Linux-Misc Digest #730, Volume #27 Fri, 27 Apr 01 21:13:02 EDT
Contents:
graphical login screen (Ted Shoemaker)
Web Server ("WaCkOkId")
Re: comprehensible help on backspace and delete... (Paul Kimoto)
Re: Tired of XEMACS, moving to VIM (The Ghost In The Machine)
no core dump ("Ozymandias")
Re: cant reset terminal settings ("Peter T. Breuer")
Re: found how pppd is shutdown ----Re: cant reset terminal settings ("Peter T.
Breuer")
Re: find errors? ("Peter T. Breuer")
Re: help with partitioning ("Peter T. Breuer")
delected file in Linux ("Bill Zhao")
XMMS Kernel 2.4.3 (Mandrake 8.0) (Bigby)
Re: delected file in Linux ("Peter T. Breuer")
Re: delected file in Linux (William Kendrick)
Building rpc.statd exploit honeypot(port 111)--need help ("AC")
Re: AOL on Linux (Dave Uhring)
Re: no core dump (Dances With Crows)
Computer case badges with 3D effect with your own logo. Order (SecurisysAgency)
Re: cant reset terminal settings (John Hasler)
Re: cant reset terminal settings (David Efflandt)
----------------------------------------------------------------------------
Crossposted-To: comp.os.linux.setup,de.comp.os.unix.linux.misc
From: Ted Shoemaker <[EMAIL PROTECTED]>
Subject: graphical login screen
Date: Fri, 27 Apr 2001 21:12:00 GMT
Hello,
Concerning the graphical login screen on SuSE 7.1 ...
I changed the default icon
(the silhouette of a face, about 80 x 80 pixels) to other images,
using the methods available on the root KDesktop.
2 reasons:
1. I wanted to personalize my computer.
2. One of my children doesn't read yet, but she can recognize a
picture of her face -- if it's large enough.
That's right, folks. Linux has graduated to the "GUI for kids" stage.
And all is fine. Except now, the new icons are too small -- about half the size
of the default icons. I have tried using images of a variety of
sizes, even up to 600 x 600 pixels, in this application, but
they all end up about 40 x 40 on the login screen.
How do I make my own icons the same size as the SuSE icons?
Thanks,
Ted Shoemaker
[EMAIL PROTECTED]
------------------------------
From: "WaCkOkId" <[EMAIL PROTECTED]>
Subject: Web Server
Date: Fri, 27 Apr 2001 21:11:27 GMT
Which distribution would be the best for running a dedicated webserver?
It will be running on a Pentium 120 with 64mb of Ram and a 1gig hard drive.
------------------------------
From: [EMAIL PROTECTED] (Paul Kimoto)
Subject: Re: comprehensible help on backspace and delete...
Date: 27 Apr 2001 17:21:43 -0500
Reply-To: [EMAIL PROTECTED]
In article <9cc29t$77n$[EMAIL PROTECTED]>, Frank da Cruz wrote:
> Add this to the list. Hopefully it's a bit more comprehensible:
>
> http://www.columbia.edu/kermit/backspace.html
Oh, if only all we had to worry about were terminal emulators. Alas,
many problems arise because of X, e.g., with web browsers or fancy
X-aware editors.
--
Paul Kimoto
This message was originally posted on Usenet in plain text. Any images,
hyperlinks, or the like shown here have been added without my consent,
and may be a violation of international copyright law.
------------------------------
From: [EMAIL PROTECTED] (The Ghost In The Machine)
Crossposted-To:
gnu.emacs.help,alt.religions.vim,alt.religion.emacs,fj.editor.vi,comp.os.linux.advocacy
Subject: Re: Tired of XEMACS, moving to VIM
Date: Fri, 27 Apr 2001 21:39:50 GMT
In comp.os.linux.advocacy, those who know me have no need of my name
<[EMAIL PROTECTED]>
wrote
on 27 Apr 2001 09:07:21 GMT
<[EMAIL PROTECTED]>:
><9c9g2j$cfh$[EMAIL PROTECTED]> divulged:
>
>>: real men use punch cards.
>>
>>And use a bowie knife to cut the holes; you just have to be careful
>>not to cut yourself when you square off the corners.
>
>careful of the chad!
Indeed. It might be pregnant.
[.sigsnip]
--
[EMAIL PROTECTED] -- probably because it had such nice dimples... :-)
EAC code #191 11d:15h:29m actually running Linux.
[ ] Do you want this message to be private? Oops, too late.
------------------------------
From: "Ozymandias" <[EMAIL PROTECTED]>
Subject: no core dump
Date: Fri, 27 Apr 2001 23:47:34 +0200
Hi all,
How do I prevent programs from core dumping when a sigfault signal arrives
?? is there an enviroment option for this, or can a program set this as an
option ?? Pleas, need to know but can't find it anywhere... I know it is
possible for programs to sigfault without coredumps, but I don't know how to
setup this option, ..... anyone ??
greetz Ozymandias
------------------------------
From: "Peter T. Breuer" <[EMAIL PROTECTED]>
Subject: Re: cant reset terminal settings
Date: Fri, 27 Apr 2001 23:24:29 +0200
Glitch <[EMAIL PROTECTED]> wrote:
> In article <[EMAIL PROTECTED]>, "Peter T. Breuer"
> <[EMAIL PROTECTED]> wrote:
>> Glitch <[EMAIL PROTECTED]> wrote:
>> reconfigure the port with setserial. And possibly also rmmod serial to
>> take the serial driver out of the kernel.
> of course if the serial driver is built in that won't be an option. I can
> try setserial but i doubt it can reset a modem/serial port
That's precisely what it does do, in the sense of resetting the kernel
drivers state at least. The port itself is dependent on the modem for
the wellbeing of its state, since you have an internal modem and it's
its job to emulate a real serial port. Note that being a pci modem, its
irq and io gate will be at "unusual" locations.
>> Don't do the "something stupid" in the first place! But obviously, yes.,
>> see above!
> if u are referring to me rebooting i cant help it. Since i havent figured
No, I was referring to killing the connection inappropriately. Use
"killall pppd".
> out a way to fix it w/o rebooting it is my only option.
Well, you could use setserial after rmmodding and insmodding the serial
module, as suggested. Where does the word "only" derive from?
>> If your modem is internal, then you have discovered that resetting the
>> modem is only possible by restting the PC. That's a feature of internal
>> modems.
> yes, i know that
Good .. that realization's practically progress.
>> I also would bet that you are not telling us that you have a
>> winmodem, and that you are thus blaming us for something for which you
>> should blame yourself. Yes, winmodems don't work in themselves. You KNEW
>> that. They have a state that is dependent on the computer. Intrrupting a
>> software process that deals with them can leave them in strange states
>> (well, I guess that's how the "driver" leaves them, anyway!).
> uh, no, i dont have a winmodem. Good job of assuming. It worked out well,
> just like they said it does. I have an Actiontec PCI modem. In Windows
Is that a ...
> Communication controller: Lucent Microelectronics Venus WinModem (V90,
> 56KFlex)
.. yes lucent venus. I _believe_ that those are indeed full modems.
Don't quote me. Go look it up to make sure.
> however I need no driver to use it in Linux. A simple setserial command
How do you know?
> configured the port and irq for me and i was up and running with it.
And what is that setserial command and what happens when you do it
again, when the modem is in a funny state?
>> If it does, then that is the correct thing that it should be doing.
> what else would it be doing?
It might be sending "+++" down the line for all I know! Whatever, the
result is that whatever held the port dies without removing its
lockfile, and the whatever should have been pppd.
Peter
------------------------------
From: "Peter T. Breuer" <[EMAIL PROTECTED]>
Subject: Re: found how pppd is shutdown ----Re: cant reset terminal settings
Date: Fri, 27 Apr 2001 23:26:55 +0200
Glitch <[EMAIL PROTECTED]> wrote:
> The program just executes /etc/ppp/ppp-down when the 'disconnect' button
> is clicked. The contents of that script are below. Is this the wrong way
> to do it? Should I replace it with killall pppd?
Yes.
> kill `cat /var/run/ppp0.pid`
Are you an expert or not? A few moments ago you were blathering on
about setserial and serial modules as though you knew what you were
doing, yet here you ask a patently absurd question. It would be helpful
if you indicated your level of expertise so that people can help you at
the appropriate level, without engaging in backtracks or doubletakes.
Peter
------------------------------
From: "Peter T. Breuer" <[EMAIL PROTECTED]>
Subject: Re: find errors?
Date: Fri, 27 Apr 2001 23:31:47 +0200
Alex Lazarevich <[EMAIL PROTECTED]> wrote:
> I'm using Linux 6.1 web server machine. I'm having a very strange
> problem.
> Some of my java programs have been unable to find the classes. After
> checking all the obvious problems (CLASSPATH, typos, etc.) I discovered
> something bizzare. When I use the 'find' command to find the file, it
> can't find it! However, the file is there, there is no mistake, check
> out this output:
> From my home directory, I issue the command:
> find /usr/java/xerces-j-1.3.1/src/org/apache/xerces/parsers SAXParser.class
Well, that's the wrong syntax for a start. I would expect very little
sense from that!
> but check out the result:
[meaningless drivel snipped]
> Any ideas as to if my "find" and "java class not found" problems are
Yes - in both case you seem to have skipped reading the instructions.
In the case of find, it is the man page that you have not read. In the
case of the class problem, it's likely because you do have a typo
(upper lower/case) or you are working on a file system with different
case senistivity to what you think it has (without more data nobody can
say).
"man find". If that page's too difficult, "man locate" instead.
Peter
------------------------------
From: "Peter T. Breuer" <[EMAIL PROTECTED]>
Subject: Re: help with partitioning
Date: Fri, 27 Apr 2001 23:39:31 +0200
Sudhakar R. <[EMAIL PROTECTED]> wrote:
> Currently my fstab is as follows
> [root@matrixuc local]# cat /etc/fstab
> LABEL=/ / ext2 defaults 1 1
> LABEL=/home /home ext2 defaults 1 2
What? Where does LABEL= come from? lvm?
> /dev/hda1 /mnt/c vfat defaults 0 0
> /dev/cdrom /mnt/cdrom iso9660 noauto,users,ro 0 0
> /dev/hda8 /mnt/d vfat defaults,uid=500,umask=007,rw 0 0
> /dev/hda9 /mnt/e vfat defaults 0 0
> /dev/fd0 /mnt/floppy auto noauto,owner 0 0
> none /proc proc defaults 0 0
> none /dev/pts devpts gid=5,mode=620 0 0
> /dev/hda7 swap swap defaults 0 0
> My / partition is full. so i'd like to move my /usr/local/ (290MB) to
You might also consider /usr/share and /usr/X11R6 as candidates.
> another new partition. I intend to repartition my /dev/hda9 and use the
> second half to mount /usr/local. Could someone please give me detailed
> steps on what to do and what NOT TO DO. also if you think there might be
Well, no. You seem to have the right idea, and that's all anybody can
tell you. You can't do any harm by hiving off /usr/local. Look at some
of the howto's (Partition?) if you really feel you need handholding
through what is essentially a simple copy, delete, and unmount, mount.
> better way of accomplishing the same (freeing up space on my / ) kindly
> let me know.
It's better to have had /usr on a separate partition to start with!
Then you could play as much as you like.
I'd tell you to make backups, but I don't see why, since your
operation is essentially a backup of /usr/local to your new
partition. When it's done ok, delete the original.
Your only problem would have been logical partition renumbering, but
as hda9 is last, you don't have the problem.
Peter
------------------------------
From: "Bill Zhao" <[EMAIL PROTECTED]>
Subject: delected file in Linux
Date: Fri, 27 Apr 2001 10:31:02 +0800
Hi, good morning
I deleted a detectory by my mistake.
I remember there is a unix command that can be used to recover the deleted
file.
I also did use it once sometime ago
please help me !
Bill Zhao
2001.4.27
------------------------------
From: Bigby <[EMAIL PROTECTED]>
Subject: XMMS Kernel 2.4.3 (Mandrake 8.0)
Date: Fri, 27 Apr 2001 22:30:06 -0000
I just installed Mandrake Linux 8.0 with the kernel 2.4.3-20mdk and XMMS
1.24 (i think). When I play a CD using XMMS, it works. I even have it set
to use the OSS mixer to play the CD. I am having problems with other
things, like .wav files and .mp3 files. In those cases the song advances
at about 20x the normal rate and no sound is output. I don't see any wav
writer thing (and why would it do this to wave files too). I need some
advice, thanks in advance.
Brian
--
Posted via CNET Help.com
http://www.help.com/
------------------------------
From: "Peter T. Breuer" <[EMAIL PROTECTED]>
Subject: Re: delected file in Linux
Date: Sat, 28 Apr 2001 00:56:16 +0200
Bill Zhao <[EMAIL PROTECTED]> wrote:
> Hi, good morning
> I deleted a detectory by my mistake.
> I remember there is a unix command that can be used to recover the deleted
> file.
No there isn't.
> I also did use it once sometime ago
You are imagining things.
> please help me !
You should umount the partition (or remount it readonly) and get out
debugfs and use its lsdel command to look at the least of recently
deleted inodes. Then you should use its link command to put them back
in the FS tree.
Or you can use mc to do much the same thing, I hear.
Peter
------------------------------
From: William Kendrick <[EMAIL PROTECTED]>
Subject: Re: delected file in Linux
Date: Fri, 27 Apr 2001 23:26:30 GMT
Bill Zhao <[EMAIL PROTECTED]> wrote:
: Hi, good morning
: I deleted a detectory by my mistake.
: I remember there is a unix command that can be used to recover the deleted
: file.
: I also did use it once sometime ago
: please help me !
There's a program called "recover" (and a GTK+ frontend called "grecover")
which you can use. It takes a while to collect all inode information,
and is obviously not guaranteed to work.
http://recover.sourceforge.net/linux/recover/
Good luck! BTW, you might want to alias some of your more dangerous
commands to have "-i" switches.
cp -i <-- won't copy over an existing file unless you approve
mv -i <-- won't move/rename over an existing file ...
rm -i <-- won't delete the file unless you approve
Another useful thing to set up is "noclobber" mode in your shell.
So, when you do this:
grep junk file1 > file2
...the command will not execute if "file2" already exists.
(The shell will complain, if 'noclobber' mode is on)
Unfortunately, I don't have much time right now to give you good
references for all this, but it's pretty FAQ, I'm sure. :)
-bill!
------------------------------
From: "AC" <$-xenite9@my-dejanews-com-$>
Subject: Building rpc.statd exploit honeypot(port 111)--need help
Date: Fri, 27 Apr 2001 19:27:47 -0400
I am trying to build a simple honeypot to mimic the rpc.statd (port 111)
vulnerablility so I can trick these script kiddies (that have scanned my box
hundreds of times in the last few months) into thinking they have "cracked"
my machine. I don't have any interest in getting anyone in trouble, I just
want to watch offenders for my amusement, and maybe scare them for added
fun. I don't understand how this exploit works. I've read all the CERT
info and that is too unspecific. My questions will follow.
Before I build my server I used netcat to listen on port 111 with this
command:
nc -L -p 111 -v -v >> log.txt
The console output after someone connects from netcat is like so (IPs
changed to protect the guilty):
listening on [any] 111 ...
connect to [x.x.x.x] from hostnameofmachine [x.x.x.x] 3096 sent 0, rcvd 0
So far nobody has done anything but connect, send nothing, and disconnect.
Does the exploit occur over TCP port 111? Should I be running a service
over UDP also? Thanks.
------------------------------
From: Dave Uhring <[EMAIL PROTECTED]>
Subject: Re: AOL on Linux
Date: Fri, 27 Apr 2001 18:39:19 -0500
Claus Atzenbeck wrote:
> Arctic Storm wrote on Thu, 19 Apr 2001 20:17:14 GMT:
>
>> > I just bought Mandrake's Linux for Windows and want to use AOL to get
>> > online. I found and installed the wine-windows emulator but it still
>> > does not work.
>>
>> I confirmed solution that I personally know of is this.
>> Run Windows in VMWare for Linux, then run AOL within VMWare virtual
>> computer.
>
> I have Mandrake an VMware running on it. As guest I have Win98 available.
>
> My ISP will cut its flatrate. The only (cheap) flatrate provider left in
> my region is AOL.
>
> Is anyone here with some experiences concerning this? How can I do ip
> forwarding with Win98? How can I tell Win98/VMware to use my external ISDN
> adapter?
>
> Thanks for any hint!
> Claus
>
If you have Win98SE, install and configure Internet Connection Sharing.
------------------------------
From: [EMAIL PROTECTED] (Dances With Crows)
Subject: Re: no core dump
Reply-To: [EMAIL PROTECTED]
Date: 27 Apr 2001 23:37:08 GMT
On Fri, 27 Apr 2001 23:47:34 +0200, Ozymandias staggered into the Black
Sun and said:
>Hi all, How do I prevent programs from core dumping when a sigfault
>signal arrives ?? is there an enviroment option for this, or can a
>program set this as an option ?? Pleas, need to know but can't find it
>anywhere... I know it is possible for programs to sigfault without
^^^^^^^^
.sigmentation fault--appending core file to end of message...
>coredumps, but I don't know how to setup this option, ..... anyone ??
man bash , paying attention to the -c option of "ulimit". I have found
that a good thing to do is to put "ulimit -c0" into /etc/profile so that
every user's shells default to generating no corefiles. If a user
really wants corefiles, they should know how to turn them back on. SuSE
puts that ulimit command into /etc/profile by default; don't know why
RedHat hasn't.
--
Matt G|There is no Darkness in Eternity/But only Light too dim for us to see
Brainbench MVP for Linux Admin / Workin' in a code mine, hittin' Ctrl-Alt
http://www.brainbench.com / Workin' in a code mine, whoops!
=============================/ I hit a seg fault....
------------------------------
Subject: Computer case badges with 3D effect with your own logo. Order
From: SecurisysAgency <[EMAIL PROTECTED]>
Date: Sat, 28 Apr 2001 00:20:37 GMT
Computer case badges with 3D effect with your own logo. Order securely
Online for FAST delivery:
http://www.securisysagency.com/shop/shop.html
We are offering quality custom PC badges printed your own design or choose
from our range of "off-shelf" labels, tamper-proof warranty labels, domed
labels, Linux, BSD badges and merchandize.
Online ordering. Stock orders are normally despatched within 48-hours to
anywhere in the world.
FREE shipping to anywhere in the world!
www.securisysagency.com
Disclaimer:
This posting is made in understanding that this newsgroup allowes
advertising material.
If you are a moderator of the channel and you believe that posting of this
nature doesn't comply with the one of the channel, please write to:
[EMAIL PROTECTED] and subsequent postings will stop.
------------------------------
From: John Hasler <[EMAIL PROTECTED]>
Subject: Re: cant reset terminal settings
Date: Fri, 27 Apr 2001 23:15:46 GMT
Peter writes:
> Whatever, the result is that whatever held the port dies without removing
> its lockfile, and the whatever should have been pppd.
I've lost track of what software the OP is using to control pppd, but a
common problem with wvdial and possibly other packages is that it sends
pppd a TERM signal, waits two seconds, checks to see if pppd is still
alive, and if so, sends a SIGKILL (i.e., kill -9). Trouble is, pppd tries
to get the other end to acknowledge a TermReq before it exits. If the ISP
ignores TermReq's (as mine does) pppd eventually gives up and exits anyway,
but this takes more than two seconds.
--
John Hasler
[EMAIL PROTECTED]
Dancing Horse Hill
Elmwood, Wisconsin
------------------------------
From: [EMAIL PROTECTED] (David Efflandt)
Subject: Re: cant reset terminal settings
Date: Sat, 28 Apr 2001 01:05:28 +0000 (UTC)
Reply-To: [EMAIL PROTECTED]
On Fri, 27 Apr 2001 04:55:57 -0400, Glitch <[EMAIL PROTECTED]> wrote:
> Every once in a while my modem stops sending/receiving. I hit disconnect
> (I'm using an app called Simpleppp since kppp wouldn't work right) and
> the line goes down like it is supposed to.
>
> However, when I try to connect again but the modem doesn't dial out. I check and
>pppd is
> running which means the modem is locked. So i wait for the modem to be
> unlocked. When it is I hit 'connect' again and then in /var/log/messages
> I'm told
>
> chat[315]: Can't get terminal parameters: Input/output error
>
> Because of this I can't use the modem to dial out, unles I reboot to
> reset the serial port I guess. Now, maybe there is a way to fix it w/o
> rebooting but I don't know what it is.
I had a similar problem when my ISP forced us to use pops with Livingston
(then Lucent) Portmaster (which had never passed beta testing). I had an
internal Zoom PnP K56flex modem (flashed to V.90) that would totally lock
up at times in Linux or Windows. Other users had similar problems. The
only way to reset it when it locked up solid was to reboot.
When it failed 3 times attempting to ftp install FreeBSD, I got a
SupraExpress 56K external that flawlessly did the 11 hr ftp install,
even with the bad Portmaster.
Lucent never could fix the Portmaster, so they ended its lifecycle. I
recently replaced a Winmodem with the old Zoom and it works fine now that
the Lucent Portmasters at my ISP have been retired (hopefully retired and
not sold to another ISP).
--
David Efflandt (Reply-To is valid) http://www.de-srv.com/
http://www.autox.chicago.il.us/ http://www.berniesfloral.net/
http://cgi-help.virtualave.net/ http://hammer.prohosting.com/~cgi-wiz/
------------------------------
** 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
******************************
