Linux-Networking Digest #118, Volume #10 Fri, 5 Feb 99 18:13:44 EST
Contents:
Modem Problem (Red Hat Linux User)
Re: PPP problems under 2.2.1 (Alan Jaffray)
Sendmail and Agent (Matthew Willcock)
Re: Accessing a HTTP proxy from Linux (Paul A. Cheshire)
Re: ???:Odd Masqu "Problem"??? (Paul A. Cheshire)
Ipmasqurading (Ron Logan)
Re: Serial overruns/collisions (incoming only oddities) (Edmund H. Ramm)
Re: Romote "root" login ("Vikram.V.Asrani")
Re: ncpfs -- slist -- no server found in ncp_open (Frank Miles)
Re: DNS forwarding ("Chris Richards")
Re: using diald (Tony Schlemmer)
please don't make me keep using this stupid Win NT server (Daddy Rabbit)
Re: Yes folks, another newbie ppp problem (Clifford Kite)
PPP link takes over as a router? ("Bernard, Greg")
RedHat 5.2 & 2.2.0 & yp (Kirk Vogelsang)
Re: Mascerading & Ip Adresses (Doug Goldstein)
Help, I'm at my wits end over PPP (The Scot)
dctrl for diald only reports "not configured" (Terry Lecander)
Standard host name for standalone box? (Jay)
eth0+eth1+ route how to ("toni")
Re: How do you install BRCFG? (Ray Willis)
----------------------------------------------------------------------------
From: Red Hat Linux User <[EMAIL PROTECTED]>
Subject: Modem Problem
Date: Fri, 05 Feb 1999 03:24:07 +0000
Hi, everyone, I have one problem with my modem. The modem is ROCKWELL
V.90 PCI F/M/V. It works well under Wondows 95 (on COM2), but under
Linux 2.0.36, no response from the modem when dailing out with minicom.
Any suggestions and comments are welcome! Thanks.
------------------------------
From: [EMAIL PROTECTED] (Alan Jaffray)
Subject: Re: PPP problems under 2.2.1
Date: Fri, 5 Feb 1999 10:21:57 GMT
Sami I Makelainen <[EMAIL PROTECTED]> wrote:
>After upgrading to 2.2.1 (from 2.0.32), the PPP connection stopped
>working. All necessary software has been upgraded with the kernel (the
>system is RH5.2 based). Login and PAP authentication apparently succeed,
>but nothing goes through.
I am having the same trouble. Curiously, DNS lookups succeed, but
nothing else seems to...
>The "bad frame" errors (see log) keep coming about every 30
>seconds. Pinging any host also causes a bad frame message.
I'm not getting these messages; perhaps I don't have the debug
level set high enough.
>The modem is actually a ZyXel external ISDN connected via ttyS2 with
>high-speed serial card - in essence acting as a very high speed modem.
In my case, it's a Ricochet wireless modem connected via ttyS2.
Not especially high-speed, but like your ISDN, not a phone modem.
>Any help would be greatly appreciated.
Same here.
Alan
--
Alan Jaffray <[EMAIL PROTECTED]>
there's nothing you can do that can't be done
------------------------------
From: [EMAIL PROTECTED] (Matthew Willcock)
Subject: Sendmail and Agent
Date: Fri, 05 Feb 1999 11:33:54 GMT
I have configured sendmail to queue all outgoing mail on my RH5.2 box.
but have hit a problem when sending from Forte Agent from Win95. I
click send, with the smtp server configured for the linux box, and
after a delay the email gets sent, but, an outgoing udp packet is sent
which triggers diald to dial, Diald reports that the packet is from
the win95 machine to the linuxbox and on netbios-ns port.
This doesn't happen if I telnet to the linuxbox on the smtp port and
enter an email that way.
looking at the email headers seems to indicate a problem, this is from
Agent:
Received: from unknown (claudia.house [192.168.0.1])
by jasmine.house (8.8.7/8.8.7) with SMTP id KAA06041
for <[EMAIL PROTECTED]>; Fri, 5 Feb 1999 10:39:45 GMT
This was when I telneted to smtp port
Received: from jasmine.house (claudia.house [192.168.0.1])
by jasmine.house (8.8.7/8.8.7) with SMTP id KAA06036
for [EMAIL PROTECTED]; Fri, 5 Feb 1999 10:36:12 GMT
Any ideas? The whole point was to stop diald dialing when sending
mail :-)
Matthew
Matthew Willcock
[EMAIL PROTECTED]
http://www.mwillc.free-online.co.uk
------------------------------
From: [EMAIL PROTECTED] (Paul A. Cheshire)
Subject: Re: Accessing a HTTP proxy from Linux
Date: Fri, 5 Feb 1999 01:32:56 +0000
Reply-To: [EMAIL PROTECTED]
On Fri, 22 Jan 1999 19:53:04 GMT,
Francois Jaccard <[EMAIL PROTECTED]> wrote:
>Hi,
>I have a cable-modem on my Win98 machine and I would like to be able to access
>internet through it from Linux (Red Hat 5.2). I am running Wingate 2.1d on the
>Win98 machine.
>
>How can I use the Wingate machine as HTTP Proxy for Linux? I can browse from
>another win98 machine via wingate so it is configured correctly but how do I
>use a HTTP proxy on Linux with wget?
>I put the IP of the wingate machine (192.168.0.1) as gateway but it does not
>work.
^
|
I believe you cannot use zeroes or 255 in normal IP addresses. They are used
for network and netmask addreses only.
HTH
>
>Thanks!
--
Windows: I can play Doom | Paul A. Cheshire
Linux: I can be a file, web | mailto: [EMAIL PROTECTED]
and news server AND play Doom. | url: http://www.nader.demon.co.uk
------------------------------
From: [EMAIL PROTECTED] (Paul A. Cheshire)
Subject: Re: ???:Odd Masqu "Problem"???
Date: Fri, 5 Feb 1999 01:27:02 +0000
Reply-To: [EMAIL PROTECTED]
On Wed, 03 Feb 1999 04:21:09 GMT,
Jim Harper <[EMAIL PROTECTED]> wrote:
>"Douglas E. Mitton" wrote:
>>
>> I am experiencing an odd "problem" in my IP Masq operation ... it
>> happened with V2.0.36 and ipfwadm and is still occuring with V2.2.1
>> and ipchains.
>>
>> The problem is that when I access a web page with graphics (?) on the
>> same subnet as me and my ISP and only from the PC behind my "firewall"
>> some graphic images refuse to complete loading.
>
>I get the same thing on my Win95 machine which is behind a firewall.
>
>If your situation is like mine, my ISP has local "members only" pages
>that reference http://www/somepage.html. The problem stems from the fact
>that when your browser sees a request for a page on the host "www", DNS
>assumes that it is on the local network (198.168.x.x in my case) behind
>the firewall. Of course if you don't have a local host on your local lan
>named www, you'll never get the page.
>
>In my situation, I'm able to get the pages by typing in the FDQN (Fully
>Qualified Domain Name) as www.myprovider.com/somepage.html, or by using
>the ip address. However, like your situation, I don't see any graphics
>due to the way in which the page HTML was written.
>
>If they enter an IMG tag like:
>
><IMG SRC="http://www/graphics/some.gif">
>
>Then once again your browser will not be able to find the host "www" on
>which the image is located.
>
>Running a proxy server to retrieve the pages would probably solve the
>problem. That way the ISP side of your firewall would make the request
>for the page and forward it to your local host.
>
>I don't have much experience with proxy servers, so I you'd have to dig
>for more information on them than I can provide.
Apache and Squid can perform this function. I run Apache on my firewall Linux
box which also acts as dial-on-demand router for my 192.168.x.x internal LAN.
AS all requests are to/from the ISP provided *real* IP address on the dial up
side all 3 of my LAN PCs (1 Linux, 2 win98) can access all web pages with no
trouble.
HTH
Paul
>
>Anyone else have a solution to this problem?
>
>-Jim
>
>Remove the obvious portion of my address to respond
--
Windows: I can play Doom | Paul A. Cheshire
Linux: I can be a file, web | mailto: [EMAIL PROTECTED]
and news server AND play Doom. | url: http://www.nader.demon.co.uk
------------------------------
From: Ron Logan <[EMAIL PROTECTED]>
Subject: Ipmasqurading
Date: Fri, 05 Feb 1999 01:10:56 -0700
Reply-To: [EMAIL PROTECTED]
I can't seem to find the how-to on IPmasqurading would some one please
give me a url.
Thanks
Ron
------------------------------
From: [EMAIL PROTECTED] (Edmund H. Ramm)
Subject: Re: Serial overruns/collisions (incoming only oddities)
Date: Fri, 5 Feb 1999 17:00:57 GMT
In <[EMAIL PROTECTED]> Robert Logan <[EMAIL PROTECTED]> writes:
> [...]
>The PC's connection is via CSLIP (PPP suffers the
>same), with the port set at 115200, the IRQ is
>not in conflict, indeed, I have been using the
>same DIP script/cfgs (accounting for upgrades)
>since my 1st kernel.
And you've specified 'send terminal flowcontrol rts\n' in your
dip-script?
> [...]
Eddi
--
email: ehramm AT dk3uz DOT hh DOT provi DOT de | AMPRNET: [EMAIL PROTECTED]
If replying to a Usenet article, please use above email address.
Linux/m68k, the best U**x ever to hit an Atari!
------------------------------
Crossposted-To: comp.os.linux,comp.os.linux.admin,comp.os.linux.help,nl.comp.os.linux
From: "Vikram.V.Asrani" <[EMAIL PROTECTED]>
Subject: Re: Romote "root" login
Reply-To: [EMAIL PROTECTED]
Date: Fri, 5 Feb 1999 16:43:08 GMT
In comp.os.linux [EMAIL PROTECTED] wrote:
=> /etc/sudoers controls which user may use su IIRC, so that means that you can
=> effecively make another wall you'll have to overcome before you even become
=> root.
I tried that but it does not seem to work. Anything that I missed
out. All I wrote in the file was the name of the user who is allowed
to use su and then I tried running su from someone not listed in
that file. This succeeded in making me root .
--
Vikram.V.Asrani.
([EMAIL PROTECTED])
------------------------------
From: [EMAIL PROTECTED] (Frank Miles)
Crossposted-To: comp.os.linux.misc,comp.os.linux.setup
Subject: Re: ncpfs -- slist -- no server found in ncp_open
Date: 5 Feb 1999 20:21:26 GMT
In article <[EMAIL PROTECTED]>,
Mike Freitas <[EMAIL PROTECTED]> wrote:
>Linux will not see my NetWare 4.11 server. I've compiled my kernel
>(2.0.36) with ipx protocol and ncp support and rebooted. Ncpfs-2.0.10
>has been compiled. IPX has been started both with the ipx_interface add
>-p eth0 802.3 and ipx_configure --auto_interface=on --auto_primary=on
>and slist still fails. I get the above error stating that 'no server
>found in ncp_out'.
For reasons that I've yet to determine, my Netware FS is invisible
until after I've telnetted or otherwise communicated with other,
non-Netware computers. Once that has been done, slist and pqlist
and other ipx/ncp utilities work fine. Have you tried that?
-frank
------------------------------
From: "Chris Richards" <[EMAIL PROTECTED]>
Subject: Re: DNS forwarding
Date: Fri, 5 Feb 1999 13:46:49 -0600
I just did this myself with success.
In /etc/named.conf (bind 8.x) add:
options {
forward [first | only];
forwarders {
ip1;
ip2;
};
};
You can list as many ip's as you want.
Cheers,
cjr
Ya Wen <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]...
>
>Hi, all:
>
>I just set up a new Red Hat 5.2 machine, I want to use it as the primary
>DNS server for my little network. But becuase my network is within my
>company's intranet, I also need to lookup all the machines in the
>intranet. I know in Solaris, we can define a DNS forwarder, just wondering
>if there is also such mechanism I can configure in /etc/named.conf to
>forward DNS requests.
>
>Thanks a bunch!
>
>-Ya
>
------------------------------
From: Tony Schlemmer <[EMAIL PROTECTED]>
Subject: Re: using diald
Date: 5 Feb 1999 20:52:29 GMT
Reply-To: [EMAIL PROTECTED]
Name <[EMAIL PROTECTED]> wrote:
[snip]
>>
>> I too am running a Linux server with IP Masquerading. My server has an IP
>> address of 198.162.1.1 but I use 127.0.0.2 for local and 127.0.0.3 for
>> remote in the diald.conf file. I'm not sure it makes sense to use the
>> same IP address for both the local and remote settings in the diald.conf
>> file.
>>
>> Tony
>>
>>
> One problem I'm having, and maybe your suggestion will fix this, is that
> diald dials the modem before any of the client machines are even logged
> in. At that time, there should be no outbound packets. I'm wondering if
> I, by putting my linux box as the local/remote ip for diald, caused diald
> to want to dial the line when something like a DHCP request comes in?
> If so, then your suggestion might help.
> Thanks
> Reggie
I've heard reports from others that Win9x boxes are notorious for causing
diald to dial the modem all of the time. I had a set of firewall rules I
use to keep all of the netbios name server lookups from going outside
of my LAN and so diald only dials when someone inside the network
starts up a web browser or email client.
I can email you my firewall rules if you want them. I run then
automatically from diald using the "ip-up", "ip-down" directives
in the diald.conf file.
Tony
--
Tony Schlemmer
Global Mobility Systems, Inc.
[EMAIL PROTECTED]
------------------------------
From: [EMAIL PROTECTED] (Daddy Rabbit)
Subject: please don't make me keep using this stupid Win NT server
Date: Fri, 05 Feb 1999 21:15:53 GMT
Reply-To: [EMAIL PROTECTED]
I have Sybase running on a linux box. I want to be able to create web
pages that are connected to Sybase on a linux box don't know how.
I also have Sybase running on an NT box. NetImpact Dynamo can as a
part of that package. It allows you to create web pages using sybase
as the sql server. I can them publish the web pages to another server.
(not sure if they will work on a linux box or not. Haven't gotten that
far)
My hope is that someone out there can tell me how I can accomplish the
same using just a linux box so I can trash this NT box one and for
all.
Any and all suggestions are most welcome.
TIA
Jim
------------------------------
From: [EMAIL PROTECTED] (Clifford Kite)
Subject: Re: Yes folks, another newbie ppp problem
Date: 5 Feb 1999 15:06:41 -0600
Eric Zwicky ([EMAIL PROTECTED]) wrote:
: (cross-posted to linux.redhat.ppp)
: I can connect to my ISP (earthlink) and pick up an IP from the dhcp
: server, but I can't "surf" nor ping the nameservers, or any other IP
: address I know by heart. The error message from ICMP says "network is
: unreachable".
<snip>
: "route" says:
: Kernel Ip routing table
: Destination Gateway Genmask Flags Metric Ref
: Use Iface
: 38.1.1.1 * 255.255.255.255 UH 0
: 0 0 ppp0
: 192.168.0.0 * 255.255.255.0 U 0
: 0 3 eth0
: 127.0.0.0 * 255.0.0.0 U 0 0
: 2 lo
You may just need to add the pppd option "defaultroute", the host route
exists but not the default route.
--
Clifford Kite <[EMAIL PROTECTED]> Not a guru. (tm)
/* I gave up on politics when no matter who I voted for, I regretted it.
* -- Pepper...and Salt, WSJ */
------------------------------
From: "Bernard, Greg" <[EMAIL PROTECTED]>
Subject: PPP link takes over as a router?
Date: Fri, 05 Feb 1999 13:26:55 -0800
Reply-To: [EMAIL PROTECTED]
Hi all,
I have a strange situation here that I need some help with. I have a
Linux box here at work running Caldera OpenLinux 1.2, kernel 2.0.33,
pppd 2.2.0f-7.
At home, I have a RedHat 5.2 system, kernel 2.0.36, pppd 2.3.5.
My box at work is set up as a pppd server and I can get in and do
everything, no problems whatsoever.
The problem is that every once in a great while after a ppp session, my
box at work will take over as a router which causes our LAN routers to
go nuts (90%+ CPU). When I down/up the eth0 interface here, the problem
goes away.
The box at work uses DHCP to get its IP address, the one at home makes
the PPP connection on a reserved IP address, so there is no collision
problems here.
Has anyone seen this problem before? Anyone know what I can do to fix
it? Any help at all would be greatly appreciated.
TIA.
Greg Bernard
------------------------------
From: Kirk Vogelsang <[EMAIL PROTECTED]>
Subject: RedHat 5.2 & 2.2.0 & yp
Date: 05 Feb 1999 16:29:23 -0500
I've just upgraded to Intel/RH5.2 and kernel 2.2.0
and am having some trouble w/ yp.
When I ls a dir, it spits out numeric uid's
and gid's instead of their corresponding names.
ypcat works. ypmatch works. domainname is correct.
+:*:0:0::: is the last line in /etc/passwd
+ is the last line in /etc/group
passwd and group are set to compat in nsswitch.conf.
(I tried 'files nis' to no avail)
id foo produces:
id: foo: No such user
ypmatch foo passwd works fine tho.
cd ~foo works fine.
foo is a user in the passwd map.
Here's what I'm using:
ypserv-1.3.5-1
yp-tools-2.1-1
ypbind-3.3-9
It's a fresh RedHat 5.2 install.
What am I missing? TIA!
=====
Kirk M. Vogelsang <[EMAIL PROTECTED]>
Northeastern University College of Computer Science
------------------------------
From: Doug Goldstein <[EMAIL PROTECTED]>
Subject: Re: Mascerading & Ip Adresses
Date: Fri, 05 Feb 1999 22:09:47 GMT
Although I'm not too familiar with that kind of setup I think I might know
the problem. I've heard of something similiar happening and it being fixed
by referencing the other internal IPs with ipfwadm. But this is on a Redhat
5.1 box. I heard that if you setup more hosts you sometimes need to add
/sbin/ipfwadm -F -a m -Sxxx.xxx.xxx.xxx/netmask -D0.0.0.0/0 xxx.xxx.xxx.xxx
being the other IPs that can't be reached.
I only have one Win98 box connected to my Linux box but to get my cable
modem and masq setup I put this in /etc/rc.d/rc.local
echo "ip_masq 192.168.1.1"
echo "1" > /proc/sys/net/ipv4/ip_forward
/sbin/depmod -a
/sbin/modprobe ip_masq_ftp.o
/sbin/modprobe ip_masq_raudio.o
/sbin/modprobe ip_masq_irc.o
/sbin/modprobe ip_masq_quake.o
/sbin/modprobe ip_masq_cuseeme.o
/sbin/modprobe ip_masq_vdolive.o
/sbin/ipfwadm -F -p deny
/sbin/ipfwadm -F -a m -S192.168.1.0/24 -D0.0.0.0/0
/sbin/ifconfig eth1 192.168.1.1
/sbin/route add -net 192.168.1.0
Hope it gives you some ideas or helps.
Doug Goldstein
------------------------------
From: The Scot <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.setup,alt.os.linux.caldera
Subject: Help, I'm at my wits end over PPP
Date: Fri, 05 Feb 1999 21:47:22 GMT
Hello again,
Thank you to everyone who has helped so far, alas I still can't get ppp to
work and I'm really lost as to what to try next. I just wnat to surf with
linux, thats all. I've tried using kppp, I set things up with netcfg but then
kppp still wouldn't work, I tried pppd but that didn't work either...
So I tried eznet. I made sure ttyS3 was linked to /dev/modem
I downloaded the binary and copied it to /usr/bin and did a chmod 04755 as
the read me instructed. I entered the info and then tried eznet tir up, it
just sits there I tried multiple times and it just sits there and doesn't
come back with anything.
Am I doomed to surf with only windoze, I really don't understand why its not
working! Please help if you can!
cheers,
Dom
============= Posted via Deja News, The Discussion Network ============
http://www.dejanews.com/ Search, Read, Discuss, or Start Your Own
------------------------------
From: Terry Lecander <[EMAIL PROTECTED]>
Subject: dctrl for diald only reports "not configured"
Date: Fri, 05 Feb 1999 12:24:35 -0600
Reply-To: [EMAIL PROTECTED]
I have been trying to use diald control 'dctrl' on a host running diald
diald-0.16.5-2d.i386.rpm
diald-config-1.2.1-1.noarch.rpm
When I try to run 'dctrl', it will usually load but is completely unresponsive. I will
end up having to kill the pid to close it. About 10% of the time it will just work out
of the blue without any changes made by me. The error log reports that FIFO could not
open
pipe <filename> : Device not configured.
Looking at '/var/log/messages' is see the
floowing.
**************
linux diald[464]: FIFO: full monitor connection to monitor /tmp/dctrl.517 requested
linux diald[464]: FIFO: could not open pipe /tmp/dctrl.517: Device not configured
linux diald[464]: FIFO: full monitor connection to monitor /tmp/dctrl.568 requested
linux diald[464]: Monitor pipe /tmp/dctrl.568 closed.
linux diald[464]: FIFO: full monitor connection to monitor /tmp/dctrl.568 requested
linux diald[464]: FIFO: could not open pipe /tmp/dctrl.568: Device not configured
***************
On the third line you can see an attempt that was successful followed by a 'closed'
when I manually close the 'dctrl' screen. Between these attemps I have done nothing
but close
'dctrl' and re-start it. I have verified that the files in the '/tmp' directory have
been created.
Any help would be appreciated.
I am a relative newbie swimming in the seas of confusion.
******************
Terry Lecaner
Red Nickel Systems
[EMAIL PROTECTED]
******************
------------------------------
From: Jay <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.misc
Subject: Standard host name for standalone box?
Date: Fri, 05 Feb 1999 14:07:14 +0000
All,
I'm curious if there is a standard hostname convention for stand
alone Linux boxes. I run Linux on my home machine and thos hostname is
localhost.localdomain. Is it customary to change that to something more
personal?
I don't have a permant IP address, just a PPP dialup address.
Thank you.
Take care,
Jay
--
[EMAIL PROTECTED]
http://www.roadrunner.com/~joconnor
"God himself plays the bass strings first when He tunes the soul"
------------------------------
From: "toni" <[EMAIL PROTECTED]>
Subject: eth0+eth1+ route how to
Date: 5 Feb 1999 13:38:40 GMT
I have a problem route table with two ethernets. can someone help me.
thanks.
------------------------------
From: Ray Willis <[EMAIL PROTECTED]>
Subject: Re: How do you install BRCFG?
Date: Fri, 05 Feb 1999 07:52:15 -0600
If you downloaded it from the site named BRCFG.tgz then just extract it and
it should have an executable in the tar. If you would like to compile it
your self type "make brcfg" in that same dir :)
Stuart Liebrecht wrote:
> I downloaded Chris Cole's bridging utility 'BRCFG' and untarred it, but
> I can't figure out how to install it. Please forgive me if this is a
> stupid question, but I just can't figure it out.
>
> Stuart
------------------------------
** 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
******************************
