Linux-Networking Digest #458, Volume #10         Thu, 11 Mar 99 15:14:13 EST

Contents:
  Re: No rlogin through a Linux Firewall? (Rodney van den Oever)
  Re: token ring sniffer (Bram Blaas)
  Re: NAT Support ("Allen")
  Re: Ethernet vs. SCSI (Erik Hensema)
  Re: WebMail Server for Linux (Izak Burger)
  Re: latest MS virus (Andy Harrison)
  Re: Logging into Samba from Windoze 98 ([EMAIL PROTECTED])
  TCP connection ("Srinivasan K.S.")
  Re: Linux-Router: pppd dialing on local connect & bootup & route (Andre Dietisheim)
  Re: smb_dont_catch_keepalive: server->data_ready == NULL ("Duarte Cordeiro")
  Re: Which program to use scsi-tape drive? (John Thompson)
  Re: Slack and DHCP (Austin Gresham)
  IPMasq Setup (3rd time is the charm) ("Stephen M. Shelly")
  Re: ppp0 -> eth0 (Lew Pitcher)
  IPCHAINS not working (Andre Giordano)
  Re: Installation of Xwindows Applications and other misc problems (William 
Burlingame)
  Re: Network Newbie Question (Frank Hahn)
  lost rsh (Sparkzz)
  Re: problem with routing ("¤ñ¥d¶W")
  Re: Linux + Novell 4.11 (Ron Flory)
  Modem: no luck w/simple AT cmd (brett russ)
  Linux 2.2.3 and Ultrix 4.3: incompatible NFS? (Georg Schwarz)

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

From: root@localhost. (Rodney van den Oever)
Crossposted-To: comp.security.firewalls
Subject: Re: No rlogin through a Linux Firewall?
Date: Thu, 11 Mar 1999 17:23:13 GMT

Habib Jalili <[EMAIL PROTECTED]> wrote:

>It works fine with telnet and ftp from the linux A to internet. But
>rloging from Linux A returns the saying error permission denied. Rlogin
>from Linux A to LAN  and rlogin from firewall (Linux B) to internet and
>to LAN does work too.
>I have tried it with -P all. No chance.

The problem is that Masquerading is in fact port-translation. All
traffic from the firewall leaves it with the IP-address of the
external interface as source-address. Because the session still has to
be unique, the source-port is changed to a value above 61000.

/*
 *      Linux ports don't normally get allocated above 32K.
 *      I used an extra 4K port-space
 */

#define PORT_MASQ_BEGIN 61000
#define PORT_MASQ_END   (PORT_MASQ_BEGIN+4096)

rlogin uses random source ports below 1024, so masquerading breaks
this.

You will have to use a proxy.



Rodney van den Oever / [EMAIL PROTECTED]
'Very funny, Scotty. Now beam down my clothes.' - Michael Finken

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

From: [EMAIL PROTECTED] (Bram Blaas)
Subject: Re: token ring sniffer
Date: Fri, 05 Mar 1999 16:38:03 GMT

Probably you can.  You will however need a promiscous token ring card.
Non-promiscous card are most IBM and Madge True Blue.


>My question: is it possible to sniff token-ring-LANs in the same manner?


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

From: "Allen" <[EMAIL PROTECTED]>
Subject: Re: NAT Support
Date: Fri, 5 Mar 1999 08:40:08 -0800

That's good to know. And I suppose it's transparent to workstations behind
it.

However, my concern is for port-mapping from outside linux (from Internet)
... should I call it reverse hosting? Does it support reverse hosting?

Thanks.
Allen



Eldir Tomassen wrote in message <[EMAIL PROTECTED]>...
>Ye sit does. Buy on linux it's often called
>masquerading instead.
>
>
>Eldir
>[EMAIL PROTECTED]



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

From: [EMAIL PROTECTED] (Erik Hensema)
Crossposted-To: comp.os.linux.hardware
Subject: Re: Ethernet vs. SCSI
Date: Thu, 11 Mar 1999 18:34:41 +0100
Reply-To: [EMAIL PROTECTED]

Christopher Howard wrote:
>I have a 386, 486, and a 586.  I was thinking of linking them with
>ethernet but then I started thinking about SCSI.  They will be stacked
>on top of each other, and I do plan on getting another system to add to
>it later.  What would be the advantages and disadvatages of using SCSI
>instead of Ethernet?  There are 2 extra bays in the 586 and the 386, 
>the 486 has 1 extra bay, but I could replace the IDE cd rom with SCSI
>(wich would be 2 bays).  Email me with your thoughts.

In theory, it's possible to have multiple scsi controllers on one bus, they
should have different ID's, that's all. All controllers can access all
devices on the bus, however, NOT simultaniously (the same device). Probably
you can manage to access different partitions on the same drive from
different controllers.
It's all very dangerous, as concurrent access of multiple controllers on the
same partition can/will damage the filesystem.
SCSI is also more expensive, and not as versatile as ethernet (you can't
telnet over scsi...). 
Don't do it if you aren't 100% sure of what you're doing. 

-- 
Erik Hensema ([EMAIL PROTECTED])
Please don't use my old address ([EMAIL PROTECTED]) anymore, it's obsolete.
Somewhere in the near future, mail to this address won't reach me anymore.

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

From: Izak Burger <[EMAIL PROTECTED]>
Subject: Re: WebMail Server for Linux
Date: Thu, 11 Mar 1999 15:47:58 +0200

Look at the ocs stuff at ftp.obsidian.co.za.  It includes a web based
email program that works like pine.

regards
Izak

=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
                                ----==-- _                
                                ---==---(_)__  __ ____  __
Microsoft is not the answer.    --==---/ / _ \/ // /\ \/ /
Microsoft is the question.      -=====/_/_//_/\_,_/ /_/\_\
Linux is the answer: Because a PC is a terrible thing to waste.
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=

Note: Mail from yahoo.com and hotmail.com domains will go to a special folder
and my not get read until much later.  I'm sorry for this inconvenience but I 
get to much spam from people in these domains.  



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

From: Andy Harrison <[EMAIL PROTECTED]>
Crossposted-To: 
linux.debian.user,dc.org.linux-users,alt.linux,alt.os.linux,comp.os.linux.setup
Subject: Re: latest MS virus
Date: Thu, 11 Mar 1999 08:53:19 -0500

Bob wrote:

>
>
> >  The thing I don't get,
> > though, is that not everyone has a network card
>
> Microsoft programmers are taught to always assign a "reasonable default".
> This is probably a random number if there is no net card. In practice it's
>
> almost the same, since that number would be common to lots of cookies
> and tied to an IP. It's still bad.

I see now.   So it's sort of like when 98 is having connectivity problems and
can't see its DHCP server, 98 just picks 162.254.##.## for an IP, something
close to that.

--

          .==,_
         .===,_`\                                Andy Harrison
       .====,_ ` \      .====,__
 ---     .==-,`~. \           `:`.__,     __
  ---      `~~=-.  \           /^^^      / /    __  _  _  _  _ __  __
    ---       `~~=. \         /         / /__  / / / \// //_// \ \/ /
                 `~. \       /         /____/ /_/ /_/\/ /___/  /_/\_\
                   ~. \____./           ...for IQs GREATER than 98...
                     `.=====
                  ___.--~~~--.__
        ___\.--~~~              ~~~---.._|/  (remove the obvious
        ~~~"                             /    to e-mail me...)
--




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

From: [EMAIL PROTECTED]
Subject: Re: Logging into Samba from Windoze 98
Date: Thu, 11 Mar 1999 17:40:25 GMT

Well you can do it that way or else just log into your Win98 box without a
password.  NT gave us the same grief even after we added all of the
authentication to it.


In article <7c24mp$fhf$[EMAIL PROTECTED]>,
  W.D. Allen <[EMAIL PROTECTED]> wrote:
> Yep, after I got the samba passwd file, smbpasswd, setup - everything went
> very well.
>
> :-)
>
> wda
> David Kirkpatrick wrote:
> > Did you setup smbpasswd?  See man smbpasswd.
> > d
> > "W.D. Allen" wrote:
> > >
> > > I have been trying to get Samba connectivity from my Linux box to my
> W98
> > >
> > > client box.
> > >
> > > From my W98 box, I can see the Linux box in the Mygroup that Samba is
> > >
> > > broadcasting.  However, when I double click the linux box, it
> identifies
> > >
> > > itself as //Linuxbox/IPC$ - and then wants a password.
> > >
> > > I followed the instructions in the Linux Network Toolkit book
> (published
> > >
> > > by IDG), but it obviously leaves out some important information.  What
> do
> > >
> > > I need to really do, is there a mini-howto that specifically addresses
> > >
> > > what should be a realatively simple Samba connectivity setup?  The
> Samba-
> > >
> > > Howto is about as clear as mud.  If someone could give me a short step-
> by-
> > >
> > > step, I would really appreciate it.
> > >
> > > Thanks,
> > >
> > > WDA
> > >
> > > ------------------  Posted via SearchLinux  ------------------
> > >                   http://www.searchlinux.com
> >
> > --
> > [EMAIL PROTECTED]
> > [EMAIL PROTECTED]
> > [EMAIL PROTECTED]
>
> ------------------  Posted via SearchLinux  ------------------
>                   http://www.searchlinux.com
>

============= Posted via Deja News, The Discussion Network ============
http://www.dejanews.com/       Search, Read, Discuss, or Start Your Own    

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

From: "Srinivasan K.S." <[EMAIL PROTECTED]>
Subject: TCP connection
Date: Thu, 11 Mar 1999 19:36:18 +0530

I have a piece of code in "C", which tries to establish a TCP connection
using Raw Sockets. But the destination says "checksum failed". Can anyone
give me pointers to the sample source, which establishes TCP connection
using raw sockets.

Thanks,
Srinivasan K.S.



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

From: [EMAIL PROTECTED] (Andre Dietisheim)
Subject: Re: Linux-Router: pppd dialing on local connect & bootup & route
Date: Thu, 11 Mar 1999 15:09:03 GMT

Thanx!!! I was on your site (nice!!) and I'll give it a try (as soon
as I read the DNS-HOWTO in order to unterstand what's going on).

>
>Andre Dietisheim wrote in message <[EMAIL PROTECTED]>...
>>Hi!
>>
>>I have installed a Linux-Machine as a Dial-out-Router for a small LAN:
>>
>==Snip
>>and the problem is that pppd dials out when I connect to the router
>>(same machine that has the modem attached to) using Telnet, Browser,
>>Mailer etc.
>==Snip
>>I guessed it to be a Lookup to the ISP's DNS
>
>
>Yes you guessed propably right
>Setup a caching name server
>- look at the DNS howto
>- or on my site
>http://www.toetsch.at/de/tips/index.html?linux/99/07_5.htm
>for an example with Bind 4.9.7.
>
>It helped for me.
>
>Also you could get HTML-Mail with some external IMG-Urls in them. If you
>want to read these, diald will dial your ISP. (I'm filtering my mail now)
>
>leo


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

From: "Duarte Cordeiro" <[EMAIL PROTECTED]>
Crossposted-To: linux.samba
Subject: Re: smb_dont_catch_keepalive: server->data_ready == NULL
Date: Wed, 10 Mar 1999 17:23:46 -0000


[EMAIL PROTECTED] wrote in message
<7c692i$voi$[EMAIL PROTECTED]>...
>Hello !!!
>
>I'm trying to smbmount our NT Server but I'm always getting this:
>
>[root@linux /mnt]# smbmount //NTS/Razvoj nts/
>Password:
>smb_dont_catch_keepalive: server->data_ready == NULL
>mount error: Invalid argument
>Please look at smbmount's manual page for possible reasons
>


If you look at the smbmount manpage...

MOUNT FAILURES:
  The mount(2) system call does not return any sensible error code, so you
have to guess what failed when you could not mount a share.[...]

Useful no ?
But, mount almost always report the error to /var/log/messages, so please
post the error here :)

Cya,

   Duarte



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

From: John Thompson <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.misc
Subject: Re: Which program to use scsi-tape drive?
Date: Wed, 10 Mar 1999 21:42:35 -0600

Ray wrote:

> Which program to use scsi-tape drive? The program taper is, what i see, only
> for Floppy-Streamer.

No. On my other linux machine I use taper with an Exabyte
2501 SCSI tape drive.  You need to have SCSI and SCSI tape
support available in linux and start taper thus: "taper -T
scsi"

I haven't managed to get taper to work on this machine yet
(Colorado 5GB IDE drive) :-(

-- 

-John ([EMAIL PROTECTED])

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

Date: Thu, 11 Mar 1999 13:45:33 -0500
From: Austin Gresham <[EMAIL PROTECTED]>
Reply-To: [EMAIL PROTECTED]
Subject: Re: Slack and DHCP

FrankB wrote:

> Anyone know of a DHCP client for Slack so my Linux box can co-exist in a
> DHCP-based Windows network???
>
> Thanks

The client is called dhcpcd.  It's available from most of the mirror sites,
like sunsite, mit, georgia tech.  I'm personally having some problems with
it, but this seems to be the only client available.

-Austin


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

Date: Thu, 11 Mar 1999 12:58:04 -0600
From: "Stephen M. Shelly" <[EMAIL PROTECTED]>
Subject: IPMasq Setup (3rd time is the charm)

I have asked this question b/4, and gotten some very thoughtful and
helpful replies, but I think I failed to ask the proper question, so
here goes again:

I have a Debian Linux box w/ 3 interfaces:
ppp0 : dials up to cable company and ONLY outbound traffic goes upstream

eth0 : connects directly to cable modem. Only inbound traffic from
internet comes this way
eth1 : connected to hub on local network

I need to use ip masquerade to hide the machine on my local lan.

Addresses:
ppp0 : real.ip.addr.ess netmask 255.255.255.0
eth0 : 10.0.0.1 netmask 255.255.255.240
eth1 : 192.168.1.1 netmask 255.255.255.0

I am having a real bear getting this to work.
I have tried ipfwadm w/2.0.36 and ipchains w/2.2.1 kernels. I think it
is a matter of the weird routing that must occur. ppp interface only
handles outbound traffic, eth0 is only incoming bound for my ppp
interface, eth1 is on local lan w/ only one client.
If anyone can give me a clue, it would be (as clues always are) greatly
appreciated.

Peace




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

From: [EMAIL PROTECTED] (Lew Pitcher)
Subject: Re: ppp0 -> eth0
Reply-To: [EMAIL PROTECTED]
Date: Thu, 11 Mar 1999 19:32:47 GMT

On Thu, 11 Mar 1999 14:21:52 -0500, "Paul" <[EMAIL PROTECTED]>
wrote:

>  Hi
>
>I have two networks I am trying to connect to each other using pppd.
>I setup the server side to assign IP's and I have configured the client side
>to call.  My connection is fine between the two servers, but machines on the
>client side can't see past the Linux client gateway.  It's as if the linux
>(client) box can't route over the ppp0 device. I know the ethernet routing
>is fine, because I have two other routers on the clinet side LAN and routing
>is fine.  If anyone knows how to get the routing working on two lans
>connected over a dialup ppp connection.  Please let me know.
>
> P.S.  Please don't send me simple routing solutions (I have tried all the
>possible routing options) it's more of an issue with pppd or arp or
>something.

Do you have IP forwarding enabled?
cat /proc/sys/net/ipv4/ip_forward should return 1
if it doesnt, then echo 1 >/proc/sys/net/ipv4/ip_forward
If /proc/sys/net/ipv4/ip_forward doesnt exist, recompile your kernel
with the IP_FORWARDING option enabled.

Lew Pitcher
System Consultant, Development Services
Toronto Dominion Bank

(Opinions expressed are my own, not my employers')

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

From: Andre Giordano <[EMAIL PROTECTED]>
Subject: IPCHAINS not working
Date: 11 Mar 1999 19:32:02 GMT

I have setup IP chains on my system and the `output` chain works great but 
I can't get ANY services working with the input chains.

This is the script I am using:

# this is the script for my firewall

/sbin/ipchains -F input
/sbin/ipchains -F output

# rules for outgoing packets
/sbin/ipchains -A output -p TCP -d 0.0.0.0/0 80 -t 0x01 0x10
/sbin/ipchains -A output -p TCP -d 0.0.0.0/0 23 -t 0x01 0x10
/sbin/ipchains -A output -p TCP -d 0.0.0.0/0 ftp-data -t 0x01 0x02

#rules for incomming packets
/sbin/ipchains -A input -i lo -j ACCEPT
/sbin/ipchains -P input DENY

/sbin/ipchains -A input -p TCP -s 0.0.0.0/0 ftp-data -d $myip 1024:5999 -j 
ACCEPT
/sbin/ipchains -A input -p TCP -s 0.0.0.0/0 ftp-data -d $myip 6010: -j 
ACCEPT
/sbin/ipchains -A input -p TCP -s 0.0.0.0/0 ftp -d $myip  -j ACCEPT
/sbin/ipchains -A input -p TCP -s 0.0.0.0/0 110 -d $myip -j ACCEPT
/sbin/ipchains -A input -p TCP -s 0.0.0.0/0 23 -d $myip -j ACCEPT
/sbin/ipchains -A input -p ICMP -j ACCEPT


==================  Posted via SearchLinux  ==================
                  http://www.searchlinux.com

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

From: William Burlingame <[EMAIL PROTECTED]>
Subject: Re: Installation of Xwindows Applications and other misc problems
Date: Wed, 10 Mar 1999 21:44:09 -0600

Found and installed 100dpi.  Good news and bad.  Helps reading a lot, some software
doesn't handle the larger fonts very well.  Not enough room on the buttons.  At
least there's hope.  I didn't want to have to resort to a poorer screen resolution
just to adjust the size to the characters.

Bill Burlingame

William Burlingame wrote:

> Thanks for the reply.  What config file do I edit for fvwm2 menu  and where can
> I find a 100dpi file?  It's not on any one the 3 CD's from RH.
>
> Bill Burlingame
>
> John Broadhead wrote:
>
> > William Burlingame wrote:
> >
> > > version 4.07 that came with the system.  I went through the procedures
> > > outlined on Netscapes web site and when I run Netscape form the menu,
> > > the old version runs.  The only way I can get the new version to run is
> > > by typing in a command line.
> >
> > In Xwindows, when you install a new application, putting itself on a windows
> > manager's menu is not one of it's jobs. These windows mamanger menus are
> > considered user's-preference, so you'll have to edit it yourself.
> >
> > > In an unrelated topic, I made an executable script to disconnect from my
> > > internet server and I always get a 'command not found' message, even
> > > when I run it from the directory where it resides.  I have set the
> > > execute attributes for the file.
> >
> > Very possibly you have this script in a directory which is not in your path.
> > If you go into the directory in which it resides and type (assuming script
> > is its name) "./script" and it still doesn't work, check to make sure the
> > permissions are right and that the path to the shell interpreter at the
> > begining of the file is right, ie #!/usr/local/bin/perl. Make sure that perl
> > is really in /usr/local/bin.
> >
> > > The font in netscape is barely readable and I would like to make it
> > > bigger.  I have changed the font on the Netscape preferences and it
> > > doesn't make any difference.  Is there a way to make the Xwindows
> > > default font size larger?
> >
> > Go into your XF86Config file (On redhat its in /etc/X11/) and swap the 75dpi
> > and 100dpi FontPath lines, so the 100dpi one comes first. Note: you have to
> > actually have the 100dpi fonts for this to work. Restart X, and Netscape
> > will use a 33% larger font.
> >
> > -John Broadhead


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

From: [EMAIL PROTECTED] (Frank Hahn)
Subject: Re: Network Newbie Question
Date: 11 Mar 1999 03:53:44 GMT
Reply-To: [EMAIL PROTECTED]

On Tue, 09 Mar 1999 18:36:13 -0600, H.G.Williams <[EMAIL PROTECTED]> wrote:
>I have a 2 machine home network, one is win '95, the other Linux. Both
>have Ethernet cards installed (the win '95 is a NE2000 compat and Linux
>is a 3Com 905). RedHat v5.2, kernel 2.0.36 is the software.
>
>I want to just configure TCP/IP        for both and have done the following:
>Linux                          Win '95
>IP=192.168.15.2                        192.168.15.4
>Netmask=255.255.255.0          255.255.255.0
>gw=192.168.15.254              192.168.15.254
>
>the whole thing is started by scripts and manually by:
>route add -net 192.168.15.0 eth0
>route add default gw 192.168.15.254
>
>They are connected via a hub, and the LED's are on.
>
>Yet, when I run 'ping 192.168.15.4' from the Linux box, there is no
>response (if I ping it's own IP, there is a reply). Same thing from the
>win '95 box. I can ping it's IP, but not the Linux box. I've changed
>cables, hubs, NIC's, all to no avail! I've read Net3 from cover to
>cover, twice, no help. I have a 'host' file on both machines (even a
>LMHOSTS on the '95 box). I can run -
>       route print
>and get the following info: (206.133.136.82 is my ISP addr).
>
I don't understand exactly what you are trying to do.  Does one
of these machine have a modem hooked up to them?

I'm not sure your line above with the "gw" in it is correct.

I have three machines networked together at home.  One is a
Linux machine, one is Windows 95 machine and one is a Sun Sparc
machine.  The Linux machine has the modem which dials my ISP.

All three machines have hosts files which have IP numbers and
machine names for all three machines.  The Windows machine has the
hosts file in c:\windows.

The Windows machine has the IP address of the Linux machine as its
gateway address in the Windows networking setup.

My startup scripts on the Linux machine are as follows:

================cut here===================
#! /bin/sh
#
# rc.inet1      This shell script boots up the base INET system.
#
# Version:      @(#)/etc/rc.d/rc.inet1  1.01    05/27/93
#

HOSTNAME=`cat /etc/HOSTNAME`

# Attach the loopback device.
/sbin/ifconfig lo 127.0.0.1
/sbin/route add -net 127.0.0.0 netmask 255.0.0.0 lo
 
# IF YOU HAVE AN ETHERNET CONNECTION, use these lines below to configure the 
# eth0 interface. If you're only using loopback or SLIP, don't include the
# rest of the lines in this file.
        
# Edit for your setup.
IPADDR="192.168.100.10" # REPLACE with YOUR IP address!
NETMASK="255.255.255.0" # REPLACE with YOUR netmask!
NETWORK="192.168.100.0" # REPLACE with YOUR network address!
BROADCAST="192.168.100.255"     # REPLACE with YOUR broadcast address, if you
                        # have one. If not, leave blank and edit below.
#GATEWAY=""     # REPLACE with YOUR gateway address!
        
# Uncomment the line below to initialize the ethernet device.
/sbin/ifconfig eth0 ${IPADDR} broadcast ${BROADCAST} netmask ${NETMASK}
        
# Uncomment these to set up your IP routing table.
/sbin/route add -net ${NETWORK} netmask ${NETMASK} eth0
================cut here===================

Since my Linux machine is the gateway for the other two machine to the
Internet, I don't point it to a gateway.  If you are trying to go
through your Windows machine to the Internet, someone else will
have to give you a hand.

Also, I assume you are fairly confident that your cards are working,
you cables are the correct ones, you have networking compiled correctly
into your kernel, you have the correct device driver configured into
your kernel, etc.  There are probably dozens of other problems also.

One thing I have found, I try and compile things directly into the kernel
until I get them figured out.  I will then go back and compile things
as modules if needed.

Good luck.

-- 
Frank Hahn

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

From: [EMAIL PROTECTED] (Sparkzz)
Subject: lost rsh
Date: 11 Mar 1999 04:05:34 GMT

Hello,

Somewhere along the road to kernel 2.0.36, I have lost the capability of
running   rsh.

If I run "rsh foo", foo asks for a password.
If I run "rsh foo ls", I get a Permission denied.

I have checked everything I know to check.

Anyone got any suggestions??

thanks,
. 
. 
....Ken

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

From: "¤ñ¥d¶W" <[EMAIL PROTECTED]>
Subject: Re: problem with routing
Date: Thu, 11 Mar 1999 12:11:08 +0800

Since I don't know which command should I enter after modify the /etc/hosts
file, so I reboot the computer.
But after rebooting, I enter route to see the route table, the table doesn't
change at all.

average ¼¶¼g©ó¤å³¹ <[EMAIL PROTECTED]>...
>yes both machines have entries in /etc/hosts as well as the localhost
>this works as when i use the nickname of the machine the correct ip
>address comes up
>
>
>
>Ron wrote:
>>
>> Did you put the hostnames in /etc/hosts?
>>
>> average <[EMAIL PROTECTED]> wrote in message
>> news:[EMAIL PROTECTED]...
>> >i have a two machine network linux & win95 on twisted pair through a
>> >hp 28684a hub
>> >the problem i have is both machines can ping them selves (nic &
>> >localhost) but cannot ping each other. at this point in time i haven't
>> >set up anything to do with samba.
>> >
>> >Kernel IP routing table
>> >Destination     Gateway         Genmask         Flags   MSS Window  irtt
>> >Iface
>> >zeus.tulku.net  *               255.255.255.255 UH     1500 0          0
>> >eth0
>> >192.168.1.0     *               255.255.255.0   U      1500 0          0
>> >eth0
>> >127.0.0.0       *               255.0.0.0       U      3584 0          0
>> >lo
>> >default         gyver.tulku.net 0.0.0.0         UG     1500 0          0
>> >eth0
>> >
>> >the hub works (it was used by a friend) i have changed the cables, all
>> >the lights that
>> >should be on are on and flashing (data lights on the cards, the hub etc)
>> >
>> >any help
>> >baz



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

From: Ron Flory <[EMAIL PROTECTED]>
Subject: Re: Linux + Novell 4.11
Date: Wed, 10 Mar 1999 22:19:49 -0600

hi-

 Wer'e using ipx_configure and ncpmount- works ok, but slower then NFS.

ron

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

From: [EMAIL PROTECTED] (brett russ)
Crossposted-To: comp.os.linux.hardware
Subject: Modem: no luck w/simple AT cmd
Date: 11 Mar 1999 19:33:42 GMT

I'm running RH5.2 on 486/66.  There is one serial port on back of
machine which I am using for my mouse.  It is mapped to /dev/ttyS1
(COM2) and is IRQ 3.  I am installing a Gateway (USR) Telepath 550
14400 internal ISA modem with jumpers configured for IRQ4, COM1.  My
BIOS does not provide access to serial port info (enabling/disabling)
so I assume I can use this port.  It's a SystemSoft BIOS and the PC is
a VTech.

My problem is that I've followed all of the instructions I've found in
the modem manual, PPP HOWTO, and serial HOWTO to simply issue an AT
command to talk to this modem and I have no luck.  Nothing shows up on
the screen when I type and nothing shows up from the modem in
response.  It feels like I'm talking to a wall.  I have used both
Kermit and minicom, setting it to look for the modem at /dev/ttyS0, to
no avail.  In Kermit I've tried setting modem type to both usrobotics
and telepath.  I'm running them as root so permissions shouldn't be
the problem.  I've run setserial on the port and it reports a 16450
UART and the address and IRQ look fine.  Neither Kermit or minicom
complained when initializing/connecting to the modem.  I've looked at
/proc/interrupts before and during the modem programs.  Before nothing
was shown for IRQ 4, during it showed activity for this interrupt.

Once I typed something and got a dialtone and some activity on the
line which I heard from the speaker on the modem but I typed another
key and it stopped.  I had some hope for a second but I could not
repeat it.  And once I got "trapped" in Kermit while connected to the
modem and could not issue the escape sequence to get out to the Kermit
prompt.  I did Ctrl-z to suspend, and when I typed 'fg' to return, I
saw an "OK" as if the modem had replied.  This too happened only once.

I am wondering if the modem takes forever to reply, but why wouldn't
my commands show up?  At this point in the installation, does it even
matter what the modem's settings are?  I tried 'stty <the string to
enable CTS/RTS HW flow control> < /dev/ttyS0' but this changed
nothing.  My modem has all dip switches in default locations.  Next
I'll try a different COM port/IRQ (I suppose I could use 5...) but I
don't know what's going on with my current config.  Is the modem
toast?  I've tried swapping ISA slots.

Thanks--
br


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

From: [EMAIL PROTECTED] (Georg Schwarz)
Crossposted-To: comp.os.linux.misc,comp.unix.ultrix
Subject: Linux 2.2.3 and Ultrix 4.3: incompatible NFS?
Date: 11 Mar 1999 19:39:47 GMT

I've just upgraded a PC with RedHat Linux 5.2 installed from kernel 2.0.36
to 2.2.3. I've noticed the following problem:

when I access (by loging in as a user) /home, which is actually a symbolic
link to /net/itp4/home, with /net being the amd automount directory and
itp4 being an Ultrix 4.3 box serving as NFS server for /home, I get

NFS: server itp4, readdir reply truncated
NFS: nr=43, slot=-1, len=0

and then the login shell (tcsh) hangs.
syslog says the same.

Now, when I stop the login script with ^c, I get into the linux machine.
When I then type "pwd"
I get the following console message on the Ultrix NFS server:

xdr_bytes: bad size FAILED
svcheckudp_send: xdr_replymsg failed
rfs_dispatch: sendreplay failed

which keeps repeating all over.
In the meantime, the linux shell hangs indefinitely.
Syslog says:
nfs: itp4 server not responding

manually mounting /home from the server works fine (as root).

When I change (as root) into /var/spool/mail, which is also amd mounted
from that same Ultrix server, I also get that "truncated" message when
that directory is mounted. Subsequently doing an ls only gives me the
first douzend or so mailboxes, with the other files being seemingly
nonexistant (they are still present on the server, but I don't know what
would happen if one tried an NFS write to that directory).

I have taken the amd src rpm and rebuilt it after upgrading to 2.2.3,
but this only made go away that "kernel is newer than nfs" (or similar)
warning message. It's am-utils-6.0a16-4.
Is this a known problem? Is my configuration at fault? Is it the
kernel's fault or amd's?
-- 
Georg Schwarz ([EMAIL PROTECTED], [EMAIL PROTECTED], PGP 2.6ui)
Institut für Theoretische Physik  +49 30 314-24254   FAX -21130  IRC kuroi
Technische Universität Berlin            http://home.pages.de/~schwarz/

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


** 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