Linux-Misc Digest #563, Volume #26 Sat, 16 Dec 00 18:13:03 EST
Contents:
Re: blindly accepted net advice--> blew away filesystem (Rob MacGregor)
Re: my list of RH7.0 bugs ("David")
Install Prlblem ("Buck Turgidson")
Re: Install Prlblem ("Dan White")
Re: Mouse problem Loading X ("Dan White")
Re: Mouse problem Loading X ("Dan White")
Re: Route Command? What is wrong here? (David Efflandt)
Re: Gnome Libs Installation ("Dan White")
X -windows libraries in RH 7.0 ("Generic User")
Re: Radeon + linux Question (Stanislaw Flatto)
Re: my list of RH7.0 bugs (Chris Ahlstrom)
Re: my list of RH7.0 bugs (Hal Burgiss)
Re: X -windows libraries in RH 7.0 (Hal Burgiss)
my @felder doppelt deklariert (Bernhard Kuemel)
Re: fetchmail problem (David Efflandt)
Re: Logging the rc.local? (David Efflandt)
Re: dependency question (Ken Dozhier)
Re: Pass character to a program? (David Efflandt)
Sound Blaster not working ("Chris")
Re: pine and spam mails (David Efflandt)
----------------------------------------------------------------------------
From: Rob MacGregor <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.security,comp.security.misc,comp.unix.misc
Subject: Re: blindly accepted net advice--> blew away filesystem
Reply-To: "rob macgregor hotmail com _ @ ."
Date: Sat, 16 Dec 2000 20:18:23 GMT
green wrote:
> I once did this and in a paniced leap i hit power (what the hell it was gone
> any way.)
> tured out linux hadn't sync 'ed the disks and nothing was lost!. fluke.
> teach me to work as root.
My favourite, done on a sun:
thishost# rm -fr thisdir
rm: internal syncronisation lost.
Disk rattles away merrily in the background until I ^C it, and discover the
magic message translated to "rm: trashing all file systems" :-(
Only happened twice (within 10 minutes of eachother, on different machines),
never seen it since.
--
Rob MacGregor (MCSE) [PGP key ID 0x1F5239DD]
The light at the end of the tunnel is an oncoming dragon.
Real email address (just add @ . ): rob_macgregor hotmail com
------------------------------
Reply-To: "David" <[EMAIL PROTECTED]>
From: "David" <[EMAIL PROTECTED]>
Crossposted-To: alt.os.linux,linux.redhat
Subject: Re: my list of RH7.0 bugs
Date: Sat, 16 Dec 2000 12:43:19 -0800
The change from inetd to xinetd is one of Red Hat's attempts to patch up a
security problem from previous versons of Red Hat. This has nothing to do
with them attempting to be "different"...
> But I would like to HEARTILY echo your comment on xinetd. As a teenager,
I
> was a spike-haired punk rocker, and just like most of my friends we were
> desperately trying to distinguish ourselves from the Izod-wearing masses.
> This change of inetd to xinetd seems to be the same effort to say "but
> we're different!"
>
> Aside from the socialogical observation, this change makes me wonder what
> other subsystems changed there configuration files. This is especially
> anoying when trying to build a "hardened" server, while also trying to
> identify the differences that might change my "hardening" procedure. The
> addition of the /etc/xinetd.d directory only makes my efforts more
> difficult, and allows another area for a hacker to exploit.
>
------------------------------
From: "Buck Turgidson" <[EMAIL PROTECTED]>
Subject: Install Prlblem
Date: Sat, 16 Dec 2000 20:39:43 GMT
I just installed a RH 6.1 workstation, using the GUI installer. After installing, it
told me to remove the "boot media" so it could
reboot.
It apparently rebooted, but the monitor is blank. Some random graphical characters
appeared as it was booting, but now it is blank.
I guess I need to get in using a terminal session and perhaps run Xconfigurator.
Anyone got any suggetions about what to do now, and perhaps what I did wrong?
Thanks for any help.
------------------------------
From: "Dan White" <[EMAIL PROTECTED]>
Subject: Re: Install Prlblem
Date: Sat, 16 Dec 2000 20:57:59 GMT
In article <jqQ_5.797$[EMAIL PROTECTED]>, "Buck
Turgidson" <[EMAIL PROTECTED]> wrote:
> I just installed a RH 6.1 workstation, using the GUI installer. After
> installing, it told me to remove the "boot media" so it could reboot.
>
> It apparently rebooted, but the monitor is blank. Some random graphical
> characters appeared as it was booting, but now it is blank.
>
> I guess I need to get in using a terminal session and perhaps run
> Xconfigurator.
>
> Anyone got any suggetions about what to do now, and perhaps what I did
> wrong?
>
> Thanks for any help.
It sounds like you are booting into run level 5, which tries to start a
grphical session on bootup, and your X configuration is bad. Try booting
into single user mode by 'linux single' at the lilo prompt. Then rerun
your X configuration by typing "Xconfigurator". Afterwards, you can boot
into normal multiuser mode by type "init 3", or graphical boot mode by
typing 'init 5'.
- Dan White
------------------------------
From: "Dan White" <[EMAIL PROTECTED]>
Subject: Re: Mouse problem Loading X
Date: Sat, 16 Dec 2000 21:02:02 GMT
In article <rDN_5.638$[EMAIL PROTECTED]>, "Buck
Turgidson" <[EMAIL PROTECTED]> wrote:
> A little more info: It appears to be looking for /dev/mouse, which
> isn't there (it exists on my other linux box and appears to be a
> symbolic link to ttys0).
>
>> I haven't used X in a number of months. Now I need to install a new
>> version of Oracle. When I run X, I > get a message saying :
>>
>> Fatal server error: Cannot open mouse (no such file or directory).
>>
>> I checked /var/log/messages, and it appears to be detecting the mouse
>> (or at least the port) upon boot. Can anyone suggest any other places
>> to look? Thanks for any help.
>>
>> Dec 16 11:06:55 localhost kernel: Detected PS/2 Mouse Port. Dec 16
>> 11:06:55 localhost kernel: Serial driver version 4.27 with MANY_PORTS
>> MULTIPORT SHARE_IRQ enabled
>>
Within the X configuration file /etc/X11XF86Config, it specifies which
mouse device you want to use. You can change /dev/mouse to /dev/psaux.
/dev/mouse is typically use just as convenient pointer to your actual
mouse device. Another alternative is to just make a symbolic link:
ln -sf /dev/psaux /dev/mouse
- Dan White
------------------------------
From: "Dan White" <[EMAIL PROTECTED]>
Subject: Re: Mouse problem Loading X
Date: Sat, 16 Dec 2000 21:04:00 GMT
In article <eLQ_5.70519$[EMAIL PROTECTED]>, "Dan White"
<[EMAIL PROTECTED]> wrote:
>
> Within the X configuration file /etc/X11XF86Config, it specifies which
That should be /etc/X11/XF86Config
- Dan White
------------------------------
From: [EMAIL PROTECTED] (David Efflandt)
Crossposted-To: comp.os.linux.networking
Subject: Re: Route Command? What is wrong here?
Date: Sat, 16 Dec 2000 21:10:03 +0000 (UTC)
Reply-To: [EMAIL PROTECTED]
On Fri, 15 Dec 2000 13:17:17 GMT, Quad <[EMAIL PROTECTED]> wrote:
>
>
>Hi,
>
>I have 3 nics all on diff networks on a linux box. Its redhat 6.2 no
>updates (yet as I wanted to test, then patch OS/Kernel). Here is the
>route command I am trying to implement:
>
>route add -net 175.23.0.0 netmask 255.255.0.0 eth1
>route add -net 24.231.45.0 netmask 255.255.255.0 eth2
>
>It just scrolls the route command (help) as though im not typing it
>right...
I have had route fail if I neglect to put 'dev' in front of the device.
As root try:
route add -net 175.23.0.0 netmask 255.255.0.0 dev eth1
This may require that eth1 be up with a 175.23. IP on it or as an alias.
>I cant even seem to add a default gateway with the route command?
>
>Do I need a newer kernel/patch/utility to make this work?
Maybe just a more complete commandline.
--
David Efflandt [EMAIL PROTECTED] http://www.de-srv.com/
http://www.autox.chicago.il.us/ http://www.berniesfloral.net/
http://cgi-help.virtualave.net/ http://hammer.prohosting.com/~cgi-wiz/
------------------------------
From: "Dan White" <[EMAIL PROTECTED]>
Subject: Re: Gnome Libs Installation
Date: Sat, 16 Dec 2000 21:13:12 GMT
In article <91fp3o$quo$[EMAIL PROTECTED]>, [EMAIL PROTECTED] wrote:
> Hi,
>
> I have a problem when I try to install gnome-libs when I run the
> ./configure script.
>
> First, it complained about imlibs and then I installed that at first and
> set
>
> export IMLIB_CONFIG=/usr/local/imlib-1.9.8.1/imlib-config
>
When you configure gnome related libs, it's a good idea to do:
./configure --prefix=/usr --sysconfdir=/etc
Which will install the *-config scripts into /usr/bin, which is easily
found in your path. Another alternative is to include
/usr/local/imlib-1.9.8.1, etc. into your path.
- Dan White
------------------------------
From: "Generic User" <[EMAIL PROTECTED]>
Subject: X -windows libraries in RH 7.0
Date: Sat, 16 Dec 2000 20:57:02 GMT
Since Red Hat support was clueless on this one, I'd thought
I'd ask you folks.
I have the X11R6 directory under /usr so I'm assuming I have
the complete X-windows libraries. Apparently not since I'm
installing an application and it can't find the libXt* libraries,
which should be in the directory. Which packages should I
install to get the complete set of X-windows libraries?
-chris
------------------------------
From: Stanislaw Flatto <[EMAIL PROTECTED]>
Subject: Re: Radeon + linux Question
Date: Sun, 17 Dec 2000 08:31:41 +0000
Michael Powe wrote:
>
>
> I tried running it with the generic VGA server, that was horrible &
> hosed my video besides. I had to reboot to get back to normal, even
> "reset" would not work.
>
> Thanks for any help.
Hi Michael.
After upgrading(sic!) to AGP card I found myself in your situation.
The possible solutions were, throw the card in garbage bin and admit
failure or be stubborn and make it work. (OK in such situation some
flexibility is needed).
So reluctantly I went fb. It works well but I cannot change resolutions
on the fly as this is given to lilo on boot. So I am stuck with 800x600
24bpp.
Not that it matters, it was my prefered resolution when I could play
with it anyway.
HTH
>
> Michael Powe Portland, Oregon USA
--
Stanislaw
Linux counter No.162760
Slak user from Ulladulla.
------------------------------
From: Chris Ahlstrom <[EMAIL PROTECTED]>
Crossposted-To: alt.os.linux,linux.redhat
Subject: Re: my list of RH7.0 bugs
Date: Sat, 16 Dec 2000 21:35:55 GMT
David wrote:
>
> The change from inetd to xinetd is one of Red Hat's attempts to patch up a
> security problem from previous versons of Red Hat. This has nothing to do
> with them attempting to be "different"...
>
> > But I would like to HEARTILY echo your comment on xinetd. As a teenager,
> I
> > was a spike-haired punk rocker, and just like most of my friends we were
> > desperately trying to distinguish ourselves from the Izod-wearing masses.
> > This change of inetd to xinetd seems to be the same effort to say "but
> > we're different!"
> >
> > Aside from the socialogical observation, this change makes me wonder what
> > other subsystems changed there configuration files. This is especially
> > anoying when trying to build a "hardened" server, while also trying to
> > identify the differences that might change my "hardening" procedure. The
> > addition of the /etc/xinetd.d directory only makes my efforts more
> > difficult, and allows another area for a hacker to exploit.
Actually, my latest favorite book, "Hacking Exposed", 2nd ed., talks about
xinetd as preferred over inetd because it is more secure.
Chris
------------------------------
From: [EMAIL PROTECTED] (Hal Burgiss)
Crossposted-To: alt.os.linux,linux.redhat
Subject: Re: my list of RH7.0 bugs
Reply-To: Hal Burgiss <[EMAIL PROTECTED]>
Date: Sat, 16 Dec 2000 22:16:32 GMT
On Sat, 16 Dec 2000 21:35:55 GMT, Chris Ahlstrom <[EMAIL PROTECTED]> wrote:
>
>Actually, my latest favorite book, "Hacking Exposed", 2nd ed., talks about
>xinetd as preferred over inetd because it is more secure.
Not only that it seems more flexible all the way around. I can't see
this as anything but a positive move. The configuration seems pretty
straightforward AFAICT. Being Linux, if someone doesn't like it for
whatever reason, rpm -e and then install something else.
--
Hal B
[EMAIL PROTECTED]
[EMAIL PROTECTED]
[EMAIL PROTECTED]
--
------------------------------
From: [EMAIL PROTECTED] (Hal Burgiss)
Subject: Re: X -windows libraries in RH 7.0
Reply-To: Hal Burgiss <[EMAIL PROTECTED]>
Date: Sat, 16 Dec 2000 22:20:15 GMT
On Sat, 16 Dec 2000 20:57:02 GMT, Generic User <[EMAIL PROTECTED]>
wrote:
>Since Red Hat support was clueless on this one, I'd thought I'd ask you
>folks.
>
>I have the X11R6 directory under /usr so I'm assuming I have the
>complete X-windows libraries. Apparently not since I'm installing an
>application and it can't find the libXt* libraries, which should be in
>the directory. Which packages should I install to get the complete set
>of X-windows libraries?
[hal@feenix bs]$ rpm -qf /usr/X11R6/lib/libXt.a
XFree86-devel-4.0.1-1
[hal@feenix bs]$ rpm -qf /usr/X11R6/lib/libXt.so.6.0
XFree86-libs-4.0.1-1
Also, RH7.
If the problems are installing from source, it is usually the devel
packages that gto overlooked.
--
Hal B
[EMAIL PROTECTED]
[EMAIL PROTECTED]
[EMAIL PROTECTED]
--
------------------------------
From: Bernhard Kuemel <[EMAIL PROTECTED]>
Subject: my @felder doppelt deklariert
Date: Sat, 16 Dec 2000 22:22:49 GMT
"my" variable @felder masks earlier declaration in same scope at
boerse.pm line
279.
--
Bernhard Kuemel Tel: +43/1/505-99-65
Adr: Brahmsplatz 7/11 [EMAIL PROTECTED]
1040 Wien BEKU1-RIPE
Austria/EUROPE
------------------------------
From: [EMAIL PROTECTED] (David Efflandt)
Subject: Re: fetchmail problem
Date: Sat, 16 Dec 2000 22:26:40 +0000 (UTC)
Reply-To: [EMAIL PROTECTED]
On Sat, 16 Dec 2000 16:07:12 +0100, Michael Heiming <[EMAIL PROTECTED]> wrote:
>[EMAIL PROTECTED] wrote:
>
>> I'm trying to use fetchmail on a dial-up
>>connection. Everything's set up correctly, but when it downloads the
>>messages and deletes them from the server, the mail never shows up on my
>>system-at least not anywhere I can find it. Am I missing something in
>>.fetchmailrc? Where should it be dropping these messages. I assumed
>>/var/spool/mail/joeuser, but it just ain't there. I'm running SuSE
>>6.3. Help!
>>
>> ef
>
>Hello,
>
>AFAIK you have to run sendmail -q
>That will drop/deliver the mail from your /var/spool/mqueue/ to
>/var/spool/mail/<username>
Or run sendmail or other smtp server as a daemon, assuming it is
configured properly. Can you send mail from one user to another user on
your own system? Check the mail logs (something in /var/log?).
--
David Efflandt [EMAIL PROTECTED] 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/
------------------------------
From: [EMAIL PROTECTED] (David Efflandt)
Subject: Re: Logging the rc.local?
Date: Sat, 16 Dec 2000 22:40:57 +0000 (UTC)
Reply-To: [EMAIL PROTECTED]
On Fri, 15 Dec 2000 19:37:16 -0800, Flej Ling <[EMAIL PROTECTED]> wrote:
>How can I log what's happening in rc.local when it runs? I put a script into
>it but the script doesn't seem to run. I want to see if there are any error
>messages or see how far the script is getting. "cat ./var/log/messages"
>doesn't show much - it just shows ppp trying to connect, ie: the end result
>of the commands. What I want is like a DOS "echo" to display things as they
>are happening. Also, does rc.local run as user "root" when you are booting?
rc.local runs as root but only outputs what you tell it to. So you need
to echo a message first saying what you are about to do or enable a more
verbose mode in the command you are executing. If you are trying to
figure out what pppd is doing you should be able to see more in the logs
by enabling debug or kdebug 7, but check /etc/syslog.conf to see if or
where that would be logged. Debug messages are not enabled in some
distros or in one case they piled up in something called kern.log instead
of debug.
Also the enviroment may be minimal, so you should use full paths to any
programs or data files.
--
David Efflandt [EMAIL PROTECTED] 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/
------------------------------
From: Ken Dozhier <[EMAIL PROTECTED]>
Subject: Re: dependency question
Date: Sat, 16 Dec 2000 22:44:19 GMT
Reply-To: [EMAIL PROTECTED]
Agh! It was late when I posted that.
I'm running RedHat 5.2 and the install program was RPM. I've been
getting that dependency error alot when I try installing programs such
as instant messengers and mail clients.
What I don't certainly understand is: can two versions of a library be
installed on one system? Yes or no would answer a myriad of other
questions.
In article <[EMAIL PROTECTED]>,
[EMAIL PROTECTED] wrote:
> Ken Dozhier wrote:
> >
> > Hi,
> >
> > I recently attempted to install a program and got a dependency error
> > message listing libc.so.6 (GLIBC 2.0) and libc.so.6 (GLIBC 2.1) as
being
> > needed by the program. I've ran into this dependency error quite a
bit.,
> > in spite of having version 2.0.7 already installed on my system.
> >
> > My question is: does this mean that I need libc.so.6 from both
versions
> > of GLIBC or just from either 2.0 or 2.1?
> >
> > TIA
> >
> > Sent via Deja.com
> > http://www.deja.com/
>
> Most likely you need just one. What you didn't mention is what kind of
> install tool is giving you that error. Or if it is a runtime error. If
> the tool is rpm, and you already have those glibc's, probably your rpm
> database doesn't list them. Try with --nodeps and/or --force. Other
> tools behave differently.
>
Sent via Deja.com
http://www.deja.com/
------------------------------
From: [EMAIL PROTECTED] (David Efflandt)
Subject: Re: Pass character to a program?
Date: Sat, 16 Dec 2000 22:55:20 +0000 (UTC)
Reply-To: [EMAIL PROTECTED]
On Fri, 15 Dec 2000 19:43:03 -0800, Flej Ling <[EMAIL PROTECTED]> wrote:
>I am running a program that does some stuff, then prompts "Press CTRL Z to
>exit". It works great.
>
>Now I want to run it via a bash script with the sleep command so that it
>starts up 45 seconds after boot up. How do I pass a CTRL Z to the program
>from within a script after the program finishes. It will be running
>invisibly in the background, then I want to stuff a CTRL Z into it to make
>it exit.
If you run a program in the background it has no keyboard, so you would
have to find some other means to stop it. Is there some reason you cannot
have it just do its thing and exit?
There are ways to communicate with background processes. The simplest
from a bash script might be a named pipe (fifo) that appears to be, and
can be written to like a regular file. Then you could simply echo
something to that fifo. Your program monitors the fifo and when the
string comes that you are looking for, it exits.
'perldoc perlipc' (interprocess communication) may give you some ideas.
--
David Efflandt [EMAIL PROTECTED] 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/
------------------------------
Reply-To: "Chris" <e e z e e 7 @ h o m e . c o m>
From: "Chris" <e e z e e 7 @ h o m e . c o m>
Crossposted-To:
alt.os.linux,comp.os.linux,comp.os.linux.hardware,comp.os.linux.help,comp.os.linux.questions,comp.os.linux.redhat,linux.redhat.misc
Subject: Sound Blaster not working
Date: Sat, 16 Dec 2000 22:58:08 GMT
I am running Redhat Linux 6.2, and I'm having trouble with my Sound Blaster
clone sound card. When I try to use sndconfig, I get the following error
messages (these also appear when the computer is booting up):
/lib/modules/2.2.14-5.0/misc/sb.o: init_module: Device or
resource busy
/lib/modules/2.2.14-5.0/misc/sb.o: insmod
/lib/modules/2.2.14-5.0/misc/sb.o failed
/lib/modules/2.2.14-5.0/misc/sb.o: insmod sound-slot-0 failed
I have no idea what the trouble is, and I am hoping someone here can resolve
this. One thing to note however; my sound card DID work properly running an
older version of RH Linux (I think it was RH5 or something).
=====
Chris
------------------------------
From: [EMAIL PROTECTED] (David Efflandt)
Subject: Re: pine and spam mails
Date: Sat, 16 Dec 2000 23:05:54 +0000 (UTC)
Reply-To: [EMAIL PROTECTED]
On Sat, 16 Dec 2000 16:52:42 -0500, Sudhakar R. <[EMAIL PROTECTED]> wrote:
>Is there anyway to block addresses in pine or do we have to use a mail
>filter program. If so, can you please send me directions on how to get
>started with mail filtering and blocking undesireable addresses.
>
>-sudhakar
The 'procmail' program can be automatically run from your ~/.forward as
any mail is received to do various filtering or dumping. See 'man
procmail' and related manpages procmailrc, promailsc and procmailex.
It also shows you how to set up a safety net (backup of last 32 msgs) in
case your filtering rules are not quite right. But what I need is a
filter for chinese spam which comes from a multitude of IPs with no
reverse lookup and is just gibberish.
--
David Efflandt [EMAIL PROTECTED] 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
******************************
