Linux-Misc Digest #117, Volume #27 Thu, 15 Feb 01 12:13:02 EST
Contents:
Re: Strange phenomena ("Puchta Milos")
Re: Linux Error: 23: Too many open files in system (Mark Hahn)
ipmasqadm forwarding question (Warren Bell)
Re: Optimize for Speed? (Mark Hahn)
Re: MS to Enforce Registration - or Else (Dan Mercer)
Re: MS to Enforce Registration - or Else (Dan Mercer)
Re: MS to Enforce Registration - or Else (Robert Surenko)
Re: MS to Enforce Registration - or Else (Dan Mercer)
Solved - slow xinetd service connection (The Archimage)
Re: MS to Enforce Registration - or Else (Robert Surenko)
Re: Solved - slow xinetd service connection (The Archimage)
Re: First impressions of RH7.1 beta ([EMAIL PROTECTED])
Re: Keyboard mapping and sterling symbol... (Jem)
----------------------------------------------------------------------------
From: "Puchta Milos" <[EMAIL PROTECTED]>
Subject: Re: Strange phenomena
Date: Thu, 15 Feb 2001 16:18:13 +0100
Solved by video card change.
"Puchta Milos" <[EMAIL PROTECTED]> wrote in message
news:96etmd$lmn$[EMAIL PROTECTED]...
> I have just installed host with 4 NICs. When opening Netscape I can see
> strange random picture in the vicinity of netscape windows.
> Sometimes I obtain a message on the network error.
>
> Any idea?
>
> TIA
> Milos
>
>
------------------------------
From: Mark Hahn <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.development.system,comp.os.linux.help,comp.os.linux.setup
Subject: Re: Linux Error: 23: Too many open files in system
Date: 15 Feb 2001 15:12:59 GMT
>> Linux Error: 23: Too many open files in system
...
>> Is the number of open files adjustable? How?
> You can change the limit in /usr/src/linux.../include/linux/limits.h
> file and recompile the kernel. Default in Redhat 6.2 is 256.
on modern kernels, "echo bignumber > /proc/sys/fs/file-max".
on obsolete 2.2 kernels, /proc/sys/kernel/file-max
------------------------------
From: Warren Bell <[EMAIL PROTECTED]>
Subject: ipmasqadm forwarding question
Date: Thu, 15 Feb 2001 07:32:28 -0800
I have a main server machine running ipchains and want to forward all
requests to a certain host to an internal machine. I use a dynamic dns
service and have different hostnames assigned to my IP. Is it possible
with ipmasqadm to forward all request to foo.dyndns.org to an internal
machine while all requests to bar.dyndns.org go to the main machine? I
was thinking somthing like:
ipmasqadm portfw -f
ipmasqadm portfw -a -P tcp -L foo.dyndns.org -R 192.168.0.2
ipmasqadm portfw -a -P udp -L foo.dyndns.org -R 192.168.0.2
I haven't been able to get this working. Am I doing it right or can it
even be done; forwarding all ports by hostname?
------------------------------
From: Mark Hahn <[EMAIL PROTECTED]>
Crossposted-To: alt.os.linux.mandrake,alt.os.linux.redhat,comp.os.linux.help
Subject: Re: Optimize for Speed?
Date: 15 Feb 2001 15:23:29 GMT
> Mandrake 7.2 with the new KDE2 is painfully slow! so slow in fact, that
> I backed out the complete installation, went back to Mandrake 7.1 and
uh, don't confuse Linux with all the silly GUI crap people slather on top.
> I run an AMD K6-3 400Mhz with 256Mb of RAM. not really a
> supercomputer... but quite functional for linux.
plenty. you just need to avoid the eye-candy. socket7 machines are
dreadfully bandwidth-starved, so don't do well on code that's been
written on p6/k7-class machines.
------------------------------
From: [EMAIL PROTECTED] (Dan Mercer)
Crossposted-To: comp.os.linux.advocacy
Subject: Re: MS to Enforce Registration - or Else
Date: 15 Feb 2001 15:55:12 GMT
Their is certainly a strong element of faith in science. We
accept the existence of that we have no direct knowledge (muons,
for instance) based upon the assurances of people we have no
direct knowledge. Is it really that far a stretch to believe
Christ existed based on the works of Matthew, Mark, Luke and John
than to believe black holes exist.
Just as the religious rely on the collective experiences of those
who have gone before, so does science. You certainly do not
perform experiments to prove every article of science you encounter,
you rely on faith that your predecessors performed their experiments
correctly. Following the cold fusion debate, you can witness the
uproar tha ensues when experiments appear to challenge the preheld
beliefs. The reaction of physicists is to deny and attack the new
evidence just as fundamentalists attack evolution. If cold fusion
yet proves out and is not the likely result of poorly conducted
experimentation, the howls from physicists will equal the howls
of those who originally shouted down the germ theory of Pasteur or
the works of Charles Darwin.
In article <[EMAIL PROTECTED]>,
"Peter T. Breuer" <[EMAIL PROTECTED]> writes:
> In comp.os.linux.misc John Hasler <[EMAIL PROTECTED]> wrote:
>> Robert Surenko writes:
>>> It also takes faith to believe the Universe is as appears to the 5
>>> senses.
>
>> I don't.
But the universe isn't as our senses report it. For instance,
you and I and a European bee may look at flower and think it is black.
The European bee, in fact, will ignore it. But an African or Africanized
bee sees a wider spectrum than either humans or their European
counterparts. So the African bee will visit that flower.
There are those who can "hear" radio waves - they hear the Northern
Lights as waves booming onto the shore - which may explain the
large number of people who began having problems hearing a background
noise after the ELF arrays began their work. Their are additional
senses we do not have (the electric field sense shared by electric eels
and some sharks, echolocation) and some (orientation) which may be
shared by only a percentage of humans (in experiments in which people
were blindfolded and soundproofed, then driven around in circles
to deliberately disorient them. They were then asked to poin to their
point of origin. Most pointed in random directions. But s significant
minority were able to point in the correct direction better than 80%
of the time).
>
> Well, there are issues of sanity involved here. Doubting the evidence of
> your own senses leaves you in a difficult position.
Then, to your mind the blind should not believe in light, nor the
deaf ins sound?
Insanity is a
> probable outcome (although that is a sane response to the predicament).
>
>>> Because of this it also takes great faith to not believe ( or believe
>>> not) in God.
>
>> Nonsense.
>
>>> Science and logic are a religion.
>
>> More nonsense.
>
> Agreed. It is after all, very difficult to program a computer using
> religious beliefs as a basis for your programming.
Faith is essential in programming a computer. Unless you are actually
programming the microcode of the CPU, you rely upon the belief that
what you write will do what you want, a belief that is all to often
shaken. Eveen the assemby writer must interact with other people's
work if only the BIOS, and must have faith that the work they did
was correct. Quite simply, we cannot confirm every postulate
we use in life. We cannot even confirm that ever postulate is ultimately
confirmable.
I tend to view that
> as evidence that scientific belief is qualitatively different, since
> believing in scientific principles like observation, no-interpretation,
> experiment, hypothesiis formation and refutation, does help you program
> a computer.
>
> On the other hand, so does alcohol and coffee.
>
> Peter
--
Dan Mercer
[EMAIL PROTECTED]
Opinions expressed herein are my own and may not represent those of my employer.
------------------------------
From: [EMAIL PROTECTED] (Dan Mercer)
Crossposted-To: comp.os.linux.advocacy
Subject: Re: MS to Enforce Registration - or Else
Date: 15 Feb 2001 15:59:40 GMT
In article <[EMAIL PROTECTED]>,
John Hasler <[EMAIL PROTECTED]> writes:
> Peter T. Breuer writes:
>> Well, there are issues of sanity involved here. Doubting the evidence of
>> your own senses leaves you in a difficult position.
>
> Not quite what I took him to mean, but I _do_ doubt the evidence of my own
> senses in that I doubt single observations and try to rely on a
> preponderance of evidence. Experience has taught me that my senses are
> easily fooled and my memory unreliable. It has also taught me that this
> is true of other people, though most deny it.
--
John Hasler
[EMAIL PROTECTED]
Dancing Horse Hill
Elmwood, Wisconsin
So, if you had personal experience of something inexplicable by science,
would you be more likely to believe that Science doesn't have the answers
for everything? I know that there are at least 3 incidents in my life
that can't be explained by any Physical laws I know.
--
Dan Mercer
[EMAIL PROTECTED]
Opinions expressed herein are my own and may not represent those of my employer.
------------------------------
From: Robert Surenko <[EMAIL PROTECTED]>
Subject: Re: MS to Enforce Registration - or Else
Crossposted-To: comp.os.linux.advocacy
Date: Thu, 15 Feb 2001 16:13:49 GMT
In comp.os.linux.misc John Hasler <[EMAIL PROTECTED]> wrote:
> Robert Surenko writes:
>> It also takes faith to believe the Universe is as appears to the 5
>> senses.
> I don't.
Good, Materialist bore me.
>> Because of this it also takes great faith to not believe ( or believe
>> not) in God.
> Nonsense.
Perhaps, given your answer above.
>> Science and logic are a religion.
> More nonsense.
The only way that the statement, "Science and logic are a religion"
can be argued about is to claim that they are merly tools. They
do not lead to "Truth" but to a organized way to think of things.
Warning, when I use the word sophist I use the original meaning,
not the insult in the dictionary. :-)
A sophist would claim that they have no understanding of the truths
of either the spiritual world or the material world. They have built
a model of what appears around them that works. Smart sophist have
a better model than less-smart sophists. It has no relationship
with knowlege or truth, just what appears to work.
Given that there are very few writers that are "true" ( sorry )
sophists it is a extremly difficult phylosophy to study.
With all my reading I've only been able to identify 3 ways
of thinking of it all. Spiritualism, Materialism and Sophistry.
All phylosophies and religions are a mixture of these.
Is this what you are saying?
> --
> John Hasler
> [EMAIL PROTECTED] (John Hasler)
> Dancing Horse Hill
> Elmwood, WI
--
=============================================================================
- Bob Surenko [EMAIL PROTECTED]
- http://www.fred.net/surenko/
=============================================================================
------------------------------
From: [EMAIL PROTECTED] (Dan Mercer)
Subject: Re: MS to Enforce Registration - or Else
Date: 15 Feb 2001 16:12:11 GMT
In article <_jMi6.33327$[EMAIL PROTECTED]>,
[EMAIL PROTECTED] writes:
>
> At any rate, the point here is that in order to claim that all people
> with religious belief must be "anti-scientific," you have to reject a
> significant body of scientists responsible for some _good_ science.
Gregor Mendel and Roger Bacon come immediately to mind.
--
Dan Mercer
[EMAIL PROTECTED]
Opinions expressed herein are my own and may not represent those of my employer.
------------------------------
From: The Archimage <[EMAIL PROTECTED]>
Crossposted-To:
alt.linux,alt.os.linux,comp.os.linux.networking,comp.os.linux.security,comp.os.linux.setup
Subject: Solved - slow xinetd service connection
Date: Thu, 15 Feb 2001 10:33:03 -0600
History:
I installed RH7 and activated imap. Checking mail from work (behind a
firewall), imap connections took 45-60 seconds to establish, and then
after a few minutes, dropped. I had to re-establish the connection, and
it again took 45-60 seconds. I thought it was a DNS issue, but all
hosts could resolve each other forward and backwards.
A co-worker was working a similar issue with sendmail, and we discovered
the problem was that the server was trying to do an ident lookup. Since
I was behind a firewall, this took 45-60 seconds to timeout before I
could get to my mail.
The solution is to set up an ipchains rule on the server, blocking
outbound ident lookups:
Kernel 2.2.x with ipchains:
ipchains -A output -p tcp -d 0.0.0.0/0.0.0.0 113 -j REJECT
Kernel 2.4.x with iptables:
iptables -i OUTPUT -p tcp -d 0.0.0.0/0.0.0.0 113 -j REJECT
Hope this helps.
--
===================================================================
-- Thomas D. Cameron --
Red hat Certified Engineer -- TurboLinux Certified Instructor
-- Certified NetWare Engineer --
Microsoft Certified Systems Engineer -- Microsoft Certified Trainer
===================================================================
deja search keywords:
red hat 7 imap pop3 telnet talk wu-ftpd slow xinetd connection
------------------------------
From: Robert Surenko <[EMAIL PROTECTED]>
Subject: Re: MS to Enforce Registration - or Else
Crossposted-To: comp.os.linux.advocacy
Date: Thu, 15 Feb 2001 16:44:23 GMT
In comp.os.linux.misc Peter T. Breuer <[EMAIL PROTECTED]> wrote:
> In comp.os.linux.misc John Hasler <[EMAIL PROTECTED]> wrote:
>> Robert Surenko writes:
>>> It also takes faith to believe the Universe is as appears to the 5
>>> senses.
>> I don't.
> Well, there are issues of sanity involved here. Doubting the evidence of
> your own senses leaves you in a difficult position. Insanity is a
> probable outcome (although that is a sane response to the predicament).
Although I am frequently accused of insanity...
But really, although I don't nessesarily agree with it...
It can be argued that all truth, knowlege or belief be suspended.
Look at how radically different the view of what the Universe is has
changed in the last 100 years. Watch how a scientist can switch
rapidly between newtonian physics and quantium.
Every chemist I know will tell you that the Borh model is totally
false yet amazingly accurate. (Sorry about the spelling)
All without too much insanity.
>>> Because of this it also takes great faith to not believe ( or believe
>>> not) in God.
>> Nonsense.
>>> Science and logic are a religion.
>> More nonsense.
> Agreed. It is after all, very difficult to program a computer using
> religious beliefs as a basis for your programming. I tend to view that
> as evidence that scientific belief is qualitatively different, since
> believing in scientific principles like observation, no-interpretation,
> experiment, hypothesiis formation and refutation, does help you program
> a computer.
2 points. First, programming can be argued to be a religious belief.
All you must have faith in is that this machine will do the same thing
given the same input. Now I have faith that this is so, but how would
you prove that repeatability is a fact or a truth. It is convenient
but not proven.
I've heard many programers not believe this when they trash their
linux hard drive for the 10th time.
Secondly, re-read your last paragraph. You say that because it helps
you program, scientific belief is qualitatively different. Yes,
your faith has helped you in times of trouble, so has mine.
I've seen no evidence that Science is not a religion, only that
it's a more convienent religion when programming a computer.
> On the other hand, so does alcohol and coffee.
Agreed!
> Peter
--
=============================================================================
- Bob Surenko [EMAIL PROTECTED]
- http://www.fred.net/surenko/
=============================================================================
------------------------------
From: The Archimage <[EMAIL PROTECTED]>
Crossposted-To:
alt.linux,alt.os.linux,comp.os.linux.networking,comp.os.linux.security,comp.os.linux.setup
Subject: Re: Solved - slow xinetd service connection
Date: Thu, 15 Feb 2001 10:46:18 -0600
AACK!! I brain cramped! The correct syntax for iptables is:
iptables -A OUTPUT -p tcp --destination-port 113 -d 0.0.0.0/0.0.0.0 -j
REJECT
My most humble apologies!
Thomas
The Archimage wrote:
>
> History:
>
> I installed RH7 and activated imap. Checking mail from work (behind a
> firewall), imap connections took 45-60 seconds to establish, and then
> after a few minutes, dropped. I had to re-establish the connection, and
> it again took 45-60 seconds. I thought it was a DNS issue, but all
> hosts could resolve each other forward and backwards.
>
> A co-worker was working a similar issue with sendmail, and we discovered
> the problem was that the server was trying to do an ident lookup. Since
> I was behind a firewall, this took 45-60 seconds to timeout before I
> could get to my mail.
>
> The solution is to set up an ipchains rule on the server, blocking
> outbound ident lookups:
>
> Kernel 2.2.x with ipchains:
> ipchains -A output -p tcp -d 0.0.0.0/0.0.0.0 113 -j REJECT
>
> Kernel 2.4.x with iptables:
> iptables -i OUTPUT -p tcp -d 0.0.0.0/0.0.0.0 113 -j REJECT
>
> Hope this helps.
> --
> -------------------------------------------------------------------
> -- Thomas D. Cameron --
> Red hat Certified Engineer -- TurboLinux Certified Instructor
> -- Certified NetWare Engineer --
> Microsoft Certified Systems Engineer -- Microsoft Certified Trainer
> -------------------------------------------------------------------
> deja search keywords:
> red hat 7 imap pop3 telnet talk wu-ftpd slow xinetd connection
------------------------------
From: [EMAIL PROTECTED]
Subject: Re: First impressions of RH7.1 beta
Date: Thu, 15 Feb 2001 13:17:57 GMT
>2. JDK
>
>Supposedly JDK1.3 works with kernel 2.4, but I'm having problems.
>I'm using the same Jdk1.3 rpm file I was using on my 6.2 setup (and
>same classpath) but when I just try to compile Helloworld, I get:
>
>% javac Helloworld.java
>/usr/java/jdk1.3/bin/javac: /usr/bin/cut: No such file or directory
>/usr/java/jdk1.3/bin/i386/native_threads/javac: error while loading
>shared libraries: libjvm.so: cannot load shared object file: No such
>file or directory
>
>..and yet if I do: "which javac"
>I get: /usr/java/jdk1.3/bin/javac
>
Simple solution. Turns out "cut" was at /bin/cut instead of
/usr/bin/cut so it only needed a symlink (or make a copy to
/usr/bin)
The ppp problem (below) is still strange. Looking at
/var/lib/messages it's conecting and even getting assigned a dynamic
IP address by my ISP.... but the local browser and terminal can't get
through to it. Anyone else having this?
>3. ppp
>
>Weird. I used the RH Dialup Config tool (frontend to WvDial) the same
>as before. It connects.... and a check of /var/lib/messages shows
>that it's connecting to my ISP, but Mozilla just waits after putting
>in a URL, and trying to telnet to my shell account from the Terminal
>just waits until it eventually gives back: "Host name lookup failure"
>It's as if it's connected, but nothing will go through. The last 2
>lines in /var/lib/messages shows this:
>
>Feb 13 21:36:57 localhost pppd[1568]: Exit.
>Feb 13 21:36:57 localhost modprobe: modprobe: Can't locate module ppp0
>
------------------------------
From: Jem <[EMAIL PROTECTED]>
Crossposted-To:
comp.os.linux.development.system,comp.os.linux.embedded,comp.os.linux.setup
Subject: Re: Keyboard mapping and sterling symbol...
Date: Thu, 15 Feb 2001 16:49:39 -0000
OK... I'm going to follow up my own post as noone else seems to have a
clue about it and see if this extra hint helps...
My application is fired up by scripts in rc.d using "open" so as it
opens in a new console window. I have just noticed that pressing the UK
Sterling symbol in this window results in the pound sign being correctly
displayed.
Also if I'm quick enough between typing exit and the next login prompt
appearing I can make a pound symbol appear there too.
BUT As soon as the login prompt appears, and from then on, I cannot get
pound symbols at all (sorry, in case this isn't clear, I can still get
them in the other console windows).
Anyone? Anywhere? Any ideas? What extra processing is performed by the
shell that could possibly be filtering out this single character from
the keymap? I have been using the ash shell and have also tried bash but
neither makes a difference.
Jeremy
In article <[EMAIL PROTECTED]>,
[EMAIL PROTECTED] says...
> Argh! Major hair loss time ;-)
>
> I am trying desperately hard to link up a UK keyboard to my Embedix
> based system. Embedix defaults to a US keyboard (not surprising) and
> doesn't seem to provide any tools (e.g. loadkeys) for configuring as a
> UK keyboard.
>
> Me being a smart-arse took the consoleutils rpm from RedHat 6.2,
> stripped out the documentation and merged the rest into my Embedix
> system and then typed "loadkeys uk"... Yippeee! It all seems to work...
> all of it that is, except the "�" (uk sterling symbol) key. It produces
> nothing. Not a thing.
>
> Why? How do I fix it? what's the matter? I've copied all the files that
> I use on my full RedHat 6.2 system, double checked loads of settings and
> yet nothing, no quid symbol.
>
> Anyone any hints?
>
> Jeremy
>
------------------------------
** 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
******************************
