Linux-Misc Digest #188, Volume #26 Mon, 30 Oct 00 19:13:04 EST
Contents:
Re: X consumes constant 98% CPU usage (Jean-David Beyer)
Re: How to find empty directories? ("John W. Krahn")
Possible to route between two virtual interfaces? (Andrew J. Perrin)
Re: SU Problem...Any ideas? (Andreas =?iso-8859-1?Q?K=E4h=E4ri?=)
Re: problem with rpm installs (Steve)
new york linuworld expo free/cheap!? ([EMAIL PROTECTED])
Re: gcc refuses cpp files; Library error ?? ("Chris Field")
Re: how to start tn5250 (Rick)
Re: decent synaptics driver? ("Cameron Jay Erens")
Re: Redhat, Slackware, SUSE, FreeBSD, Help... (Stanislaw Flatto)
Removing drive from RAID1 ("Steve Wolfe")
Re: Something wrong with my terminal (cursor) (Juergen Heinzl)
Re: daylight savings. (Juergen Heinzl)
Re: Stock Red Hat 7 kernel compile fails. (repost) (Chris Pelton)
Linux Dual-Server Fault-Tolerant Setup ("Scott M. Navarre")
Re: after recompiling kernel boot errors ([EMAIL PROTECTED])
How to use the floppy drive in Linux ([EMAIL PROTECTED])
mount swap in /tmp (* Tong *)
----------------------------------------------------------------------------
From: Jean-David Beyer <[EMAIL PROTECTED]>
Subject: Re: X consumes constant 98% CPU usage
Date: Mon, 30 Oct 2000 17:16:46 -0500
"J.Smith" wrote:
> Hi there.
>
> I just downloaded and installed XFree86 4.01 from sources. However, X is
> very slow. Since I have a Athlon 600, 128 MB, and a Matrox Millenium G400
> 32Mb, I decided it couldnt be slow hardware, and I ran 'top'. 'top' shows X
> as using a constant 90-99% CPU usage. Even if I replace my window manager by
> twm, X still uses 90-99% CPU, so its not my window manager. Now I have been
> running X on my system before (and other OS'es), and that ran fine. So I
> guess its not faulty hardware either. Only if I switch from X to a virtual
> console using ALT-Fxx, does the cpu usage of X drop down again to (almost)
> zero. And then as soon as I switch back to my X session, it goes all the way
> back up again. I am running the mga driver that comes with XFree86 401 and
> not the driver that Matrox supplies themselves for X.
>
> Does anyone know what could be going on here? I have no idea on where to
> start looking, or how to even start to troubleshoot this situation. If
> someone could give me a few pointers here, that would be greatly
> appreciated. Thanks.
I was wondering about this on this board last week, I think it was. I assume it
is a bug in X somewhere.
I have dual 550MHz Pentium IIIs, so I cannot get it up to 100% CPU useage, but
it can go up to about 30%, which is silly if there is just top, xosview, and an
xterm (doing nothing) running using the display. In the background, I have a
whole lot of stuff running. X normally runs at about 6% or less (right now it
is at 1.5%). When it pops up to 30%, it suffices to just type a newline into an
xterm and it will drop back down for a while. If I run a job that hogs a CPU
(i.e., 100%), and that CPU is #0, X behaves itself, but if it is running on
CPU #1 with just setiathome running on CPU #0, X gets greedy.
I have no explanation other than my theory that there is a bug in X.
I have a Matrox G200 AGP (or whatever than fancy slot is). X points to
/usr/X11R6/bin/XF86_SVGA, whatever that proves.
--
.~. Jean-David Beyer Registered Linux User 85642.
/V\ Registered Machine 73926.
/( )\ Shrewsbury, New Jersey
^^-^^ 5:10pm up 6 days, 5:29, 2 users, load average: 2.25, 2.16, 2.10
------------------------------
From: "John W. Krahn" <[EMAIL PROTECTED]>
Subject: Re: How to find empty directories?
Date: Mon, 30 Oct 2000 14:18:19 -0800
Retro Grouch wrote:
>
> I have an automated file cleaner that deletes files older than 3 days in
> my network scratch directory. I'm using find -type f -mtime +3 to find
> the files; this works great.
>
> My problem is that directories get left behind. I am having a problem
> finding empty directories.
>
> Any ideas? I have not found any magical combination of commands that
> would produce a list of empty directories.
Delete the directory with rmdir, if the directory is empty it will be
deleted, if it is not empty then rmdir will return an error message and
not delete it.
John
------------------------------
From: [EMAIL PROTECTED] (Andrew J. Perrin)
Subject: Possible to route between two virtual interfaces?
Date: 30 Oct 2000 16:52:42 -0500
Greetings. I'm considering getting a DSL connection in my home, and
would like to use an old Sun IPC running Debian as a router to the
network of the rest of the machines in my house. My only concern is
that the IPC has only one ethernet interface, and since it's an SBus
box it's prohibitively expensive to add a second. I believe, however,
that it's possible to define two virtual interfaces (something like
eth0:0 and eth0:1, I think....?). Am I right about this and, if so,
can I route between them by having only one routable to the outside
world?
Thanks in advance.
--
======================================================================
Andrew J Perrin - Ph.D. Candidate, UC Berkeley, Dept. of Sociology
Chapel Hill, North Carolina, USA - http://demog.berkeley.edu/~aperrin
[EMAIL PROTECTED] - [EMAIL PROTECTED]
------------------------------
From: [EMAIL PROTECTED] (Andreas =?iso-8859-1?Q?K=E4h=E4ri?=)
Subject: Re: SU Problem...Any ideas?
Date: 30 Oct 2000 23:20:11 +0100
In article <8tklnk$54u$[EMAIL PROTECTED]>, [EMAIL PROTECTED] wrote:
>No, it doesn't work even over SSH. This is very frustrating.
>Any other ideas?
>
What did you change between "it used to work" and "it doesn't
anymore"?
>ps. What is wrong with SSHing with Root, doesn't the password get
>encrytped before sending?
Yes, it is. But don't you think it's safer to disallow root to login
via ssh and in that way force the occasional cracker to have to
guess/crack two passwords instead of just one?
You shouldn't feel "safe" just because you use ssh.
/A
>
>Thanks for your help!
>Aaron.
>
>
>
>
>In article <[EMAIL PROTECTED]>,
> [EMAIL PROTECTED] (Andreas =?iso-8859-1?Q?K=E4h=E4ri?=)
>wrote:
>> In article <8tkdoh$tef$[EMAIL PROTECTED]>, [EMAIL PROTECTED]
>wrote:
>> >I used to be able to telnet into my box as a regular user and "su -"
>> >for root privileges. However, now I can not...it says the Password is
>> >incorrect...
>>
>> That's probably a good thing. Telnet is too insecure to do root stuff
>> over.
>>
>> >but I know the password for root and I can login as root
>> >via SSH.
>>
>> Hmmm, that's not at all good. One usually does not allow root to
>> login directly from any connection (instead, use ssh to an ordinary
>> user account and then 'su -' from there).
>>
>> >
>> >Can anyone give me an idea how I can enable the su to work as it did
>> >before?
>> >
>>
>> Does 'su -' work when connecting with ssh?
>>
>> /A
>>
>> --
>> Andreas K�h�ri, Uppsala University, Sweden (until 1:st of Dec. 2000)
>> ----------------------------------------------------------------------
>--
>> "If you leave now, you're going to miss the real experience."
>> -- Richard M. Stallman, Stockholm 1986. Visit www.gnu.org
>>
>
>
>Sent via Deja.com http://www.deja.com/
>Before you buy.
--
Andreas K�h�ri, Uppsala University, Sweden (until 1:st of Dec. 2000)
========================================================================
"If you leave now, you're going to miss the real experience."
-- Richard M. Stallman, Stockholm 1986. Visit www.gnu.org
------------------------------
From: [EMAIL PROTECTED] (Steve)
Crossposted-To: alt.os.linux.mandrake
Subject: Re: problem with rpm installs
Date: 30 Oct 2000 22:27:23 GMT
Reply-To: [EMAIL PROTECTED]
On Sun, 29 Oct 2000 13:37:14 -0500, Rick wrote:
>Steve wrote:
>>
>> On Sat, 28 Oct 2000 18:29:58 -0400, Rick wrote:
>> >
>> >Could you please explain the last part.... dir rpm --frephen rpm.n.n.rpm
>> >
>>
>> Oops, some bad typing there sorry. I did:
>>
>> # rpm --freshen rpm.n.n.n.rpm
>>
>> as root where "n" just represents the vearious version numbers.
>>
>
>I tried it and got the same error.
Ok, I originally had the problem when I was trying to use some RPMs
that had come from RH7.0, and I'm only on RH6.2, so I got the updated
rpm package for RH6.2 and did
# rpm --freshen rpm.the_new_rpm_version_number.rpm
where the updated rpm file of the rpm package is in the
current directory.
Then I went and tried to upgrade the bits and pieces that I needed from
RH7.0 and it worked fine.
PS: I did the same with gnomerpm (or whatever it's called) too.
--
Cheers
Steve email mailto:[EMAIL PROTECTED]
%HAV-A-NICEDAY Error not enough coffee 0 pps.
web http://www.zeropps.uklinux.net/
or http://start.at/zero-pps
9:13pm up 19 days, 23:34, 2 users, load average: 1.00, 1.02, 1.00
------------------------------
From: [EMAIL PROTECTED]
Subject: new york linuworld expo free/cheap!?
Date: Mon, 30 Oct 2000 22:25:11 GMT
The registration page for the Linuxworld conference in New York City
has the discount codes hard coded into the javascript source code (at
the very bottom). I guess the focus is not on security this year!
http://register.rcsreg.com/regos-1.0/linuxny2001/adv/ga/live/mail.html
Sent via Deja.com http://www.deja.com/
Before you buy.
------------------------------
From: "Chris Field" <[EMAIL PROTECTED]>
Subject: Re: gcc refuses cpp files; Library error ??
Date: Sat, 28 Oct 2000 21:11:04 -0400
try g++ filename.cpp -o whatever.x
"Christian Schubert" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]...
> The system runs on Suse 6.2 Kernel 2.2.10
> My aim is to do some progamming under KDE.
> This was the first time to utilize the compiler with my own code.
> Compiling a new Kernel works o.k.
> The compiler accepts files with a .c suffix e.g. hello.c but then
> interprets them as C-code.
>
> In the man-files cpp suffix should be accepted by the compiler but the
> compiler reacts as you can see
> the lines below.
>
>
> gcc/home/..../hello.cpp -o hello -I$QTDIR/include
> gcc/home/..../hello.cpp file not recognized: File format not recognized
>
> Getting the version of the compiler causes the following lines:
>
> gcc -v reading specs from /usr/lib/gcc-lib/i486-linux/2.7.2.3/specs
> gcc version 2.7.2.3
>
>
> What do I do wrong ? Is there some confusion in the C-Libraries ? I did
> already changes from egcs to gcc and vice versa. Nothing helps. Didn't
> expect to have such difficulties.
>
>
> Thanks for help in advance
>
> Christian
------------------------------
From: Rick <[EMAIL PROTECTED]>
Crossposted-To: alt.os.linux.mandrake,comp.os.linux.portable
Subject: Re: how to start tn5250
Date: Mon, 30 Oct 2000 17:56:08 -0500
Harold Stevens ** PLEASE SEE SIG ** wrote:
>
> Searched http://google.com/linux with "tn5250" and it returned a FAQ as the
> 2nd hit:
>
> http://perso.libertysurf.fr/plinux/tn5250-faq.html
>
> Not sure if this gets you what you need, but it should be a start.
>
No, it doesnt tell me how to get started. Thanks anyway.
--
Rick
* To email me remove NOSPAM from my address *
------------------------------
From: "Cameron Jay Erens" <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.portable,alt.os.linux.mandrake
Subject: Re: decent synaptics driver?
Date: Mon, 30 Oct 2000 08:34:46 -0500
The syaptics.com website is down...what's wrong?
"Alexander Clouter" <[EMAIL PROTECTED]> wrote in message
news:8t8t3f$1pu$[EMAIL PROTECTED]...
> On Mon, 23 Oct 2000 16:20:42 -0400, Rick wrote:
> >
> > Unfortunately, I dont have a /etc/gpm.conf. I have a /etc/gpm-root.cont,
> > so Im still at a loss. But, there does seem to be some third party
> > drivers, and my 3 button mouse is now working fine.
> >
> Thats probably because you are not using Debian :) Fire up grep (or if
> you don't know the syntax use 'mc') to find the file you need to
> edit.
>
> BTW I used all those 3rd party tools and after some time I found it
> easier to just go back to gpm. However its your call :)
>
> Alex
>
------------------------------
From: Stanislaw Flatto <[EMAIL PROTECTED]>
Subject: Re: Redhat, Slackware, SUSE, FreeBSD, Help...
Date: Tue, 31 Oct 2000 10:06:26 +1100
Rudy wrote:
> The books of O'Reilly are really very good books.
> e.g. Linux network administrator's guide
> Olaf Kirch // O'Reilly
> ISBN: 1-56592-087-2
>
> Rudy
> [EMAIL PROTECTED]
>
> Stanislaw Flatto <[EMAIL PROTECTED]> wrote in message
> news:[EMAIL PROTECTED]...
> >
> >
I know, I know!
For about four years I have the "Essential System Administration" by Aeleen
Frisch.
Old, simple, depending on "elbow grease" for doing things - but few times
pulled me out of deep s...,
That's life.
Have fun.
Stanislaw.
------------------------------
From: "Steve Wolfe" <[EMAIL PROTECTED]>
Subject: Removing drive from RAID1
Date: Mon, 30 Oct 2000 15:54:15 -0700
I am currently working on a system that was set up by another
individual to use software RAID1 (mirroring) across two IDE drives. The
machine is being migrated to a larger setup, with hardware RAID.
In the interest of least downtime, the data needs to be transferred
quickly. Since the machines are fifty miles apart, it's tough to transfer
several gigs quickly. I'm considering simply removing the second drive from
the original machine, driving it to the new server, and copying the
information from it. My question:
If I simply remove the second drive from the first computer, will it
keep working with just the first drive, or will I have to interact with it
in any way?
steve
------------------------------
From: [EMAIL PROTECTED] (Juergen Heinzl)
Subject: Re: Something wrong with my terminal (cursor)
Date: 30 Oct 2000 23:18:53 GMT
In article <3dDADJ$[EMAIL PROTECTED]>, ����linux wrote:
>
>I can not recall what I have done.
[-]
It's always like that.
>But wright now, I can not use elm and vi.....
>Every time I enter elm, some messages appear:
>
>'
>Your terminal does not support the "clear screen" function (cl).
>Your terminal does not support the "clear to end of line" function (ce).
>Your terminal does not support the "clear to end of display" function (cd).
>Your terminal does not support the "cursor motion" function (cm).
>Your terminal does not support the "move cursor up" function (cm).
>Your terminal does not support the "move cursor right" function (nd).'
>
>I also can not move my cursor in vi !!!
>What should I do to fix this problem...
[-]
You seem to use the termcap library and make sure your TERM setting
is correct, export TERM=linux at the console, and that /etc/termcap
isn't corrupt for some reason.
Hope it helps,
Juergen
--
\ Real name : J�rgen Heinzl \ no flames /
\ EMail Private : [EMAIL PROTECTED] \ send money instead /
------------------------------
From: [EMAIL PROTECTED] (Juergen Heinzl)
Crossposted-To: alt.os.linux
Subject: Re: daylight savings.
Date: 30 Oct 2000 23:18:52 GMT
In article <8tk0qo$5ud$[EMAIL PROTECTED]>, Samuel Irlapati wrote:
>Is there a way to change the time automatically for daylight savings for
>Linux? I had to swallow my pride and boot into windoze98 to get the right
>time. I know there is also a Unix command to change time. Does anyone know
>what is that command?
[-]
It ought to get it right ?! Is there a symbolic link localtime in /etc
like this one ?
lrwxrwxrwx [...] /etc/localtime -> /usr/share/zoneinfo/MET
Can assure you my machine went back from 3 o'clock in the morning
to 2 o'clock Sunday morning %-)
Cheers,
Juergen
--
\ Real name : J�rgen Heinzl \ no flames /
\ EMail Private : [EMAIL PROTECTED] \ send money instead /
------------------------------
From: Chris Pelton <[EMAIL PROTECTED]>
Subject: Re: Stock Red Hat 7 kernel compile fails. (repost)
Date: Mon, 30 Oct 2000 16:39:27 -0700
i don't know if this is too late, but this is what i did, and it will at least
compile. change the /usr/src/linux/include/linux/kernel_stat.h file, line 47, replace
smp_num_cpus with the number of cpus on the machine
(i.e. 1 or 2 or more) i'm still having a lot of problems with the kernel, but i can
get a new one to compile now.
good luck.
Jeff Workman wrote:
> Due to lack of response to this, I am reposting it with the hopes that
> somebody can help me this time.
>
> All,
>
> OK, I have read the gotchas and caveats about RH 7 and it's "kgcc" but it
> still doesn't help the problem I'm having.
>
> This is the errors I am getting on 2 different boxes. One of these is a
> RH 6.2 server install upgraded to RH 7 and the other is a clean RH 7
> workstation class install. No modification to the kernel configuration
> has been made or anything, just a simple "make menuconfig" then immediate
> save, then a "make dep; make clean; make bzImage" which gives the
> following compile errors on both machines. Can somebody tell me what the
> fix is for this?
>
> TIA,
> Jeff
>
> kgcc -D__KERNEL__ -I/usr/src/linux/include -Wall -Wstrict-prototypes -O2
>-fomit-frame-pointer -fno-strict-aliasing -pipe -fno-strength-reduce -m386 -DCPU=386
>-DUTS_MACHINE='"i386"' -c -o init/version.o init/version.c
> make -C kernel
> make[1]: Entering directory `/usr/src/linux-2.2.16/kernel'
> make all_targets
> make[2]: Entering directory `/usr/src/linux-2.2.16/kernel'
> kgcc -D__KERNEL__ -I/usr/src/linux/include -Wall -Wstrict-prototypes -O2
>-fomit-frame-pointer -fno-strict-aliasing -pipe -fno-strength-reduce -m386 -DCPU=386
> -DEXPORT_SYMTAB -c ksyms.c
> In file included from /usr/src/linux/include/linux/modversions.h:50,
> from /usr/src/linux/include/linux/module.h:19,
> from ksyms.c:14:
> /usr/src/linux/include/linux/modules/i386_ksyms.ver:6: warning: `cpu_data' redefined
> /usr/src/linux/include/asm/processor.h:96: warning: this is the location of the
>previous definition
> /usr/src/linux/include/linux/modules/i386_ksyms.ver:28: warning: `smp_num_cpus'
>redefined
> /usr/src/linux/include/linux/smp.h:77: warning: this is the location of the previous
>definition
> /usr/src/linux/include/linux/modules/i386_ksyms.ver:118: warning:
>`smp_call_function' redefined
> /usr/src/linux/include/linux/smp.h:83: warning: this is the location of the previous
>definition
> In file included from /usr/src/linux/include/linux/interrupt.h:51,
> from ksyms.c:21:
> /usr/src/linux/include/asm/hardirq.h:23: warning: `synchronize_irq' redefined
> /usr/src/linux/include/linux/modules/i386_ksyms.ver:138: warning: this is the
>location of the previous definition
> In file included from /usr/src/linux/include/linux/interrupt.h:52,
> from ksyms.c:21:
> /usr/src/linux/include/asm/softirq.h:75: warning: `synchronize_bh' redefined
> /usr/src/linux/include/linux/modules/i386_ksyms.ver:142: warning: this is the
>location of the previous definition
> /usr/src/linux/include/linux/kernel_stat.h: In function `kstat_irqs':
> In file included from ksyms.c:17:
> /usr/src/linux/include/linux/kernel_stat.h:47: `smp_num_cpus' undeclared (first use
>in this function)
> /usr/src/linux/include/linux/kernel_stat.h:47: (Each undeclared identifier is
>reported only once
> /usr/src/linux/include/linux/kernel_stat.h:47: for each function it appears in.)
> make[2]: *** [ksyms.o] Error 1
> make[2]: Leaving directory `/usr/src/linux-2.2.16/kernel'
> make[1]: *** [first_rule] Error 2
> make[1]: Leaving directory `/usr/src/linux-2.2.16/kernel'
> make: *** [_dir_kernel] Error 2
>
> --
> "For competitive reasons we can't tell you the location of our fiber."
> -- An anonymous representative of a very large telco
> "For competitive reasons we can't tell you the location of our backhoe."
> -- An anonymous representative of a contractor.
------------------------------
From: "Scott M. Navarre" <[EMAIL PROTECTED]>
Subject: Linux Dual-Server Fault-Tolerant Setup
Date: Mon, 30 Oct 2000 23:58:40 GMT
Hello,
How can Linux be set-up so that there are two servers (a primary and a
backup) which share a RAID-10 disk array (using software RAID), and the
backup server takes over (automatically) when the primary one dies?
I currently have a SCO OpenServer system setup this way using 1776 Disk
Array software. So we have two servers (DEC Prioris XL's), one with SMP
(the primary) and one without (the backup). Each server then has two
Adaptec AHA-1740 SCSI controllers which connect to a shared RAID cage in
which there are 6 SCSI drives [2 sets of 3 RAID-0'd (striped) drives, which
are then RAID-1'd (mirrorred) together].
So, I would like to change the system over to Linux while being able to
keep this current hardware setup.
It would also be nice to mirror the boot/root drive internally. Is this
also possible?
I understand that Linux has built-in software RAID capabilities, so I
would just need software with the ability to detect when the primary machine
goes down and takes over the RAID filesystem and network connection. Right?
What is currently available to do this, and what do you recommend?
Thanks in advance,
Scott
------------------------------
From: [EMAIL PROTECTED]
Subject: Re: after recompiling kernel boot errors
Date: Mon, 30 Oct 2000 23:47:57 GMT
There's a kernel compiling howto at HardcoreLinux. You can find it here:
http://www.hardcorelinux.com/kernel-howto.htm
I think this will help. Good luck.
> I'm having a bit of trouble with a new kernel that I have just
compiled,
> in that I'm getting an interesting error when booting...
>
> attempt to access beyond end of device
> 03:06: rw=0, want=2, limit=0
> dev 03:06 blksize-1024 blocknr=1 sector=2 size=1024 count=1
> EXT2-fs: unable to read superblock
> FAT bread failed
>
> What I have I not/compiled into my kenel?? I am attempting this with
RH7
> and with kernel 2.2.16
>
> Thanks in advance...
>
> yowsr
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.
>
Sent via Deja.com http://www.deja.com/
Before you buy.
------------------------------
From: [EMAIL PROTECTED]
Subject: How to use the floppy drive in Linux
Date: Mon, 30 Oct 2000 23:45:55 GMT
I�m new in Linux world; I need to know how to use the floppy drive.How
do I copy files from hard drive to floppy and how from floppy to hard
drive. What commands should I use?
Any suggestion is appreciated.
Thank you
Sent via Deja.com http://www.deja.com/
Before you buy.
------------------------------
From: * Tong * <[EMAIL PROTECTED]>
Subject: mount swap in /tmp
Date: 30 Oct 2000 20:06:50 -0400
Hi,
How can I mount swap in /tmp?
I used to have a linuxswap partition for the swap, but now since I
noticed that my swap utilization is extremly low, I want to change it for
/tmp volume. Please also provide some comments or tips on this.
Thanks!
PS. My linux: RedHat6.2,
--
Tong (remove underscore(s) to reply)
http://members.xoom.com/suntong001/
- All free contribution & collection & music from the heavens
------------------------------
** 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 (and comp.os.linux.misc) via:
Internet: [EMAIL PROTECTED]
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
******************************
