Linux-Networking Digest #342, Volume #11         Mon, 31 May 99 04:14:57 EDT

Contents:
  Clear out the route table? ("Geoffrey L. Goetz")
  WD 8013 NIC: Strange I/O address problem (Rohan Oberoi)
  Re: ftp problem (Gilford Wimbley)
  Re: Call me clueless... (buggy g. riphead)
  Re: Is there a way to send AT commands to a modem during PPP connexion ? (Gilford 
Wimbley)
  Re: Auto-dial for IP Masq ("George Georgakis")
  Re: Req: Samba configuration help.... (Gilford Wimbley)
  Re: Best Free X Windows Server for Win95/98 Box on Samba/Linux Network? ("Jeff 
Grossman")
  Re: What are drawbacks to using an ISA NIC? (Frank Sweetser)
  Re: Multiple IP accounts (Paul Schultz)
  Re: Minicom Help Requested (William Park)

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

From: "Geoffrey L. Goetz" <[EMAIL PROTECTED]>
Subject: Clear out the route table?
Date: Mon, 31 May 1999 00:05:11 -0400

Is there anyway to clear out the entire route table in order to re-build
it?  Something like a route -CA (Clear All)?

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

From: [EMAIL PROTECTED] (Rohan Oberoi)
Subject: WD 8013 NIC: Strange I/O address problem
Reply-To: [EMAIL PROTECTED]
Date: Mon, 31 May 1999 05:23:41 GMT

I have a WD8013W card in my Compaq Prolinea 486DX2/50 (RedHat
5.2/2.0.36) that is behaving very strangely -- surprisingly, because I
have had two Intel EtherExpress NICs working more or less well in that
system for months, and the WD is supposed to be an older and better
tested Linux driver.  Perhaps this is a bug in the driver with this
hardware? 

At boot time, the card is recognised properly, as seen in dmesg:

eth0: WD80x3 at 0x280,  00 00 C0 E5 38 27 WD8013, IRQ 5, shared memory
at 0xd0000-0xd3fff.

However, at ifconfig the card comes up with the last line reporting:

Interrupt:5 Base address:0x290 Memory:d0000-d4000

It looks like the base I/O address and the upper memory range are both
being bumped up.  When I put another WD in the system (which is
reported by the SMC EZSETUP program as I/O 300 IRQ 10) it shows up at
"Base address:0x310".

I searched on Deja(News) and found one report of a similar problem,
from 1997, which was fixed through the PnP bios settings: I append
that report.

If anyone knows anything about this "I/O +10" bug/feature, and has any
idea about a workaround, I'd appreciate hearing about it.  Please
remove the obvious from my address if replying by email.

Regards,
Rohan.

========================================================================

   Subject: Q: Turn off ethernet card autoprobe on boot?
   Date: 1997/10/05
   Author: rcattig # mindspring O com (Craig Attig)
   
   I have an SMC Ultra ethernet card. Running ISAPNP.SYS on NT, it comes
   in as IRQ 10, Addr 0x300, Mem 0xCC000. And this config works for
   talking to another nearby machine.
   
   Booting into RH 4.2, however, the card is autoprobed and comes back as
   IRQ 10, >>> Addr 0x310 <<<. Mem 0xCC000. The card then does not seem
   to work -- no big surprise -- as ifconfig shows 0 packets for both TX
   and RX.
   
   I've tried to boot with 'linux ether=10,0x300,0,0,eth0' but this
   doens't seem to get around the autoprobe'd settings.
   
   How do I turn off autoprobing for this card?
   
========================================================================

   Subject: Re: Q: Turn off ethernet card autoprobe on boot?
   Date: 1997/10/05
   Author: rcattig # mindspring O com (Craig Attig)
   
   Lars,
   
   >> I have an SMC Ultra ethernet card. Running ISAPNP.SYS on NT, it
   comes
   >> in as IRQ 10, Addr 0x300, Mem 0xCC000. And this config works for
   >> talking to another nearby machine.
   >>
   >> Booting into RH 4.2, however, the card is autoprobed and comes back
   as
   >> IRQ 10, >>> Addr 0x310 <<<. Mem 0xCC000. The card then does not
   seem
   >> to work -- no big surprise -- as ifconfig shows 0 packets for both
   TX
   >> and RX.
   
   >It sounds like you're using a module for your card...
   >options smc-ultra io=0x300
   
   I've looked more closely into what's going on. SMC-ULTRA.C does
   correctly identify the card as running at addr 0x300 using autoprobe
   at boot, but upon 'ifconfig' it still comes up as set to addr=0x310.
   
   I've tried adding the line 'options smc-ultra io=0x300' to my
   /etc/conf.modules file, but this doesn't do anything (the card *is*
   correctly id'd, after all), and booting with the similar option 'linux
   options smc-ultra io=0x300', but to no avail. I've also tried to take
   down the card manually, 'ifconfig eth0 down' and bring it back up
   ('ifconfig eth0 inet 192.168.... bla bla bla) with ifconfig's
   'io_addr' option, but ifconfig seems to choke on that option.
   
   Any ideas?
   
========================================================================

   Subject: Re: Q: Turn off ethernet card autoprobe on boot?
   Date: 1997/10/05
   Author: Craig Attig <[EMAIL PROTECTED]>
   
   I've finally mamanged to get things to work!!! I want to thank
   everybody who lent a hand and pitched in with advice. It helped --
   thank you!!
   
   Ultimately, I needed to change the PNP BIOS setting on my motherboard
   such that the IRQ's I know were taken were set aside as "Legacy ISA"
   IRQ's. Every IRQ I knew about I set up that way. Strnage thing is that
   'ifconfig' still reports the eth0 device as being at 0x310 when it's
   clearly not -- autoprobing comes up 0x300, and NT runs it as 0x300.
   But, hey, it works now, and that's all I care about. :)

========================================================================


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

From: [EMAIL PROTECTED] (Gilford Wimbley)
Subject: Re: ftp problem
Date: Mon, 31 May 1999 05:25:25 GMT

I had the exact same problem.  As I recall, the way I fixed it was by
installing the wu-ftpd rpm included (but not installed) on my machine.
Why don't  you try installing wu-ftpd using the upgrade switch on the
rpm.  I am betting that it is not already installed, and that when you
install it, everything will work.  

on my machine, it was like this
% rpm -U /mnt/cdrom/RedHat/RPMS/wu-ftpd-2.4.2b18-2.i386.rpm

but it will probably be different for you.
good luck!
GW


On Sun, 30 May 1999 18:09:34 +0200, Simon Gouder <[EMAIL PROTECTED]>
wrote:

>
>--------------A794DC0AF5E1B5F42E3E7E17
>Content-Type: text/plain; charset=us-ascii
>Content-Transfer-Encoding: 7bit
>
>Hi,
>
>I have a small network consisting of 2 Windows 98 machines and 1
>Red Hat
>Linux machine (5.1).
>
>My problem is that I cannot use the ftp service between the Windows
>machine and the Linux box.
>However I can ping to and from all machines and can also telnet from the
>W98 machines to the Linux.
>The /etc/services file is all right as specified int the HOW TOs
>literature.
>
>I have looked into the /var/log/secure file and found the following
>message :
>
>     in.ftpd[631]: error : cannot execute /usr/sbin/in.ftpd : No such
>file or directory.
>
>Please note that the ftp package is installed, however I cannot find the
>above file.
>
>Can somebody help me,
>
>Simon
>
>
>
>
>
>--------------A794DC0AF5E1B5F42E3E7E17
>Content-Type: text/html; charset=us-ascii
>Content-Transfer-Encoding: 7bit
>
><!doctype html public "-//w3c//dtd html 4.0 transitional//en">
><html>
>Hi,
><p>I have a small network consisting of 2 Windows 98 machines and 1
><br>Red Hat
><br>Linux machine (5.1).
><p>My problem is that I cannot use the ftp service between the Windows
>machine and the Linux box.
><br>However I can ping to and from all machines and can also telnet from
>the W98 machines to the Linux.
><br>The /etc/services file is all right as specified int the HOW TOs literature.
><p>I have looked into the /var/log/secure file and found the following
>message :
><p>&nbsp;&nbsp;&nbsp;&nbsp; <i><font size=+1>in.ftpd[631]: error : cannot
>execute /usr/sbin/in.ftpd : No such file or directory.</font></i>
><p>Please note that the ftp package is installed, however I cannot find
>the above file.
><p>Can somebody help me,
><p>Simon
><br>&nbsp;
><br>&nbsp;
><br>&nbsp;
><br>&nbsp;</html>
>
>--------------A794DC0AF5E1B5F42E3E7E17--
>


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

From: buggy g. riphead <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.setup
Subject: Re: Call me clueless...
Date: Sun, 30 May 1999 23:50:49 -0400

Private IP addresses, I believe, are not directly routable to the internet. 
The firewall can ping all hosts and the subnets because it is physically
connected to all three subnets via the the ethernet interfaces.  For the hosts
on the private network to ping hosts on the internet I think you may have to
set up IP masquerading, the details of which can be found in the IP
Masquerading Howto.


On Tue, 25 May 1999, Greg Franks wrote:
>>>>>> "Greg" == Greg Franks <[EMAIL PROTECTED]> writes:
>
>    Greg> I am trying to use a RH 5.2 box as a firewall with three
>    Greg> ethernet cards...  There are three subnets, one connected to
>    Greg> the big bad internet (call it eth0, 123.123.123.23, its
>    Greg> connected to 123.123.123.22 at the isp), one connected to a
>    Greg> public local net with external IP address (say eth1
>    Greg> 123.123.124.1) and one connected to a private local net
>    Greg> (eth2 172.20.201.1).  The firewall can ping to hosts on each
>    Greg> of the three subnets, and vice versa.  However, I cannot get
>    Greg> hosts on the private net to ping hosts on the public local
>    Greg> net (ie. 172.20.201.10 -> 123.123.124.5), nor vice versa
>    Greg> (couldn't test the big bad internet because the ISP hasn't
>    Greg> connected the wire in yet :-( ).
>
>    Greg> In /etc/sysconfig/network on the firewall machine.
>    Greg> FORWARD_IPV4=true GATEWAY=123.123.123.22 GATEWAYDEV=eth0
>
>    Greg> Clearly, I am missing something, but I can't figure out
>    Greg> what.  Perhaps it's because I have three cards installed on
>    Greg> the machine and that's a touch beyond the standard RH
>    Greg> install scripts (I guess I should have stuck with
>    Greg> slakware...)
>
>    Greg> Signed...  ...perplexed.
>
>Here's what route -n says...
>
>% /sbin/route -n
>Kernel IP routing table
>Destination     Gateway         Genmask         Flags Metric Ref    Use Iface
>216.58.91.168   0.0.0.0         255.255.255.252 U     0      0        2 eth0
>216.58.91.172   0.0.0.0         255.255.255.252 U     0      0        1 eth1
>172.20.201.1    0.0.0.0         255.255.255.0   U     0      0        1 eth1
>127.0.0.0       0.0.0.0         255.0.0.0       U     0      0        4 lo
>0.0.0.0         216.58.91.169   0.0.0.0         UG    1      0        1 eth0
>% 
>
>The ISP has assigned the block 168-175 which I have subnetted into
>168-171 (ISP side of firewall) and 172-175 (Public local net side of
>firewall).  216.58.91.174 (the public machine on the inside of the
>firewall) can ping the firewall, but not beyond. 
>
>My guess is that IP forwarding is not enabled in the kernel (Redhat
>5.2 `custom' install).  Is there a way to tell one way or the other?
>
>2.0.35 can handle subnetting of class C networks n'est pas?
>
>Or, maybe I have done something totally clueless.
>
>signed...
>   ...frustrated
>
>
>-- 
>   __@            Greg Franks, (613) 520-5726         <|       _~@ __O 
> _`\<,_    Systems Engineering, Carleton University,   |O\   -^\<;^\<, 
>(*)/ (*)       Ottawa, Ontario, Canada  K1S 5B6.       (*)--(*)%---/(*)
>          "Where do you want to go today?"   Outside.


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

From: [EMAIL PROTECTED] (Gilford Wimbley)
Subject: Re: Is there a way to send AT commands to a modem during PPP connexion ?
Date: Mon, 31 May 1999 05:09:10 GMT

On Thu, 27 May 1999 14:08:40 GMT, [EMAIL PROTECTED] wrote:

>Hello,
>
>Is there a way to send AT commands to a modem
>during PPP connexion ?
>
>For the moment I've managed to do that :
>- start PPP connexion
>- stop pppd daemon with ##$kill -19 `cat
>/var/run/ppp0.pid`##
>- listen the modem with ##$cat /dev/modem##, what
>is not possible without stopping pppd daemon
>
>If I send commands to the modem, AFTER PPP
>connexion, with ##$echo -ne '+++' > /dev/modem##
>for example, these commands are encapsulated ! So
>I can see, thanks to the ##"cat /dev/modem"##,
>that the modem is recieving :
>###############
>)o    �d    @pH?�889�) ccc 5 x[��r   ` �b � +++
>?^
>@pe��88:���
>cc 5 x{%�r   `        a `fN_1anbbrsrockwellbcom  `
>`   (b  P�(o    $�!oi    �t�b�!*� � +++ ?^
>@p�C�88:�$?pcc 5 x���r   `        a `
>tN_1anbbrsrockwellbcom  ` `       (b  P�(o
>$�!oi  x*̷S)SN�@�?"?m?y�XVp�d\-#�C
>�\n��f�.I�t�b�!*�w
>################
>
>All these garbage instead of +++
>So I CAN'T return in on-line command mode, to send
>AT commands .....
>
>Does anybody have an idea to achieve this ?
>
>
>One more thing ! Do you know how make the pppd
>daemon continue, after being stopped by ##$kill
>-19 `cat /var/run/ppp0.pid`## .... I've managed to
>do this too
>...
>
>FX
>
>
>--== Sent via Deja.com http://www.deja.com/ ==--
>---Share what you know. Learn what you don't.---
Well, normally people don't do what you are trying to do.  The reason
is that normally you can use a program called "chat" to send whatever
is neccessary to the modem.  Chat is a simple program that looks at
standard input.  It searches for a sequence of input strings.  After
it detects each one, it sends a possibly null response, also to
standard output.  Chat is not used directly, but with pppd's connect
option.  You ought to read up on this stuff and see if it can do
whatever you are trying to do.    I can almost guarantee that there is
an easier way than stopping and restarting pppd in the foreground.

good luck.
GW


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

From: "George Georgakis" <linuxstart.com@geegee>
Subject: Re: Auto-dial for IP Masq
Date: Mon, 31 May 1999 04:59:30 GMT

Do a LDP search on diald  
http://amelia.db.erau.edu/Excite/AT-LDPquery.html

Masqdialer also has some potential   http://cpwright.villagenet.com/mserver

George 
===========================================================================
I never reply by email as a) I don't give out my real email address freely,
and b) it stops other NG users from reading the solutions to problems
If necessary, however, I can be contacted thru linuxstart.com@geegee. 
(Swap "geegee" and "linuxstart.com").
===========================================================================

Brian Witowski <[EMAIL PROTECTED]> wrote in article
<[EMAIL PROTECTED]>...
> Greetings,
> 
> I just got (finally!) IP masquerading to work.  Now...how do I get my
> Linux box to automatically dial my ISP when I 'hit' it with my Win98
> box?
> 
> Thanks!
> Brian
> 
> 

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

From: [EMAIL PROTECTED] (Gilford Wimbley)
Subject: Re: Req: Samba configuration help....
Date: Mon, 31 May 1999 04:54:38 GMT

I have seen a lot of posts complaining about problems between samba
and both windows98 and later versions of NT.  The answer, in most
cases, was that windows98 and later versions of NT use encrypted
passwords by default, so you might have to enable password encryption
in smb.conf.  Also, that file DIAGNOSIS.txt that another post
mentioned *is* helpful.

Try doing a subject search on "windows98" and "samba" and "password
encryption".  Dejanews would be a good place to start.  I bet you will
have it working in no time.

best of luck!
GW


On Sun, 30 May 1999 19:01:35 -0700, dlwright <[EMAIL PROTECTED]>
wrote:

>I am currently trying to get samba installed too.   www.samba.org has a
>"Diagnosing Samba"
>document with 8 troubleshooting tests that I found useful. You can get it at
>
>                 http://us1.samba.org/samba/docs/DIAGNOSIS.html
>
>If you haven't already read this it might help.
>
>lp
>
>brett
>
>> I am fairly new to Linux, and not having much luck. Perhaps someone can
>> suggest where I am going wrong.
>>
>> Here are my computer configurations:
>>
[snip]
>> AT this point I restarted the Samba and tried to connect. I log into my
>> Windows NT or windows 98 machines using bretts and d3b0rchary. I am able to
>> see my primate in my browse lists but when I connect, I am promped to login.
>> Access is always denied.
>>
>> What am I overlooking ?Am I creating my guest account correctly.
>


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

From: "Jeff Grossman" <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.x,alt.os.linux,comp.os.linux.misc
Subject: Re: Best Free X Windows Server for Win95/98 Box on Samba/Linux Network?
Date: Sun, 30 May 1999 22:24:03 -0700

Okay, thanks.  I downloaded the Windows side, but I guess I need to go get
the Linux side and load it up.

Thanks,
Jeff

--
Jeff Grossman ([EMAIL PROTECTED])


Ted Staberow <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]...
> Hi Jeff,
>
>     You can use VNC or an X server for Windows.  Check out download.com
for
> some free / shareware X servers for Windows.  The X server on your Linux
box
> will not help you here because the server must run on the machine where
the
> application is to appear.  The app (client) runs remotely but the server
runs
> locally.  This seems a bit backwards but it is really not.  Consider that
your
> first network interface is localhost and the client/server relationship
will
> make more sense.  With VNC you will need the Linux and Windows versions.
VNC
> is a good choice for what you intend to do.
>
>
> --
> Ted Staberow
> Prairie Networking, Inc.
>
>
>
>
> Jeff Grossman wrote:
>
> > I just downloaded this version.  I want to be able to run my X sessions
from
> > my Linux box to my Win98 box.  What do I need?  Can I just use the
built-in
> > Xfree that is part of Redhat for the Linux side, and use this VNC
program
> > for the Win98 side?  And if so, how would I go about setting that up?
> >
> > Thanks,
> > Jeff
> >
> > --
> > Jeff Grossman ([EMAIL PROTECTED])
> >
> > Ursa_M <[EMAIL PROTECTED]> wrote in message
> > news:[EMAIL PROTECTED]...
>



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

From: Frank Sweetser <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.hardware
Subject: Re: What are drawbacks to using an ISA NIC?
Date: 31 May 1999 01:16:50 -0400

Tim Moore <[EMAIL PROTECTED]> writes:

> ISA is clocked at 8.3MHz, PCI at 33MHz.  ISA has more CPU overhead too.

yep, i'd forgotten about the vast difference in speed.

> This card is 10bT if I recall correctly, so it won't move more than
> about 850KB/s.  If you want 3MB/s you'll have to go 100bT which means
> PCI.

knowing the cable modem speeds i've seen, he probably meant 3Mbits, not
3Mbytes. 

> Are you sure the slot is defective?  My PCI slots 4&5 share an IRQ which
> could cause weird behavior under some circumstances.  Other boards that
> have built-in SCSI, sound, etc are known to share IRQ's with particular
> PCI slots.

other random thought - perhaps you're trying to put a bus mastering card
into a non-busmastering slot?  try swapping cards around, it may
mysteriously start working again.

-- 
Frank Sweetser rasmusin at wpi.edu fsweetser at blee.net  | PGP key available
paramount.ind.wpi.edu RedHat 5.2 kernel 2.2.5        i586 | at public servers
If this were Ada, I suppose we'd just constant fold 1/0 into

    die "Illegal division by zero"
             -- Larry Wall in <[EMAIL PROTECTED]>

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

From: [EMAIL PROTECTED] (Paul Schultz)
Subject: Re: Multiple IP accounts
Date: 30 May 1999 16:04:17 -0500

On Sat, 29 May 1999 22:25:40 , Alfredo Todini <[EMAIL PROTECTED]> wrote:
>I have two PPP dialup accounts. How should I set up the /etc/resolv.conf
>file to handle multiple accounts? (They have different domains).
>Thank you.

One way to do it would be to just have whatever script you use to dial your
ISP overwrite /etc/resolv.conf with the settings that you need.  

For example if you are using redhat:

Make /etc/resolv1.conf and /etc/resolv2.conf.

in /etc/sysconfig/network-scripts/ifcfg-ppp0 add
DNSCONFIG=1

in /etc/sysconfig/network-scripts/ifcfg-ppp1 add
DNSCONFIG=2

in /etc/sysconfig/network-scripts/ifup-ppp add these in

if [ ${DNSCONFIG} = 1 ] ; then
  cp /etc/resolv1.conf /etc/resolv.conf
fi

if [ ${DNSCONFIG} = 2 ] ; then
  cp /etc/resolv2.conf /etc/resolv.conf
fi


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

From: William Park <[EMAIL PROTECTED]>
Subject: Re: Minicom Help Requested
Date: 30 May 1999 04:59:03 GMT

Arthur <[EMAIL PROTECTED]> wrote:
> It just occurred to me that I forgot to describe the error msg I'm getting
> which is: "Cannot open /dev/modem. Permission denied. Also, I can open a
> minicom connection as root but not as a regular user. 
> Thanks.

> On 29 May 1999 06:50:31 GMT, Arthur <[EMAIL PROTECTED]> wrote:
>>I am running Redhat 6.0 and am having some problems connecting to my ISP
>>thru Minicom.
>>
>>I'm new to this so I'm not sure what it all means, but /dev/modem is a
>>symlink to /dev/ttyS1 and shows owner root group root and permissions
>>rwx rwx rwx. /dev/ttyS1 shows owner root group tty and permissions
>>rw- rw- and ---.
>>
>>For what it's worth, I am able to establish a ppp connection to my ISP
>>using either the Usernet utility or from the command line. If anyone
>>could point me in the right direction on how to establish one using
>>Minicom, I'd appreciate it.
>>
>>Thanks 

You are close to tracking down what's wrong.  /dev/ttyS1 should have
666 mode, that is, read/write for everyone.

William

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


** 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.networking) 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-Networking Digest
******************************

Reply via email to