Linux-Misc Digest #121, Volume #28 Sat, 16 Jun 01 11:13:02 EDT
Contents:
Re: acces to log files as nonroot (David Efflandt)
Visual Network Monitoring Tools?? (Subba Rao)
Re: put a running console application into background (Helmut Haefner)
Visual Network Monitoring Tools?? (Subba Rao)
Re: bad python display (David Efflandt)
Re: just screwed mandrake 7.2...... (Steve Bradley)
Re: Help on getting disconnected problem (dial-in getty) (David Efflandt)
Re: Workaround to get HP Colorado IDE tape drives to work under RH7.1 (2.4
kernels). (Robert Davies)
env variables ("Tom Edelbrok")
Re: Kernel problems ("Tom Edelbrok")
Re: How to stop runaway print jobs? (David Efflandt)
Re: copiling tared software (John Thompson)
Re: What to use to write my thesis? (John Thompson)
Re: bluewave ("Dave Stanton")
Re: How to regenerate RPM database? (Michael Heiming)
Where can i find a linux newsgroup with the topic open source? ("sliver")
How do I get the module files for a new kernel build? (wb0gaz)
Configure Sun A1000 under Linux? (John Forsythe)
Re: Linux servers in corporate networks (David Efflandt)
Re: need module alias name (David Efflandt)
----------------------------------------------------------------------------
From: [EMAIL PROTECTED] (David Efflandt)
Subject: Re: acces to log files as nonroot
Date: Sat, 16 Jun 2001 13:10:42 +0000 (UTC)
Reply-To: [EMAIL PROTECTED]
On Fri, 15 Jun 2001, Stefan Rauch <[EMAIL PROTECTED]> wrote:
>
> Hi,
> I'm using a viewer, that shows me some logfiles (/var/log/messages) on the
> background of my desktop (like root-tail). Therefore I have to set the
> read-bit for others (instead of -rw-r----- I us -rw-r--r--). But there is a
> cronjob running on SuSE, that zips that file from time to time. After that
> cronjob, the accesrights are set to -rw-r----- again.
> Where can I change this behaviour?
SuSE (7.1 anyway) splits messages into 'messages' which are only readable
by root and 'localmessages' which are readable by anybody. Rather than
open your possibly private logs to all (security risk), edit
/etc/syslog.conf to log what you want to /var/log/localmessages and
monitor that.
Then run: /etc/init.d/syslog restart
--
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/
------------------------------
From: Subba Rao <[EMAIL PROTECTED]>
Subject: Visual Network Monitoring Tools??
Date: Sat, 16 Jun 2001 13:11:20 GMT
Are there any visual tools that will generate network maps for network admins?
Most of the tools are commercial and are specific to some platform (OpenView
from HP and NetView from IBM). I am looking for something that could be used on
Linux or OpenBSD.
Any information or pointers appreciated.
Thank you in advance.
Subba Rao
[EMAIL PROTECTED]
------------------------------
From: [EMAIL PROTECTED] (Helmut Haefner)
Subject: Re: put a running console application into background
Date: Sat, 16 Jun 2001 13:15:38 GMT
Hello baum
On Fri, 15 Jun 2001 20:49:51 +0200, "baum" <[EMAIL PROTECTED]> wrote:
>hello
>
>is there any way to turn a normal console application into a background
>application, just like it has been started with "./application&" ?
>
If bash is your shell running the application, then you can type
ctrl-z
it will pause your application, after that the command
bg
will let it work in background, the command
fg
will let it work on foreground again.
type
man bash
for exact syntax and options.
Greetings Helmut
------------------------------
From: Subba Rao <[EMAIL PROTECTED]>
Subject: Visual Network Monitoring Tools??
Date: Sat, 16 Jun 2001 13:22:21 GMT
Are there any visual tools that will generate network maps for network admins?
Most of the tools are commercial and are specific to some platform (OpenView
from HP and NetView from IBM). I am looking for something that could be used on
Linux or OpenBSD.
Any information or pointers appreciated.
Thank you in advance.
Subba Rao
[EMAIL PROTECTED]
------------------------------
From: [EMAIL PROTECTED] (David Efflandt)
Crossposted-To: comp.lang.python
Subject: Re: bad python display
Date: Sat, 16 Jun 2001 13:27:41 +0000 (UTC)
Reply-To: [EMAIL PROTECTED]
On 15 Jun 2001 17:59:36 GMT, Holland King <[EMAIL PROTECTED]> wrote:
>
> sorry for the cross-posting but i am not real sure what is causing the
> problem.
> i have a python program using the tkinter module to create a gui. my
> local system is running woody debian with kde. i am running the program
> on a soloris box. on both my system and the solaris system i am using the
> same version of ssh, python, and tkinter lib. the problem is when i try
> to export the display from the solaris box to my local machine all i get
> is the frame and half the boxes and buttons. when i run it locally i get
> the correct display and when i try to export the display on other machines
> i get the correct display. i think it has something to do with ssh, but i
> have tried everything i can think of. can anyone think of what might be
> causing the display on my machine to be unusable?
> ssh -V == 2.5.2p1
> python -v == 1.5
> tkinter -v == 1.127
> thank you.
Are you running X under ssh-agent (I have "ssh-agent startx" as an alias
to startx), then ssh-add from an xterm? That should automatically set up
your DISPLAY variable on the other end to tunnel everything back through
ssh. Not quite sure how to do that from GUI login (runlevel 5).
If not tunnelling through ssh, maybe there is a port permission problem
on either end or you may need to use xhost to allow it. I don't really
know anything about python or tkinter, but do other X programs run
remotely automatically display locally (xterm, netscape, etc.)?
--
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/
------------------------------
From: Steve Bradley <[EMAIL PROTECTED]>
Subject: Re: just screwed mandrake 7.2......
Crossposted-To: comp.os.linux.questions
Date: Sat, 16 Jun 2001 12:37:11 GMT
> Now i have redhat 7.1 in hand.. but i still wanna give mandrake another
> try..worth it?
>
> any thoughts��folks ???
>
> James
Well, on a strictly personal note, I didn't like MD..not 7.1, 7.2 or the
8.0 betas. I've heard that they're easy..but they do a lot of strange
things to make themselves that way - then when something goes wrong it's
hard to track down (JMHO).
I got rid of Mandrake and went to RH7.0 (and later online-updated myself to
7.1) and I much prefer it.
--
Steve Bradley
[EMAIL PROTECTED]
------------------------------
From: [EMAIL PROTECTED] (David Efflandt)
Subject: Re: Help on getting disconnected problem (dial-in getty)
Date: Sat, 16 Jun 2001 13:52:04 +0000 (UTC)
Reply-To: [EMAIL PROTECTED]
On Fri, 15 Jun 2001 14:49:37 -0700, Yasuki Izaki <[EMAIL PROTECTED]> wrote:
> Hello,
>
> I have a problem on dial-in port. I can log in through the port, but after a
> few seconds
> my line is dropped (disconnected). I'm using ttyS1 (COM2) on Redhat 6.1.
> Anybody knows what I should look for? Thank you in advance.
Are you logging into a shell, or trying to do AutoPPP? There should not
be any problem doing a dialup shell login. But for AutoPPP you have to
set up /etc/pap-secrets properly with 4 fields in order to use 'auth' and
'login' options for user authentication. The 4th field is allowed IP's
(or * for any).
Using the +pap option (per mgetty examples) can trip you up, because that
tries to authenticate the other machine (peer is not clearly defined).
The auth and login options take care of PAP authentication of the user.
Also for pppd temporarily enable more extensive debugging until you figure
it out (kdebug 7 and maybe debug), but check /etc/syslog.conf to see where
it would log debug messages (you might have to set something up for that
because it can fill up a file fast).
http://www.de-srv.com/linux/dialin.txt might give you some pointers, but
it was sort of old from RH 5.0(?) (I now use a modem on a Cisco router).
--
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/
------------------------------
From: Robert Davies <[EMAIL PROTECTED]>
Subject: Re: Workaround to get HP Colorado IDE tape drives to work under RH7.1 (2.4
kernels).
Crossposted-To: comp.os.linux.setup,redhat.general,comp.os.linux.hardware
Date: Sat, 16 Jun 2001 13:29:57 +0100
Denis Leroy wrote:
> Leonard Evens <[EMAIL PROTECTED]> wrote in message
> news:<[EMAIL PROTECTED]>...
>> I think that as long as I don't try to access the tape drive first
>> as /dev/ht0 (and thus load the ide-tape module), then Denis's
>> fix works fine. I think I could just put this is rc.local
>> or just load the modules when I need them to use the tape drive.
>>
>> I know several people had complained about not being able to use
>> the HP Colorado IDE tape drives under RH7.1. I've been trying
>> to track them down without success. I just came upon Denis's
>> solution by luck, and I think it should be better advertised.
>
>
> To do that, you can add the line
>
> alias ide-tape off
>
> in the /etc/modules.conf file
>
> Also, adding
>
> below st ide-scsi
>
> will force the auto-load of ide-scsi when you access /dev/st0
Now that is a nice way of handling it, but is less general than the method
I use as it will probably interfere with SCSI tape drives you have.
I like the ide-tape off, I wonder if ide-tape works for anyone? My drive
seemed to work, but positioning commands failed completely and I gave up on
it. IDE tape drives have a bad reputation, I'm wondering if it's the
driver's fault (ide-tape) because they seem fine once you've got them
operating correctly.
Rob
------------------------------
From: "Tom Edelbrok" <[EMAIL PROTECTED]>
Subject: env variables
Date: Sat, 16 Jun 2001 14:05:45 GMT
What is the normal way of making environment variables permanent? Where do
the default ones get set?
I have a way of doing it - I just put any new ones into /etc/bashrc. For
example:
DISPLAY=blade:0.0 ; export DISPLAY
This works great. Is this the proper place to do it? If so then why do a
whole bunch of environment variables show up at boot time and they are not
even found in bashrc?
Thanks,
Tom
------------------------------
From: "Tom Edelbrok" <[EMAIL PROTECTED]>
Subject: Re: Kernel problems
Date: Sat, 16 Jun 2001 14:09:33 GMT
You can also do a "make bzdisk". This will wipe and write a floppy disk from
which you can boot and test the new kernel. This doesn't solve your hard
disk space problems but it will permit you to continue testing your new
kernel while you're figuring out what to erase from your hard drive to free
up space.
Tom
mike <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]...
> I tried to install a new kernel all went well up to the point where
> you create a new bzImage. When I tried to make this image it said no
> space left on device. What can I do to fix this. any help would be
> graciously apreciated.
------------------------------
From: [EMAIL PROTECTED] (David Efflandt)
Subject: Re: How to stop runaway print jobs?
Date: Sat, 16 Jun 2001 14:13:39 +0000 (UTC)
Reply-To: [EMAIL PROTECTED]
On Fri, 15 Jun 2001, Andrew Nesbit <[EMAIL PROTECTED]> wrote:
> hi... sometimes when I'm printing big postscript files, I notice that
> after one page or so has come out, the formatting or something similar is
> wrong, so I want to cancel it so that I can fix stuff up and recompile,
> etc. The problem is that usually, by this stage, lpq shows that there
> are no entries (so I think this means that the job is further down the
> line somewhere), and so I have no way of stopping it! I am using an
> HP LaserJet III, and magicfilter which formats the output with gs.
>
> Taking the printer off line and resetting it doesn't help. Neither
> does turning it off then on again. The only way I can get things back
> to a sane state is by rebooting, because I don't want to have to waste
> 200 pages of paper! Ugh.
>
> Can somebody please tell me a way of stopping these runaway print jobs?
Does 'lpc status' show anything related to that print job? If you start
lpc on the commandline and type 'help' or '?' it gives you a list of
commands. But usually when I try to stop or abort a printer, it stops all
printing to that printer (even after reboot) and I have to experiment with
the sequence of 'start', 'up' and 'enable' commands to get it running
again. I have an hplj4l (which used hplj3 filter before it had its own).
I forget if you also have to clean any files out of the spool dir.
--
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/
------------------------------
From: John Thompson <[EMAIL PROTECTED]>
Subject: Re: copiling tared software
Date: Sat, 16 Jun 2001 07:11:35 -0500
gabriel wrote:
> when i use the commmand "./configure" i get an error "no no valid cc
> avalible in path, or something likethat.. first it checks for cc with the
> return of no and then gcc with the same return.
Do you have the gcc compiler installed? Try typing "gcc
--version" at a shell prompt. If it tells you gcc can't be
found, then install it. If it tells you "permission denied"
check the permissions on the gcc package.
--
-John ([EMAIL PROTECTED])
------------------------------
From: John Thompson <[EMAIL PROTECTED]>
Crossposted-To: comp.os.ms-windows.apps.word-proc,comp.text.tex
Subject: Re: What to use to write my thesis?
Date: Sat, 16 Jun 2001 07:14:30 -0500
Federico Bravo wrote:
> I'm using LyX and learning LatTeX and I can tell you it is definitely worth
> it! It seems more difficult than other word processing systems, but it just
> takes a little time to get into it. Moreover I did use Word and I remember
> always being angry with HIS OWN decisions about where to place stuff in the
> page!!!! Moreover I remember just loosing important files because Word
> wouldn't open them anymore. At the end I think you will spend less time
> learning and using LaTeX than trying to recover Word errors.... and the
> results are not comparable.
> If you do use Windows there's a w.p. called Scientific Word which is totally
> WYSIWYG and gives you same results as LaTeX.
Another one to try is "TeXmacs" which gives an emacs-like
front-end to TeX. It seems pretty comparable to lyx's equation
editing facilities, but (IMHO) does a better job of typesetting
text (ie, uses ligatures when appropriate, etc.).
--
-John ([EMAIL PROTECTED])
------------------------------
From: "Dave Stanton" <[EMAIL PROTECTED]>
Subject: Re: bluewave
Date: Sat, 16 Jun 2001 15:31:55 +0100
In article <8vzW6.12$[EMAIL PROTECTED]>, "faeychild"
<[EMAIL PROTECTED]> wrote:
> That was a memory trip, "Blue wave" Hell I haven't heard that for a long
> time.
> I think fido is limping along, I know the os2bbs is still up.
>
> --
> -
> -
> Regards
> RonN
Exactly what I thought Ron, thats why I asked. I always thought there was
a bit more adventuring into the unknown on BBS.
Cheers
Dave
------------------------------
From: Michael Heiming <[EMAIL PROTECTED]>
Subject: Re: How to regenerate RPM database?
Date: Sat, 16 Jun 2001 16:49:56 +0200
Jean-David Beyer wrote:
>
> Dave Uhring wrote:
> >
> > Jean-David Beyer wrote:
> >
> > > Dave Uhring wrote:
> > >>
> > >> Jean-David Beyer wrote:
> > >>
> > >> > I tried to upgrade my Linux distribution from one release to the next.
> > >> > It was so difficult to do the configuration of the new release that I
> > >> > just restored everything back to the previous release. Or so I
> > >> > thought. In any case, the RPM database is no good: it thinks I have
> > >> > the new release installed, when I do not.
> > >> >
> > >> > I tried
> > >> >
> > >> > rpm --rebuilddb
> > >> >
> > >> > and it rummages around for a little while, but the database still
> > >> > thinks I have the new release in there and I can tell that I do not.
> > >> >
> > >> > What must be done for it to REALLY rebuild the database; i.e., look at
> > >> > everything on the machine and use the right RPMs in there? Or is this
> > >> > impossible?
> > >> >
> > >>
> > >> rpm(8)
> > >>
> > >> rpm -i --rebuilddb
> > >
> > > I told you I had already done that (actually, I did it correctly).
> > > When I do it your way, I get:
> > >
> > > valinux:root[/home/jdbeyer]# rpm -i --rebuilddb
> > > rpm: only one major mode may be specified
> > > valinux:root[/home/jdbeyer]#
> > >
> > > It is not enough to copy the manual page; you must understand it.
> > >
> > > In any case, this does not work.
> > >
> >
> > Since I am using the newest version of rpm, perhaps the manpage does not
> > apply in your case.
>
> The man page I have says the same as yours, though, so I would think
> it applies:
>
> REBUILD DATABASE OPTIONS
> The general form of an rpm rebuild database command is
>
> rpm --rebuilddb
>
> To rebuild a new database, do:
>
> rpm --initdb
>
> The only options for these modes are --dbpath and --root.
>
> I took the default for --dbpath and --root. I infer that it thought it
> was working because I tried again yesterday and the files now say:
>
> valinux:jdbeyer[/var/lib/rpm]$ ls -l
> total 11038
> 16384 Jun 15 15:25 conflictsindex.rpm
> 2605056 Jun 15 15:25 fileindex.rpm
> 16384 Jun 15 15:25 groupindex.rpm
> 28672 Jun 15 15:25 nameindex.rpm
> 8960072 Jun 15 15:25 packages.rpm
> 45056 Jun 15 15:25 providesindex.rpm
> 90112 Jun 15 15:25 requiredby.rpm
> 16384 Jun 15 15:25 triggerindex.rpm
> valinux:jdbeyer[/var/lib/rpm]$
>
> > Perhaps you can execute the command
> >
> > # rpm --rebuild (options)
> >
> > where (options) are --dbpath and --root as explained in "Linux in a
> > Nutshell" from O'Reilly.
>
> I do not have that O'Reilly book. I am running:
>
> valinux:jdbeyer[/boot]$ rpm --version
> RPM version 3.0.6
> valinux:jdbeyer[/boot]$
>
> If I examine the documented options for rpm --rebuild, I see only:
>
> rpm --rebuild <source_package_file>+
>
> When invoked this way, rpm installs the named source pack-
> age, and does a prep, compile and install. In addition,
> --rebuild builds a new binary package. When the build has
> completed, the build directory is removed (as in --clean)
> and the the sources and spec file for the package are
> removed.
>
> I hesitate to try it with the --dbpath and --root options, since I do
> not know what they would do.
>
> --
> .~. Jean-David Beyer Registered Linux User 85642.
> /V\ Registered Machine 73926.
> /( )\ Shrewsbury, New Jersey http://counter.li.org
> ^^-^^ 8:20am up 8 days, 21:15, 4 users, load average: 3.01, 3.05, 2.92
Hello Jean-David,
I'm not sure, my man page (RPM version 3.0.4 - SuSE 7.0) mentions this:
REBUILD DATABASE OPTIONS
The general form of an rpm rebuild database command is
rpm --rebuilddb
To rebuild a new database, do:
rpm --initdb
The only options for these modes are --dbpath and --root.
[...]
But it isn't explained what actually the difference between those
two options are, but backing up /var/lib/rpm/* and trying, shouldn't
be a big risk...:-)
Good luck
Michael Heiming
------------------------------
From: "sliver" <[EMAIL PROTECTED]>
Subject: Where can i find a linux newsgroup with the topic open source?
Date: Sat, 16 Jun 2001 16:48:03 +0200
I want to join a newsgroup that discuss the advantages and dissadvantages of
open source an linux.
Does anyone know such a newsgroup?
------------------------------
From: [EMAIL PROTECTED] (wb0gaz)
Subject: How do I get the module files for a new kernel build?
Date: 16 Jun 2001 07:54:01 -0700
I'm trying to move from kernel 2.2.14 to 2.2.18 on a redhat 6.2
system (necessary to get the combination of peripherals I want and
the version of Xfree I require all to work together.)
I downloaded the 2.2.18 kernel .gz and modutils 2.2.x from ftp.kernel.org
and was able to compile and generate a kernel that ran, however,
PCMCIA (and other stuff) does not start because there are no
corresponding module files in the directory (specific to the
kernel) that contain the modules files and modules.dep file.
The 2.2.14 version module files do not work with 2.2.18 (I would
certainly expect this). I answered the configuration questions
posed by "make config" as best I could but there were many
questions and perhaps I did something wrong that prevented the
module files from appearing in the required directory...
As I've never tried this (kernel change) before I'm a bit stuck,
and a pointer (to more detailed help) or advice would sure help.
Tnx,
Dave [EMAIL PROTECTED]
------------------------------
From: [EMAIL PROTECTED] (John Forsythe)
Crossposted-To: comp.os.linux.hardware
Subject: Configure Sun A1000 under Linux?
Date: 16 Jun 2001 07:56:59 -0700
We have a Sun A1000 (hardware RAID) that we have to hook
up to a Linux box. Is there a way to configure/manage it
under Linux? It does have a serial port at the back,
but the manual says it is for "factory/future use".
Any ideas, help would be appreciated.....
Thanks!
------------------------------
From: [EMAIL PROTECTED] (David Efflandt)
Subject: Re: Linux servers in corporate networks
Date: Sat, 16 Jun 2001 14:58:31 +0000 (UTC)
Reply-To: [EMAIL PROTECTED]
On 15 Jun 2001 07:34:57 -0700, nag rao <[EMAIL PROTECTED]> wrote:
> We hear about Linux replacing a lot of UNIX networks in Corporate
> networks. Does anyone have practical experiences on this? In such an
> environment, how are user logins handled to multiple machines? Is
> NIS/NFS the solution Network administrators prefer?
>
> Want this info to propose to our IS dept.
Our company is relatively small (< 500 employees). And we just use Linux
where it excels, as an inexpensive internet firewall (adsl) for smtp, pop3
and squid proxy on an old P166 box. Our factory uses Netware, but our
branch office connected by frame relay does not, so we can hook any box or
modem to the WAN without being a security risk. There is no internal DNS,
so you have to know the IP of our mail/proxy box or corporate computer
(HP3000 which requires special software and triple password).
Maybe I should consider myself fortunate that our MIS dept does not have
any control over my computer. So we are still up and running when the
Netware server crashes.
--
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/
------------------------------
From: [EMAIL PROTECTED] (David Efflandt)
Subject: Re: need module alias name
Date: Sat, 16 Jun 2001 15:08:18 +0000 (UTC)
Reply-To: [EMAIL PROTECTED]
On Thu, 14 Jun 2001 23:59:17 -0600, D. Stimits <[EMAIL PROTECTED]> wrote:
> I have a 2.4.x kernel, with the following modules for a joystick:
> joydev
> gameport
> ns558
> analog
>
> I've added the following to /etc/modules.conf so I can simply run
> "modprobe analog" and ahve it install all related modules:
> pre-install analog /sbin/modprobe ns558
> pre-install ns558 /sbin/modprobe gameport
> pre-install gameport /sbin/modprobe joydev
>
> What I'm having a problem with, despite having the kerneld running for
> module autoloading, is that I must still manually modprobe analog for it
> to work. If /dev/js0 (a sym link to /dev/input/js0) is accessed, is
> there an alias I can add to modules.conf which will allow it to
> autoload? I don't see any messages in /var/log/messages, so I have no
> clue what it wants...is there an alias or other name I can alias to
> "analog" to get it to autoload?
If you want it to autoload when something tries to access js0 maybe:
alias js0 analog
or the preferred method might be to consult
/usr/src/linux/Documentation/devices.txt and use the major:
alias char-major-15 analog
--
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
******************************
