TCLUG Archive
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Unsent Message Returned to Sender



Notice to Sender
================

This message was received by this installation but could not be
delivered to its intended cc:Mail recipient(s).

  Original subject: tclug-list Digest 22 Oct 1999 23:00:10 -0000 Issue 443

Intended recipient(s) who DID NOT receive this message:

  kyle.johnson..@ccmail.reliastar.com

The following cc:Mail error(s) were recorded:

  ***  Message recipient is unknown  ***


-------- Original Message Text --------

-------------------------Embedded message follows:-----------------------


tclug-list Digest 22 Oct 1999 23:00:10 -0000 Issue 443

Topics (messages 9358 through 9368):

686 code generation
	9358 by: Christopher McKinley <lamfada@lugh.net>

cursor freezing
	9359 by: Chris Schumann <whizkid@dwave.net>

pppd with demand dialing / keeping packet that opens connection
	9360 by: Nate Carlson <natecars@real-time.com>
	9363 by: ^chewie <chewie@wookimus.net>
	9364 by: Nate Carlson <natecars@real-time.com>
	9365 by: "John J. Trammell" <trammell@nitz.hep.umn.edu>

Problems...
	9361 by: "Kyle Johnson" <kylejohnson@uswest.net>

Local Linux training resources
	9362 by: Carlos Sabo <carlos@coffeebean.net>

dual ethernet module loading
	9366 by: "Carl Wilhelm Soderstrom" <carls@agritech.com>
	9367 by: Peter Lukas <peter@math.umn.edu>

NIS problems on a Sparc64 running RedHat
	9368 by: Nate Carlson <natecars@real-time.com>

Administrivia:

To subscribe to the digest, e-mail:
	tclug-list-digest-subscribe@mn-linux.org

To unsubscribe from the digest, e-mail:
	tclug-list-digest-unsubscribe@mn-linux.org

To post to the list, e-mail:
	tclug-list@mn-linux.org


----------------------------------------------------------------------


-------------------------Embedded message follows:-----------------------

Message-ID: <380F7C89.6D9CF69D@lugh.net>
Date: Thu, 21 Oct 1999 15:50:17 -0500
From: Christopher McKinley <lamfada@lugh.net>
MIME-Version: 1.0
To: tclug-list@mn-linux.org
Subject: Re: [TCLUG:9357] 686 code generation

Try -mpentiumpro and/or -march=pentiumpro.

Bob Tanner wrote:
> 
> Do I need to get a special version of gcc to generate 686 code?
> 
> What is the command I put into the CFLAGS to get 686 code? I thought it was
> -mpentium, but man gcc hit nothing with that, but compiling with a -mpentium
> seems to work.
> 
> But is pentium = 686?
> 
> gcc pukes on -mpentiumII, so is pentium the best I can get?
> 
> --
> Bob Tanner <tanner@real-time.com>       | Phone : (612)943-8700
> http://www.real-time.com                | Fax   : (612)943-8500
> Key fingerprint =  6C E9 51 4F D5 3E 4C 66 62 A9 10 E5 35 85 39 D9
> 
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: tclug-list-unsubscribe@mn-linux.org
> For additional commands, e-mail: tclug-list-help@mn-linux.org

-- 
*--------------------------*------------------*
| Chris McKinley           | http://lugh.net  | 
| Unix & Linux Consulting  | lamfada@lugh.net |
| C/C++ & Perl programming | 612-623-0586     |
*--------------------------*------------------*


-------------------------Embedded message follows:-----------------------

Date: Thu, 21 Oct 1999 16:36:46 -0500 (CDT)
From: Chris Schumann <whizkid@dwave.net>
To: tclug-list@mn-linux.org
Message-ID: <Pine.LNX.3.96.991021163430.25917A-100000@home.dwave.net>
MIME-Version: 1.0
Subject: Re: cursor freezing

My system locked up under X and Win95 and WinNT every now and again, until
I unplugged the mouse and plugged it back in with a big nudge to make sure
it's secure. No problems since.

We computer types sometimes overlook the simple problems, especially if it
makes you go buy a new toy. (I really like my IBM ScrollPoint, BTW!)

Chris Schumann <whizkid@dwave.net>

"I am a cipher, wrapped in an enigma, smothered in secret sauce."
  - Jimmy James, News Radio



-------------------------Embedded message follows:-----------------------

Date: Thu, 21 Oct 1999 16:53:09 -0500 (CDT)
From: Nate Carlson <natecars@real-time.com>
To: Twin Cities Linux User Group <tclug-list@mn-linux.org>
Message-ID: <Pine.GSO.4.10.9910211650420.21753-100000@wizard.real-time.com>
MIME-Version: 1.0
Subject: pppd with demand dialing / keeping packet that opens connection

I'm setting up a machine with pppd demand dialing, and I seem to recall
someone on the list awhile ago mentioning that there was a /proc/... file
you could echo 1 > to that would tell the kernel to save the packet that
caused the connection to come up. Anyone remember what this file was? My
goal is to allow the connection attempt that brings the modem up to go
through without the user having to hit reload. I searched the list
archives, and could not find this message. Thanks!

-- 
Nate Carlson <carlson@real-time.com>    | Phone : (612)943-8700
http://www.real-time.com                | Fax   : (612)943-8500



-------------------------Embedded message follows:-----------------------

Date: Fri, 22 Oct 1999 09:09:28 -0500 (CDT)
From: ^chewie <chewie@wookimus.net>
To: Twin Cities Linux User Group <tclug-list@mn-linux.org>
Message-ID: <Pine.LNX.4.10.9910220813050.11673-100000@guinness.urw.org>
MIME-Version: 1.0
Subject: Re: [TCLUG:9360] pppd with demand dialing / keeping packet that
 opens connection

On Thu, 21 Oct 1999, Nate Carlson wrote:

NC >I'm setting up a machine with pppd demand dialing, and I seem
NC >to recall someone on the list awhile ago mentioning that there
NC >was a /proc/... file you could echo 1 > to that would tell the
NC >kernel to save the packet that caused the connection to come
NC >up. Anyone remember what this file was? My goal is to allow
NC >the connection attempt that brings the modem up to go through
NC >without the user having to hit reload. I searched the list
NC >archives, and could not find this message. Thanks!

echo 1 > /proc/sys/net/ipv4/ip_forward

Later!

    ^chewie

+----------------------------------------------------+
| Chad Walstrom           mailto:chewie@wookimus.net | 
| ICQ: 9985127           http://wookimus.net/~chewie |
+----------------------------------------------------+
 Need a new truck?  Check out my '97 Explorer 2-door
   Sport at http://wookimus.net/~chewie/truck.html




-------------------------Embedded message follows:-----------------------

Date: Fri, 22 Oct 1999 10:28:07 -0500 (CDT)
From: Nate Carlson <natecars@real-time.com>
To: Twin Cities Linux User Group <tclug-list@mn-linux.org>
Message-ID: <Pine.GSO.4.10.9910221019430.27182-100000@wizard.real-time.com>
MIME-Version: 1.0
Subject: Re: [TCLUG:9360] pppd with demand dialing / keeping packet that
 opens connection

That's ip forwarding, I've already got that working. Somebody mentioned
something about there being a different file that will actually save the
packet that brings up the diald (well, in this case, pppd dial on demand)
connection and will send that packet when the connection comes up, instead
of just forgetting it as usually happens. This way, whatever request
brought the connection up will be completed. I just have to deal with some
clients that can't understand the concept of having to do one connection
attempt to the web site / e-mail server to bring the connection up, and
then resending that connection attempt to actually establish the
connection. Any ideas?

-- 
Nate Carlson <carlson@real-time.com>    | Phone : (612)943-8700
http://www.real-time.com                | Fax   : (612)943-8500

On Fri, 22 Oct 1999, ^chewie wrote:

> On Thu, 21 Oct 1999, Nate Carlson wrote:
> 
> NC >I'm setting up a machine with pppd demand dialing, and I seem
> NC >to recall someone on the list awhile ago mentioning that there
> NC >was a /proc/... file you could echo 1 > to that would tell the
> NC >kernel to save the packet that caused the connection to come
> NC >up. Anyone remember what this file was? My goal is to allow
> NC >the connection attempt that brings the modem up to go through
> NC >without the user having to hit reload. I searched the list
> NC >archives, and could not find this message. Thanks!
> 
> echo 1 > /proc/sys/net/ipv4/ip_forward
> 
> Later!
> 
>     ^chewie
> 
> +----------------------------------------------------+
> | Chad Walstrom           mailto:chewie@wookimus.net | 
> | ICQ: 9985127           http://wookimus.net/~chewie |
> +----------------------------------------------------+
>  Need a new truck?  Check out my '97 Explorer 2-door
>    Sport at http://wookimus.net/~chewie/truck.html
> 
> 
> 
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: tclug-list-unsubscribe@mn-linux.org
> For additional commands, e-mail: tclug-list-help@mn-linux.org
> 



-------------------------Embedded message follows:-----------------------

Date: Fri, 22 Oct 1999 03:36:37 -0500
From: "John J. Trammell" <trammell@nitz.hep.umn.edu>
Message-Id: <199910220836.DAA06344@nitz.hep.umn.edu>
To: tclug-list@mn-linux.org
Subject: Re: [TCLUG:9360] pppd with demand dialing / keeping packet that opens connection

Hello all:

> From: Nate Carlson <natecars@real-time.com>
> Subject: Re: [TCLUG:9360] pppd with demand dialing / keeping packet that
>  opens connection
>
> That's ip forwarding, I've already got that working. Somebody mentioned
> something about there being a different file that will actually save the
> packet that brings up the diald (well, in this case, pppd dial on demand)
> connection and will send that packet when the connection comes up, instead
> of just forgetting it as usually happens. This way, whatever request
> brought the connection up will be completed. I just have to deal with some
> clients that can't understand the concept of having to do one connection
> attempt to the web site / e-mail server to bring the connection up, and
> then resending that connection attempt to actually establish the
> connection. Any ideas?

I don't know about saving the acutal packet, but there's a debugging
option to turn on the logging of matched filter rules...

Check out:

    http://www.r-senior.demon.co.uk/linux/diald.html

John



-------------------------Embedded message follows:-----------------------

Message-ID: <002c01bf1c1f$e0e04020$0301a8c0@mclean.com>
From: "Kyle Johnson" <kylejohnson@uswest.net>
To: <tclug-list@mn-linux.org>
Date: Thu, 21 Oct 1999 18:56:21 -0500
MIME-Version: 1.0
Subject: Problems...

I was just wondering if there were any problems with the server, or if it was on my end.    I haven't seen anything come through the past two days.

Kyle



-------------------------Embedded message follows:-----------------------

Message-Id: <3.0.6.32.19991021220423.0093a7f0@minn.net>
Date: Thu, 21 Oct 1999 22:04:23 -0500
To: tclug-list@mn-linux.org
From: Carlos Sabo <carlos@coffeebean.net>
Mime-Version: 1.0
Subject: RE: [TCLUG:9335] Local Linux training resources

I'm told that Total Solutions Group has a class.  It doesn't seem to be
mentioned on their web site though.

http://www.tsg-usa.com


-------------------------Embedded message follows:-----------------------

Message-Id: <2.2.16.19991022140622.45577dc0@smtp.agritech.com>
Mime-Version: 1.0
To: tclug-list@mn-linux.org
From: "Carl Wilhelm Soderstrom" <carls@agritech.com>
Date: Fri, 22 Oct 1999 12:15:24 -0700
Subject: dual ethernet module loading

        at the recent installfest, I got my new firewall-to-be set up with 2
ethernet cards (3c509s). it took several advanced people some time to figure
out that we needed to put '/sbin/modprobe 3c509' at the end of
/etc/rc.d/rc.sysinit, to make sure the modules were loaded.
        I decided to upgrade the box to RH 6.1. (foolish me).
        since then, I can't get the modules to load at boot time.
        I put '/sbin/modprobe 3c509' in rc.sysinit; and at boot time I see
"can't locate module 3c509"
        trying the same command from the command line after boot, also gives
"can't locate module 3c509".
        trying '/sbin/modprobe /lib/modules/2.2.12-20/net/3c509.o' gives 'no
dependendcy information for module /lib/modules/2.2.12-20/net/3c509.o"'
        so I tried doing a 'depmod -a'; which generated a modules.dep file
(just like the /lib/modules/2.2.5 directory had). still gives the same error.

        I can get the modules to load if I do:
insmod 3c509
ifup eth0
ifup eth1
        but I tried doing this at boot time (putting those lines at the end
of the rc.sysinit file); and it always gives a "resource busy" error (or
something like that) at boot time.
        
        what's wrong here?
        would putting some boot parameters into lilo help? (I wouldn't think
so, but...)  if so, what should they be?

thanks,
Carl Soderstrom
System Administrator	307 Brighton Ave. 
Minnesota DHIA		Buffalo, MN	
carls@agritech.com	(612) 682-1091



-------------------------Embedded message follows:-----------------------

Date: Fri, 22 Oct 1999 15:14:24 -0500 (CDT)
From: Peter Lukas <peter@math.umn.edu>
To: tclug-list@mn-linux.org
Message-ID: <Pine.LNX.4.10.9910221503510.485-100000@paoli.math.umn.edu>
MIME-Version: 1.0
Subject: Re: [TCLUG:9366] dual ethernet module loading

What does `cat /etc/conf.modules` reveal about your adapters?
At boot time, the RedHat init scripts de several module checks including
a `depmod -a`.  Because RedHat places their modules into a nonstandard
location (kernel-version-RPM_package_release), traditional depmod searches
don't always work (just take a look at /etc/rc.d/rc.sysinit).  In any
event, you should be alright as long as the depmod-a can successfully
update /lib/modules/`uname -r`/modules.dep. 

A quick fix would be to search for replaced files from the upgrade
(provided of course that you did an actual upgrade, not a freah install of
6.1).  Look for anything ending in .rpmsave.  Do a diff against the
replaced version to see if your modprobe magic was somehow messed up by
the upgrade.

Your best bet is to compile a dedicated kernel (this is good practice
for production boxes, not to mention general Linux use anyway).  Compile
the drivers for your adapters into the kernel directly and use lilo
parameters to initialize both adapters (Note: you will still need to `make
modules` though.  There are several masquerading kernel modules that must
be arbitrarily loaded).

Good Luck

Peter Lukas

On Fri, 22 Oct 1999, Carl Wilhelm Soderstrom wrote:

>         at the recent installfest, I got my new firewall-to-be set up with 2
> ethernet cards (3c509s). it took several advanced people some time to figure
> out that we needed to put '/sbin/modprobe 3c509' at the end of
> /etc/rc.d/rc.sysinit, to make sure the modules were loaded.
>         I decided to upgrade the box to RH 6.1. (foolish me).
>         since then, I can't get the modules to load at boot time.
>         I put '/sbin/modprobe 3c509' in rc.sysinit; and at boot time I see
> "can't locate module 3c509"
>         trying the same command from the command line after boot, also gives
> "can't locate module 3c509".
>         trying '/sbin/modprobe /lib/modules/2.2.12-20/net/3c509.o' gives 'no
> dependendcy information for module /lib/modules/2.2.12-20/net/3c509.o"'
>         so I tried doing a 'depmod -a'; which generated a modules.dep file
> (just like the /lib/modules/2.2.5 directory had). still gives the same error.
> 
>         I can get the modules to load if I do:
> insmod 3c509
> ifup eth0
> ifup eth1
>         but I tried doing this at boot time (putting those lines at the end
> of the rc.sysinit file); and it always gives a "resource busy" error (or
> something like that) at boot time.
>         
>         what's wrong here?
>         would putting some boot parameters into lilo help? (I wouldn't think
> so, but...)  if so, what should they be?
> 
> thanks,
> Carl Soderstrom
> System Administrator	307 Brighton Ave. 
> Minnesota DHIA		Buffalo, MN	
> carls@agritech.com	(612) 682-1091
> 
> 
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: tclug-list-unsubscribe@mn-linux.org
> For additional commands, e-mail: tclug-list-help@mn-linux.org
> 
> 



-------------------------Embedded message follows:-----------------------

Date: Fri, 22 Oct 1999 17:57:03 -0500 (CDT)
From: Nate Carlson <natecars@real-time.com>
To: Twin Cities Linux User Group <tclug-list@mn-linux.org>
Message-ID: <Pine.GSO.4.10.9910221754490.1798-100000@wizard.real-time.com>
MIME-Version: 1.0
Subject: NIS problems on a Sparc64 running RedHat

Hey,

I'm trying to get NIS to authenticate a user on a Sparc64 running RH6.
Basically, it authenticates all the users currently in the NIS tree, but
when I add a new user, it will not authenticate on the Sparc, but
authenticates on every other machine. There is nothing in /etc/hosts.deny
that would deny it, and all the pam and login configuration files look
fine. The weird thing is if I delete a user from NIS and add them again,
that user can still log in. Any ideas?

-- 
Nate Carlson <carlson@real-time.com>    | Phone : (612)943-8700
http://www.real-time.com                | Fax   : (612)943-8500