Linux-Misc Digest #426, Volume #27               Fri, 23 Mar 01 03:13:02 EST

Contents:
  Re: Linux for a 486? (Mike Flournoy)
  Re: Weird(?) magic word for sh to invoke perl under Linux (Michael Wang)
  Re: pdf (was: Best E-mail Client?) ("Matt O'Toole")
  fopen error (Joe)
  Re: new kernel 2.4.1 ("BetrOffDed")
  No in.ftpd after install ([EMAIL PROTECTED])
  Re: RH 6.2 does'nt recognize 2nd NIC? (Wayne Pollock)
  same inode?? ("Chowalit Tinnagonsubout")
  Re: Best E-mail Client? (Christopher Booth)
  Re: Weird /proc/mtrr ("Eric")
  Re: same inode?? (David)
  Re: Best E-mail Client? (Christian Garms)
  Re: ALMOST SOLVED! I did 'dd if=/boot/mbr.b of=/dev/hda' :-((( ("Eric")
  Re: Best E-mail Client? (Christian Garms)
  Re: No in.ftpd after install ("Quiney, Philip [HAL02:HH00:EXCH]")

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

Subject: Re: Linux for a 486?
From: Mike Flournoy <[EMAIL PROTECTED]>
Date: Fri, 23 Mar 2001 04:56:43 GMT

in article [EMAIL PROTECTED], Hartmann Schaffer at
[EMAIL PROTECTED] wrote on 3/22/01 10:10 PM:

> In article <[EMAIL PROTECTED]>, Mike Flournoy wrote:
>> I am ignorant to Linux but have a old 486 I wanted to use to learn on. It
>> appears most newer versions are aimed at Pentiums. Is it possible to run
>> linux with a GUI on a 486 or is that just too slow?
> 
> i used to run it on a 486dx2/66 with first 16, then 32 mb and 1GB, then 2,
> then 4 without any problems.
> 
> the base system should install on <100mb disk (actually, there are
> distributions that fit on a floppy) and 8mb ram, but in this case there isn't
> much you can do with it.  all major distributions come with tons of
> applications, and esp. the major ones (like redhat) tend to install every-
> thing in sight, so you would most likely run out of disk space.  afaik each
> distribution lets you do a custom install which is time consuming (you have
> to pick the packages or package groups, and with the number of available
> packages that takes a while).  personally, i found the custom install
> relatively easy with debian and slackware, not so pleasant with redhat, no
> experience with other distributions.
> 
> many years ago the common wisdom was that you would need at least 8mb to run
> X.  i suspect that even if that is still possible it would be painfully
> slow, so go with at least 16, better 32mb (unfortunately you didn't specify
> how much memory you have, but with less than 16 mb i would stick with a non
> X installation, or upgrade the memory).  on a 486 i would stay away from the
> big desktops like gnome or kde, just stick to one of the early window mangers
> like fvwm2 or fvwm95 (if you crave the windows feeling).  they provide much
> of the same gui functionality, but with less flash.
> 
> the rest is a question of picking the applications you install wisely, keeping
> available ram and disk space in mind
> 
> hs
Thanks for all the input. I was gifted a NeXT machine which is really nice
but the screen is going dim ( a common problem with them apparently )and is
only B & W. It works so well on a motorola 68040 & 16mb ram that I thought
Linux on a 486 might work as well and give me a color screen and better
printer support.  But then when I researched everything points to pretty
heavy hardware. Oh well.
 It sounds like Afterstep might be my best option but I'll need more ram and
maybe a bigger HD. I have a Digital 486/50 with 16mb and either 200mb or
500mb hd, I'll have to fire it up and check.

       Mike


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

Crossposted-To: comp.unix.shell,comp.lang.perl.misc
Subject: Re: Weird(?) magic word for sh to invoke perl under Linux
From: [EMAIL PROTECTED] (Michael Wang)
Date: Fri, 23 Mar 2001 05:17:44 GMT

>You mean mere users can't modify their PATH in their own shell profile or rc
>files? If people are smart enough to write scripts, wouldn't they be smart
>enough to change PATH? Assuming neither . not ~/bin is in their default
>path, they're either going to get very sick of typing ./ or similar before
>every command or they're going to change their path.

For non-interactive use, cron as an example, you expect people to use

PATH=/usr/local/bin /whatever/path/perl-program-with-#!/usr/bin/env-perl?


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

From: "Matt O'Toole" <[EMAIL PROTECTED]>
Subject: Re: pdf (was: Best E-mail Client?)
Date: Thu, 22 Mar 2001 21:20:21 -0800


"Jan Schaumann" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]...

> Treating intelligent people like they're dumb is disrespectful.  It's a
> matter of how you communicate the message:  yelling "You dumb asshole,
> shove this MS-Word Doc up your ass" will certainly not work.  Being
> polite and willing to help, will.

Oh, I'm polite and willing alright, enough to keep a machine running with MS
software installed.  It's easier than trying to educate a VP with an
attitude problem, of which there are many, ie,   "Whatsa matter with you,
you dunno howta use Word?  What is this Yoo-nix shit anyway?  I thought we
got ridda that in the seventies!  And I thought you were the computer
genius...  Jeezus!"

Matt O.




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

From: Joe <[EMAIL PROTECTED]>
Subject: fopen error
Date: Fri, 23 Mar 2001 05:30:19 -0000

I get an fopen error on all scripts that I use on my server. 

example:
failed to open log file
fopen: Permission denied
[Fri Mar 23 00:16:28 2001] [error] [client 33.44.44.44] Premature end of 
script headers: /usr/local/plesk/apache/vhosts/domain.com/cgi-
bin/quirex.cgi

I see the commands for fopen in the scripts, but I can't figure how to fix 
it. Installing something extra? Perl Module?

I use Redhat 7.0, Apache, w/ Plesk

Thank you for any help!

Joe

--
Posted via CNET Help.com
http://www.help.com/

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

From: "BetrOffDed" <[EMAIL PROTECTED]>
Subject: Re: new kernel 2.4.1
Date: Fri, 23 Mar 2001 06:01:10 GMT

In article
<[EMAIL PROTECTED]>, "Mohammed
Khalid Ansari" <[EMAIL PROTECTED]> wrote:




> Hi,
> I compiled linux kernel 2.4.1 on my machine which has IDE dirve and made
> it run. Now I compiled the same kernel on another machine with SCSI
> drive but it doesn't work. It is using initrd to boot the old kernel
> 2.2.41 using initrd while in the first case above it doesn't boot using
> initrd. Please help me out.
> 
initrd is "initial ramdisk" and contains the applicable modules for
booting from a scsi device in your case. If you haven't already, you need
to make a new initrd for the new kernel on the scsi machine.

Read "man initrd" for more info.

You didn't mention what distro, but you may want to check if mkinitrd is
on your system. It makes creating a new initrd simple. Somehting like:
mkinitrd /boot/initrd-2.4.1.img 2.4.1

Then edit lilo.conf to point to the initrd: "initrd=/boot/initrd-2.4.1.img"
in the appropriate "image" section should do the trick.

"man mkinitrd" for more info if its installed.

HTH...

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

From: [EMAIL PROTECTED] <[EMAIL PROTECTED]>
Subject: No in.ftpd after install
Date: Fri, 23 Mar 2001 05:33:41 +0000

I use Redhat 6.2. For some reason I don't have in.ftpd:

# pwd
/usr/sbin
# ls in.*d
in.fingerd  in.ntalkd  in.rlogind  in.talkd    in.tftpd
in.identd   in.rexecd  in.rshd     in.telnetd

What caused this during installation? I don't recall selecting it to be NOT installed. 
Which CD contains in.ftpd? I wonder if I'm missing other important utilities.

Thanks for help.

Yong Huang
[EMAIL PROTECTED]
_______________________________________________
Submitted via WebNewsReader of http://www.interbulletin.com


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

From: Wayne Pollock <[EMAIL PROTECTED]>
Subject: Re: RH 6.2 does'nt recognize 2nd NIC?
Date: Fri, 23 Mar 2001 01:22:50 -0500

Try to manually  configure it with "ifconfig".  If this fails then
the hardware isn't being seen.  Use "kudzu" program to detect hardware,
but this is supposed to be run at every boot.  Still I don't think it
can hurt to run it manually.

If the ifconfig solved your problem, then reboot and try this:
After you system comes up, as root type:
   /etc/rc.d/init.d/network restart

In RH6.2 (which I have, and ran into a similar problem), the brain-dead
network script *deliberately* ignores all but the primary network
connection if invoked with "start".  It will do a full network setup
only when invoked with "restart".

If this solves your problem, you need to hack the network script.

Hope this helps!

-Wayne Pollock

Teeitup816 wrote:
> 
> I've successfully installed Redhat 6.2 on a P75 server. At the time of install,
> only one NIC was installed. I've added a second NIC, because I want the Linux
> server to act like a router. I believe I've installed the Linux drivers
> succesfully on the server, but Linux does not list my second ethernet device.
> 
> When I executed "netstat -i"  I expected to see eth0 and eth1 listed, but I
> only see eth0.
> 
> Since I'm fairly new to Linux, I'm sure I missed a step. Can anyone help me
> with this problem?
> 
> Thank you in advance for responding so quickly.
> 
> Vince

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

From: "Chowalit Tinnagonsubout" <[EMAIL PROTECTED]>
Subject: same inode??
Date: Fri, 23 Mar 2001 13:48:55 +0700

Dear all
    Server of my friend(RedHat 6.2) was attacked by some kind of
rootkits(attack by wu-ftp BOF). It change many things. One of changing is
copy new script to command netstat and ps . My friend try to delete that
script but he can't. Because that scripts have copy inode from other file .
How do hacker do? And How to delete that script file ?
Thanks




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

From: Christopher Booth <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.setup,comp.os.linux.networking
Subject: Re: Best E-mail Client?
Date: Fri, 23 Mar 2001 16:30:15 +1100

Check out tradeclient

It looks very good and professional, the client is free
IMAP, POP, SMTP etc...

Chris

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

From: "Eric" <[EMAIL PROTECTED]>
Subject: Re: Weird /proc/mtrr
Date: Fri, 23 Mar 2001 08:52:15 +0100

> > > The mtrr entry in /proc looks odd to me
> > >
> > > [root@localhost /root]# ls -l /proc/mtrr
> > > -rw-r--r--   1 root     root           66 Mar 21 21:04 /proc/mtrr
> > > it contains
> > > reg00: base=0x00000000 (   0MB), size= 128MB: write-back, count=1
> > > reg01: base=0xe2000000 (3616MB), size=  32MB: write-combining, count=1
> > >
> > > every other entry (except kcore) has a size of 0
> > > Is it possible to create a normal file in /proc ?
> > >
> >
> > /proc is *not* a normal directory, but a peephole into the kernel
instead.
> > The 'files' and 'directories' in it are created by the kernel parts on
the
> > fly. The directory contents is nowhere on the disk, so it is not
possible to
> > create a file or subdirectory on it in the usual way.
> >
> > It is perfectly normal that the file lengths are reported as zeroes, as
the
> > file system does not know the read response length before the read is
> > performed. The 'contents' of a file is created by a suitable printf()
inside
> > the kernel at the time of reading the file.
> >
>
> That's why I thought it odd that /proc/mtrr has a size of 66, which is the
number of chars
> output by cat.

Look in the mtrr code, if you need to know more about it.
You can give a proc entry any (apparent) size you like.

Eric




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

From: David <[EMAIL PROTECTED]>
Subject: Re: same inode??
Date: Fri, 23 Mar 2001 07:46:18 GMT

Chowalit Tinnagonsubout wrote:
> 
> Dear all
>     Server of my friend(RedHat 6.2) was attacked by some kind of
> rootkits(attack by wu-ftp BOF). It change many things. One of changing is
> copy new script to command netstat and ps . My friend try to delete that
> script but he can't. Because that scripts have copy inode from other file .
> How do hacker do? And How to delete that script file ?
> Thanks


Try "chattr -i /path/to/filename" then delete it.
But once the system has been "Cracked" it is highly recommended to do a
format and reinstall. That is the only way to be sure that you don't
miss a back door somewhere. Then secure the system before putting it on
line again.

-- 
Confucius say: He who play in root, eventually kill tree.
Registered with the Linux Counter.  http://counter.li.org
ID # 123538
Completed more W/U's than 99.128% of seti users. +/- 0.01%

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

From: Christian Garms <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.setup,comp.os.linux.networking
Subject: Re: Best E-mail Client?
Date: 23 Mar 2001 08:13:40 +0100

[EMAIL PROTECTED] (Carlos D. Garza) writes:

>     I don't care much for Java either but how is it a security risk?
> You probably correct and I'm not argueing. How is Java a 
> security risk?

In theory there should be none. But both 'bigger' browsers (Netscape, 
Internet Explorer) have crappy implementations, so the secure sand box is 
weakened against perfomance and functionallity. But Java is safer then 
JavaScript.

-- 
regards,
        Christian               mailto:[EMAIL PROTECTED]

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

From: "Eric" <[EMAIL PROTECTED]>
Subject: Re: ALMOST SOLVED! I did 'dd if=/boot/mbr.b of=/dev/hda' :-(((
Date: Fri, 23 Mar 2001 08:38:34 +0100

> > > does it mean that my problem was the "image above cyl #1024"? AFAIK
> > > the image was under cyl #1024 even using the CHS detected by the
> > > BIOS (58168/16/63);
> >
> > I don't think so.
> > My guess is that you never ran LILO with the correct settings/file
>
> but it was working perfectly in the old HD!

But it must be run from the new HDD.
using the new kernel_image (even though it is just a copy)
LILO uses the physical address of those files, so if you copy a
file it is no longer at the same (physical) location.

> FYI: 3 days ago my old HD started failing; I bought this one,
> partitioned it more-or-less like the old one, copied everything with
> 'cp -ax' and ran lilo;
>
> > (cnp'ed your lilo.conf)
> >
> > root@queluz:~# cat /etc/lilo.conf
> > boot=/dev/hda
> > root=/dev/hda2
> >
> > this one was never correct.
> > Or was / on hda2 instead of /boot
>
> yes! / has allways been on /dev/hda2; /dev/hda1 was unused before I
> moved /boot to it;

It *should* not have been necessary.
Don't know why it did matter.

> > It's a very small partition for /
>
> but it is just 28% used! look:

I see, you stripped everything of :-)

> $ df
> Filesystem  1k-blocks    Used Available Use% Mounted on
> /dev/hda2       66373   18427     47946  28% /
> /dev/hda1       66341    2824     63517   4% /boot
> /dev/hda3      257705  140822    116883  55% /usr
> /dev/hda5      128812   55825     72987  43% /var
> [snip]
>
> > > Partition 1 does not end on cylinder boundary:
> > >      phys=(16, 127, 63) should be (16, 254, 63)
> > > /dev/hda2             9        18     68544   83  Linux
> > > Partition 2 does not end on cylinder boundary:
> > >      phys=(33, 127, 63) should be (33, 254, 63)
> > > /dev/hda3            18        51    266112   83  Linux
> > > Partition 3 does not end on cylinder boundary:
> > >      phys=(99, 127, 63) should be (99, 254, 63)
> > > /dev/hda4            51      3650  28913472    5  Extended
> > > Partition 4 does not end on cylinder boundary:
> > >      phys=(1023, 127, 63) should be (1023, 254, 63)
> > >
> > > should I worry?
>
> > not unless you let another OS touch that table.
> >
> > As you have a lot of free space on this disk,
> > what are you planning to do with it? install an MS OS?
>
> what? MS? definitely not!
> as I told you I needed another HD and the best deal I could find was
> this 30G;

Then you won't have problems.
(Still a lot is unused)

> > > if yes how do I fix it?
> >
> > That could be done by recreating the entire partitiontable.
>
> I understand you mean:
> - saving everything to another HD
> - recreating the entire partitiontable
> - loading everything back
> or is there an easier "automagic" way?

No automagical way.
The table exists, and was made with different CHS settings then the
one you're using now. You cannot safely change it, unless you know
*exactly* what you're doing.

Eric



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

From: Christian Garms <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.setup,comp.os.linux.networking
Subject: Re: Best E-mail Client?
Date: 23 Mar 2001 08:18:40 +0100

[EMAIL PROTECTED] (H.Bruijn) writes:

> On Thu, 22 Mar 2001 10:32:04 +0000, John Beardmore allegedly wrote:
> >
> >Yes, it's very cute, but I don't know if I'd like to do the bulk of my 
> >authoring in tex, though I did contemplate it a few years ago.
> 
> For me the main advantadge is that tex files themselves are in plain
> text, which anyone can read. That also makes them very compact. 
> Second using latex and other macro's forces you to structurise your texts,
> which you can also do in in Word, except that hardly anyone does, most
> people end up using a larger bold font size to designate chapters and
> number them manually.

Hehe, but M$-Word is ok, I've written my whole PhD-Thesis in Word. No 
offense to you, but Word is not that bad. Especially if you have EndNote, 
so the endnotes for references really work nifty. 

> There is though a much steeper learning curve to be able to create
> documents in latex, the results is generally much better as well. The
> moment that one wants to do things outside of the supplied macro's, is
> generally more difficult.

If you use LyX (graphical WYSIWIG frontend for LaTeX), the learning 
curve isn't that steep. My recommendation: Try LyX, and then switch to 
LaTex (if you want).

-- 
regards,
        Christian               mailto:[EMAIL PROTECTED]

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

From: "Quiney, Philip [HAL02:HH00:EXCH]" <[EMAIL PROTECTED]>
Subject: Re: No in.ftpd after install
Date: Fri, 23 Mar 2001 07:43:10 +0000

"[EMAIL PROTECTED]" wrote:
> 
> I use Redhat 6.2. For some reason I don't have in.ftpd:
> 
> # pwd
> /usr/sbin
> # ls in.*d
> in.fingerd  in.ntalkd  in.rlogind  in.talkd    in.tftpd
> in.identd   in.rexecd  in.rshd     in.telnetd
> 
> What caused this during installation? I don't recall selecting it to be NOT 
>installed. Which CD contains in.ftpd? I wonder if I'm missing other important 
>utilities.
> 
> Thanks for help.
> 
> Yong Huang
> [EMAIL PROTECTED]
> _______________________________________________
> Submitted via WebNewsReader of http://www.interbulletin.com
Hi,

I would suspect you chose a 'workstation' installation as this leaves
out the ftp server.

On my machine if I ask rpm to specify which package contains
/usr/sbin/in.ftpd I get..

[root@weasel /root]# rpm -qf /usr/sbin/in.ftpd
wu-ftpd-2.6.1-6

If you install this package from your CD with rpm then you will fix it.

Regards

Phil Q

-- 

Phil Quiney                             CSIP Demonstrator
[EMAIL PROTECTED]              Nortel Networks,
Telephone: +44 (0)1279 402363           London Rd, Harlow,
Fax:       +44 (0)1279 402885           Essex CM17 9NA,
                                        United Kingdom.

"This message may contain information proprietary to Northern 
Telecom so any unauthorised disclosure, copying or distribution
of its contents is strictly prohibited."

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


** 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