Linux-Misc Digest #891, Volume #18 Thu, 4 Feb 99 00:13:17 EST
Contents:
Re: Consumer Poll Says Microsoft Is Good For Consumers (William Wueppelmann)
Re: How to make it run faster? (Dave Philips)
configuring zip drive. (Pupeno)
Environment variables and C (Greg Cannon)
Re: Linux is not even in Windows 9X's class. ("Jim Ross")
Re: Consumer Poll Says Microsoft Is Good For Consumers (Zoltan Kocsi)
Re: syslogd and high cpu loads (Frank J. Dziuba)
Re: Clock Skew (Christopher Browne)
Re: How to attach a file on command line (Wonko The Sane)
Re: Sybase ASE for non-RedHat Linux distributions (Christopher Browne)
Re: > 64MB RAM (Christopher Browne)
Re: CD-RW as backup alternative (Christopher Browne)
Re: Test - Please Ignore ("Wael Sedky")
Re: upgrade to RedHat 5.2 and now my tape backup doesn't work (Stuart R. Fuller)
----------------------------------------------------------------------------
From: [EMAIL PROTECTED] (William Wueppelmann)
Crossposted-To: alt.destroy.microsoft,comp.os.linux.advocacy,comp.os.os2.advocacy
Subject: Re: Consumer Poll Says Microsoft Is Good For Consumers
Date: Wed, 03 Feb 1999 17:51:00 GMT
In an interview, the artist formerly known as Craig Kelley said:
>
>You make some pretty good points, but I think your target is off-
>center a bit. The US government is not trying to define 'what' an OS
>is, but whether or not Microsoft's immense OS power is allowing it to
>hedge other markets.
>
>If the Internet Explorer is so much better than Netscape and Microsoft
>is really "COMPETING in the free market" then why didn't they compete
>toe-to-toe with Netscape instead of "integrating" it into Windows? It
>seems to be conveniently coincidental that the OS needed an internet
>browser right when IE was at a fraction of the marketshare it enjoys
>today. The obvious answer to this question is that Microsoft
>integrated Windows into it's OS to strangle Netscape -- not very fair
>play, now is it.
The problem isn't so much that Microsoft is including Internet Explorer with
Windows. That shouldn't be a big deal in and of itself. Nor is it reall the
issue that Microsoft is linking the browser to Windows in innovative* new ways
to frustrate and annoy users who spent three years getting used to the Windows
95 interface after it changed radically from the Windows 3 interface. But I
digress.
The real problem, vis a vis the browser, is that Internet Explorer is not a
value-added feature -- it's a weapon. Microsoft has spent hundreds of
millions of dollars developing, promoting and distributing a Web browser that
it never expects to see any revenue, much less profit, from. To develop a
product solely for the purpose of destroying a rival product, or to make your
product comparatively better by artificially hampering a competing product is
contrary to the purpose of having market competition. If Microsoft faced
serious competition from, say, Apple, they might argue that developing an
integrated browser into Windows is a tactic designed to make Windows a more
attractive choice than a Macintosh, but to the best of my knowledge, they have
never argued anything to this effect (arguing that IE will provide enough
value to cause people to choose Windows over an alternative OS). If they
don't expect to make any money on the browser itself, and it isn't going to
induce Linux, OS2 or Mac devotees to switch to Windows, then why are they
spending so much money developing and promoting it?
It seems to me that the answer is that controlling the browser market would
give Microsoft enormous control over the World Wide Web. Once everyone is
using Internet Explorer, how long before IE supports innovative* new features
that you can take advantage of using Front Page 2000 in conjunction with a
Microsoft IIS 5 Web Server running on Windows 2000? Maybe IE will even just
not be able to quite understand the HTTP used by Apache, CERN, and the rest of
the world, so if you risk putting up your Web site on one of those out-of-date
Unix servers, people might run into difficulties viewing your site. That
can't be good for your business (in the M$ Web world, commercial Web sites
would be the only important part of the Web, I assume.) Moreover, once you
have IE hooked up with online auto-updating, upggrading your browser will
become automatic. That means when IE 6 is released, millions of people will
have the new browser in a matter of weeks. Maybe your company would like to
purchase a button on the toolbar that takes users to your Web site? And by
the way, IE 6 supports great new Microsoft HTML extensions, so you'll almost
certainly want to buy Front Page 2002 so you can update your site to take
advantage of these innovative* new features, especially since support for some
of the old features was dropped by IE6.
If Microsoft wants to create a proprietary Internet, I don't suppose there is
anything wrong with that, aside from the fact that it would be horrible. But
look at how this would happen: Microsoft is leveraging its Windows system and
its influence over computer retailers in order to get every computer shipped
with IE so firmly integrated into the system that users have to use it all the
time, and to make it harder to acquire and use an alternative Web browser.
Ultimately, Microsoft will be using it's control over the OS market in order
to lock the world into it's own proprietary network. Once this is done, the
holdouts (the ones who don't use Windows) will have to switch, or else be left
out, because there won't be Internet Explorer for Linux, and MSHTTP will not
be a published open standard. That's not giving consumers choice. That's
manipulating the market so that consumers have no choice.
And if that's not what Microsoft intends to do, then *why* have they spent so
much money and effort and legal headaches trying to *give* you software?
Something to think about, anyway.
* Innovative here is used in the sense that Microsoft uses the term, and not
the sense that the rest of the English-speaking world understands the term.
--
William
------------------------------
From: Dave Philips <[EMAIL PROTECTED]>
Crossposted-To:
alt.linux,alt.os.linux,alt.os.linux.redhat,comp.os.linux.setup,comp.os.linux.x,linux.redhat.misc
Subject: Re: How to make it run faster?
Date: Wed, 3 Feb 1999 23:09:19 -0500
| that site doesnt seem to exist!
i believe he meant http://www.ssc.com/lg/
dave
------------------------------
From: Pupeno <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.help,comp.os.linux.setup,comp.os.linux.questions
Subject: configuring zip drive.
Date: Thu, 04 Feb 1999 04:19:00 +0000
I'm trying to configure a non plus parallel zip drive.
I'm using 2.2.0
I have parport on the kernel and in my lilo:
append="parport=auto"
here's the log when I try to load ppa or imm modules:
~# insmod ppa
ppa: Version 2.03 (for Linux 2.0.0)
WARNING - no ppa compatible devices found.
As of 31/Aug/1998 Iomega started shipping parallel
port ZIP drives with a different interface which is
supported by the imm (ZIP Plus) driver. If the
cable is marked with "AutoDetect", this is what has
happened.
scsi : 0 hosts.
/lib/modules/2.2.0ac2/scsi/ppa.o: init_module: Device or resource busy
~# insmod imm
imm: Version 2.03 (for Linux 2.0.0)
scsi : 0 hosts.
/lib/modules/2.2.0ac2/scsi/imm.o: init_module: Device or resource busy
~#
If you have any idea of what's wrong, please tell me.
Thank you.
--
Pupeno:[EMAIL PROTECTED]
www.pupeno.com
PS: With 2.0.35 I used just to load the ppa module and then mount
/dev/sda4 and that's all. No it doesnt work with 2.2.0 nor 2.0.35. It
gave me the same error, but I'm not secure that the 2.0.35 kernel is
very well compiled.
------------------------------
From: Greg Cannon <[EMAIL PROTECTED]>
Crossposted-To: comp.lang.c,gnu.gcc.help
Subject: Environment variables and C
Date: Mon, 01 Feb 1999 11:33:36 +1100
People,
I'm writing a Linux application that needs to find certain files
in its install directory. I've noticed that many programs (e.g.
Netscape) solve this problem by using an environment variable.
Great. So how do I access an environment variable from within C?
I have a feeling there's some scheme similar to the argc/argv
method of passing command line parameters, but I don't know the
details. Can some kind person point me in the right direction?
Does it differ depending on your shell? (please say no)
Replies by email preferred. Ta.
- Greg.
--
Greg Cannon, Email: [EMAIL PROTECTED]
CSIRO Land and Water, Phone: +61-2-6246-5917
Black Mountain Labs, "It is better to have loved and lost than to
Canberra, Australia. have your lip stuck under a manhole cover."
------------------------------
From: "Jim Ross" <[EMAIL PROTECTED]>
Crossposted-To: alt.linux,alt.os.linux,comp.os.linux.advocacy
Subject: Re: Linux is not even in Windows 9X's class.
Date: Wed, 3 Feb 1999 23:30:28 -0500
[EMAIL PROTECTED] wrote in message <799v94$ldo$[EMAIL PROTECTED]>...
>I'm not going to get to crazy on this, but yes you are right "Linux is not
>even in Windows 9X's class", it far out performs Windows on every level and
>should not even be used in the same sentence.
You forgot to say "just kidding." Linux device support is not in Windows
9X's class. Nor is apps support for X. Not specifically a "Linux problem",
but nonetheless not yet in Windows class. Ease of use, nope. Linux is not
yet spoonfeeding as is Windows. Being powerful is ok, but just as important
to most is ease of use. I can appreciate choice and a good GUI and GUI apps
help me there. PPP works in Windows, in Linux for me it does not. Maybe I
could do even more to fix it, but the need is why Linux is not ready. I do
not want to have to troubleshoot it. I shouldn't have to.
Jim
------------------------------
From: Zoltan Kocsi <[EMAIL PROTECTED]>
Subject: Re: Consumer Poll Says Microsoft Is Good For Consumers
Date: 01 Feb 1999 19:37:54 +1100
Mayor Of R'lyeh wrote:
> I'll go along with that if you admit that it is a valid competitor for
> Windows. Many of the Linux guys want to have it both ways. When
> discussing price they want to claim that it is competition and want to
> include it but when you cite it as competition for Windows they claim
> that it doesn't count since it isn't a monolithic product from a
> single company. All I'm looking for is some consistency.
This *is* consistent. Linux is not a competitive product. It is an
alternative product. There is a major difference between the two.
Linux does not compete with Microsoft for Linux is not sold. It is
freely available. Linux support contracts are sold, just like I could
set up a business selling Microsoft Windows support. I would not
compete with Microsoft for what I sell is a service, what they sell
is a software license. For the very same reason RedHat does not
compete on the OS market, they compete on the OS support market.
They happen to specialise on Linux. As a courtesy service, they give
you a copy of the *free* Linux when you buy their support service.
Linux is *not* a product. It has no commercial value. Its existence
can hurt Microsoft for it is an *alternative*. However, the DoJ
can discount it because it is not on the *market* and that's what
the trial is about.
If I was selling petrol and I killed or tried to kill every other
petrol refiner company I would be dragged to court for monopolistic
practices. If a uni student builds a device that made the cars run
on water instead on petrol and gives the device away for free,
this will not change the fact that I was tried and probably found
guilty on the trial. The uni student may completely piss me off but
he does not compete on the petrol market even if he shrinks it.
Zoltan
--
+------------------------------------------------------------------+
| ** To reach me write to zoltan in the domain of bendor com au ** |
+--------------------------------+---------------------------------+
| Zoltan Kocsi | I don't believe in miracles |
| Bendor Research Pty. Ltd. | but I rely on them. |
+--------------------------------+---------------------------------+
------------------------------
From: [EMAIL PROTECTED] (Frank J. Dziuba)
Subject: Re: syslogd and high cpu loads
Crossposted-To: comp.os.linux.help,comp.os.linux.networking
Date: Thu, 04 Feb 1999 00:41:48 GMT
J. S. Jensen ([EMAIL PROTECTED]) wrote:
: "Frank J. Dziuba" wrote:
:
: > The syslogd process stays at about 70% cpu load!
:
: Is that CPU utilization or CPU load? Load is not expressed in percentages,
: rather abosolute values. Is your load average staggering around 1.0? Or do you
: just see 70% as the ps output. This is the time used over the time the process
: has been running (or in the run-queue), and does not indicate it is directly
: using 70% of the machines resources. It almost means that 30% of the time it
: sits in the run queue, something else is being scheduled.
>From top:
4:38pm up 3 days, 1:12, 19 users, load average: 3.30, 3.48, 3.12
125 processes: 120 sleeping, 4 running, 1 zombie, 0 stopped
CPU states: 10.7% user, 82.3% system, 8.9% nice, 7.4% idle
Mem: 127784K av, 123972K used, 3812K free, 78612K shrd, 35484K buff
Swap: 96384K av, 212K used, 96172K free 45524K cached
�
PID USER PRI NI SIZE RSS SHARE STAT LIB %CPU %MEM TIME COMMAND
8980 root 20 0 488 488 404 R 0 76.9 0.3 1325m syslogd
:
: > When we do some analysis on it we see what appears to be a lot or DNS lookups
: > on the
: > sending hosts.
:
: Place the IP address instead of the hostname in the the syslog.conf file.
Sorry, I meant the _receiving_ server is looking up the names of the _sending_
hosts.
Here is some vmstat output:
vmstat 5
procs memory swap io system cpu
r b w swpd free buff cache si so bi bo in cs us sy id
1 0 0 212 4480 35740 45772 0 0 3 21 127 50 6 78 16
2 0 0 212 4468 35740 45768 0 0 0 14 134 40 2 90 8
2 0 0 212 4772 35740 45772 1 0 1 18 137 65 7 85 7
3 0 0 212 4928 35740 45768 0 0 0 21 138 23 1 89 10
1 0 0 212 4864 35804 45768 0 0 0 13 127 44 17 75 8
2 0 0 212 4864 35804 45768 0 0 0 9 125 24 2 93 6
1 0 0 212 4864 35804 45768 0 0 0 18 135 34 8 83 9
3 0 0 212 4864 35804 45768 0 0 0 9 126 33 7 90 3
2 0 0 212 4864 35804 45776 0 0 0 12 129 26 4 89 6
This machine is being used for e-mail by about 10 people and for syslogs
from 2 other hosts. That's it.
If I kill syslogd _or_ remove the syslog forwarding from the _other_ two
hosts the load goes way down.
thanks
frank
------------------------------
From: [EMAIL PROTECTED] (Christopher Browne)
Subject: Re: Clock Skew
Reply-To: [EMAIL PROTECTED]
Date: Thu, 04 Feb 1999 00:50:18 GMT
On Thu, 04 Feb 1999 00:29:37 +0800, Yap Yin Onn <[EMAIL PROTECTED]> wrote:
>Hi,
>I keep getting the below message when ever I try to compile my kernel no
>matter v2.2 or v2.0
>"make[1]: *** Warning: Clock skew detected. Your build may be
>incomplete."
Apparently something is dropping trash into your clock, with the result
that the system realizes that since the clock time predates compile
times, it's not necessarily going to correctly compile all that is
"new."
Two things to do:
a) Sync your clock with something correct. xntp is a real good
candidate for this; see:
<http://www.google.com/search?q=xntp#=20>
b) Do a "make clean" to get rid of all the existing object code so that
a make can start afresh with file dates that are guaranteed to be up to
date.
You probably ought to do this anyways.
--
'There is no substitute for good manners, except, perhaps, fast
reflexes.' -- random unix fortune
[EMAIL PROTECTED] <http://www.hex.net/~cbbrowne/lsf.html>
------------------------------
From: [EMAIL PROTECTED] (Wonko The Sane)
Subject: Re: How to attach a file on command line
Date: Wed, 03 Feb 1999 22:05:14 GMT
On Tue, 02 Feb 1999 17:24:11 GMT, [EMAIL PROTECTED] (Topher)
wrote:
>Or use "ELM" option "-A"
>
>
>[EMAIL PROTECTED] wrote:
>
>>Take a look at http://alge.anart.no/linux/scripts/
>>
>>[EMAIL PROTECTED] wrote:
>>:I need to send a file to some users by email under certain circumstances, so I
>>:am looking for some utility or script that could send an attached file to some
>>:email address.
>>--
>>Anders Gulden Olstad @ Jeeves
>>RedHat 5.2 Linux kernel 2.0.36
>>
>>"Penguins are generally nice creatures"
I'm not sure if the syntax is exactly the same in Linux (I'm at work
and can't try it right now...) but in Solaris, I do:
uuencode filename filename | mailx -s "Subject" [EMAIL PROTECTED]
Note, you need to specify the filename twice for uuencode
Hope this helps,
Pete
------------------------------
From: [EMAIL PROTECTED] (Christopher Browne)
Crossposted-To: comp.databases.sybase
Subject: Re: Sybase ASE for non-RedHat Linux distributions
Reply-To: [EMAIL PROTECTED]
Date: Thu, 04 Feb 1999 00:50:17 GMT
On Wed, 3 Feb 1999 09:28:40 -0500, Scott Michels <[EMAIL PROTECTED]> wrote:
>Tarfile? Slackware? Grok? I gotta find out more about Linux some time if
>for no other reason than to find out what some of these terms mean.
>Something else to put on my ever growing to-do list...
- Tarfile
In the same way that a "zip file" is an archive generated by "Zip," and
a "Word file" is a file generated by Microsoft Word, a "Tarfile" is an
archive produced by the utility "tar."
"tar" stands for "tape archive," although since it is implemented as a
typical UNIX program, it may be used for other diverse purposes
associated with assembling together files into a single archive file.
- Slackware
Slackware was the first really famous Linux "distribution." It postdates
SLS, and thus is not the *first* distribution, but greatly popularized
the idea, which is one of the "Linux distinctives."
It was and is a commercial package just like Caldera, Red Hat, and
SuSE's respective distributions; people tend to rave either positively
or negatively about the way that it uses not-terribly-thinly-veiled
"tarfiles" as its packaging scheme. (Which is greatly less
sophisticated than RPM or dpkg, assortedly to benefit or detriment
depending on your point of view.)
See:
<http://www.slackware.com>
<http://www.slackware.net>
<http://www.slackware.org>
- Grok
A word coined by Robert A Heinlein in the book "Stranger in a Strange
Land," where it is a Martian word meaning literally "to drink" and
metaphorically "to be one with."
It connotes intimate and exhaustive knowledge. When used of programs,
it may merely connote "sufficient understanding."
--
"If Bill Gates had a dime for every time a Windows box crashed...
.... Oh, wait a minute, he already does."
[EMAIL PROTECTED] <http://www.ntlug.org/~cbbrowne/lsf.html>
------------------------------
From: [EMAIL PROTECTED] (Christopher Browne)
Subject: Re: > 64MB RAM
Reply-To: [EMAIL PROTECTED]
Date: Thu, 04 Feb 1999 00:50:22 GMT
On 03 Feb 1999 11:07:36 -0800, Michael Powe <[EMAIL PROTECTED]> wrote:
>-----BEGIN PGP SIGNED MESSAGE-----
>Hash: SHA1
>
>>>>>> "Ben" == Ben Russo <[EMAIL PROTECTED]> writes:
>
> >> Kernels 2.0.36 and higher should automatically find the extra
> >> RAM. Mine did when I added an extra 64 MB (you did check that
> >> it was being found in the BIOS settings didn't you?) .
>
> Ben> For 2.0.x kernel's only finds up to 1023 MB, (the kernel
> Ben> won't boot if more than that!!) for 2.2.x kernel's only
> Ben> finds up to 2GB.
>
> Ben> Sorely limited if you ask me. I have HP-UX and SOLARIS boxes
> Ben> that are limited to 4GB or 2TB respectively.
>
> Ben> Linux needs to move to 64 bit!!
>
>Hmm, and how would that be useful on a PC, using a 32-bit processor?
As I understand it, Linux has an inherent limitation on IA-32 to address
2GB of RAM.
Reports suggest that Linus is aware of this, and has indicated that he's
not planning to change this;
"Go for a 64 bit system if you need that much RAM."
At this point, that's not too much of a limitation, as few IA-32
motherboards seem to support more than about a half GB of RAM.
Does it freak anyone else out that L2 caches these days have more
storage than we used to have as the combined RAM and disk space on
machines of yesteryear? My first computer had 32K of RAM, and 120K
floppies...
--
Where do you *not* want to go today? "Confutatis maledictis, flammis
acribus addictis" (<http://www.hex.net/~cbbrowne/msprobs.html>
[EMAIL PROTECTED] <http://www.hex.net/~cbbrowne/oldcomp.html>
------------------------------
From: [EMAIL PROTECTED] (Christopher Browne)
Crossposted-To: comp.unix.solaris
Subject: Re: CD-RW as backup alternative
Reply-To: [EMAIL PROTECTED]
Date: Thu, 04 Feb 1999 00:50:28 GMT
On 3 Feb 1999 16:54:37 -0600, Leslie Mikesell <[EMAIL PROTECTED]> wrote:
>In article <[EMAIL PROTECTED]>,
>Ben Russo <[EMAIL PROTECTED]> wrote:
>>"Kerry J. Cox" wrote:
>>> Howdy,
>>> I was wondering what the feasibility would be to install a CD-RW on my
>>> Linux box and have it be used to backup all my Solaris machines here at
>>> work? How would I go about backing up critical files from off the Sun
>>> Solaris machines then onto a CD-RW? Would you recommend using ufsdump
>>> or tarring them up or setting up a selective process to back up certain
>>> files?
>
>>Not enough capacity, it would take 3 CD's to back up every 2Gig disk.
>>Network latency would kill your CDwrites, so you would have to copy
>>an entire disks contents to the writer system, then build the CD image,
>>then write it. Way too slow.
>
>This really depends on how much is critical and how often it needs
>to be saved. 600K is a lot of data if you are only concerned about
>what you have typed yourself. It it very easy to let a linux
>box NFS mount the data directory (although I've always had some
>trouble with Solaris on the other end...) and just make iso9660
>images with both the unix and MS long file extensions so you could
>get to the files from just about anything. It is also a good way
>to archive logfiles and the like that just accumulate forever.
That's 660MB, and it is reasonably possible to come up with useful
subsets that *can* be backed up.
>>CD's work ok for backing up a source tree on
>>major version number or for backing up log files once a month. But
>>definitely
>>not a nightly or weekly solution for lot's of boxes.
>
>You could do a CD-RW nightly cycling through several before erasing
>and re-using. It is slow, but a person is only needed to change
>the disk.
Quite frankly, I'd rather go with the CD-R approach. In order to
establish a decently safe set of CD-RW's to represent a proper schedule,
you need at least 10-15 of 'em, at which point you've spent $300, at
which point you could start getting decent economies of scale on the
disposables.
>>Also, you won't find any archive software solutions that work off of CD's.
>
>If you write in a standard format and can organize things to fit, you
>don't need anything special to get the files back.
Indeed. If at all possible, write out CD data in a *useful* form.
And if you're burning CD-R coasters, it might as well be as useful as
possible.
--
"take USABLE from UNSTABLE and you get NT"
[EMAIL PROTECTED] <http://www.hex.net/~cbbrowne/lsf.html>
------------------------------
From: "Wael Sedky" <[EMAIL PROTECTED]*>
Subject: Re: Test - Please Ignore
Date: Wed, 3 Feb 1999 23:46:01 -0800
I never did this before, but usually when i do something like that i get
tons of emails telling me about the million test newsgroups out there and
they complain about some bandwidth nonsense.
Sorry I couldn't help it.
>Testing please ignore .....
------------------------------
From: [EMAIL PROTECTED] (Stuart R. Fuller)
Subject: Re: upgrade to RedHat 5.2 and now my tape backup doesn't work
Reply-To: [EMAIL PROTECTED]
Date: Thu, 04 Feb 1999 05:00:02 GMT
Daddy Rabbit ([EMAIL PROTECTED]) wrote:
: After upgrading from Redhat 5.1 to 5.2, my scsi tape quit working. The
: message I get is that the 'device nst0 is not working'. My question
: is:
:
: How do I get the device up and running again?
Post the command(s) you entered and the error message(s) you received. Post
them exactly as entered and displayed.
The type of SCSI tape might also be of interest.
Stu
------------------------------
** 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
******************************
