall, Cambridge |  Excession: http://excession.ucam.org
"For every complex problem, there is a solution that is simple,
neat, and wrong."  (H. L. Mencken)

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

From: [EMAIL PROTECTED] (Rick Walker)
Subject: Re: Umount won't unmount /usr
Date: 10 Feb 1999 01:14:13 GMT

: My rc.6 script that runs to do a system shutdown does a umount -a to
: unmount all filesystems.  This used to work just fine.  Recently, however,
: I am unable to unmount /usr because it is busy.  I've made a lot of system
: changes recently, including upgrading to glibc2.06.  Can someone suggest a
: way to diagnose why this is happening and propose a fix?

Some process is cd'd into /usr.  You can take a look in /proc after the
umount fails and find the current working directory for all the
processes.  Once you find the culprit, you will have to make sure that
it is killed prior to the umount command.  It's probably some little
background script that you've forgot about like:

    (cd /usr/doc/HOWTO/html; netscape index.html) &

--
Rick Walker

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

From: [EMAIL PROTECTED]
Subject: Re: VFAT floppy for fstab
Date: Fri, 29 Jan 1999 06:22:20 GMT

Well, I snooped around some more on my own and if anyone cares, here's what I
did. I edited linux/fs/filesystems.c and simply moved this portion:

#ifdef CONFIG_VFAT_FS
        init_vfat_fs();
#endif

in front of the similar msdos fs init. I recompiled the kernel and now in
/proc/filesystems vfat is before msdos and gets properly autodetected. Ext2
and HFS still autodetect OK too.

Does anyone know if I can expect any problems down the road or is there no
reason this shouldn't be stable?

-G.B.Smith
In article <78p755$nfa$[EMAIL PROTECTED]>,
  [EMAIL PROTECTED] wrote:
> I'm having trouble getting floppies automatically detected for VFAT.
> Currently I have this line in my /etc/fstab:
>
> /dev/fd0        /mnt/floppy     auto    rw,user,noauto  0       0
>
> and any user can mount floppies of several types (ext2, HFS, MSDOS) with:
>
> mount /mnt/floppy
>
> Great. The problem is when I mount a DOS floppy this way, it automatically
> decides to use MSDOS mode giving me crappy 8.3 filenames when there are long
> filenames on the disk. It seems to pick MSDOS over (or before) VFAT. Now, as
> root (and ONLY root) I can do a:
>
> mount -t vfat /dev/fd0 /mnt/floppy
>
> which works for long filenames but is exactly what I wanted to avoid with the
> fstab method.
>
> What I would like is to be able to type 'mount /mnt/floppy' and mount as VFAT
> (when appropriate), as well as ext2, HFS, etc. seamlessly with this short
> command by any user at that same mountpoint (vs having a separate /mnt/vfat or
> whatever). Is this even possible? If it matters /proc/filesystems is:
>
>         ext2
>         msdos
>         vfat
> nodev   proc
> nodev   nfs
> nodev   smbfs
>         iso9660
>         hfs
> nodev   autofs
>
> Is this a factor? Does mount use this ordering for fs detection? If so can
> VFAT be moved ahead of MSDOS?
>
> Any advice, wisdom, lectures, songs, chants would be greatly appreciated...
>
> -G.B.Smith
>
> -----------== Posted via Deja News, The Discussion Network ==----------
> http://www.dejanews.com/       Search, Read, Discuss, or Start Your Own
>

============= Posted via Deja News, The Discussion Network ============
http://www.dejanews.com/       Search, Read, Discuss, or Start Your Own    

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

From: Ludger Solbach <[EMAIL PROTECTED]>
Subject: Re: glibc2.1 and egcs1.1.1 problem
Date: Wed, 10 Feb 1999 05:18:01 +0100

Charles Mulks wrote:
> 
> In article <[EMAIL PROTECTED]>, [EMAIL PROTECTED]
> says...
> >
> >Hello folks!
> >
> >I have a problem installing glibc 2.1 with egcs 1.1.1.
> >I thought it would be time to get my old libc 5 based SuSE 5.0
> >installation up
> 
> may I ask where did you get the glibc2.1
> 
> I've been looking for it and haven't managed to track it down.

follow the link at http://www.linuxhq.com/

Bye,
      soulman
-- 
Ludger Solbach                          Phone: +49 5251 640070
Grimmestra�e 5                          email: [EMAIL PROTECTED]
D-33098 Paderborn                              [EMAIL PROTECTED]
Germany                                 http://home.pages.de/~soulman/

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

From: [EMAIL PROTECTED] (Richard Steiner)
Subject: Re: Antivirus
Reply-To: [EMAIL PROTECTED]
Date: Wed, 10 Feb 1999 21:47:38 -0600

Here in comp.os.linux.misc, [EMAIL PROTECTED] (Stan)
spake unto us, saying:

> On 25 Jan 1999 14:53:28 GMT, [EMAIL PROTECTED] (Aurelien
> Jarno) wrote:
>
> >Hello,
> >
> >I'm looking an antivirus for Linux.
>
>There are no viruses in linux.
>Viruses exist in an OS where:
>       There are no user access permissions.
>       The user/admin installs shamelessly binaries from X.screensaver.com

Just keep in mind that a DOS bootsector virus can still infect the MBR
of a Linux box if an infected diskette is left in the floppy drive and
the box is powered up with floppy disk booting enabled in the BIOS.

-- 
   -Rich Steiner  >>>--->  [EMAIL PROTECTED]  >>>---> Bloomington, MN
       OS/2 + Linux (Slackware+RedHat+SuSE) + FreeBSD + Solaris +
        WinNT4 + Win95 + PC/GEOS + Executor = PC Hobbyist Heaven!
                          Who is John Galt?

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

From: [EMAIL PROTECTED] (Andrei A. Dergatchev)
Crossposted-To: gnu.gcc.help
Subject: [Off-topic][Q] How to get swapping times ?
Date: Wed, 10 Feb 1999 13:40:29 GMT

Hi,

Sorry for offtopic, I didn't find any gprof specific newgroups :-(

I have a problem. I have a program which uses large arrays which
usage depends on a defined task. When cumultative sizes of active
arrays become larger than my RAM, the calculation becomes
a continous HDD swapping in fact. I determined by trials critical
parameters with which that happens and ran the program with
profiling options on. However to my great disappointment
the data I've got does not contain real execution time but CPU
effectively used time only. The problem is that during running with
small actively used arrays top report 99.5% CPU usage and when
those sizes become large then during execution HDD is contiously
running and top reports 1% of CPU usage and 98% time idle !
I want to find out which subroutines are responsible for these time
delays. Is there any way to find it or GETTIME is the only way
to go for me ?

Thanks for all responses and flames,
Rgds,

Andrei

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

From: [EMAIL PROTECTED] (Julian T. J. Midgley)
Crossposted-To: alt.linux,alt.os.linux,comp.os.linux.advocacy
Subject: Re: Advice for Microsoft-haters
Date: 10 Feb 1999 04:24:39 -0000

In article <y29v2.6940$[EMAIL PROTECTED]>,
mz001 <[EMAIL PROTECTED]> wrote:
>>
>>Anyone who thinks the U.S. isn't the most technologically advanced
>>country in the world is living in a vacuum.
>
>I guess that's what happens when war is the driving force for the
>advancement of technology in a country.  You think that microwaves, radar,
>and infrared would be used by policemen and telecommunications companies if
>they hadn't already been invented for use in war?
>
>There are so many things the US is working on that the government doesn't
>want us to know about.  By now, they have to be able to clone humans, cure
>AIDS and build hovercars because some so called 3rd world countries can cure
>many cases of AIDS and cancer.  The technology exists but it is either too
>costly or if it is too costly, it's used for war.
>
>Peking scientists in China were able to take DNA fragments of a dinosaur
>from a fossilized egg back in 1994.  By now, there has to be some genetic
>splicing going on between humans and certain animals that we don't know
>about.  It's a really sick world we live in.
>

Another fine American export... the conspiracy theorist :-)



-- 
Julian T J Midgley      |                    [EMAIL PROTECTED]
Trinity Hall, Cambridge |  Excession: http://excession.ucam.org
"For every complex problem, there is a solution that is simple,
neat, and wrong."  (H. L. Mencken)

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

From: [EMAIL PROTECTED] (Julian T. J. Midgley)
Crossposted-To: alt.linux,alt.os.linux,comp.os.linux.advocacy
Subject: Re: Advice for Microsoft-haters
Date: 10 Feb 1999 04:23:21 -0000

In article <[EMAIL PROTECTED]>,
Paul Doherty  <[EMAIL PROTECTED]> wrote:
>Ciaran Dunn wrote:
>
>> Well no. Unless you also consider India, South Africa or even America
>> a colony of GB. They are all ex-colonies.
>
>My mistake...  Australia was let go?  Recently?

A fine example of American ignorance :-)
For an unusual value of "recently" (circa 90 years), the answer is yes.
But I suspect you were born when Australia was already independent.

-- 
Julian T J Midgley      |                    [EMAIL PROTECTED]
Trinity Hall, Cambridge |  Excession: http://excession.ucam.org
"For every complex problem, there is a solution that is simple,
neat, and wrong."  (H. L. Mencken)

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

From: Greg Evans <[EMAIL PROTECTED]>
Subject: (newbie) shadow help
Date: Wed, 10 Feb 1999 22:59:01 -0500
Reply-To: [EMAIL PROTECTED]

First thanks for listening to me, I hate being new.....

Second, the problem.

After setting up RedHatLinux 5.2 and getting qualcomms qpopper up and
running successfully, I decided it was time to shadow the password
file.  After running 'pwconv' I can no longer get mail via pop3 though
telnet seems to work fine as does everything else relating to shadowed
passwords that I have tried.

The question is can qpopper-2.5.3 deal shadowed passwords or do I need
to get a different popper application?

or are there patches out there somewhere to allow qpopper to deal with a
shadowed passwd file?  I looked in the FAQ at qualcomms site, but could
not find anything :/

Please reply via email as I don't get a chance to come here much
currently though I hope to be able to frequent the group and contribute
to it in the future.

thanks

Greg Evans
[EMAIL PROTECTED] or [EMAIL PROTECTED]


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

From: Dan Harris <[EMAIL PROTECTED]>
Subject: Re: screwed up fstab
Date: Fri, 29 Jan 1999 04:59:47 +0000

Gary Momarison wrote:
> 
> [EMAIL PROTECTED] writes:
> 
> > I was running redhat 2.0.35 kernel with a hard drive running on a promise
> > ultra-33 card.        The card wasn't supported at the time (with out a patch) so i
> > used a successful work around where a passed the correct parameters to the
> > kernal via lilo.  I decided to upgrade to the 2.2.0 kernel which supports
> > this card by automatically probing for the correct paramaters.        When I booted
> > the 2.2.0 kernel it detected the card but instead of detecting it as hda and
> > hdb it detected it as hde and hdf. I loaded up my old kernel and changed my
> > fstab to the newly detected hd's and forgot to change my lilo to point to the
> > hdf root point.  I tried to tell lilo manually where the mount point is
> > (linux2.2 root=/dev/hdf) and was succesfull at getting to the root maintance
> > mode but when I ran vi everything was set to read-only.  I am wondering how I
> > go about writing to my fstab in some way.
> >
> > Wish I had made that boot floppy....
> 
> So make it now.
> 
> -- Get to your read-only maintance mode.
> -- Insert floppy in fd0.
> -- cat your-kernel-file >/dev/fd0H1440
> -- rdev /dev/fd0 /dev/your-root-partition
> 
> From the English Police:
> You'd cause your readers to stumble less often if you'd use
> "without", "I", and "work-around" instead of "with out", "i",
> and "work around".
> 

Agreed... and thanks



The problem was that my root=/dev/hdf command was added to the wrong
kernel.

Adding the command to my 2.2.0 kernel booted.  The command 

rdev /boot/bzImage  /dev/hdf6 followed by rdev /boot/bzImage  pointed
the kernel to the correct root point.

Editing the lilo.conf and running /sbin/lilo gave me this error:

Sorry, don't know how to handle device 0x2146

lilo.conf reads as follows:

boot=/dev/hde    
map=/boot/map
install=/boot/boot.b
prompt
timeout=50
image=/boot/vmlinuz
        label=linux
        root=/dev/hdf6
#       append="ide0=0x6100, 0x6206, 10"
        read-only
other=/dev/hde1
        label=95
        table=/dev/hde
image=/boot/vmlinuz-2.0.35-2
        label=lin35
        root=/dev/hdf6
#       append="ide0=0x6100, 0x6206, 10"
        read-only
image=/boot/bzImage
        label=2.2
        root=/dev/hdf6
        read-only
 
Reverting back to the original:
 boot=/dev/hda
..
..
..
other=/dev/hda1
        label=95
        table=/dev/hda

gives me this error:

open /dev/hda: Operation not supported by device 

I am only able to boot when giving the <my-kernel> root=/dev/hdf6
command.

To make it more clear my hard drives were set as hda(win95) and
hdb(linux).
The card automatically sets the hard drives at hde and hdf now.

thanks again

Dan 

[EMAIL PROTECTED]

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

From: Tom Ballinger <[EMAIL PROTECTED]>
Crossposted-To: dc.general
Subject: Re: Linux InstallFest -- DC -- 20 Feb 99
Date: Wed, 10 Feb 1999 08:43:12 -0500

Charles Packer wrote:
> 
> In article <[EMAIL PROTECTED]>, Ben Russo wrote:
> >I think we should relax, be friendly, let Linux sell itself.
 
<snip>
 
> I say this as a Linux fan and user for the last four years:
> If Linux is ever going to sell itself, the hackers that created it
> are going to have to take time off from tweaking the kernel and
> get cracking on _successfully_ automating the video configuration
> portion of the installation process.

I agree whole heartedly.  If Linux wants to compete on the desktop with
Windows, it does need to become less user knowledge dependent.  
Windows, for better or worst makes it's success on the fact that people
can install it, and most times get it running with little computer
knowledge. RH 5.2 goes a long way to improving configurations with
detection of sound, video, NICs etc.. and well as many admin functions
now in GUI (e.g. linuxconf), it still requires a user to run them or see
them running.  It needs to do this detection and as much of the
configuration without the user knowing it is happening and just ask for
preference information (resolution, background color).  

Don't get me wrong here, I use Linux and am somewhat anti M$, but if
Linux is to gain large market on the desktop, it needs to get to a point
to where the non technical user "feels comfortable" with it, and that
required dummying it down to where Windows is.

Tom

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

From: Chad M. Townsend <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.admin,comp.os.linux.networking
Subject: Re: hacked login
Date: Thu, 11 Feb 1999 03:53:09 GMT

They most likely got in through another service like 'named' or 'imap' etc..
and the rootkit replaced 'login', so they did'nt hack login, they prob. hack
another service in /etc/inetd.conf, (or maybe not), be sure you ONLY have
services you need on and the rest off, and the ones you keep on, make sure
you know exactly what the are doing.  Keeping you software uptodate and
patched will help.

-chad


  "Mitchell Maltenfort" <[EMAIL PROTECTED]> wrote:
> >It sounds like some one used ROOT-kit on you computer, tehey did the same
> >on my PC, i had to reinstall all becouse he also planted a backdoor
> >(check) he also replaced a hacked inetd so he can log in not to be
> >detected he can also replace ls, ps, etc...  check
> >http://www.rootshell.com/ for the rootkit and unpack it en read the readme
> >and you be dazzeled ehat it can do...
> >
>
> I checked out rootshell and was surprised, although at my end of the
> learning curve it looked more like an ammo dump than a bomb shelter.
>
> I did a search on rootkit and was surprised to only see ways to download it.
> Is there any way to block it?
>
>


========================================================
Chad M. Townsend         Virtual Community Network, Inc.
Chief Technical Officer  Your Local Community Online!

============= Posted via Deja News, The Discussion Network ============
http://www.dejanews.com/       Search, Read, Discuss, or Start Your Own    

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

From: Kevin Currie <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.networking
Subject: Circumventing my ISP
Date: Mon, 08 Feb 1999 17:43:18 -0500

I am wonderring if anyone can help me find a way to circumvent some
firewalling my ISP is doing.  Here is the situation (names have been changed
to protect the guilty):

~ I have a linux box on xxx.yyy.zzz.0/24 (say xxx.yyy.zzz.e/32)
~ Incoming traffic to this subnet can only come from xxx.yyy.0.0/16
  without going though some ip manipulation to prevent servers on my
  subnet.
~ I have access to a linux machine on xxx.yyy.0.0/16 which has two ip
  addresss (lets say xxx.yyy.a.a/32 and xxx.yyy.a.b/32).

Now this setup the ISP has is all fine and dandy for Windows users, but I'd
love to be able to telnet/ftp to my machine from remote computers to do
things like get my mail and transfer files and such.

I would like to be able to take all traffic coming to xxx.yyy.a.b/32 and
send it to my machine on xxx.yyy.zzz.e/32.  Is there a way to do this with
ipportfw, ipchains, ipfwadm, or anything?  I've looked at ipautofw and it is
not acceptable because it would take traffic from both xxx.yyy.a.a _AND_
xxx.yyy.a.b and send it though.  xxx.yyy.a.a must be able to get its own
traffic.

Can anyone help me with a solution?  Is this even possible?
Please post and mail if possible.

Thanks,
Kevin Currie

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

Crossposted-To: comp.os.linux.setup
From: [EMAIL PROTECTED] (Kevin Martin)
Subject: Re: kernel too big?
Date: Wed, 10 Feb 1999 04:33:45 GMT

Followups to col.misc only, since this isn't going to be on-topic for 
col.setup by the time I get done....  :-)

In article <79qgbd$rtb$[EMAIL PROTECTED]>, it says "David A. Frantz" 
<[EMAIL PROTECTED]> wrote:
>Mark;
>
>l ran into the same problem late last night, while I did change some of the
>default items the too large kernel came as a surprise. 

Happened to me for the first time.  Seems 2.2.1 has added a bit of muscle.   
It's not fat, though -- it's too fast for that. ;-)

"How" it knows it's too big is a function of "why."

>The other thing I'm wondering about is why the limit in the first place?

Intel CPU's.  The kernel has to fit under the 640K architectural limit, 
UNLESS you use bzImage, which is smart enough to load high.  Took a bit of 
fiddling to get it right, but it's pretty reliable now.

>If anybody has pointers to an explanation of the hows and whys of this
>limitation a pointer would be very helpful.

How about 'Bill "640K is enough for anyone" Gates is an idiot'?   The 
original IBM PC development team had a chance to buy the Atari 800, which 
would have given them a flat memory model and actual device drivers, which 
means we'd all be using fourth-gen Amigas by now, and Nolan Bushnell might 
be applying the biggest private fortune on earth to the development of 
true artificial intelligence and humanoid robots instead of buying off the 
Justice Department.  Makes you weep, doesn't it?

>  The problem I have is that
>modern hardware and optimizing compilers seem to be at odds with this size
>limit.   Comments?

Yep, that's why 'make bzImage' is such a painless solution to the problem.  
If you get the "too big" message and IMMEDIATELY issue a 'make bzImage'  it 
should only take about 20 seconds.  You've already done the heavy lifting; 
as long as you don't undo it by running "make clean" again, all it has to do 
is relink the object modules a bit differently.  That's why it hardly 
seems worth getting so upset about.  Just my $.02....


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

From: Keon-woo Hong <[EMAIL PROTECTED]>
Subject: Re: How to make a ISA PNP modem work in Linux?
Date: Tue, 09 Feb 1999 23:03:12 -0500

I forgot to say, the modem isn't a winmodem.

-KH


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

From: Thomas Frese <[EMAIL PROTECTED]>
Subject: APM/DPMS harddisk spindown?
Date: Tue, 09 Feb 1999 23:28:28 -0500

Would anybody know how to Linux to spin down 
the hard disk after a specified inactivity time?
My machine has a APM bios and I'm using 

'xset dpms 0 600 0'

which switches my screen to pwr-save after 600secs...
How do you switch off the HD? The 'Battery Powered mini-HOWTO'
doesn't really give conclusive info on this and neither does
the apmd documentation that I found. I don't want to use the bios
inactivity timer since it doesn't work with my graphics
card under X, i.e. it doesn't switch off the screen. So running
both the Linux dpms timer as well as the bios timer is not
recommended...
Any suggestions?

Thanks 

Tom

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

From: Ed Finch <[EMAIL PROTECTED]>
Subject: Re: Simple but sweet. a.out is not my friend
Date: Wed, 10 Feb 1999 22:09:42 -0500

josh richard wrote:
> 
> I have compiled a c program using the gcc compiler in red hat 5.2.  After
> the compilation is complete, I have an executable created called a.out.  I
> type a.out to run the program and get the following error
> "bash - command a.out not found"

You don't have the current-working-directory (".") in your PATH
environment variable. It's ok to add it - but don't add . to 
root's path for security reasons.


Regards,
Ed
--
   Q: Why do PCs have a reset button on the front?
   A: Because they are expected to run Microsoft operating systems.

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


** 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