Linux-Misc Digest #242, Volume #26 Sun, 5 Nov 00 22:13:02 EST
Contents:
Re: Linux on a dual processorboard (Jean-David Beyer)
Re: pump and AT&T Broadband ([EMAIL PROTECTED])
Email question ("Jay")
Help with error messages... (Jean-David Beyer)
HELP: HelixGnome with Slackware 7.1 (p e a r c e)
Re: Me tryin' convince my company to use Samba (Steve)
Re: KDE vs GNOME: specific issues ([EMAIL PROTECTED])
Re: eOne and Linux - modem question (Rob Clark)
Re: Tape drive recommendations? ("Andrew E. Schulman")
Where to place IDE tape drive? ("Paul Steckler")
Re: Looking for "clean disk" utility (Dances With Crows)
Re: Mandrake install problems ("Daniel Lenski")
Problem with Linux IP-Masq'ing and Win2K ("Don Whitlow")
need help with kernel patching 2.2.16 -> 2.2.16-1 (Erik Myllymaki)
Re: KDE vs GNOME: specific issues (John Obenauer)
----------------------------------------------------------------------------
From: Jean-David Beyer <[EMAIL PROTECTED]>
Subject: Re: Linux on a dual processorboard
Date: Sun, 05 Nov 2000 20:24:57 -0500
dick dijk wrote:
> Hello,
>
> I am thinking about buying a new PC with two processors in it, does anyone
> have a suggestion for a dual processor motherboard. If so what processors do
> you use? Do you experience good performance?
>
> Dick
I have two 550MHz Pentium IIIs, 512Megabytes 100Mhz ECC SDRAM running on a Tyan
S1832DL Tiger 100 mother board. It works as expected, though some things are
counterintuitive.
As expected, if I run setiathome in the background, it can get up to 50% of the
total CPU capacity, since it has not been designed to run specifically on and
SMP machine, and Linux does not choose to examine a single process to see if it
could be split up onto multiple parallel processors. (I can run two instances
and get over 98% of the total CPU power.
As unexpected, if I run IBM's DB2 V6.1 UDB database management system, I can
tell it to run on just one, or to use both CPUs. If I tell it to run on both
CPUs, they really do that, and run three servers (one sort-of watches the other
two) instead of the usual one. However most jobs actually run slower that way.
The one that benefits the most runs in about 43 seconds with one CPU and 33
seconds on two CPUs. (This is for one iteration of a lenghthier program.)
Is that the kind of thing you want to know?
--
.~. Jean-David Beyer Registered Linux User 85642.
/V\ Registered Machine 73926.
/( )\ Shrewsbury, New Jersey
^^-^^ 8:15pm up 12 days, 8:33, 3 users, load average: 3.31, 3.33, 3.23
------------------------------
From: [EMAIL PROTECTED]
Subject: Re: pump and AT&T Broadband
Date: Mon, 06 Nov 2000 01:32:23 GMT
Thank you for your prompt reply. I have reboot the computer and did a
ifconfig eth0, it returned with the info for the loopback. I then
typed pump -h mother and it failed. I then typed ifconfig again and
it returned with eth0 information but lacking IP address. I can tell
you that it transmitted 28 packets and didn't receive any.
I did type route -n and only got back the desitnation 127.0.0.0.
When I type dmesg, I see that the tulip driver is loaded and then
information on eth0 the two lines report:
eth0: Lite-On 82c168 PNIC rev 32 at 0x8c00, 00:A0:CC:62:5B:7D
eth0: MII transceiver #1 config 3000 status 7829 advertising 01e1
Any other insight would be appreciated.
John
On Sun, 05 Nov 2000 23:15:44 GMT, [EMAIL PROTECTED] (Rod
Smith) wrote:
>[Posted and mailed]
>
>In article <[EMAIL PROTECTED]>,
> [EMAIL PROTECTED] writes:
>> I a running RH 7.0. I just installed it and was able to get connect
>> to the internet, after some work. I then rebooted and now it won't
>> connect. I have a 3Comm external modem with little indicators. For
>> these indicators I can tell the computer is sending information a
>> couple of times. I can see the NIC card go up and down (by the green
>> light on the modem).
>>
>> I learned from a couple of HOWTOs about the -h foobar on the end of
>> pump. I added it to the ifcfg-eth0. Any one have any ideas when I am
>> doing wrong.
>
>Immediately after a failed attempt to use pump, try issuing the
>following commands:
>
>/sbin/ifconfig eth0
>/sbin/route -n
>tail /var/log/messages
>
>The first shows the status of eth0 (change this if you've got more than
>one NIC and are using something other than eth0 for the cable modem
>connection). If all is well, you'll see a status message and an IP
>address (called "inet addr"). If you get nothing at all, it's probably a
>matter of a driver for the Ethernet card not being loaded.
>
>The second command shows the route that's set for the system. Assuming
>you got an IP address when you tried ifconfig, it's possible that the
>routing table is wrong. You should see a line along these lines in the
>output:
>
>0.0.0.0 192.168.45.1 0.0.0.0 UG 1 0 0 eth0
>
>This shows the default route (the first 0.0.0.0) going through the
>gateway system at 192.168.45.1 (yours will be a different number, of
>course).
>
>The third command should show you pump's status messages when it tried
>to obtain an IP address from the DHCP server. There may be clues there
>about what went wrong.
>
>--
>Rod Smith, [EMAIL PROTECTED]
>http://www.rodsbooks.com
>Author of books on Linux & multi-OS configuration
------------------------------
From: "Jay" <jayp@*spamfree*datainn.co.nz>
Subject: Email question
Date: Mon, 6 Nov 2000 14:32:30 +1300
I know how to set-up pops and aliases for virtual email domains...
A client wants an email system (for his 12 email accounts) whereby their
email router thing will clear all their accounts and distribute them to the
relevant inboxes within their set-up.
Any idea what I need to do at my end?
Set up a wildcard email and have that cleared by them? if so can someone
tell me how to set this up please.
Many thanks
Jay
------------------------------
From: Jean-David Beyer <[EMAIL PROTECTED]>
Subject: Help with error messages...
Date: Sun, 05 Nov 2000 20:34:38 -0500
This morning, I found the following error message on my console
(the thing you get when you press control-alt-pf1):
sym53c810a-1:3: ERROR (0:18) (1-21-ff) (8/13) @ (script
830:1900069a).
sym53c810a-1: script cmd = 88080000
sym53c810a-1: regdump: da 10 80 13 47 08 03 1e 80 01 83 21 80
01 01 00.
st0: Error with sense data: [valid=0] Info fld=0x0, Current
st09:00: sense key Unit Attention
Additional sense indicates Power on, reset, or bus device reset
occurred
I have two questions about this.
1.) I do not normally leave console #1 logged in, so normally,
I would never have seen this message. There is nothing like
this in /var/log/messages where I would normally expect to find
it. Is there a way to make this appear there? I assume it is
the driver for the sym53c810a, the narrow SCSI controller that
has only my DDS-2 tape drive on it.
2.) Does anyone know how to interpret what this message means?
I do know that it was doing a verification of a backup it had
just made from disk to tape, and got a read error on the
device. It seems to have been totally unrecoverable and all the
blocks after the one with the read error had incorrect
checksums. Yet I had no trouble re-running the backup job
correctly (with the same, fairly new, tape).
I had this problem once before, and someone advised me that the
tape drive, less than about 6 months old, was failing. My
vendor replaced the drive, and I get the same thing on the new
one that I just installed August 8. Tape Drive is HP C1599A
DDS-2 SCSI DAT Drive.
--
.~. Jean-David Beyer Registered Linux User 85642.
/V\ Registered Machine 73926.
/( )\ Shrewsbury, New Jersey
^^-^^ 8:25pm up 12 days, 8:43, 3 users, load average: 2.24, 2.89, 3.09
------------------------------
From: [EMAIL PROTECTED] (p e a r c e)
Crossposted-To: comp.os.linux.hardware
Subject: HELP: HelixGnome with Slackware 7.1
Reply-To: [EMAIL PROTECTED]
Date: Mon, 06 Nov 2000 01:35:04 GMT
I have just recently tried Slack 7.1 having previously used Stormix.
I like Slack a ton, but miss HelixGnome. Has anyone installed
HelixGnome into Slack 7.1 yet? If so, how does one go about doing
that?
Thanks in advance!
p e a r c e
REPLY TO: pearce AT orangedelta DOT com
------------------------------
From: [EMAIL PROTECTED] (Steve)
Crossposted-To:
comp.protocol.smb,linux.samba,comp.os.linux.network,mailing.unix.samba-technical,micosoft.public.windowsnt
Subject: Re: Me tryin' convince my company to use Samba
Date: 6 Nov 2000 01:36:06 GMT
Reply-To: [EMAIL PROTECTED]
On Sun, 05 Nov 2000 13:33:49 -0800, YY Lee wrote:
>Is there a way to create Word97 document with mactro to display
>downloading time from a NT file server and Linux running Samba so to
>prove which is really faster?
>
Use a web browser to d/l the file, they tell you how long it's going to
take (Netscape does anyway).
--
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
1:24am up 26 days, 3:46, 2 users, load average: 1.03, 1.04, 1.01
------------------------------
From: [EMAIL PROTECTED]
Crossposted-To: comp.os.linux.advocacy,linux.redhat,alt.os.linux
Subject: Re: KDE vs GNOME: specific issues
Date: Mon, 06 Nov 2000 01:33:33 GMT
In article <jeff_jeffries-0511001347260001@sdn-ar-
002nyprinp217.dialsprint.net>,
[EMAIL PROTECTED] (Jeff Jeffries) wrote:
> I need to choose either GNOME or KDE. I will be doing computationally
> intensive C++, with very heavy disk I/O. Results will be displayed in
> 3D preferrably with OpenGL.
>
> 1. Are GNOME and KDE C++ and/or object oriented? How will this affect
> developing with C++?
KDE is built with QT and C++, so there's a certain amount of object-
orientedness built in. I think QT has recently worked out its weird
licensing schemes, so I believe you don't have to worry about licensing
fees anymore (I think, there was a recent ruckus on www.linuxtoday.com
about it...). Do your research first, just to make sure. For a while,
QT was available only for non-commercial use -- if you wanted to
distribute, there were weird licensing schemes. I think this is fixed
with the move to GPL, but I'm not completely certain.
If you decide that you'd like to try C++ coding in GNOME, look up the
GTK-- widget set (it works with C++, they're just trying to be clever
with the naming). If you decide to go with C itself, look into the GTK+
widget set. GTK+ isn't exactly object-oriented, although there's been a
lot of work to try to make it FEEL object oriented. The biggest pain
will come from trying to create your own widgets -- the widget
constructor in C/GTK+ is not as easy as you'd hope. But, it's doable.
You're also going to find that in most ways you can build a GTK+
program and run it in KDE, or build a QT program and run it in GNOME,
so don't worry about making a high-level choice based on the XWindow
environment you think will be the most-used. It's a moot point.
> 2. I know GNOME has gtkglarea; does KDE?
Dunno.
> 3. What else should a C++ developer know?
Well, KDevelop 1.3 (for KDE) just came out, so if you need an IDE to
work within that might be your best choice. GNOME has glade, but I
found it less confusing to just code the stuff from scratch myself.
-andrew
Sent via Deja.com http://www.deja.com/
Before you buy.
------------------------------
Subject: Re: eOne and Linux - modem question
From: [EMAIL PROTECTED] (Rob Clark)
Date: Mon, 06 Nov 2000 01:32:18 GMT
In article <[EMAIL PROTECTED]>,
Jon <[EMAIL PROTECTED]> wrote:
[..]
>As far as I can tell from their web site and my "documentation"
>this machine has a winmodem (I know, I know) of this flavor:
>HSP MicroModem 56K PCI set up on COM 4 IRQ 3
>That's all I can find.
>
>Has anyone succeeded in getting this sucker hooked to the web via Linux?
>If so, how?
I just received an email from someone who is successfully using this modem
with the PCTel "Linmodem" driver. Please see:
http://www.o2.net/~gromitkc/winmodem.html
for details.
Rob Clark, [EMAIL PROTECTED]
------------------------------
From: "Andrew E. Schulman" <[EMAIL PROTECTED]>
Subject: Re: Tape drive recommendations?
Date: Sun, 05 Nov 2000 20:44:33 -0500
> I've been backing up stuffs on 4mm tape drive
> with BRU 15.1 (RedHat 6.X) unattended. Hard
> disks are getting bigger and bigger, but my
> trusted old TLZ-06 can only do 2G per tape.
>
> Any recommendations and experiences with
> newer tape drives, either SCSI[2,3] or IDE,
> with higher capacity per tape?
I have a Seagate Travan STT220000N-MC, an internal SCSI drive that writes
10/20 GB per tape, at 1 MB/s. It has hardware compression. Cost $350, and
tapes are about $35 each. Works very well under Windows and Linux. But,
even 10 or 20 GB isn't what it used to be.
------------------------------
From: "Paul Steckler" <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.hardware,linux.act.tape
Subject: Where to place IDE tape drive?
Date: Sun, 5 Nov 2000 19:45:23 -0600
I have two PC's connected via Ethernet,
a Win98 box and a RedHat 6.2 box. I'd like
to back up both of them with an IDE tape drive.
Of course, I don't want to have to get two tape
drives. Using Samba, I think I can mount the volumes
from one machine on the other, and backup both
machines with a single drive.
Samba will let me do the mounting in either direction.
I can Linux drives visible on the Win98 box, or
vice-versa (at least this is what I'm told).
So is there any reason to prefer having the tape drive
on one machine vs. the other?
A conjecture: The backup software for Windows may
be easier to use, hence it might be nicer to have the
drive there. Also, installation on Windows may be
more straightforward.
Any help on this appreciated.
-- Paul
------------------------------
From: [EMAIL PROTECTED] (Dances With Crows)
Crossposted-To: comp.security.unix
Subject: Re: Looking for "clean disk" utility
Date: 6 Nov 2000 01:57:02 GMT
Reply-To: [EMAIL PROTECTED]
On Sun, 05 Nov 2000 20:03:30 GMT, [EMAIL PROTECTED] wrote:
>Can anyone direct me to a Linux utility like (Windoze) "Clean Disk";
>essentially a utility that writes (and re-writes) unused space in the
>filesystem?
Clarify what you mean, please. Do you wish to overwrite all unused
space in a filesystem with zeroes or random data, so no one can
"undelete" anything? If so, you can use a quick hack like the ones
below, or look into various "secure deletion" programs/options on
freshmeat.net :
dd if=/dev/urandom of=/mountpoint/bigfile bs=8192 && rm
/mountpoint/bigfile
(repeat that a couple of times for better security, replacing
"mountpoint" with the mount point of the filesystem you wish to remove
deleted files from.)
Keep in mind that undeleting files under Linux is quite a bit more
complex than undeleting files in WinXX, MacOS, and some Netware
setups. True undeletion generally requires access to the raw disk,
which normal users should never have in most cases.
If you wish to "defrag" a Linux filesystem, you don't need to. ext2
filesystems do fragment, but Linux is built to handle this fragmentation
efficiently through aggressive disk caching. HTH,
--
Matt G|There is no Darkness in Eternity/But only Light too dim for us to see
Brainbench MVP for Linux Admin / Workin' in a code mine, hittin' Ctrl-Alt
http://www.brainbench.com / Workin' in a code mine, whoops!
=============================/ I hit a seg fault....
------------------------------
From: "Daniel Lenski" <[EMAIL PROTECTED]>
Crossposted-To:
alt.linux,alt.os.linux,alt.os.linux.mandrake,comp.os.linux,comp.os.linux.hardware,comp.os.linux.help,comp.os.linux.questions
Subject: Re: Mandrake install problems
Date: Sun, 05 Nov 2000 21:25:09 +0500
Linux sometimes has a problem detecting how much memory your computer
has. This causes it to try to access RAM that isn't there! How much
memory do you have?
When you install 7.1, try pressing F1 at the first
screen and typing in 'linux ram=???M', where ??? is the number of
megabytes of RAM that you have. This has worked for me many times.
> Im installing mandrake 7.1 on an Asus CUSL2 Intel 815E Atx 133
> motherboard with a 733 processor. Using the Drakx install program, when
> I reach the partition format proceedure, the system hangs or sometimes
> reboots on its own, usually at the latterportion of the hard drive. I
> also get an error saying that
> "seems like memory is missing as install crashes'
>
> does anyone have any ideas?
>
>
>
--
Daniel Lenski
[EMAIL PROTECTED]
"If we couldn't laugh at things that didn't make sense,
we couldn't react to a lot of the world around us."
--Calvin and Hobbes
------------------------------
From: "Don Whitlow" <[EMAIL PROTECTED]>
Subject: Problem with Linux IP-Masq'ing and Win2K
Date: Sun, 5 Nov 2000 20:27:59 -0600
Hey All,
I'm currently running a Red Hat 6.2 box as an IP-Chains Firewall / Ip Masq
gateway for my DSL line. On my side of the network, I have 2 Win2K
workstations setup, one of which dual boots Win98 which is my primary
bootup. Win98 accesses the internet great and works perfectly. I'm using
private 192.168.0.* addresses of which the Win98 / Win2K machine uses
192.168.0.2 for both OSes. My Linux box uses 192.168.0.1.
When I'm in Win2K on either machine, I can't get to the Internet. I've
configured the network as close to the same as the Win98 box as possible,
including setting my gateway to the address of my Linux box (192.168.0.1).
Everybody sees everybody else ok on the internal network, only the 2 Win2K
boxes can't see out onto the internet. Both will do Nslookups for internet
sites just fine, and both at least seem to start trying to make the
connection before simply timing out when browsing using IE. I can even ping
from both boxes out to sites on the internet. Any idea what is going on
here? I'm stumped.
Thanks in advance for any light anyone can shed on this.
Sincerely,
Don
------------------------------
Subject: need help with kernel patching 2.2.16 -> 2.2.16-1
From: Erik Myllymaki <[EMAIL PROTECTED]>
Date: Mon, 06 Nov 2000 02:36:26 GMT
Hello,
I am trying to get a clean 2.2.16-1 kernel source so I can do some LRP
development. I have downloaded the kernel-2.2.16-1.i386.srpm from
sourceforge and installed it. Inside of that I have untarred the 2.2.16
clean source. The problem I am having is that applying the patches produces
many errors. I have no idea about which order to apply them in and I think
this is the reason.
Can anyone tell me how to apply all the patches in the SRPM in the correct
order? Is there a shell script or something I am missing?
Thanks.
Erik Myllymaki
[EMAIL PROTECTED]
------------------------------
From: John Obenauer <[EMAIL PROTECTED]>
Subject: Re: KDE vs GNOME: specific issues
Crossposted-To:
comp.os.linux.advocacy,linux.redhat,alt.os.linux,comp.unix.solaris,alt.os.linux.mandrake
Reply-To: [EMAIL PROTECTED]
Date: Mon, 06 Nov 2000 02:40:11 GMT
Jeff Jeffries wrote:
> I need to choose either GNOME or KDE. I will be doing computationally
> intensive C++, with very heavy disk I/O. Results will be displayed in 3D
> preferrably with OpenGL.
>
> 1. Are GNOME and KDE C++ and/or object oriented? How will this affect
> developing with C++?
>
> 2. I know GNOME has gtkglarea; does KDE?
>
> 3. What else should a C++ developer know?
>
> Thanks!
I read that the Qt libraries on which KDE is based are written in C++,
while the GTK+ libraries on which GNOME is based are written in C. So KDE
is probably the more object-oriented desktop.
I'm personally a KDE fan. I have the newly-released KDE 2.0 running (as
part of Linux Mandrake 7.2, released a few days ago), and it's just dreamy.
Everything can be user-configured, and I like the new applications
(Konqueror, KOffice, KNode). I haven't tried doing any development with it
yet, but the KParts components it has are expected to speed up lots of
development projects.
Johno
------------------------------
** 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
******************************
