Linux-Networking Digest #164, Volume #11 Sat, 15 May 99 12:13:36 EDT
Contents:
Re: Tape Backup software ([EMAIL PROTECTED])
Re: Telnet/ftp question (mist)
Re: get client machine's IP-address (Iond Research Srl)
RH 5.2 + WU-FTP: Can't connect from 9x client (Frederic Faure)
Re: Linux <--> Linux ("Curt")
Re: Null Cable For Ethernet (Ben Whalley)
How do I setup ip accounting with ipchains? (Caveman)
ms-chap + callback connection to NT with pppd problem ("Ofir Israel")
IPX, 2.2.5 kernel and SuSE 6.1 (Arturo)
Re: ppp: Protocol-Reject for unsupported protocol? (Clifford Kite)
Re: Apache on Large Corporate Network ("Chris Knapp")
Re: Red Hat 6.0 upgrade - TAR PROBLEM (LBrown1698)
Re: can't locate module ppp-compress-21 (Clifford Kite)
Re: Fetchmail and Netscape ("Gero H. Marten")
Re: can't ping, ftp etc "Unable to connect, unknown host" ("Eriksson")
Null Modem Connection Linux <> Win95? (Alex Manuk)
Re: NE2000 won't ping ("Jeffrey A. Bell")
Re: Masquerading security ("Kai Krakow")
----------------------------------------------------------------------------
From: [EMAIL PROTECTED]
Crossposted-To:
comp.os.linux.misc,comp.os.linux.setup,hk.comp.os.linux,tw.bbs.comp.linux
Subject: Re: Tape Backup software
Date: 13 May 1999 22:23:28 GMT
Johannes Niess <[EMAIL PROTECTED]> wrote:
[...]
> Amanda is my backup program of choice. It is GPL, with a very helpfull
> mailing list, it handles dozends of client computers with a lot of
> operating systems (Linux, other Unixes, Windows via Samba), it
> distributes your load (one tape might contain a level 0 from one disk
> and a level 2 from another), output of a selfcheck is mailed and it
> handles tape changers. And Idiot can handle it after setup.
I had a quick view on the package description of Amanda, but there's
still a question left: Does Amanda support quick file access (QFA)?
What I need is the ability to choose some files to restore from
a backup catalog and then have Amanda restore them using QFA
(not(!) scanning through the whole archive like tar does).
Selecting those files to backup/restore through a nice user-interface
is highly appreciated. What about those two points?
--
Daniel Dorau [EMAIL PROTECTED]
<< If a train stops at a trainstation, what happens at a workstation? >>
PGP key available, send mail with 'Subject: send pgp key'
fingerprint: 8D7E0B2F9E2E5338 DB7B24742E8B2EAE
------------------------------
From: mist <[EMAIL PROTECTED]>
Subject: Re: Telnet/ftp question
Date: Sat, 15 May 1999 11:35:04 +0100
Reply-To: mist <new$[EMAIL PROTECTED]>
Guy Dillen <[EMAIL PROTECTED]> scribed to us that -
>I'm running RedHat Linux 5.2 on both a server and a workstation (both
>Intel). A third machine is a client running under Win95 (all connected by a
>hub). My problem is when i try to ftp or telnet from the Win95 or Linux
>client/workstation to the Linux server it takes several minutes before I
>get the login prompt.
>
The machine is waiting for name resolution from the Linux box. Add the
windows box IP and name to the /etc/hosts file on the Linux box. Look
up slow Telnet on Deja News if you want more information.
--
Mist.
------------------------------
From: Iond Research Srl <[EMAIL PROTECTED]>
Crossposted-To:
comp.os.linux.development.apps,comp.os.linux.development.system,comp.programming,comp.unix.programmer
Subject: Re: get client machine's IP-address
Date: Fri, 14 May 1999 18:13:49 +0200
Mark Hahn wrote:
> getpeername. this would be obvious if you had read even basic
> network programming books.
Thanks Mark for your answer, but I'm not sure getpeername can solve my
problem.
I'll try to explain myself better with an example:
You are connected from a linux machine (client) to a solaris machine
(server) with
a telnet session during which you want to execute this c-program
(complex.c)
main()
{
printf("Hi, Mark\n");
/* Now I wish to know the IP-address of the machine on which
the string "Hi, Mark" is displayed"
*/
}
This program runs in the server but thru the pseudo-tty its output is
displayed
at the client machine.
No sockets are used so we haven't the first arg to pass to the
getpeername.
You can try to pass 0 as socket-fd to getpeername but it returns the
local-host
IP-address (127.0.0.1).
So what's the matter ?
Maybe work around the slave part of the pseudo-tty ?
Sorry for this silly question but I'm not a network programmer, really
this is
my first networking problem, and also reading Stevens' I haven't yet
found a
solution.
Anyway, thanks a lot
Lucio
[EMAIL PROTECTED]
------------------------------
From: [EMAIL PROTECTED] (Frederic Faure)
Subject: RH 5.2 + WU-FTP: Can't connect from 9x client
Date: Fri, 14 May 1999 17:18:22 GMT
Reply-To: [EMAIL PROTECTED]
All,
Although ping works between that Linux test server and a 98 PC, and I
can connect with FTP locally on the Linux server, when trying it from
the 98 machine, I get:
"C:\WINDOWS>ftp 192.168.0.1
Connected to 192.168.0.1. "
... and no login. Do the default settings need to be changed to allow
network connections in FTP mode?
Thanks,
FF.
--
The system required Windows 95 or better, so I installed Linux!
------------------------------
Reply-To: "Curt" <[EMAIL PROTECTED]>
From: "Curt" <[EMAIL PROTECTED]>
Subject: Re: Linux <--> Linux
Date: Sat, 15 May 1999 05:58:53 -0500
Please post the results of 'ifconfig' and 'netstat -nr'
Amir Malik <[EMAIL PROTECTED]> wrote in message
news:7hit9u$g4e$[EMAIL PROTECTED]...
> I have one 486 running RedHat Linux 5.1 and another Intel Pentium 2 440MHz
> with MMX computer also running RedHat Linux 5.1. They are connected via a
> hub which is hooked up to the 486.
>
> Both machines can recognize their network cards, but they cannot ping each
> other. Using ifconfig, should I set it to a Point-to-Point link or just
> give the Broadcast adderss? What do I have to set up in my routing table?
No, an ethernet connection does not use Point-to-Point.
>
> ------------------ Posted via SearchLinux ------------------
> http://www.searchlinux.com
------------------------------
From: Ben Whalley <[EMAIL PROTECTED]>
Subject: Re: Null Cable For Ethernet
Date: Fri, 14 May 1999 19:32:16 +0100
mike wrote:
> I have heard that one can modify a standard twisted pair ethernet
> cable to allow two computers to connect with each other without
> using a hub. Can one purchase such a cable. Is a " patch cable "
> such a cable?
> Mike
yeah, you can, i am trying to do it at the moment, but am having a few
difficulties, the Ethernet HOWTO has a great section on how to
build/identify a patch cable...all the details are there
ben
--
Hey, wait a minute!! I want a divorce!! ... you're not Clint Eastwood!!
------------------------------
From: [EMAIL PROTECTED] (Caveman)
Subject: How do I setup ip accounting with ipchains?
Date: Sat, 15 May 1999 11:45:52 GMT
Hi,
I've got 3 computers here. 2 windows 95 machines and one Linux
internet gateway:
192.168.10.2 (Win 95)
192.168.10.25 (Win 95)
192.168.10.1 (Linux)
The Linux-box masquerades all packets going to the internet and uses
ip forwarding. I'm connected using dynamic ip and ppp0 as device.
Now I want to setup ip accounting so I could see the amount of data
that the machines (192.168.10.2 and 192.168.10.25) produce, both
incoming and outgoing (if possible I would like these split into two
seperate parts).
I've read the manual but I can't figure out how to do it. The manual
only deals with a single computer going to the internet and not with
an entire LAN. Can anyone send me an example for ip accounting with
ipchains. I only need the accounting rules, the rest is already
working.
I already tried making seperate chains with "ipchains -N pc1_input"
"ipchains -N pc1_output" etc. and then adding the rules there.
It's also not completely clear to me how the packets get to my local
machine, for example 192.168.10.25.
Let's say my ip at the current moment is 195.96.110.8 and I'm
connected to a site with the ip 193.43.1.45.
Is this the way the packets will go before they reach the destination?
195.96.110.8 --> 192.168.10.1 --> 192.168.10.25
And after they reach 192.168.10.1, I cannot use ppp0 in my rule when I
want to filter out the packets but must I use eth0 instead?
And because I use ip masquerading, can't I use the input and output
chain but must I use the forward chain instead?
I'm very confused. Maybe someone can help me on this one.
Dennsi van der Meer
------------------------------
From: "Ofir Israel" <[EMAIL PROTECTED]>
Subject: ms-chap + callback connection to NT with pppd problem
Date: Sat, 15 May 1999 15:47:55 +0300
Hi
I recently read the PPP-NT HOWTO. It showed me how to connect to my
ISP's NT server using chap and callback. What it didn't say exactly was what
I needed to write as the name in the pppd's options
file.
Should I write "name 'domain\\my_DUN_username'" or "name
'domain\\my_ComputerName_inTheNTserver'".
Actually, both didn't work. Are there any known bugs in connecting to a
NT server with mschap and callback? Please help me, now I have to connect to
the net using W95 with an ISP that shows ADS and is VERY VERY slow.
------------------------------
From: Arturo <[EMAIL PROTECTED]>
Subject: IPX, 2.2.5 kernel and SuSE 6.1
Date: Sat, 15 May 1999 00:49:21 +0200
<HTML>
I have just upgraded from SUSE 5.3, with 2.0.35 kernel, to SUSE 6.1 with
2.2.5 kernel
<P>The IPX support in SUSE 6.1 is configured as a module, not as a permanent
kernel resource.
<P>And it does not load automatically when needed. If I do it by "modprobe
ipx", it seems to go fine, but after that I get lots of OOPS error messages
with many commands: ifconfig, ipx_configure, nwserv, etc.
<P>Can anybody tell me what is wrong ? How is ipx.o expected to be loaded
?
<P>Thanks a lot.
<P>Arturo
<BR>[EMAIL PROTECTED]
<BR> </HTML>
------------------------------
From: kite@NoSpam.%inetport.com (Clifford Kite)
Subject: Re: ppp: Protocol-Reject for unsupported protocol?
Date: 14 May 1999 23:51:51 -0500
Carsten G Brautigam ([EMAIL PROTECTED]) wrote:
: After upgrading to RedHat 6.0, I have problms with PPP.
: The lgo file says:
: May 13 22:48:17 unicorn pppd[6890]: pppd 2.3.7 started by carsten, uid
: 501
: May 13 22:48:17 unicorn pppd[6890]: Using interface ppp0
: May 13 22:48:17 unicorn pppd[6890]: Connect: ppp0 <--> /dev/modem
: May 13 22:48:23 unicorn modprobe: can't locate module ppp-compress-21
: May 13 22:48:24 unicorn modprobe: can't locate module ppp-compress-26
: May 13 22:48:24 unicorn modprobe: can't locate module ppp-compress-24
These are CCP compression modules that needed aliasing according to many
other posts regarding them. I don't do RH or modules (well I did use
insmod to load a couple of modules once, but not with kmod) and can't
say offhand what to do to get rid of them.
: May 13 22:48:25 unicorn pppd[6890]: Protocol-Reject for unsupported
: protocol 0x7263
: May 13 22:48:25 unicorn pppd[6890]: Protocol-Reject for unsupported
: protocol 0x7263
: May 13 22:48:26 unicorn pppd[6890]: Protocol-Reject for unsupported
: protocol 0x7263
: May 13 22:48:26 unicorn pppd[6890]: Protocol-Reject for unsupported
: protocol 0x5072
: May 13 22:48:26 unicorn pppd[6890]: Protocol-Reject for unsupported
: protocol 0x5072
Some or all of the PPP link negotiation messages are likely being sent
to the modem's serial port somehow. Notice that 5072 is hex for `Pr'
(from ProtRej or Protocol) and 7263 is hex for `rc' (from rcvd [LCP.. ).
Adding the pppd option kdebug 7 may show what's happening better - one of
the few times where kdebug 7 is useful. The log file for link negotiation
messages varies with the distribution; /etc/syslog.conf is where the log
files are configured.
The recent ppp-2.3.7 has changed and, for reasons that are not at all
clear to me, kppp can cause (and *has* caused) this type of behavior, and
perhaps other programs can cause it as well. You can get the latest kppp
when it is suspect and it should cure the problem. You can also *try*
specifying to pppd the modem's device file ttySx as /dev/ttySx rather
than as the symbolic link /dev/modem, but I'm not sure this will work.
--
Clifford Kite <kite@inet%port.com> Not a guru. (tm)
/* The signal-to-noise ratio is too low in many [news] groups to make
* them good candidates for archiving.
* --- Mike Moraes, Answers to FAQs about Usenet */
------------------------------
From: "Chris Knapp" <[EMAIL PROTECTED]>
Subject: Re: Apache on Large Corporate Network
Date: Sat, 15 May 1999 09:50:27 -0400
Thx for the feedback.
The PDC is the primary domain controller for the network. It suprised me
when I rebooted after the install and the commandline prompt was something
like 'root.kanatasms1#' The machine correctly detected who was "boss" on
the network (the PDC is called 'kanatasms1')
Anyway, you are right, I need to figure out how to 'name' the Linux box so
that it can be known to all of the NT machines it will cooexist with. I
think I have some work to do....
Cameron Spitzer <[EMAIL PROTECTED]> wrote in message
news:7hj0ik$qv7$[EMAIL PROTECTED]...
> In article <7hicsr$fph$[EMAIL PROTECTED]>,
> Chris Knapp <[EMAIL PROTECTED]> wrote:
> >Hello,
> >
> >I recently installed redhat 5.2 on an older Pentium Pro at work. The
install
> >went flawlessly, I selected DHCP as the networking protocol, and when I
> >rebooted, the machine logged me in as root on the PDC.
>
> What's a PDC?
>
>
> >Questions: I want to use this machine as a temporary webserver on a
> >300-node NT network- how do I get Apache to work?
>
> The only hard problem will be telling the 300 NT machines where
> you are. DHCP is likely to give you a new IP number each time
> you connect. You have to arrange for somebody to serve that
> number when one of the 300 NT machines asks for an address
> for www.linux.yourcompany.com. Hint: ask whoever is responsible
> for the DHCP server. Chances are they'll give you a static
> IP number from their secret stash.
>
>
> > A lot of the HowTo's
> >explain how to use a Linux machine as the sole server, but we have fully
> >functioning network and simply want to add a web-server. Can someone
point
> >me to info on how to do this or give me some general tips?
>
> Red Hat comes with Apache. But if you're serious, go to
> www.apache.org and follow the "Download!" and "Server Documentation"
> links. Apache is very well put together, and it will install and
> run for you with no problems, despite its complexity
> and zillions of installation options. Just read the instructions.
>
> Cameron
> (still baffled by suEXEC though.)
>
------------------------------
Date: 13 May 1999 09:40:25 GMT
From: [EMAIL PROTECTED] (LBrown1698)
Subject: Re: Red Hat 6.0 upgrade - TAR PROBLEM
Bryan,
I had the same problem (skipping files) with a tar backup I did using NFS
as root. Turns out root has limited permissions when using NFS, so might be a
permission problem.
LLB
------------------------------
From: kite@NoSpam.%inetport.com (Clifford Kite)
Crossposted-To: alt.os.linux.slackware
Subject: Re: can't locate module ppp-compress-21
Date: 15 May 1999 08:54:16 -0500
Eriksson ([EMAIL PROTECTED]) wrote:
: When I connect to my ISP I get an IP address and everything seems to be OK.
: But I can't do anything. (like ping, ftp, browsing etc) "WTF!" Was my
: thought cause according to /var/log/messages everything SHOULD be ok.
: But when I checked "ifconfig" there is no ppp0 interface.
: My /etc/resolv.conf should be ok.
: Is there anything else I've missed??
Have you read the linux/Documentation/Changes file in the kernel source
tree? If not then you may have missed more than one thing.
--
Clifford Kite <kite@inet%port.com> Not a guru. (tm)
/* The wealth of a nation is created by the productive labor of its
* citizens. */
------------------------------
From: "Gero H. Marten" <[EMAIL PROTECTED]>
Subject: Re: Fetchmail and Netscape
Date: Sat, 15 May 1999 07:36:48 +0200
Leopold Toetsch wrote:
> You could duplicate your mail with procmail to a different folder and then
> read it over telnet.
Good idea. How would you do that?
--
Gero H. Marten
<http://www.provi.de/gmarten/index.html>
--
------------------------------
From: "Eriksson" <[EMAIL PROTECTED]>
Crossposted-To: alt.linux.slakware
Subject: Re: can't ping, ftp etc "Unable to connect, unknown host"
Date: Sat, 15 May 1999 16:20:27 +0200
No.. That wasn't my problem. I'm using Slackware 3.6 and kernel 2.2.4 btw.
Any other ideas?
/Martin
>Hi,
>Beware advice from the very-newbie (like me)....
> Do you have a ethernet card? And if so, do you remember the point
in
>your installation when you set it up --there's a line in the setup screen,
the
>third i believe, that reads "default gateway". The redhat install program
will
>automatically put a 'reasonable' number in here for you if you simply enter
a
>IP for your nic on the first field. Unfortunately, this will screw-up your
PPP
>connection to Internet , if you make that connection directly from your
Linux
>box without going through a gateway host on your LAN. You would
deliberately
>have to zero out this field during initial setup to avoid having unusable
PPP.
>
> If I have guessed your problem correctly, in the output of the route
>command you will see some # like 192.168.1..254 flagged with the UG
'gateway'
>tag in the output--but PPP.is also flagged as UG! This means that your box
>dials up and authenticates itself to the ISP but all your TCP requests and
>such are mis-routed to the eth interface. Study the man page for route to
>figure out how to delete this spurious gateway. Or reinstall and be sure to
>avoid having the eth0 interface become a gateway device. Ask someone for
help
>with route, it's a complicated pain in the neck to use correctly.
>
>I don't really know what I am doing --your situation sounds exactly like
one I
>had, though.
------------------------------
From: [EMAIL PROTECTED] (Alex Manuk)
Subject: Null Modem Connection Linux <> Win95?
Date: Sat, 15 May 1999 06:35:36 -0800
Hi all... Trying DESPERATELY to interconnect two machines.. RH Linux to
Win95 over a null
modem cable. The two machines are capable of talking over Win95's "direct
cable connection",
but when I try to set up pppd on the Linux box, it won't talk to the Win
machine. Using the proper
null modem serial cable, Com1 (ttys0) to Com1. Is there additional software
that's required on
either box to make this work? The eventual goal is to "modem-share" between
the two boxes, so
pppd is kind of necessary, but I have no clue what authentication procedure
Win95's direct
cable connection follows, so I don't know what to send to the serial port
during the connection
process.
Another option (I think) is to use Win95's "dial up networking" to establish
the connection, and use
the Linux box as a server, but I'm pretty lost as to how to acheive even
THAT goal. Any assistance
would be greatly appreciated.
Alex
WD8JMM
**** Posted from RemarQ - http://www.remarq.com - Discussions Start Here (tm) ****
------------------------------
From: "Jeffrey A. Bell" <[EMAIL PROTECTED]>
Subject: Re: NE2000 won't ping
Date: Sat, 15 May 1999 17:28:31 +0000
Reply-To: [EMAIL PROTECTED]
You might want to look at /etc/hosts,
Here is a snippet of what mine looks like.
192.168.1.1 pegasus.zbadlands.org pegasus
192.168.1.2 raptor.zbadlands.org raptor
[EMAIL PROTECTED] wrote:
> Hi,
>
> Marco Teeuwen <[EMAIL PROTECTED]> wrote:
> > Hello everybody,
>
> > I have been using SuSE 6.0 for a few months now, but I haven't been able to
> > get my network cards working. They are NE2000 compatible(Realtek 8019
> > clones). In the howto's I was instructed to get the cards off PnP, and give
> > them an IRQ and IO, which I did.
>
> > Now, Linux finds the card, but I'm still not able to ping the machine from
> > my other Windows machines.
>
> > IP adress=192.168.100.3, should be ok, I guess.
>
> > Am I overseeing something?
> If you only did this, you forgot to add the ip addresses to the routing
> tables.. try 'man route' - or one of the networking-howto's :)
>
> Robert/Emphyrio
--
Jeffrey A. Bell
-------------------------------------------------------------------------
Research is what I'm doing when I don't know what I'm doing.
-- Wernher von Braun --
------------------------------
From: "Kai Krakow" <[EMAIL PROTECTED]>
Subject: Re: Masquerading security
Date: Sat, 15 May 1999 17:09:55 +0200
Reply-To: "Kai Krakow" <[EMAIL PROTECTED]>
> My question is, can linux set the ip-masquerading to a specific device
> instead of a local net. Say to route eth0 <-> ppp0. I've read the how-
> to and they seem to point to a routing of [local net] <--> [internet]
> before i decided to instantiate the changes i'd like to know if its
> possible to spoof a packet as one of the allowed addresses from the
> local net, send it to the linux machine in question and have it forward
> it out to the internet. This would be a real easy way to blindly
> spoof/attack another machine.
Shouldn't be possible if you bind the masquerading to a special device,
means:
If you allow the kernel to only forward packets from the local net (say
192.168.1.0/24 or similar) when they came in on device eth0. Then spoofing
the packet to source 192.168.1.* but sending it in thru another device like
ppp0 will not forward this packet...
ciao
Kai
------------------------------
** 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
******************************
