Linux-Misc Digest #328, Volume #20 Mon, 24 May 99 09:13:26 EDT
Contents:
Re: Can't communicate through 2nd NIC (Mike)
Re: Commercially speaking....? (Paul Black)
DUMP errors (ger)
Re: Linux or linux? ("D. Vrabel")
Re: Linux's Last Chance (Mark Tranchant)
Hard drive flipping bits! (Andy Bianchi)
Re: Linux or linux? (jane chav)
Re: Will a SupraExpress 56i modem run under linux? (Rob Clark)
Re: Root Password lost... (Mark Forsyth)
SETI CPU/OS Comparisons (Jason Jordan)
Re: NT the best web platform? (Matthias Warkus)
Re: good, free ORB w/ C++ & Java lang support (Chris Sherlock)
----------------------------------------------------------------------------
From: Mike <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.hardware,comp.os.linux.networking
Subject: Re: Can't communicate through 2nd NIC
Date: Sun, 23 May 1999 05:10:50 +0000
Steve Snyder wrote:
>
> On Fri, 21 May 1999 23:53:51 -0700, Harley Waagmeester wrote:
>
> >IPADDR="24.4.162.173" <-------
> >
> >You have assigned your internet /cable modem address to your eth1 network card,
> >that won't work.
> >Give your eth1 card a local ip like you have for the eth0 card.
> >I'll give a tough sketch of what needs to happen:
> >if you give eth1 an ip address of 192.168.0.13,
> >Then you need :
> >route add 24.4.162.173 gw 192.168.0.13
> >route add default gw 24.4.162.173
> >
> >I'm probably wrong about the syntax
> >Just give the eth1 a local ip address and leave the gateway address as
> >24.4.162.173,
> >and maybe the startup scripts will set the default route up correctly
> >
> >I hope someone explains this better, or gives the right numbers to plug into the
> >config files :))
> >
> >The point is that you want a local ip for the eth1 interface card and use that as
> >the gateway out
> >of the machine, and the default route is a "logical route" that flows through the
> >hardware route.
> >
> >The 24.4.162.173 is the address of the cable modem device
>
> More info:
>
> In my last message I wrote that I changed my config as recommended above,
> but saw no difference in behavior (ping/telnet still didn't work.) However
> there *is* a difference in the output of tcpdump.
>
> Output with previous config:
> ----------------------------
> # /usr/sbin/tcpdump -i eth1
> tcpdump: listening on eth1
> 15:17:58.005410 arp who-has 128.63.2.53 tell ct52636-a
> 15:17:58.005529 arp who-has 128.9.0.107 tell ct52636-a
> 15:17:58.008410 arp who-has 198.41.0.4 tell ct52636-a
> 15:17:58.026120 arp who-has 128.63.2.53 tell 24.4.162.129
> 15:17:58.035820 arp who-has 128.9.0.107 tell 24.4.162.129
> 15:17:58.045582 arp who-has 198.41.0.4 tell 24.4.162.129
> 15:17:59.005240 arp who-has 198.41.0.4 tell ct52636-a
> 15:17:59.005254 arp who-has 128.9.0.107 tell ct52636-a
> 15:17:59.005275 arp who-has 128.63.2.53 tell ct52636-a
> 15:17:59.025473 arp who-has 198.41.0.4 tell 24.4.162.129
> 15:17:59.036064 arp who-has 128.9.0.107 tell 24.4.162.129
> 15:17:59.045111 arp who-has 128.63.2.53 tell 24.4.162.129
> 15:18:00.005232 arp who-has 128.63.2.53 tell ct52636-a
> 15:18:00.005242 arp who-has 128.9.0.107 tell ct52636-a
> 15:18:00.005264 arp who-has 198.41.0.4 tell ct52636-a
> 15:18:00.027275 arp who-has 128.63.2.53 tell 24.4.162.129
> 15:18:00.038154 arp who-has 128.9.0.107 tell 24.4.162.129
> 15:18:00.045851 arp who-has 198.41.0.4 tell 24.4.162.129
> 15:18:02.005334 arp who-has 128.9.0.107 tell ct52636-a
> 20 packets received by filter
> 0 packets dropped by kernel
>
> With new configuration:
> -----------------------
> # /usr/sbin/tcpdump -i eth1
> tcpdump: listening on eth1
> 11:50:49.001747 arp who-has 24.4.162.173 tell ct52636-a
> 11:50:50.001746 arp who-has 24.4.162.173 tell ct52636-a
> 11:50:53.001844 arp who-has 24.4.162.173 tell ct52636-a
> 11:50:54.001733 arp who-has 24.4.162.173 tell ct52636-a
> 11:50:55.001739 arp who-has 24.4.162.173 tell ct52636-a
> 11:50:56.001900 arp who-has 24.4.162.173 tell ct52636-a
> 11:50:57.001739 arp who-has 24.4.162.173 tell ct52636-a
> 11:50:58.001734 arp who-has 24.4.162.173 tell ct52636-a
> 11:50:59.021801 arp who-has 24.4.162.173 tell ct52636-a
> 11:51:00.021729 arp who-has 24.4.162.173 tell ct52636-a
> 11:51:01.021733 arp who-has 24.4.162.173 tell ct52636-a
> 11:51:04.001822 arp who-has 24.4.162.173 tell ct52636-a
> 11:51:05.001742 arp who-has 24.4.162.173 tell ct52636-a
> 11:51:06.001733 arp who-has 24.4.162.173 tell ct52636-a
> 11:51:07.381810 arp who-has 24.4.162.173 tell ct52636-a
> 11:51:08.381734 arp who-has 24.4.162.173 tell ct52636-a
> 11:51:09.381737 arp who-has 24.4.162.173 tell ct52636-a
> 17 packets received by filter
> 0 packets dropped by kernel
>
> Note that during this output I am not explicly addressing eth1. If may be
> just a coincident that the tcpdump has changed, but there you go.
>
> Although there apparently is activity on eth1, the RX seen with ipconfig
> are not increment regularly during this activity, though the TX count does
> increment. Hmm. This is an example of output:
>
> # /sbin/ipconfig eth1
> eth1 Link encap:Ethernet HWaddr 00:60:97:C8:01:C8
> inet addr:192.168.0.18 Bcast:192.168.0.255 Mask:255.255.255.0
> UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
> RX packets:44 errors:0 dropped:0 overruns:0 frame:0
> TX packets:3014 errors:0 dropped:0 overruns:0 carrier:0
> collisions:0 txqueuelen:100
> Interrupt:10 Base address:0xe800
>
> Any thoughts?
>
> ***** Steve Snyder *****
try "man route"
------------------------------
From: Paul Black <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.advocacy,uk.comp.os.linux
Subject: Re: Commercially speaking....?
Date: Mon, 24 May 1999 12:27:29 +0100
Reply-To: Paul Black <[EMAIL PROTECTED]>
Jamie <[EMAIL PROTECTED]> wrote:
>
> Iain Georgeson wrote:
> >
> >
> > Iain, forced to use DOS every working day.
>
> It is amazing the number of people that do not realise that Win 95 is
> running on top of DOS just like 3.x did. They just put a (not so)
> pretty picture up at the begining to hide the DOS stuff at boot time.
It's the subject of Caldera's suit against MS (along with the other
dubious tactics used to shut DR-DOS out of the market).
Paul
------------------------------
From: ger <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.help
Subject: DUMP errors
Date: Mon, 24 May 1999 12:02:54 +0100
Folks,
I've been trying to get an Archive Python (90M, 4-8GB, DDS, SCSI-II)
tape drive working with dump on my Redhat 5.2 system (kernel currently
at 2.2.6). The drive is connected externally to an AHA1542, and all are
recognised quite happily by the kernel. The device is accessible and
responds to 'mt' manipulations, and also responds to 'restore -i'.
When I try to dump to it, however, using the following line:
/sbin/dump 0usfd 5000 /dev/nst0 42500 / (I know the values look odd, but
they were obtained from a reliable source)
I get the following error:
DUMP: estimated 843315 tape blocks on 0.65 tape(s).
DUMP: dumping (Pass III) [directories]
DUMP: master/slave protocol botched.
The strange thing is that this command worked quite well until I issued
an 'mt erase' on one of the tapes. Subsequent dump attempts have
yielded this error, regardless of parameter changes.
Any suggestions would be most gratefully received.
Regards, Ger.
------------------------------
From: "D. Vrabel" <[EMAIL PROTECTED]>
Crossposted-To: aus.computers.linux
Subject: Re: Linux or linux?
Date: Mon, 24 May 1999 12:07:36 +0100
On Mon, 24 May 1999, jane chav wrote:
> D. Vrabel wrote:
> >
> > On Sat, 22 May 1999, jane chav wrote:
> >
> > > eloki wrote:
> > > >
> > > > Ah, so you'd say that there's lots of users running Symantec/Windows or
> > > > who used to be running Quarterdeck/Windows? I like the GNU utils and all,
> > > > and I'm not saying that "Linux" as a whole system could do without binutils,
> > > > fileutils etc. But it's just not warranted as a name. People out there
> > > > aren't running Symantec/Netscape/Winzip/Installshield/Windows 95. They're
> > > > just running Windows 95, with various utilities.
> > >
> > > agreed, kernel is the core of an OS, the OS provides interfaces for the
> > > user applications. Without the utilities, the computer can still boot,
> > > but without the kernel, no utilities can run. utilities are important
> > > but they are not part of an OS. Linux is licensed under GNU (ie.
> > > Copyleft), just like MS-DOS has its own Copyright, but we don't really
> > > call it "whatevery license/Windows". So, it is suffice to call it Linux.
> > The license is the GNU General Public Licence not GNU (GNU stands for
> > GNU's not UNIX).
> >
> > Both the kernel and the utilities are required to run any user programs
> > therefore they are considered an OS when combined.
> >
> > GNU/Linux is used to indicate that it is a GNU system running with a Linux
> > kernel.
>
> Sorry, I made a mistake on the GNU part. but what is the definition of
> utilities? I interprete it as a tool to accomplish something, e.g.
> emacs. I might be wrong on that one too. but I thought user programs
> make use of C library functions.
>
> Do you mean since GNU GCC is a utility and without this compiler, there
> will be no other programs?
I would say that gcc was a utility and therefore a part of the OS. emacs
is not a utility but a user program because there are alternative and
smaller editors.
David
--
David Vrabel
Engineering Undergraduate at University of Cambridge, UK.
------------------------------
From: Mark Tranchant <[EMAIL PROTECTED]>
Crossposted-To: uk.comp.os.linux
Subject: Re: Linux's Last Chance
Date: Mon, 24 May 1999 11:37:14 +0100
Reply-To: [EMAIL PROTECTED]
"Aaarrrgh! Light...too bright...to code...NOO!"
http://www.userfriendly.org/cartoons/archives/99apr/19990423.html
Gareth Owen wrote:
> [EMAIL PROTECTED] (Juergen Heinzl) writes:
> > >
> > >And as an encore - how many Usenet posters does it take to change a
> > >light-bulb...? ;)
> >
> > We do not use light bulbs. Our genius shines on itself.
>
> To be honest though, when it comes to the provision of light, even Linux
> users prefer Windows.
>
------------------------------
From: Andy Bianchi <[EMAIL PROTECTED]>
Reply-To: [EMAIL PROTECTED]
Crossposted-To: comp.os.linux.hardware,comp.os.linux.setup
Subject: Hard drive flipping bits!
Date: Mon, 24 May 1999 12:40:55 +0100
Really annoying problem:
I have the following setup:
- PCChips TXProII (with built in VGA/Sound, I use neither of
these at the moment) motherboard
- Pentium P90
- 32Mb memory, 2.1Gb Quantum Hard drive
- RedHat5.2
[ This is just a cheapo system that I cobbled togther as a
compliment to the old trusty 486 running linux (I wanted
more HDD space and something with PCI for more extras!) ]
Unfortunately the system appears to be suffering from
an intermittent problem where a few bits (it is usually
just one or two) in a file can become corrupted.
To make this fault appear I'm running a shell script
which repeatedly compiles the kernel (in fact I have two
of these running concurrently).
After about 5hours or so one of the compiles will fail due
to some parse error or something similar. Then it may
compile the whole kernel successfully for a few times and
then fail again (each compile is less than an hour).
I'm logging all errors and normal compile output to text
files so the fact that one compile has gone wrong can
easily be identified by comparing the size of the logs.
=======================================================
I've been looking into this problem for a while now and
I've tried the following things:
* On the above system I've tried different cables and also
making the cable shorter (down to 12").
* I've changed the PIO mode in the BIOS to 2 (I don't care
about UDMA or anything fancy).
I've also changed various bits of hardware:
* I changed motherboard, CPU, and memory (all at the same time,
I was testing a new base system) so the only things that
were common were the case/PSU and HDD (and cable).
The new motherboard was a Gigabyte 5AA (Uses the Aladdin V
chipset I believe) with a P120 and 32Mb RAM
This gave the same errors but more of them. I tried to change
PIO mode in the BIOS but there doesn't appear to be a way to
do that.
* I've now gone back to the TXPro and got a new HDD from the
shop since I was convinced that the HDD was faulty - well
I tried changing everything else in the system!
None of the above have erradicated the problem completely
although some may have lessened it - specifically dropping
PIO mode I *think* had some effect (it was a while since
I tried that and I've been running at PIO mode every since).
===============================================================
My only remaining theory now is that the Linux/BIOS/chipset
combination does work due to the ide driver not setting up
some timing or other such parameters correctly, hence I
get weird intermittent faults:
* The Gigabyte definitely uses Aladdin V chipset (M1543). The
TXPro reports it's using an SiS 85C5513 although, if my
memory serves me correctly I've seen posts saying that
"... can't use UDMA on TXPro ... it uses Aladdin chipset ..."
* Reading some FAQs about ATA/IDE etc it appears that these new
fancy chipsets have all sorts of config registers to change
timing params. Something must set these up and I can't see
any evidence that Linux is doing this.
* Looking at the kernel source there is a ali14xx.c file which
claims to setup the ide drivers for Aladdin 14xx chipsets and
appears to do some tuning, like setting timing registers.
Since there is no ali15xx.c file and I see no messages saying
that anything special has been done for my IDE I've got to
assume that Linux is just using some default IDE driver.
* I notice that the 2.0.37preXX and also some others aimed at
getting UDMA working mention Aladdin chipsets etc and I'm
going to have a look at these at some point - I assume that
there will be an ali15xx.c file which may help.
Anyone got any ideas/suggestions/comments etc?
Any kernel/driver gurus out there had these problems or
could contradict any of what I've said above!
PLEASE HELP!
Andy
--
"You're not too tired for this life, and it's not going
to matter if you fall down twice" - Lisa Loeb 'Snow Day'
------------------------------
From: jane chav <[EMAIL PROTECTED]>
Crossposted-To: aus.computers.linux
Subject: Re: Linux or linux?
Date: Mon, 24 May 1999 21:12:09 +1000
Jason Stokes wrote:
>
> >Sorry, I made a mistake on the GNU part. but what is the definition of
> >utilities? I interprete it as a tool to accomplish something, e.g.
> >emacs. I might be wrong on that one too. but I thought user programs
> >make use of C library functions.
>
> "Utility" is a very vague term, like "operating system" itself. The Linux
> kernel, for example, is only part of the story -- a lot of Linux's
> "Unix-like" behavior is down to glibc and other shared libraries.
>
> >Do you mean since GNU GCC is a utility and without this compiler, there
> >will be no other programs?
>
> That's not the argument. The argument is only that GNU libraries, utilities
> etc. compose such a large proportion of the Linux "platform" (as opposed to
> the Linux *kernel*) that we should give credit where credit is due. FSF
> favours a particular nomenclature for GNU-based systems -- GNU first,
> operating system "kernel" last. For example, you could port glibc over to
> solaris (I believe this has actually been done) and end up with a
> GNU/Solaris system.
I agree that Linus doesn't deserve all the credit to himself, GNU would
be a term that implicitly (or explicitly, whatever you may think) gives
credits to all the other programmers out there who have made the Linux
platform possible. But it is a known fact that Linus recieved help from
people all of the world.
Also I thought it was Linus who agreed to license the kernel under GPL
AFTER it sort of worked, so I assume the kernel doesn't use any of the
GNU libraries (haven't had a thorough look at the kernel source code
yet). So, all those GNU utilities and programs are just running on top
of the kernel. (you see, The thing is that I think an OS can simply
consits of the kernel and a shell maybe, something like the old MS-DOS )
If this was true, then we could have a Linux platform running a lot of
non-GNU software e.g. free/commercial. If I am wrong again then I'll
give up, it should be *written* formally as "GNU/Linux", but I think
most of the people still would like to verbally refer to it as Linux.
It is just too complicated to explain to people "What's GNU stand for"
------------------------------
Crossposted-To:
comp.os.linux.hardware,comp.os.linux.help,comp.os.linux.networking,comp.os.linux.questions,comp.os.linux.setup
Subject: Re: Will a SupraExpress 56i modem run under linux?
From: [EMAIL PROTECTED] (Rob Clark)
Date: Mon, 24 May 1999 11:47:35 GMT
In article <2A523.14487$[EMAIL PROTECTED]>,
Ozzy <[EMAIL PROTECTED]> wrote:
>Anybody running this modem under linux?
>successfully?
Please check your model number on the list at
http://www.o2.net/~gromitkc/winmodem.html
There are several modems with very similar names, some which do work, and
some which do not.
Rob Clark, [EMAIL PROTECTED]
------------------------------
From: Mark Forsyth <[EMAIL PROTECTED]>
Crossposted-To: linux.redhat.misc
Subject: Re: Root Password lost...
Date: Mon, 24 May 1999 17:09:57 +1000
NF Stevens wrote:
>
> Mark Forsyth <[EMAIL PROTECTED]> wrote:
>
> >
> >
> >NF Stevens wrote:
> >>
> >> [EMAIL PROTECTED] (Thomas Veach) wrote:
> >>
> >> [snip]
> >>
> >> >Is it just me, or is it realy this easy to break into a linux machine
> >> >and take over as root?
> >>
> >> It depends on how you have your machine set up. On mine using
> >
> >No it DOESN'T. If you have complete physical access to the machine
> >security
> >IS compromised. Excepting things like encrypted disks. I don't give a
> >rats
> >what sort of boot hinderances you provide. Give me a system and I'll
> >read what's
> >on the disks therein..!
>
> I am not disputing that if you take the machine to pieces
> you can gain access to its data. The question is whether
> you can gain root access without taking the machine apart.
> My contention is that it is possible to maintain security
> in this situation.
Alright then WITHOUT pulling it apart.
Mark F...
>
> Norman
------------------------------
From: Jason Jordan <[EMAIL PROTECTED]>
Subject: SETI CPU/OS Comparisons
Date: Mon, 24 May 1999 19:53:00 +0800
Reply-To: [EMAIL PROTECTED]
http://www.penguinpowered.com/~setiathome
lists lots of different architectures and OSes and the various rates
some folk are reporting.
------------------------------
From: [EMAIL PROTECTED] (Matthias Warkus)
Crossposted-To: comp.infosystems.www.servers.unix,comp.os.linux.advocacy
Subject: Re: NT the best web platform?
Date: Mon, 24 May 1999 01:28:17 +0200
Reply-To: [EMAIL PROTECTED]
It was the Sun, 23 May 1999 17:38:38 GMT...
..and Anthony Ord <[EMAIL PROTECTED]> wrote:
> Why? Apache and Squid (sounds like a duo on a Saturday
> morning cartoon...)
Today on cartoon network: Cow and Chicken, followed by Apache and
Squid at ten and J. Bravo Linux User at half past ten.
Yay!
mawa
--
[...] there was no region where American capital did not support local
labour. Moreover the American press, gramophone, radio, cinematograph
and televisor ceaselessly drenched the planet with American thought.
-- W. Olaf Stapledon, _Last_and_First_Men, 1931
------------------------------
From: [EMAIL PROTECTED] (Chris Sherlock)
Crossposted-To: alt.os.linux,comp.os.linux,linux.redhat.misc
Subject: Re: good, free ORB w/ C++ & Java lang support
Date: 24 May 1999 12:18:38 GMT
Reply-To: [EMAIL PROTECTED]
What the *heck* did you do? I think we get the picture already!
Chris S.
On Tue, 18 May 1999, Salman Ahmed <[EMAIL PROTECTED]> wrote:
:Robert Lynch wrote:
:
:> Salman Ahmed wrote:
:> >
:> > I am looking to learn CORBA, and want to know if someone
:> > can recommend to me a good, free implementation of
:> > CORBA for Linux (RedHat 5.2 w/ kernel 2.0.36).
:> >
:> > I intend to write CORBA apps in both C++ and Java, and
:> > won't be using any of the advanced features of the CORBA
:> > spec.
:>
:> I'm reading an article in the latest Linux Journal, part 1 of a 3 part
:> series.
:>
:
:I ended up buying that issue of LJ. And that article looks good.
:
:Thanks.
:
:--
:Salman S. Ahmed
:ssahmed AT interlog DOT com
:
:Remove the "nospam." portion from my email address
:to reply to this message.
:
:
------------------------------
** 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
******************************
