Linux-Networking Digest #919, Volume #10 Tue, 20 Apr 99 05:13:56 EDT
Contents:
Re: 2 ISA Ethernet cards (John Engel)
Re: please help! apache problem with numeric IP. (David Efflandt)
ISDN ("Alex")
Re: NFS WITHIN AN IP MASQ FIREWALL (fwd) (Paul Rusty Russell)
Re: Linux as NetWare Server? (Tom)
Re: VERY URGENT !!! : Samba networking (Fabrice Coudert)
Modem doesn't work. ("Xinxin Shao")
Re: pppd error explanation needed (Clifford Kite)
nis+ (mamo)
kppp: Got it working with NT RAS & MS-CHAP ! ([EMAIL PROTECTED])
Re: No way to avoid "denial of service attacks"? ("donoli")
fetchmail shall wait (Christophe Zwecker)
Re: Configure NE2000 compatible ethernet NIC ("Morten")
Re: Portable - Home/Work IP Configuration ([EMAIL PROTECTED])
Re: ISDN-TA configuration, need help! (Lim Chuan Wee)
NFS client ("martin")
Re: SIOCADDRT: Invalid argument (Javier Prieto)
----------------------------------------------------------------------------
From: John Engel <[EMAIL PROTECTED]>
Subject: Re: 2 ISA Ethernet cards
Date: Tue, 20 Apr 1999 03:53:42 GMT
The problem turned out to be an I/O confilct. I configured the Linksys card
to use 0x380 instead of 0x340 and everything is working fine.
kel wrote:
> I just installed RH Linux 2.0.36 with an Intel EtherExpress and went
> through hell with it. Had to go to 2.2.5 kernel where there is an
> improved eexpress.o module. Might need to configure the card with c800
> and 32k. It seemed the module was looking for those specs at boot.
> Dunno about irq 11 but 0x300 is a preferred setting.
>
> Still cant get samba working but at least the machines can talk to
> each other, and transfer large files without CRC errors.
>
> On Mon, 12 Apr 1999 05:48:05 GMT, John Engel <[EMAIL PROTECTED]>
> wrote:
>
> > I have 2 ISA ethernet cards that I want to configure on a Redhat 5.2
> >box using the 2.0.36 kernel. One card is an Intel EtherExpress card and
> >the other is a Linksys Ether16 card. I have run the software to disable
> >PNP and set the I/O address and irq's. I have both drivers compiled as
> >modules and my /etc/conf.modules looks like this....
> >
> >alias eth0 ne
> >alias eth1 eexpress
> >options -k eth0 io=0x340 irq=11
> >options -k eth1 io=0x300 irq=10
> >
> >
> >With this setup, only the Linksys (ne) card gets detected at boot and
> >not the EtherExpress (eexpress) card. I can switch it around so the
> >EtherExpress card is eth0 and the Linksys card is eth1. In this case,
> >the EtherExpress card gets detected at boot, but not the Linksys card.
> >So I know both cards work... I just can't get them to work together.
> >I've also tried to use modprobe to load the undetected card, but
> >modprobe returns without any messages. Any ideas?
> >
> > ... John
------------------------------
From: [EMAIL PROTECTED] (David Efflandt)
Crossposted-To: comp.infosystems.www.servers.unix
Subject: Re: please help! apache problem with numeric IP.
Date: 20 Apr 1999 04:32:12 GMT
Reply-To: [EMAIL PROTECTED]
On 19 Apr 1999 11:52:55 GMT, urgrue <[EMAIL PROTECTED]> wrote:
>i have apache 1.3.4 set up on a suse linux kernel 2.0.36 box. basically i need
>to get httpd working on it. it has no actual domain name yet, but it has a
>fixed IP. i assume this problem will be fixed when i get the name set up, but
>i'd like to understand this problem for future reference.
>
>i have a user named 'testi' (finnish for 'test', not any reference to male you
>know), who has a public_html/index.html file.
>
>the problem seems to be that when i try to go to http://193.65.232.198/~testi,
>apache redirects this NOT to
>193.65.232.198/home/testi/public_html/index.html",but to
>"webserver.193.65....". i assume this is because the linux box's host name is
>webserver. so how can i tell apache to leave out the hostname and tell it
>we're working with a numeric IP?
Note that when referencing a directory without a trailing slash, apache
redirects to what it thinks is a proper URL, so apparently your apache is
confused about its identity. Then your browser goes to DNS to resolve
that name which fails. The reason it succeeds when you remove the
'webmaster.' prefix is because by then the redirected URL is technically
correct with trailing slash, so apache accepts it.
As to why apache is confused, check the first name listed on the same line
as that IP in /etc/hosts and also what your machine sets for its own
hostname. Once you fix your hostname and DNS, everything should work.
Until then, you could always try: http://localhost/~testi/
>second oddity that i would like explained, though doesnt need fixing afaik, is
>that when you type the above url, and you get redirected to webserver.193..etc,
>you get an error message from the DNS server (attached below). this is fine,
>but if you THEN go and remove the "webserver." from the url window it will
>work! correct index.html displayed and all. this despite the fact that what is
>in fact in the window is exactly what you typed the first time.
>
>please help.
>freddie
(snip)
--
David Efflandt [EMAIL PROTECTED]
http://www.xnet.com/~efflandt/
------------------------------
From: "Alex" <[EMAIL PROTECTED]>
Subject: ISDN
Date: Mon, 19 Apr 1999 16:19:41 +0200
Does anyone know how to set up an ISDN card?
The card is from WISECOM, 128kbps.
Thanks.
------------------------------
Subject: Re: NFS WITHIN AN IP MASQ FIREWALL (fwd)
From: Paul Rusty Russell <[EMAIL PROTECTED]>
Date: 20 Apr 1999 15:39:49 +0930
Glenn Graham <[EMAIL PROTECTED]> writes:
> Now... When I attempt to do a mount from within my ip-masq LAN, to the
> same machine on the outter internet, it attempts a connection to
> the outter machine, but the connection is refused by the outter machine,
> due to the fact that ip-masq is assigning it an Illegal Port number.
> I note this by the /var/log/messages file from the NFS Server which reads
> NFS from xxx.xxx.xxx.xxx on ILLEGAL port number.
The masquerading box alters the source ports. You will need to alter
the nfs SERVER to allow any ports; from `man nfsd':
-n or --allow-non-root
Allow incoming NFS requests to be honored even if
they do not originate from reserved IP ports. Some
older NFS client implementations require this.
Some newer NFS client implementations don't believe
in reserved port checking. This check can be turned
off for individual hosts by specifying the insecure
export option in /etc/exports.
Rusty.
--
Tridge, Raster, DaveM, Cort, maddog... Where will you be 9-11 July 1999?
http://www.linux.org.au/projects/calu
------------------------------
From: Tom <[EMAIL PROTECTED]>
Subject: Re: Linux as NetWare Server?
Date: Mon, 19 Apr 1999 18:17:57 -0700
Yes, mars-nwe works if you do not need NDS support. You will also have to
come up with the sys:public files from a NW 3.1x installation if you want
the netware commands.
I cannot recall any trauma in getting it running, so it must have been
pretty simple.
:Tom
"David R. Christianson" wrote:
> Hi,
>
> This may sound wierd, but is there any package that will make Linux act
> like a NetWare server (like samba does for NT)? There is no productive
> reason why I want to do this, except to try something different and
> maybe do some performance comparisons.
>
> Dave
------------------------------
From: Fabrice Coudert <[EMAIL PROTECTED]>
Subject: Re: VERY URGENT !!! : Samba networking
Date: Tue, 20 Apr 1999 05:33:05 +0000
The problem comes from smbmount. This tool is not supported by the Samba
team, but is developed by the Linux team with smbfs (it is used by
smbmount) and some differences remain in the source code of Samba and
Linux about the internal structures of the protocol, depending on what
release you used. The smbfs is designed to have the ability to brute
force to upper or lower case names coming from the server. I realy dont
know why while in the spirit of samba this is managed in the server side,
not in the client side. This behavior can be controled in the source code
smbmount.c. If you take a look inside, you wilI see:
conn_options.case_handling = CASE_LOWER;
wich means that the name would be in lower case ?:( This is a bug, but
it dont occurs (normally) because of another bug: smbmount is compiled
with the samba headers in wich CASE_LOWER corresponds to SMB_CASE_DEFAULT
in the smbfs headers so, names are not modified... :=)
Maybe your source code or the binaries you used are wrong with this. Try
to modify the smbmount.c and rebuild it. Correct values are 0, 1 or 2 in
place of CASE_LOWER. If you have the source of the linux kernel,
corresponding smbfs values can be found by doing :
$ grep _CASE_ /usr/include/linux/smb.h
it will respond to something like :
SMB_CASE_DEFAULT,
SMB_CASE_LOWER,
SMB_CASE_UPPER
do not try to directly put one of this but put their sesquence number
instead: the first is 0, second is 1, last is 2.
smbmount is not built by default in the samba Makefile. Before, you have
to do a:
$ ./configure --with-smbmount
then:
$ make
And finally, copy by hand the smbmount, smbumount and smbmnt in the path
of samba binaries.
Fabrice.
Stephane POMATTO wrote:
> Hi all
>
> A Big problem for which I would like a quick answer.....
>
> I'm running Rh 5.2 with 2.0.36 Kernel. Samba is installed and works
> correctly (I can see shares with an NT client)
>
> When I use SMBMOUNT to map an NT disk on my Linux box, it works
> correctly except ONE thing : every files I transfer are in Upper Case.
>
> I would like to know how to have files copied in Lower case. I have to
> make softwar updates by FTP and updates cannot be made because don't
> find files on FTP server. If I rename all files in lower case, updates
> work perfectly
>
> Thanks for your help.
>
> Steph
------------------------------
From: "Xinxin Shao" <[EMAIL PROTECTED]>
Subject: Modem doesn't work.
Date: Tue, 20 Apr 1999 00:36:28 -0500
I use Linux redhat 5.2. I can't use modem. When I run minicom, it will lock
the computer. Would you please give me some idea about that in detail? I am
very appreciate your help.
Thanks
--
*********************************************************************
Xinxin Shao
Department of Electrical and Computer Engineering
University of Minnesota
Email: [EMAIL PROTECTED]
Tel: 612-625-5053 (Office) 651-641-1869(Home)
*********************************************************************
------------------------------
From: kite@NoSpam.%inetport.com (Clifford Kite)
Subject: Re: pppd error explanation needed
Date: 19 Apr 1999 09:03:09 -0500
Chris ([EMAIL PROTECTED]) wrote:
: After connecting (seemingly ok) I get the following messages form pppd
: LCP: Timeout sending config-requests
: Connection Terminated
: Receive Serial link is not 8 bit clean
: Problem: all had bit 7 set to 0
In almost all cases it means that the ISP didn't start PPP and there
is a text prompt reflecting what pppd sends with the 8th bit stripped.
There's a whole section on it in the PPP-HOWTO . One common thing that's
not mentioned in there is a chat expect/send CONNECT '' which sends
an extra carriage return that sometimes confuses an ISP. Replace it
with CONNECT '\d\c' , man chat for the meanings of \d and \c .
--
Clifford Kite <kite@inet%port.com> Not a guru. (tm)
------------------------------
From: [EMAIL PROTECTED] (mamo)
Subject: nis+
Date: Mon, 19 Apr 1999 09:10:53 -0800
I need to set up a 10 machines running linux.
I want to install nis+ to make not possible to the client knowing the
encripted password of the other users.
Is there some of you that setted up a nis+ server on linux.
Where to find documentation
Thanks
Mamo
**** Posted from RemarQ - http://www.remarq.com - Discussions Start Here (tm) ****
------------------------------
From: [EMAIL PROTECTED]
Crossposted-To: comp.windows.x.kde,comp.os.linux.setup
Subject: kppp: Got it working with NT RAS & MS-CHAP !
Date: Mon, 19 Apr 1999 13:56:20 GMT
I finally got kppp connecting to our NT RAS server consistently, using MS-CHAP
authentication!
In my case, it was best not to use the "CHAP" setting in kppp and instead to
choose "Script-Based" authentication. I did not enter my username and
password in kppp.
Because our NT RAS server is not a PDC (primary domain controller), it is
necessary to include the NT domain name as part of the username, separated by
a backslash. Because the "\" character has a special meaning to pppd, it is
necessary to use "\\". This is explained here:
<http://metalab.unc.edu/LDP/FAQ/PPP-FAQ.html>
Also, the NT RAS server returns "" for it's name, in the MS-CHAP challenge.
These things seemed to confuse kppp (if I used the kppp CHAP option); I was
not able to consistently connect. It might work the first time I tried it,
but not any following time. Very frustrating.
I edited the /etc/ppp/chap-secrets file to the correct format for NT RAS
(including all the quotation marks):
"NTDOMAIN\\username" "" "mypassword"
"" "NTDOMAIN\\username" "mypassword"
And I added the following options to the /etc/ppp/options file:
name "NTDOMAIN\\username"
remotename ""
Also, I added "Pause 1 sec" as the first and only script item. Without this,
sometimes there would be no response from the NT RAS server, after sending the
LCP ConfReq.
I downloaded the RPM binaries for DES library and for the patched ppp-2.3.7
here:
<http://www.xs4all.nl/~diedvdyk/PPP-NT-HOWTO/PPP-NT-HOWTO.html>
Then used kpackage to install those two RPM packages. I removed the existing
pppd package first.
They worked fine with Mandrake 5.3 (which is Red Hat 5.2 plus KDE 1.1), kernel
version 2.0.36.
That's it!
Regards,
Richard
============= Posted via Deja News, The Discussion Network ============
http://www.dejanews.com/ Search, Read, Discuss, or Start Your Own
------------------------------
From: "donoli" <[EMAIL PROTECTED]>
Crossposted-To: comp.security.misc
Subject: Re: No way to avoid "denial of service attacks"?
Date: Tue, 20 Apr 1999 01:28:31 +0100
[EMAIL PROTECTED] wrote in message <7ffofr$klb$[EMAIL PROTECTED]>...
>Hello,
>
>My domain on an Apache hosted server was under some kind of "denial of
service
>attacks". The latest version of Apache do have a mechanisim to protect
against
>such activities, so i am going to upgrade and see if it can improve the
>situation.
>
>However my question is that, is there really no way to fight against such
>attack? How easy is it to trace the source of the attack? So if anyone
wants
>to bring a site down, all he need to do is keep calling the website index
>page / CGI requests to overflooded the server? If I am a big company I
might
>have the money to fight with it, but what happen to small and home based
>Internet business?
>
>Very scary! Any comments and help on this will be very appreciated.
>
>Regards
>
>Ted
It's not really clear what you're running or exactly what the nature of the
attack was. I guess you're running linux since you posted to linux NG.
Did comment out any services you don't need in inetd.conf? Besides that,
have you considered running a firewall which would log IP addresses?
Donoli.
------------------------------
From: [EMAIL PROTECTED] (Christophe Zwecker)
Subject: fetchmail shall wait
Date: 20 Apr 1999 01:25:02 +0200
Hi,
I retrieve mail with fetchmail and scan incoming mail for
viruses. When I get like a 10 MB mail the virusscan can take 5 min or
so. Fetchmail just sits there and waits until the mail is
delivered. After that I get:
Apr 20 00:18:10 blinux fetchmail[29320]: flushed
Apr 20 00:18:10 blinux fetchmail[29320]: POP timeout
Apr 20 00:18:10 blinux fetchmail[29320]: client/server protocol error while fetching
from mail.nacamar.de
Apr 20 00:18:10 blinux fetchmail[29320]: Query status=4
Isnt it possible for fetchmail not to wait and get already the next
mail ?
Or is this an error from the popserver ?
Right now I had to disable virusscan...not good :(
--
Christophe Zwecker mail: [EMAIL PROTECTED]
Hamburg, Germany fon: +49 172 9738154
UNIX is user-friendly. It's just not ignorant-friendly and
idiot-friendly.Build a system even a fool can use,and only
a fool will want to use it.
------------------------------
From: "Morten" <[EMAIL PROTECTED]>
Subject: Re: Configure NE2000 compatible ethernet NIC
Date: Tue, 20 Apr 1999 08:37:08 +0200
> I set up a Plug-n-Pray NE2000 on a Packard Bell machine a few days
> ago, and ended up having to specify the I/O port and IRQ on the insmod
> command line. If your card is PnP, make sure you configure it (using
> isapnp) to use otherwise an idle I/O range and IRQ, then specify the
> port and IRQ when you insmod ne.o. (Even if it's not PnP, try
> explicitly specifying I/O and IRQ ... I suspect that there's something
> funky about the NE2000 autodetect code [maybe it's not possible to
> autodetect it safely and correctly?])
>
I'm having that same problem with a PnP NE2000 compatible card. I haven't
been able to make the installation program recognize my card at all. I'm
very new to Linux (and UNIX for that matter) so I'm a bit lost when you talk
about "insmod ne.o". What exactly is it that you do here??
Thanks in advance for any help.
Morten
------------------------------
From: [EMAIL PROTECTED]
Subject: Re: Portable - Home/Work IP Configuration
Date: Tue, 20 Apr 1999 06:28:37 GMT
In article <[EMAIL PROTECTED]>,
Kevin der Kinderen <[EMAIL PROTECTED]> wrote:
> This has to be simple. I don't want to use dhcp (I've had problems with
> it before). I'm using my laptop at both work and home (to an ISDN
> router). When I start up in the new location I have to run netconf and
> change the ip, netmask and default gateway.
You could always use the same IP address at home and at work. Just set up a
gateway box at each location (if one doesn't already exist), and use an IP
address from the 10.0.0.0, 192.168.0.0, or 172.16.0.0 ranges. The gateway
will take care of the rest (such as NAT'ing your private IP to a public one).
-Bill Clark
Systems Architect
ISP Channel
http://locale.ispchannel.com/
============= Posted via Deja News, The Discussion Network ============
http://www.dejanews.com/ Search, Read, Discuss, or Start Your Own
------------------------------
Date: Mon, 19 Apr 1999 23:10:26 +0800
From: Lim Chuan Wee <[EMAIL PROTECTED]>
Subject: Re: ISDN-TA configuration, need help!
"Gunnarsson, Patrik" wrote:
>
> Hi.
> I use an external ISDN terminal adapter connected to one of the serial
> ports.
> I've not been able to get it to work, though. Guess I've grown too used to
> Win plug-play.
>
> I run RH 5.2 and KDE. Started kppp and tried to configure. Bewildered!
> Got some info from TA-supplier, said to enter an AT-string (something like
> ATS118.3= etc.)
> Lost me there, I'm affraid. When connecting, got print in log window (looks
> fine to a newbie like me) and after a few seconds got a message: ppp deamon
> died unexpectedly.
>
> What to do???
>
> TIA,
> Patrik Gunnarsson
> [EMAIL PROTECTED]
Try running this script
pppd connect 'chat -v "" "[modem init string ]" "" ATDT[tel. number]
CONNECT ""' name [username in pap-secrets] /dev/modem debug crtscts
modem defaultroute noipdefault
all of the above in a single line. Just replace the words in between the
[].
Put this in a file and make it executable by chmod a+x file
The pap-secrets file is in /etc/ppp with the format
username server password
on a line. The server can usually be a * wildcard.
Hope this helps.
Cheers.
------------------------------
From: "martin" <[EMAIL PROTECTED]>
Subject: NFS client
Date: Mon, 19 Apr 1999 17:17:34 +0200
Hallo,
could someone help me with my small problem?
I need to find some NFS client for Win 95/98/NT/2000 to conect to my Linux
file server. I'd be glad, if this NFS client will be free of charge.
Thanks for your help Martin Dvorak CZ
[EMAIL PROTECTED]
------------------------------
From: Javier Prieto <[EMAIL PROTECTED]>
Subject: Re: SIOCADDRT: Invalid argument
Date: Mon, 19 Apr 1999 14:31:20 GMT
In article <[EMAIL PROTECTED]>,
Brian McCauley <[EMAIL PROTECTED]> wrote:
> > Whenever I try to add a route ("route add -net 127.0.0.1"), I get that error
> > message: "SIOCADDRT: Invalid argument".
> route add -net 127.0.0.0
I've tried that command and I get the same message.
> > Anyway the card seems to work perfectly.
> What card? We're talking about the loopback device here. The
> loopback device is a virtual device - there's no hardware.
Anyway "route add -net 192.168.200.0" gives me the same error.
It seems that any attempt to add a route displays that message.
I'd like to know who in the kernel is giving me that message... I thought it
was the NIC, but even "route add -net 127.0.0.0 lo" displays it.
Javier Prieto
Optima Technologies, Sevilla (Spain)
[EMAIL PROTECTED]
============= Posted via Deja News, The Discussion Network ============
http://www.dejanews.com/ Search, Read, Discuss, or Start Your Own
------------------------------
** 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
******************************
