Linux-Misc Digest #14, Volume #25 Sun, 2 Jul 00 03:13:01 EDT
Contents:
autoconf/automake to compile and produce an obkect module file (Young4ert)
Re: autoconf/automake to compile and produce an obkect module file (Erik de Castro
Lopo)
Re: 64 MB RAM shown instead of 128 (Markus Kossmann)
Re: Mail and Web server - Any Distributions more secure than others? (William
Wueppelmann)
Re: VMWare question (Martin Skj�ldebrand)
Re: Where Can I Find Pico Source Code? (Martin Skj�ldebrand)
Re: OSS driver for Sony Vaio Yamaha sound (David Efflandt)
anybody can tell me about mouse data format? ("jacky cui")
Compiling GnuCash 1.4.1 ("Robert L. Cochran Jr.")
Re: 64 MB RAM shown instead of 128 (Juergen Pfann)
Re: newbie questions (David M. Cook)
Re: Mail and Web server - Any Distributions more secure than others? (David M. Cook)
remote monitoring of Linux OS stats (killelea)
out of mem and INIT: PANIC segmentation problem (Enrico Arturo Zappi)
Re: Command or piped commands to show directory size...
Re: out of mem and INIT: PANIC segmentation problem (Hal Burgiss)
Re: Linux Server Security (David M. Cook)
Re: CLI Important? (Lew Pitcher)
Re: tape backup with dd (Juergen Pfann)
----------------------------------------------------------------------------
From: Young4ert <[EMAIL PROTECTED]>
Subject: autoconf/automake to compile and produce an obkect module file
Date: Sun, 2 Jul 2000 00:19:51 -0400
This message is in MIME format.
------------------------------
From: Erik de Castro Lopo <[EMAIL PROTECTED]>
Subject: Re: autoconf/automake to compile and produce an obkect module file
Date: Sun, 02 Jul 2000 04:51:03 +0000
Young4ert wrote:
>
> This message is in MIME format.
Any your quest is ....?
Erik
--
+-------------------------------------------------+
Erik de Castro Lopo [EMAIL PROTECTED]
+-------------------------------------------------+
"Don't be fooled by NT/Exchange propaganda. M$ Exchange is
just plain broken and NT cannot handle the sustained load
of a high-volume remote mail server"
-- Eric S. Raymond in the Fetchmail FAQ
------------------------------
From: Markus Kossmann <[EMAIL PROTECTED]>
Subject: Re: 64 MB RAM shown instead of 128
Date: Sun, 02 Jul 2000 06:27:17 +0200
Leonard Evens wrote:
>
> [EMAIL PROTECTED] wrote:
> >
> > When I try the free command, the memory shown is only 64 MB whereas I
> > have 128 MB. I am running kernel version 2.2.14 Linux Mandrake 7.0.
> > Is there anything I can do about it.?
> >
> > Sent via Deja.com http://www.deja.com/
> > Before you buy.
>
> You have gotten lots of advice about the append command,
> and that may solve your problem. But recent kernels should
> be able to see all the memory.
Unfortunately , that's not true. When 2.2 was developed. a spezific
BIOS call ( IIRC int 15 ax=e801h ) was used to detect memory >64 MB .
But now, BIOS/Motherboard manufactorers have switched to another call (
int15 ax=e820h) , which was specified by M$
for Windows compatibility. This call is not supported by 2.2 kernels.
And the e801 call was removed from BIOS. So 2.2 kernels will see max.
64 MB with these boards/BIOSses.
--
Markus Kossmann
[EMAIL PROTECTED]
------------------------------
From: [EMAIL PROTECTED] (William Wueppelmann)
Subject: Re: Mail and Web server - Any Distributions more secure than others?
Date: Sun, 02 Jul 2000 05:05:50 GMT
In our last episode (Sat, 1 Jul 2000 14:31:19 +0100),
the artist formerly known as Ashley Heath said:
>Are any Linux distributions more secure than others or better suited for use
>and Web and Mail servers?
>So far I have looked at Suse and Redhat, but any recommendations welcome.
They can all be made about equally secure. Some ship more secure than
others, and most need at least some adjustments. Remember that security has
as much to to with your own activities and practices as it does with the
software itself.
One report for last year put the number of security problems found in Red
Hat Linux at 30, and in Debian Linux at 3. The other distributions came
somewhere between the two.
Debian is pretty good out of the box. It does most of the sensible things
you'd want to do, like install TCP wrappers and asks you if you want to
install shadow passwords.
--
It is pitch black.
You are likely to be spammed by a grue.
------------------------------
Subject: Re: VMWare question
From: [EMAIL PROTECTED] (Martin Skj�ldebrand)
Date: Sun, 02 Jul 2000 05:29:26 GMT
This is the output of cat /dev/sndstat > sound.txt:
Build: 2.2.16-UP
Card config:
Aureal Vortex at 0xe8040000 irq 10
Software mixing (audio)
Audio devices:
0: Aureal Vortex 2 (AU8830) (QSRC)
1: SoftOSS v1.2 CH #0
2: SoftOSS v1.2 CH #1
3: SoftOSS v1.2 CH #2
4: SoftOSS v1.2 CH #3
5: SoftOSS v1.2 CH #4
6: SoftOSS v1.2 CH #5
7: SoftOSS v1.2 CH #6
8: SoftOSS v1.2 CH #7
Synth devices:
0: SoftOSS v1.2
Midi devices:
0: Aureal Vortex 2
Timers:
0: System clock
1: SoftOSS
Mixers:
0: Aureal Vortex 2 (��v5/0x83847605)
1: SoftOSS
To me it looks as if it is initialized OK.
Cheers,
M.
--
Martin Skj�ldebrand
[EMAIL PROTECTED]
Sys admin, web designer, tech writer
Hungry? Visit http://www.bahnhof.se/~chimbis/tocb
------------------------------
Subject: Re: Where Can I Find Pico Source Code?
From: [EMAIL PROTECTED] (Martin Skj�ldebrand)
Date: Sun, 02 Jul 2000 05:29:27 GMT
[EMAIL PROTECTED] writes:
> One more question. My computer has Linux and I do know it already has pine.
> Does it mean the source file of pine is already on my computer? or only the
> binary files are on the computer??
>
> Sorry for asking so elementary questions. . .
>
That depends on how you got either Pine or Linux onto your
computer. Usually only the binaries are on the computer, unless you
built them yourself.
The source files should be on the installation media.
M.
--
Martin Skj�ldebrand
[EMAIL PROTECTED]
Sys admin, web designer, tech writer
Hungry? Visit http://www.bahnhof.se/~chimbis/tocb
------------------------------
From: [EMAIL PROTECTED] (David Efflandt)
Subject: Re: OSS driver for Sony Vaio Yamaha sound
Date: 2 Jul 2000 05:32:17 GMT
Reply-To: [EMAIL PROTECTED]
On Sat, 01 Jul 2000, Kaushik Raghavan <[EMAIL PROTECTED]> wrote:
>This is a multi-part message in MIME format.
>
>------=_NextPart_000_001C_01BFE376.FA56C5A0
>Content-Type: text/plain;
> charset="iso-8859-1"
>Content-Transfer-Encoding: quoted-printable
Please just post plain text.
>After I upgraded to RH 6.2 my OSS driver for my Sony Vaio Yamaha sound =
>doesn't work properly anymore. I get the same click 8 times. Almost as =
>if there;s a 8x loop. Anybody else see this? how can I fix this?
>
>I have a Sony Vaio PCG-F430
The commercial OSS needs to be reconfigured when you install a new kernel,
and if you upgraded your RH version, probably also needs the
/etc/conf.modules entries too. The simplest way to do that is to
reinstall OSS. It should retain your existing /usr/lib/oss/license.asc
I was using RH 6.1 on my F450 and am now using Mandrake 7.0. Alsa now has
drivers for Yamaha chips, but that does not seem to work with playmidi
like the OSS drivers do.
--
David Efflandt [EMAIL PROTECTED] http://www.de-srv.com/
http://www.autox.chicago.il.us/ http://www.berniesfloral.net/
http://hammer.prohosting.com/~cgi-wiz/ http://cgi-help.virtualave.net/
------------------------------
From: "jacky cui" <[EMAIL PROTECTED]>
Subject: anybody can tell me about mouse data format?
Date: Sun, 2 Jul 2000 13:45:44 +0800
general serial mouse used in linux,i want to know:
1. how data is transmitted to mouse driver?
2. is this 3 bytes? what is the meaning of this 3 bytes?
3. there are some protocol :Microsoft ,Logitech,mousesytem ....
what difference they have ?
4. how the x-windows recognize 3 bytes?
Any help is appreciated
thanks.
--
===============================
Best wish to you
http://home.etang.com/jackyeasy
OICQ:474007
Jacky
===============================
------------------------------
From: "Robert L. Cochran Jr." <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.development.apps
Subject: Compiling GnuCash 1.4.1
Date: Sun, 2 Jul 2000 01:50:38 -0400
I am trying to compile gnucash-1.4.1 from a tarball downloaded from =
http://www.gnucash.org on a RedHat 6.2 system running the 2.2.16-3 =
kernel. (I've upgraded the kernel twice since installing this system =
from the RedHat Deluxe boxed set.)
When I run the configure script, it fails with an error message "can't =
find gtk-xmhtml.h". This header is located in my /usr/include/gtk-xmhtml =
directory. I printed the configure script and noticed this line
#include <gtk-xmhtml/gtk-xmhtml.h>
Since this is in a configure script, it looks like it is a comment line. =
I didn't think to check the config.log file to see what it had to say; I =
only noticed it later. (I'm rather new to Linux and C programming.) =
However I thought this line is actually trying to specify an include =
file for conftest.c so I altered it to look like this:
#include </usr/include/gtk-xmhtml/gtk-xmhtml.h>
and reran the configure script. This failed with the same error message =
as above, "can't find gtk-xmhtml.h". After seeing references to =
config.log in the configure script, I printed this file and found the =
following error messages from the gcc compiler:
In file included from /usr/lib/glib/include/glibconfig.h:13,
[several more lines formatted like the above, including "from =
/usr/include/gtk-xmhtml/gtk-xmhtml.h:5"]
/usr/include/limits.h:45: warning: 'MB_LEN_MAX' redefined
/usr/lib/gcc-lib/i386-redhat-linux/egcs-2.91.66/include/limits.h:26: =
warning: this is the location of the previous definition
configure: failed program was:
#line 5534 "configure"
...
...
Wht is the above error message trying to tell me, and what can be done =
to fix it? I was going to post this question to the GnuCash web site, =
which had a forum for installation help. The forum has suddenly =
disappeared from the web site.=20
I could just chicken out and install the rpm that GnuCash provides for =
their application, but I'm very interested in developing code in Linux =
and would like to see this compile succeed as a sort of introduction to =
programming in Unix. Especially since I already compiled and installed =
SWIG 1.1p5 successfully.=20
Thanks for any advice you can offer.
Bob Cochran
Beltsville, MD
------------------------------
From: Juergen Pfann <[EMAIL PROTECTED]>
Subject: Re: 64 MB RAM shown instead of 128
Date: Sun, 02 Jul 2000 07:59:16 +0200
Markus Kossmann wrote:
>
> Unfortunately , that's not true. When 2.2 was developed. a spezific
> BIOS call ( IIRC int 15 ax=e801h ) was used to detect memory >64 MB .
> But now, BIOS/Motherboard manufactorers have switched to another call (
> int15 ax=e820h) , which was specified by M$
> for Windows compatibility. This call is not supported by 2.2 kernels.
> And the e801 call was removed from BIOS. So 2.2 kernels will see max.
> 64 MB with these boards/BIOSses.
>
I guess, altogether you're right - but I want to add that some
BIOS programmers seem to do a lousy job !
For instance, I tried several BIOS revisions with my Abit BP6 -
and the original BIOS didn't have that >64 problem, but my first
upgrade did (together with different problems, so I reverted to
the original) ! The next update fixed this problem again, but
invoked some other... sigh !
BTW : in my case, this BIOS-rev. related memory detection problem
affected *both* 2.2.1x and 2.0.36 kernels.
CU
Juergen
------------------------------
From: [EMAIL PROTECTED] (David M. Cook)
Subject: Re: newbie questions
Date: 2 Jul 2000 06:02:47 GMT
On Sat, 01 Jul 2000 16:25:27 -1000, Ron Nicholls <[EMAIL PROTECTED]> wrote:
>I saw sndconfig load during install but cannot run it or find it .
The whereis command is good for this
$ whereis sndconfig
sndconfig: /usr/sbin/sndconfig /usr/share/sndconfig /usr/man/man8/sndconfig.8
>One more thing , at one time my mouse would cross into and switch too
>the next desktop if it ran into the
>edge. Where is this function enabled /disabled.
Probably under Enlightenment configuration.
Dave Cook
------------------------------
From: [EMAIL PROTECTED] (David M. Cook)
Subject: Re: Mail and Web server - Any Distributions more secure than others?
Date: 2 Jul 2000 06:05:57 GMT
On Sun, 02 Jul 2000 05:05:50 GMT, William Wueppelmann
<[EMAIL PROTECTED]> wrote:
>One report for last year put the number of security problems found in Red
>Hat Linux at 30, and in Debian Linux at 3. The other distributions came
>somewhere between the two.
Red Hat probably has a lot more uses, especially a lot more *naive* users,
so this statistic doesn't really say anything about relative security of Red
Hat vs. Debian.
Dave Cook
------------------------------
From: killelea <[EMAIL PROTECTED]>
Subject: remote monitoring of Linux OS stats
Date: Sat, 01 Jul 2000 22:52:55 -0700
I wrote a freeware program that lets you monitor CPU
usage etc remotely, without logging in. Docs and download
at:
http://patrick.net/software/rstat/rstat.html
Please let me know what you think.
Patrick
* Sent from AltaVista http://www.altavista.com Where you can also find related Web
Pages, Images, Audios, Videos, News, and Shopping. Smart is Beautiful
------------------------------
From: Enrico Arturo Zappi <[EMAIL PROTECTED]>
Subject: out of mem and INIT: PANIC segmentation problem
Date: Sun, 2 Jul 2000 02:10:35 -0400
I have a P200 with 32 ram working as a NAT to share internet connection.
When i leave the system alone for several housr and come back the mesages
show up:
Out of memory for pump.
Out of memory for klogd.
Out of memory for init.
Out of memory for syslogd.
INIT: PANIC: segmentation violation! giving up..
the real problem is that after this happens i am able to log in and run
programs but unable to reboot or halt. Also many of the programs after
they are closed show up as <defunct> when i do a ps -A. For example
mingetty so i cant log in to the same vt twice. Its also strange that
when i log in the messsage
PAM_pwdb[733]: (login) session opened for user root by (uid=0)
shows up which i never see otherwise.
Any ideas why this happens and how to fix it?
Thank you, Enrico
------------------------------
From: [EMAIL PROTECTED] ()
Crossposted-To: alt.os.linux
Subject: Re: Command or piped commands to show directory size...
Date: 2 Jul 2000 06:13:46 GMT
On Fri, 30 Jun 2000 08:32:41 -0400, Fro-Man <[EMAIL PROTECTED]> wrote:
>
>du `ls -alF | awk '{print $5 \"+ 0\" }'`
>
>
>Something crazy like this may work... In theory bc should take the output
>of that "ls | awk" command. The "ls | awk" thing shoudl print out a
It would be easier to let awk do the addition:
ls -l | awk 'BEGIN {tot=0}; tot=tot+$5; END {print tot}'
--
========================================================================
David E. Fox Census Thanks for letting me
[EMAIL PROTECTED] 2000 change magnetic patterns
Be Counted: http://www.census.gov on your hard disk.
=======================================================================
------------------------------
From: [EMAIL PROTECTED] (Hal Burgiss)
Subject: Re: out of mem and INIT: PANIC segmentation problem
Reply-To: [EMAIL PROTECTED]
Date: Sun, 02 Jul 2000 06:17:04 GMT
On Sun, 2 Jul 2000 02:10:35 -0400, Enrico Arturo Zappi <[EMAIL PROTECTED]> wrote:
>I have a P200 with 32 ram working as a NAT to share internet connection.
>When i leave the system alone for several housr and come back the mesages
>show up:
>Out of memory for pump.
>Out of memory for klogd.
>Out of memory for init.
>Out of memory for syslogd.
>INIT: PANIC: segmentation violation! giving up..
>
>the real problem is that after this happens i am able to log in and run
>programs but unable to reboot or halt. Also many of the programs after
>they are closed show up as <defunct> when i do a ps -A. For example
>mingetty so i cant log in to the same vt twice. Its also strange that
>when i log in the messsage
>PAM_pwdb[733]: (login) session opened for user root by (uid=0)
>shows up which i never see otherwise.
>Any ideas why this happens and how to fix it?
>Thank you, Enrico
Sounds like a memory leak somewhere. Set up a cron job and mail
yourself, or log, the output of 'top -b' every hour or so.
--
Hal B
[EMAIL PROTECTED]
--
------------------------------
From: [EMAIL PROTECTED] (David M. Cook)
Subject: Re: Linux Server Security
Date: 2 Jul 2000 06:25:38 GMT
On Sat, 01 Jul 2000 08:30:06 GMT, Scippie <[EMAIL PROTECTED]> wrote:
>Now my question is... can anyone that has my IP number hack the machine ?
>(please don't prove it, you probably already know my IP number), and what
>can he mess up ?
Short answer: yes, most definitely. However, you can reduce the risk by
turning off any services you don't use (you can do this with linuxconf, or a
combination of ntsysv and the scripts in /etc/rc.d/init.d). Comment out
anything you don't use in /etc/inetd.conf. Edit /etc/hosts.deny and
/etc/hosts.allow to restrict access to services (this only covers things
managed by inetd).
Short of actual root access, there are DoS attacks. For instance, a haxor
d00d once took down (caused a kernel panic) on my IP masq gateway using a
vulnerability in the kernel IP stack. Not much you can do about these
except keep up to date on kernel vulnerabilities.
If they get root access (using some kind of buffer overflow in a service run
as root), they can of course do anything.
Things to look for are a change in the size or checksum of /bin/login,
/bin/ps, /bin/netstat, etc. You can use rpm to check (assuming they haven't
modified the rpm database).
rpm -q util-linux
rpm -q fileutils
rpm -q procps
rpm -q net-tools
Make sure you stay on top of security updates. Red Hat has a mailing list
you can subscribe to to get update reports.
Dave Cook
------------------------------
From: Lew Pitcher <[EMAIL PROTECTED]>
Subject: Re: CLI Important?
Date: Sat, 01 Jul 2000 21:44:27 -0400
Winsun Hsieh wrote:
>
> Within NSD, The NetRaid-4M team is working to add Linux support for second
> release (Oct.). Can you help me determine if a CLI (command line interface)
> configuration tool is needed for Linux? That is, do users need an online
> configuration tool or are they okay with rebooting and hitting CTRL A to
> configure that way. Configuration is basically installation of a NetRAID
> card and setup of the RAID levels.
Reboot a system just to change a configuration? Sorry, but no.
You apparently don't believe in 7/24 systems, or perhaps are directing
your product development towards the MSWindows environment, where
frequent reboots are a necessity.
> The NetRAID systems team has received no request from the Linux team
What "Linux team"?
You apparently believe that there is a centralized 'team' with sole
ownership of Linux development. This is not true. Perhaps the core
Linux developers have not contacted you, but it appears that you
haven't contacted them either.
> that
> such a tool is needed. Thus, if Linux users don't need an online tool, my
> team would just as soon focus their limited resources in other directions.
--
Lew Pitcher
Master Codewright and JOAT-in-training
------------------------------
From: Juergen Pfann <[EMAIL PROTECTED]>
Subject: Re: tape backup with dd
Date: Sun, 02 Jul 2000 08:54:56 +0200
Christian Eriksson wrote:
>
> I've failed in writing a backup back to disk from tape with dd.
> I'm running Red Hat 6.0 on Intel and I have a HP Colorado 14GBe
> tape drive attached to the paralell port. What I've done is the
> following:
>
> First I wanted to backup my primary slave drive. From linux I did
>
> # dd if=/dev/hdb of=/dev/npt0
> 2504880+0 records in
> 2504880+0 records out
>
> This took some time and as I got the prompt back I assumed the backup
> operation to tape was succesful and I did not verify it in some other
> way. I then wanted to write it back so I did
>
> # dd if=/dev/pt0 of=/dev/hdb
> 0+0 records in
> 0+0 records out
>
> and as you can see, it failed. Can someone help me here?
>
> Thanks in advance!
>
> /Christian Eriksson.
Well, I wouldn't say it failed...
If you issued the 2nd command without any tape movement between writing
and reading again, dd's behaviour is *correct* :
Be aware that you used the *non-rewinding* device (/dev/npt0) for
writing, and after that the tape head was after the end of a "file" on
tape containing your "diskdump". You then tried to read from that
position on, and there was nothing on tape anymore ("EOM" resp. "EOD").
So dd didn't find anything to read.
BUT - as you used /dev/pt0 (the rewinding device) for that, the tape
probably rewound to the beginning ("BOT") afterwards; I'm pretty sure
that if you had repeated your read command, you would've succeeded.
Seems you're not yet familiar with the *sequential* nature of tape
devices ?
Again, be aware which device you use - the rewind-on-close or the
non-rewinding one. BTW, that applies analoguously to any tape driver,
be it "floppy" tape, ATAPI or SCSI or whatever...
with the rewinding devices, you can do one tape operation (read or
write) and will *always* be at the beginning of the tape again
afterwards. Watch your LEDs (or simply listen with Travan drives!)
to see what happens...
But if you want to append another backup session to an existing
tape archive, for ex., you must use the non-rewinding device -
and the "mt" command to skip the first archive ('mt -f /dev/npt0 fsf 1'
in your case). See "man mt". You should experiment with that :
write several sessions with the non-rewinding device, then jump
back and forth to get familiar with that.
I write that with "tar" in mind, which is my mostly used tape backup
utility. But I did a short test just now - this also seems to work
pretty much the same with several "dd" sessions appended to each other
on one tape. It's up to you to keep track of your "sessions",
"archives" or "(tape) files" - whatever you like to call them -
and get used to use "mt tell" or "mt status" often; it's quite easy to
overwrite one session with another, or restore the wrong archive...
Juergen
------------------------------
** 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
******************************
