Linux-Misc Digest #983, Volume #26 Wed, 31 Jan 01 14:13:04 EST
Contents:
Re: nameserver not available, Linux network goes down ("Tauno Voipio")
Re: Command-Line Editing Issues (Villy Kruse)
mandrake vs redhat? (Peter Bismuti)
Re: Networking strange problems! ("Natman")
filesystem space disappeared ([EMAIL PROTECTED])
Re: what is ld-linux.so (Jim Dennis)
Re: best internal modem? (Steve Ackman)
Re: Berkeley "r" commands (Nick Condon)
How to add new item to ntsysv? (Carfield Yim)
ftpaccess file question ("Charles")
Re: I wish to RTFM, but where is TFM I need? (Kenneth Mair)
ocr linux scanner: ocr support for linux scanners? (Robert Nagle)
setting the editor in pine?! (Peter Bismuti)
Hello, asking for help (tony)
Re: Kernel recompilation - "system too big?" (Steve Holdoway)
Re: Better wait for Linux Kernel 2.4.x (Steve Holdoway)
----------------------------------------------------------------------------
From: "Tauno Voipio" <[EMAIL PROTECTED]>
Subject: Re: nameserver not available, Linux network goes down
Date: Wed, 31 Jan 2001 17:31:39 GMT
"Dustin Puryear" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]...
> On Mon, 29 Jan 2001 20:26:19 +0100, Peter T. Breuer <[EMAIL PROTECTED]>
wrote:
> >John Hasler <[EMAIL PROTECTED]> wrote:
> >> Peter writes:
> >>> Correct. Didn't they have their own hostname in their /etc/host
> >tables?
> >
> >> I believe that he should be able to ping the IP of any interface that
> >is
> >> configured up no matter what is in /etc/hosts. IIRC the kernel
> >> short-circuits such packets.
> >
> >That's correct. But maybe his path includes an nfs mounted directory,
> >mounted hard. Or maybe he tried it from within X with the display
> >variable set to me.my.mine:0.
>
> No on all of these.
>
> >I.e. I am looking for things that would prevent him seeing what is
> >happening by locking most of his machine up.
>
> Well, if I have time and I get the greenlight I am going to try and
reproduce
> this error. I'll then come back with more information.
>
> Anyway, any more ideas, clues, or gotchas?
>
If you have a spare machine (or even a spare window to run it on) run
tcpdump and log the results.
Tauno Voipio
tauno voipio @ iki fi
------------------------------
From: [EMAIL PROTECTED] (Villy Kruse)
Subject: Re: Command-Line Editing Issues
Date: 31 Jan 2001 17:33:46 GMT
On Wed, 31 Jan 2001 15:25:07 GMT, Bob Simon <[EMAIL PROTECTED]> wrote:
>> I wrestled with this problem myself when i installed rh6.2 a couple
>> of months ago. The way i fixed it was by creating a file in my home
>> dir called ".inputrc", containing the following line:
>> set editing-mode vi
>
>Thanks for the tips, Chris. However, set editing-mode vi did
>not work for me (with bash 2.04.11). For one thing, echo
>$SHELLOPTS shows that I'm still in emacs commandline editing
>mode. Additionally, I should be able to toggle from insert
>mode to command mode with esc and have subsequent input taken
>as commands. For example, if I want to move my cursor three
>words to the left, I should be able to type: esc bbb
>
You probably need to unset INPUTRC before ~/.inputrc takes precedense
over /etc/inputrc. Depending on your situation you may also choose to
edit /etc/inputrc.
Villy
------------------------------
From: [EMAIL PROTECTED] (Peter Bismuti)
Subject: mandrake vs redhat?
Date: 31 Jan 2001 17:33:16 GMT
At my company there has been a big wave of defections from redhat to
mandrake, which is supposed to be "redhat compliant". One of the most
attractive things is that mandrake has built in IBm viavoice voice
recgonition.
My concern is that the support for Mandrake is not as good as with Redhat
as far as mailing lists, updates on their webpage, etc, etc.
Are there any opinions? Are there any compelling reasons why I should NOT
switch to Mandrake and stay with (and tolerate) Redhat7.0?
Thanks
------------------------------
From: "Natman" <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.redhat,linux.redhat.misc
Subject: Re: Networking strange problems!
Date: Wed, 31 Jan 2001 17:48:03 GMT
Personally, I know my Motorola cable modem doesn't do this. My cable modem
will talk to (supposedly) 5 ethernet cards at once. I've been able to
switch between 2 machines for testing and such. Only noticable "feature" is
that it seems to assign the IP address based on MAC address (so I start with
an IP for machine A, get a diff IP for machine B, but when I switch back to
machine A, i get the first IP back). I havn't tried with more than 2
machines.
To the origional poster:
Why are you using a hub in the first place? Try without one (and a
crossover cable, of course). I have a feeling this a hardware problem with
your NIC. Try replacing it. If itz ISA, get a PCI one if you can. If itz
PCI, try moving it away from all of the other cards (so it has a space on
each side). Sounds like linux looses the IRQ, etc. If you are really
desperate, try running a packet sniffer in linux, try enable debugging in
the kernel, check to see (before and after the NIC is out) if the proper
modules are still loaded... perhaps RH "accidentally" removes the module
while itz doing itz cleaning. (Yes, every so often a cron job runs that
removes un-needed modules). There's so much you can still do. Please post
back with results of this testing. Also, look through logs and send
anything that might be out of order. IF ALL ELSE FAILS --- my personal
solution to any unsolvable problem --- reinstall Linux, trying to just
"upgrade". Don't actually format your hard drive, just reinstall over the
existing stuff.
Natman
<[EMAIL PROTECTED]> wrote in message
news:gD6d6.22647$[EMAIL PROTECTED]...
> somebody correct me on this if I am wrong but there is a
problem(feature? )
> with some cable modems that they grab the first mac address that they see.
> and will only "talk" to the first mac address that they see. so if you are
> going between two machines for these test/s try and reset the modem
between
> them so the modem will "relearn" the correct mac address.
>
>
> Brian
>
> "Rick Devey" <[EMAIL PROTECTED]> wrote in message
> news:[EMAIL PROTECTED]...
> > I have had the same problem with RedHat7.0. What I have
> > done is just to reconfigure the network interface.
> >
> > ay it is eth0 I have just done
> >
> > ifconfig eth0 192.168.1.x up netmask 255.255.255.0
> >
> > and it seemed to work without rebooting for me.
> >
> > Rick
> >
> > Londonboy wrote:
> > >
> > > I am running Linux Redhat with 1 NIC with default setup, connecting
the
> > > RedHat to a hub, and the hub connects to my cable modem. This is my
> problem:
> > >
> > > 1. Internet connection works fine, the RJ45 cable connects to the hub,
> the
> > > light is "on" on the hub
> > > 2. Unplug the RJ45 cable which connects the Redhat server to hub
> > > 3. Wait for 1-2 hours
> > > 4. Plug the RJ-45 cable back in
> > > 5. The light goes out on the hub!!!??? and I don't have internet
> connection
> > > anymore
> > > 6. To resolve this, I need to do /etc/init.d/network restart (or
> reload)
> > >
> > > I tried it on Redhat 6.1 and 7.0, they have the same problem. Also, it
> > > doesn't matter if the NIC is running DHCP client or Static IP (using
> > > 192.168.0.x), I tried both, have the same problem too.
> > >
> > > Any idea? and how can I fix it? It doesn't happen to Microsoft windows
> NT
> > > Server /9x.
> > >
> > > Please help. help.....
> > >
> > > N.B.
>
>
------------------------------
From: [EMAIL PROTECTED]
Crossposted-To: linux.redhat.misc
Subject: filesystem space disappeared
Date: Wed, 31 Jan 2001 17:40:40 GMT
Hi,
I have installed Redhat 6.1 on two machine with the exact
configuration. There is one filesystem showing 95% full and the other
97%. Both were 95% full in the beginning. One of the increase slowly
from 95% to 96% and then to 97%. This filesystem contains sybase
database files only. All these files have pre-allocated size and will
not extend on its own. I have also checked the files and sizes, they
are exactly the same on both system. The only difference I can see so
far is when I use the "-s" option in "ls -als" command. Here is the
output ...
system 1
========
datamill@ginger:/sybdevices/sysdev> ls -als
total 340536
4 drwxrwxr-x 2 sybase sybase 4096 Feb 2 2000 ./
4 drwxr-xr-x 5 sybase sybase 4096 Feb 2 2000 ../
8796 -rw-rw-r-- 1 sybase sybase 31457280 Jan 26 15:02 master.dat
61504 -rw-rw-r-- 1 sybase sybase 62914560 Jan 25 20:22
systemprocs.dat
270228 -rw-rw-r-- 1 sybase sybase 524288000 Jan 26 15:07
tempdb_01.dat
system 2
========
datamill@maryanne:/sybdevices/sysdev> ls -als
total 125612
4 drwxrwxr-x 2 sybase sybase 4096 Feb 14 2000 ./
4 drwxr-xr-x 5 sybase sybase 4096 Feb 14 2000 ../
8792 -rw-rw-r-- 1 sybase sybase 31457280 Jan 26 15:05 master.dat
61504 -rw-rw-r-- 1 sybase sybase 62914560 Dec 16 10:00
systemprocs.dat
55308 -rw-rw-r-- 1 sybase sybase 524288000 Jan 26 15:03
tempdb_01.dat
system 1 is the one with 97% full. What's the "-s" option show? In
both case, the first column (in Kbytes) is smaller than the 6th column
(in bytes) for the file tempdb_01.dat.
I have run fsck numerous times and it didn't help.
Any help will be appreciated. And please send responses directly to my
email account [EMAIL PROTECTED]
Thanks,
Banquo.
Sent via Deja.com
http://www.deja.com/
------------------------------
From: [EMAIL PROTECTED] (Jim Dennis)
Subject: Re: what is ld-linux.so
Date: 31 Jan 2001 17:40:52 GMT
Reply-To: [EMAIL PROTECTED]
In article <[EMAIL PROTECTED]>, all wrote:
>i'm not sure is it ld-linux.so or id-linux.so
>it took 92% of my cpu usage while i'm online.
>help plzzz
ld-linux.so is the main Linux shared library loader.
All normal statically linked programs are linked against
ld-linux.so which contains the code to resolve, locate and
manage loading
On Debian ld-linux.so has its own man page:
``
ld.so(8) ld.so(8)
NAME
ld.so/ld-linux.so - dynamic linker/loader
DESCRIPTION
ld.so loads the shared libraries needed by a program, pre
pares the program to run, and then runs it. Unless
explicitly specified via the -static option to ld during
compilation, all Linux programs are incomplete and require
further linking at run time.
''
That goes on to explain how you can control Linux' dynamic
loading using LD_* environment variables and how /etc/ld.so.cache
is used by the loader/linker. You should also read about the
ldconfig command (which is used to read the /etc/ld.so.conf and
to create the /etc/ld.so.cache --- which ensures that the system
can find all the installed/configured shared libraries).
As far seeing an unusual load from ld-linux.so... I'd be very
suspicious of this activity. I would *NEVER* expect ld-linux.so
to be executing as a separate binary (so no normal process should
have that for its name nor as it's executable image).
It's possible that you've been hacked. Some cracker may have
replaced your ld-linux.so with one that performs some clandestine
function (portscanning for more targets, or acting as an agent
in DDoS --- distributed denial of service attacks).
If I were you, I'd restart the system from a rescue diskette
(write-protected) or a bootable CD (Tom's Root/Boot and the
Linuxcare bootable business card, BBC, are good choices).
Do a backup of your whole system. Do a second, separate,
backup of just your data and /etc files. If you use
cpio, dump or some other backup program, then you should
also create a tar file of the root and /usr filesystems.
Now do a re-install of Linux; from the root directory of
the new system run tar dzf specifying the file or
media (/dev/st0 for SCSI tape) on which you stored your
system backup. tar's "d" (--diff) directive will report
on differences between your installation (extent files) and
the corresponding files from your backup. Obviously some
of those differences will be accounted for by changes
you made to your system (particularly among your /etc and
other config files). Some might be a result of various
RPM or other packages that you installed or upgraded after
your previous ISL (initial system load).
However, you may see a pattern of changes to files like
/bin/login, /sbin/inetd, /bin/ps, /bin/ls, etc. That's
characteristic of a "rootkit." Obviously you'll be
particularly interested in whether files in your /lib/
and /usr/lib (including ld-linux.so) have been modified.
I should mention that your conclusion that ld-linux.so
is running might be based on erroneous data. Programs
can modify their own command tail (argument list) including
arg[0] (the "program name" as displayed in ps, top, et al).
You can look under /proc and track down the exe symlink
for your (suspect) process to determine which executable the
kernel actually loaded (exec()'d) for that process. On
older kernels this would be a device/inode pointer; but with
2.2 and later it should normally be a symlink to the executable's
filename (due to the kernel's "dcache" directory entry caching
subsystem).
Of course that assumes that your kernel is trustworthy. If you think
you've been hacked, don't trust your kernel. There was a
Linux kernel module called linspy.o that demonstrated ways that a
loadable kernel module could "hide" itself and a list of
processes and files. Similar demonstrations have been made
by direct modification of /dev/kmem or /proc/kmem, on kernels
that didn't have loadable module support enabled; and similar
exploits are possible (some have been demonstrated) under
FreeBSD and other UNIX systems.
If you've been cracked then your only safe recourse is to
reboot from a "known clean" medium and perform a full
re-installation (or at least a full system audit against
a *known clean* backup).
------------------------------
From: [EMAIL PROTECTED] (Steve Ackman)
Subject: Re: best internal modem?
Reply-To: [EMAIL PROTECTED]
Date: Wed, 31 Jan 2001 12:53:46 -0500
On Wed, 31 Jan 2001 08:01:49 -0800, <[EMAIL PROTECTED]> wrote:
>
>thanks, but to be more specific, what is the most compatible internal
>*PCI* modem? My Dell 4100 only has PCI slots.
Any modem that has a *16550 UART* is a hardware modem. They
are relatively rare compared to the plethora of winmodems, but
all you have to do is look for that key term on the box, or in
the specs.
--
Steve Ackman
http://twovoyagers.com
Registered Linux User #79430
------------------------------
From: [EMAIL PROTECTED] (Nick Condon)
Subject: Re: Berkeley "r" commands
Date: 31 Jan 2001 17:55:14 GMT
[EMAIL PROTECTED] (Tom Edelbrok) wrote in
<ozUd6.61622$[EMAIL PROTECTED]>:
>How do you disable access to Berkeley "r" commands. I presume these are
>things like rlogin, rsh, etc. I see that they are located in /usr/bin,
>so I could wipe them out or lock them up. But is there a global setting
>that says "Trash all Berkeley 'r' commands until indicated otherwise"?
Have a look at /etc/inetd.conf, you've probably got a section that looks
something like this (actually from a Solaris box):
#
# Shell, login, exec, comsat and talk are BSD protocols.
#
shell stream tcp nowait root /usr/sbin/in.rshd in.rshd
login stream tcp nowait root /usr/sbin/in.rlogind in.rlogind
exec stream tcp nowait root /usr/sbin/in.rexecd in.rexecd
comsat dgram udp wait root /usr/sbin/in.comsat in.comsat
talk dgram udp wait root /usr/sbin/in.talkd in.talkd
Comment out the ones you wish to disable and send a kill -HUP to inetd.
------------------------------
From: Carfield Yim <[EMAIL PROTECTED]>
Subject: How to add new item to ntsysv?
Date: Thu, 1 Feb 2001 02:03:12 +0800
How to add new item to ntsysv to add new boot up job?
------------------------------
From: "Charles" <[EMAIL PROTECTED]>
Subject: ftpaccess file question
Date: Wed, 31 Jan 2001 11:10:09 -0600
I have set up an FTP user on my Red Hat 7 machine. The user can FTP but the
settings in the ftpaccess for are not being applied. How do I get the users
FTP login to apply the rules in the ftpaccess file?
I am using the same setup on my Slackware machine and it works properly.
Thanks in advance,
Charles
------------------------------
From: [EMAIL PROTECTED] (Kenneth Mair)
Subject: Re: I wish to RTFM, but where is TFM I need?
Date: Wed, 31 Jan 2001 11:17:12 +0000
On Tue, 30 Jan 2001 14:51:49 -0500, Jean-David Beyer <[EMAIL PROTECTED]> wrote:
<snippetty snip snip>
>PEERDNS is not in man pppd.
>From man pppd :
usepeerdns
Ask the peer for up to 2 DNS server addresses. The
addresses supplied by the peer (if any) are passed
to the /etc/ppp/ip-up script in the environment
variables DNS1 and DNS2. In addition, pppd will
create an /etc/ppp/resolv.conf file containing one
or two nameserver lines with the address(es)
supplied by the peer.
--
Kenneth Mair | "There's no justice,
[EMAIL PROTECTED] | there's just us."
ICQ 21702236 |
------------------------------
From: Robert Nagle <[EMAIL PROTECTED]>
Subject: ocr linux scanner: ocr support for linux scanners?
Date: Wed, 31 Jan 2001 18:01:24 GMT
I'm looking for a scanner/OCR combination for linux.
I don't want to spend a bundle.
http://www.buzzard.org.uk/jonathan/scanners-usb.html
I know that vividata ocr shop can be used with linux.
http://www.vividata.com/scanshop.html
But the scanners they support seem only to be high end, and most of
them I can't even find prices for. (maybe they are out of date).
Does anyone have experiences with running ocr shop on linux? Are there
any open source ocr attempts? rj
--
Robert Nagle ([EMAIL PROTECTED])
Technical Writer
Austin, TX 78758
www.crosswinds.net/~rj2nagle/rjlinks/
"Art Anticipates Life"
Sent via Deja.com
http://www.deja.com/
------------------------------
From: [EMAIL PROTECTED] (Peter Bismuti)
Subject: setting the editor in pine?!
Date: 31 Jan 2001 18:19:45 GMT
I use elm because it is Vi-esque. Pine seems to be more popular but it is
Pico-esque. Is it possible to run Vi(m) with Pine?!
thanks
------------------------------
From: tony <[EMAIL PROTECTED]>
Subject: Hello, asking for help
Date: Wed, 31 Jan 2001 18:30:09 -0000
Hi I just want it to know, which programming language can be used to
create an operating system such as lynex, microsoft ect.
--
Posted via CNET Help.com
http://www.help.com/
------------------------------
From: Steve Holdoway <[EMAIL PROTECTED]>
Subject: Re: Kernel recompilation - "system too big?"
Date: Wed, 31 Jan 2001 19:31:07 +0100
Try make bzImage. Works for me.
Stsve
On Wed, 31 Jan 2001 09:20:52 -0500, Stefan Viljoen
<[EMAIL PROTECTED]> wrote:
>"Jens K. Thomsen" wrote:
>>
>> Hi
>>
>> Try "make bzlilo" instead.
>
>I did try this - still didn't work. Only solution was to drop some
>functionality from the kernel to get it to be smaller. Does work now -
>the rest of the stuff seems to be balanced with dynamically loaded
>modules.
>
>Thanks anyway!
------------------------------
From: Steve Holdoway <[EMAIL PROTECTED]>
Subject: Re: Better wait for Linux Kernel 2.4.x
Date: Wed, 31 Jan 2001 19:33:58 +0100
I've upgraded RedHat 6.2 to 2.4.0, but without glibc 2.2. It's running
fine, supporting Oracle 8.1.7, lvm and IBM jfs on an SMP platform.
So, what's your problem?
Steve
On Wed, 31 Jan 2001 04:18:39 GMT, Arctic Storm
<[EMAIL PROTECTED]> wrote:
>Anyone else tired of reading about complaints about Linux Kernel 2.4?!
>I was seriously thinking of upgrading to Linux Kernel 2.4, but I'm not sure
>anymore. Maybe waiting for 2.4.1, or 2.4.2, or 2.4.x maybe be prudent,
>considering the problems many people are having with 2.4.
------------------------------
** 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
******************************
