Linux-Networking Digest #309, Volume #10 Fri, 26 Feb 99 05:13:44 EST
Contents:
Re: NFS mystery (Timothy J. Lee)
Re: Slow echo from Minicom but can still dail ("lunchmeat7")
Does Samba enable page/record locking within Linux? ("Jeff R")
can Windows rovers tunnel to Linux? ([EMAIL PROTECTED])
Detecting second ethernet trouble (Stephen Anderson)
Re: Machine name themes - what do you use? (None)
Compiling ppp-2.3.5 (Mathias Jansson)
Re: testparm hangs ("Michel A. Lim")
Re: token ring pcmcia for linux? (Kazin)
Re: Token Ring locking up (Kazin)
IP to process network interface? (Phil Howard)
Re: Adding the chargen deamon ("lunchmeat7")
Mapping a Linux drive in Win 95 via TCP/IP ([EMAIL PROTECTED])
Re: Adding the chargen deamon ("lunchmeat7")
----------------------------------------------------------------------------
Crossposted-To: comp.os.linux.misc
From: [EMAIL PROTECTED] (Timothy J. Lee)
Subject: Re: NFS mystery
Reply-To: see-signature-for-email-address---junk-not-welcome
Date: Wed, 24 Feb 1999 18:45:27 GMT
prasad <[EMAIL PROTECTED]> writes:
|I'm trying to NFS mount a plain Linux 2.0.35 machine's exported
|directory on a host sporting an AFS/DFS kernel extension, and the
|mount fails with the message:
|
| Feb 24 11:25:00 wizard nfsd[115]: NFS request from
|druid.watson.ibm.com originated on insecure port,
|psychoanalysis suggested
|
|in /usr/adm/syslog. The same directory mounts without a fuss on
|other hosts.
What OS is druid.watson.ibm.com running, and what OSes are the
other hosts that don't show this problem running?
Some OSes originate NFS connections from non-privileged ports
(i.e. those numbered higher than 1024), while others used
privileged ports (those numbered lower than 1024 which only
root is supposed to be able to use). Usually, this behavior
is configurable. On the NFS server side, it is also often
configurable whether the server allows NFS connections from
non-privileged ports.
--
========================================================================
Timothy J. Lee timlee@
Unsolicited bulk or commercial email is not welcome. netcom.com
No warranty of any kind is provided with this message.
------------------------------
From: "lunchmeat7" <[EMAIL PROTECTED]>
Subject: Re: Slow echo from Minicom but can still dail
Date: Wed, 24 Feb 1999 13:49:30 -0500
piece of cake.. i had the same problem ;-)
i had a dual boot machine, between windows and suse linux. Under windows the
modem smoked, was lightning fast and worked beautifully.
under linux when trying to dial with minicomm if a sent a simple AT command
the modem would take minutes to respond. and the pppd daemon would die when
connecting to my isp.
the problem was the modems interrupt and i/o. linux is a bit more finicky
about serial ports. i suggest setting taking the bios off "auto" and setting
them up more like this.
/dev/cua0 (com1) address 0x3f8 irq 4
/dev/cua1 (com2) address 0x2f8 irq 3
and set your modem to match one of these, i had jumpers, you might have
something else. if you have an external modem even better.
you can almost always get it to work using the "setserial" program but i
liked changing the bios/hardware setting to something its going to
recognize.
HTH
Thomas Leung wrote in message <[EMAIL PROTECTED]>...
>Hi,
>
>I am using RedHat 5.1 with Minicom 1.81. After I start Minicom, the
>cursor stays at the middle of the screen. It's seems nothing happen.
>After about 10-15 second, the cursor moves to the left of the screen
>with "ATZ" printed on the same line.
>
>Everything I type seems to go through a long delay until it is finally
>display on the screen. I've tried "minicom -s" and look for time delay
>setting. I couldn't find anyone which is relevant. I can still dial
>using Minicom but the echo response is very slow.
>
>Has anyone got any clue. Thanks.
>
>Thomas
>
------------------------------
From: "Jeff R" <[EMAIL PROTECTED]>
Subject: Does Samba enable page/record locking within Linux?
Date: Wed, 24 Feb 1999 18:50:54 GMT
I'm investigating the feasibility of replacing a Novell 3.2 server with one
running Linux (RedHat 5.2). Netware 3.12 has a 2k page size and the
ability to independently set the maximum number of record locks for both an
individual connection and total locks. From within AUTOEXEC.NCF ...
set maximum record locks per connection = x
set maximum record locks = y
A critical concern: Does Samba have the ability to perform page/record
locking? If so, what is the maximum number of locks per connection as well
as overall? (Is page size 2k?)
Thanks in advance, Jeff
------------------------------
From: [EMAIL PROTECTED]
Subject: can Windows rovers tunnel to Linux?
Date: Wed, 24 Feb 1999 18:45:05 GMT
Is there a way a Windows machine could participate in the mobile IP
scheme outlined in the NET3 HOWTO (below), as the roving machine "B"
away from its home network, re-joining it from afar by gaining a local
ISP connection then tunneling to Linux host "A" back home?
==============================
>From NET3 HOWTO:
6.8.2. A tunneled host configuration.
... we have just host `B' which to want to act and behave as if it is
both fully connected to the Internet and also part of the remote
network supported by host `A':
192.168.1/24
-
| ppp0 = ppp0 =
| aaa.bbb.ccc.ddd fff.ggg.hhh.iii
|
| /-----\ /-----\
| | | // | |
|---| A |------//---------| B |
| | | // | |
| \-----/ \-----/
| also: 192.168.1.12
-
Linux router `A' would be configured with:
[snip]
Linux host `B' would be configured with:
[snip]
This sort of configuration is more typical of a Mobile-IP
application. Where a single host wants to roam around the Internet and
maintain a single usable IP address the whole time. You should refer
to the Mobile-IP section for more information on how that is handled
in practice.
------------------------------
From: Stephen Anderson <[EMAIL PROTECTED]>
Subject: Detecting second ethernet trouble
Date: 26 Feb 1999 07:33:38 GMT
I am running Linux 2.2.2 (Reddat) and having trouble detecting a second ethernet
adapter. My system (Dell OptiPlex something-or-other) has an on-board 3C509 ethernet
chip. I then placed a 3C905 ethernet card in one of the PCI slots. Linux
automatically detects and configures the 3C905 (card) but it does not see the on-board
3C509. I have compiled both 90X and 509 drivers into the kernel. I have used their
(3Com) Etherdisk utilities to detect both cards without any trouble. The 905 is IRQ
11 and IO of FC80 (or something like that). The 509 (not-working) is detected at IRQ
10 with a base address of 250. So I went into my lilo.conf and added:
append="ether=10,0x250,eth1"
And of course ran lilo to update. Unfortunately I still don't see that card.
/proc/interrupts doesn't list eth1. Nor does anything else. Does anybody have any
ideas how to go about tackling this problem? Thanks!
================== Posted via SearchLinux ==================
http://www.searchlinux.com
------------------------------
From: [EMAIL PROTECTED] (None)
Crossposted-To:
vmsnet.networks.misc,microsoft.public.windowsnt.domain,comp.unix.solaris,comp.os.os2.networking.server,comp.os.ms-windows.nt.admin.networking,comp.infosystems.www.servers.unix,comp.protocols.tcp-ip.domains
Subject: Re: Machine name themes - what do you use?
Date: 26 Feb 1999 08:52:41 GMT
I name my machines "Big Muddy", "Revenge of the Nerds", current machine is
"ancien regime".
I guess I like something different.
------------------------------
From: Mathias Jansson <[EMAIL PROTECTED]>
Subject: Compiling ppp-2.3.5
Date: Fri, 26 Feb 1999 07:47:49 +0000
Hi Everyone!
Im trying to set up my linux box (debian 2.0/ linux 221)
to use callback ppp at a NT server using ms-chap. Connecting
to the server is no problem at all, but after that i just
sits there.
I've tried to set the passive option in pppd to force any kind of
communication or get at prompt where I can send my phone number and hang
up. (see chatscript below). Then I found a notice about MS callback i
the package documentation (ppp-2.3.5) - It says that you need to
recompile ppp with the DCBCP_SUPPORT option set because MS-chap uses
its own kind of cbcp protocol (that was a surprise :)).
After spending 5 hours looking at compiler errors I have to admit
im stuck - ppp won't compile. I've tried gcc272 and egcc290 and
even pgcc291 (desperate) - no difference. The ppp package has a
script to patch/replace the ppp-kernel header but since Im using
linux-2.2.1 the kernel header are newer than the ones in ppp.
It all comes down to this:
In file included from sys-linux.c:68:
../include/linux/ppp_defs.h:97: parse error before `ext_accm'
../include/linux/ppp_defs.h:97: warning: type defaults to `int' in
declaration of `ext_accm'
../include/linux/ppp_defs.h:97: warning: data definition has no type or
storage class
../include/linux/ppp_defs.h:113: parse error before `__u32'
The line in ppp_defs.h looks like this:
/*
* Extended asyncmap - allows any character to be escaped.
*/
typedef __u32 ext_accm[8];
--followed by alot of declarations of type __u32
Why (the ****) is ppp impossible to compile? Someone must have done
it before me on a debian system since I am using the ppp-2.3.5
package. I suspected it had to do with the 2.2.1 headers so I tried
using a clean box with the 2.0.35 kernel with the same result...
Any hints, tips or other methods?
Thank you!
//Mathias
([EMAIL PROTECTED])
###############
--chatscript--
ABORT BUSY
ABORT "NO CARRIER"
ABORT VOICE
ABORT "NO DIALTONE"
ABORT "NO ANSWER"
TIMEOUT 90
"" ATZ
OK ATDT1111111
CONNECT \d\c
ABORT "Not registered"
"ack to:" 555-myhome
uthorized ""
TIMEOUT 120
RING ATA
CONNECT ""
--end chatscript--
--ppp log--
Feb 24 13:44:45 wally pppd[24493]: Serial connection established.
Feb 24 13:44:46 wally pppd[24493]: Using interface ppp0
Feb 24 13:44:46 wally pppd[24493]: Connect: ppp0 <--> /dev/ttyS0
Feb 24 13:44:46 wally pppd[24493]: sent [LCP ConfReq id=0x1 <asyncmap
0xa0000> <magic 0x5054b815>]
Feb 24 13:44:46 wally pppd[24493]: rcvd [LCP ConfReq id=0x0 <asyncmap
0x0> <auth chap 80> <magic 0x3b2d> <pcomp> <accomp>]
Feb 24 13:44:46 wally pppd[24493]: sent [LCP ConfRej id=0x0 <pcomp>
<accomp>]
Feb 24 13:44:46 wally pppd[24493]: rcvd [LCP ConfAck id=0x1 <asyncmap
0xa0000> <magic 0x5054b815>]
Feb 24 13:44:47 wally pppd[24493]: rcvd [LCP ConfReq id=0x1 <asyncmap
0x0> <auth chap 80> <magic 0x3b2d>]
Feb 24 13:44:47 wally pppd[24493]: sent [LCP ConfAck id=0x1 <asyncmap
0x0> <auth chap 80> <magic 0x3b2d>]
Feb 24 13:44:47 wally pppd[24493]: sent [LCP EchoReq id=0x0
magic=0x5054b815]
Feb 24 13:44:47 wally pppd[24493]: rcvd [CHAP Challenge id=0x4
<2f7cf6d7d1df50a6>, name = ""]
Feb 24 13:44:47 wally pppd[24493]: sent [CHAP Response id=0x4
<000000000000000000000000000000000000000000000000a2f426b6b9f581a986a798c422dbb42
d4623fc227a213c7201>, name = "lha"]
Feb 24 13:44:47 wally pppd[24493]: rcvd [LCP EchoRep id=0x0
magic=0x3b2d]
Feb 24 13:44:47 wally pppd[24493]: rcvd [CHAP Success id=0x4 ""]
Feb 24 13:44:47 wally pppd[24493]: sent [IPCP ConfReq id=0x1 <addr
0.0.0.0> <compress VJ 0f 01>]
Feb 24 13:44:47 wally pppd[24493]: rcvd [CCP ConfReq id=0x2 < 12 06 00
00 00 01>]
Feb 24 13:44:47 wally pppd[24493]: Unsupported protocol (0x80fd)
received
Feb 24 13:44:47 wally pppd[24493]: sent [LCP ProtRej id=0x2 80 fd 01 02
00 0a 12 06 00 00 00 01]
Feb 24 13:44:47 wally pppd[24493]: rcvd [IPCP ConfReq id=0x3 <compress
VJ 0f 01> <addr 10.0.1.148>]
Feb 24 13:44:47 wally pppd[24493]: sent [IPCP ConfAck id=0x3 <compress
VJ 0f 01> <addr 10.0.1.148>]
Feb 24 13:44:47 wally pppd[24493]: rcvd [IPCP ConfNak id=0x1 <addr
10.0.1.159>]
Feb 24 13:44:47 wally pppd[24493]: sent [IPCP ConfReq id=0x2 <addr
10.0.1.159> <compress VJ 0f 01>]
Feb 24 13:44:47 wally pppd[24493]: rcvd [IPCP ConfAck id=0x2 <addr
10.0.1.159> <compress VJ 0f 01>]
Feb 24 13:44:47 wally pppd[24493]: local IP address 10.0.1.159
Feb 24 13:44:47 wally pppd[24493]: remote IP address 10.0.1.148
Feb 24 13:45:17 wally pppd[24493]: Terminating connection due to lack of
activity.
--end log--
========================================================================
Human mathias 2.3 #4 Sun Jun 22 08:11:23 CET 1975 brain85 unknown
------------------------------
From: "Michel A. Lim" <[EMAIL PROTECTED]>
Crossposted-To: comp.protocols.smb
Subject: Re: testparm hangs
Date: 24 Feb 1999 18:54:13 GMT
thank you for your assistance. my linux box does not have a static ip
address. instead my NT3.51 server assigns the linux box an ip address using
DHCP. furthermore, the NT server acts as the WINS server for my network.
the ip address for the NT server is 192.168.34.1, and this static address
and name has already been added to my /etc/hosts file.
do i NEED to run DNS?
regards,
michel a. lim
associate
wong hobach lau
consulting engineers
structural/civil/construction management
www.whl-international.com
------------------------------
From: Kazin <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.hardware,comp.os.linux.portable
Subject: Re: token ring pcmcia for linux?
Date: Fri, 26 Feb 1999 04:05:04 -0500
Antoni Lapinski wrote:
>
> These are the supported pcmcia cards for Linux:
>
> 3Com 3c389 TokenLink Velocity
> 3Com 3c689 TokenLink III
> IBM Token Ring Adapter
> iBM Turbo 16/4 Token Ring
>
> Which one would you suggest?
Last I knew, all of those use the same IBM Tropic chipset, and should
be pretty much the same. I think the driver used the 'lowest common
denominator' approach. I have two cards, both the usual IBM TR Adapters
(though one says 'Dell'), and I've used the Turbo, with basically the
same results.
I heard rumours about a Madge driver for their PCMCIA RingNodes, but I
haven't actually seen such a beast. Though it'd be nice, I've got a
dozen of those...
=======================================================================
Mike Stella Software / Systems Engineer
http://www.sector13.org/kazin Thirteen Technologies, LLC
=======================================================================
Remove 'FREAK' - you know the drill...
------------------------------
From: Kazin <[EMAIL PROTECTED]>
Subject: Re: Token Ring locking up
Date: Fri, 26 Feb 1999 04:07:30 -0500
Michael Knigge wrote:
>
> [EMAIL PROTECTED] schrieb in Nachricht
> <7asgvk$oml$[EMAIL PROTECTED]>...
> >I have some IBM Auto and Turbo TR cards here, and whenever I try and do the
> >ifconfig command to start it up, it just hangs forever. I've tried
>
> When I remember right, the Turbo TR Cards are not supported. You should
> have a look at the TokenRing HOWTO and the Source code....
Actually, they are, in compatability mode. I don't remember if you
need to switch that with LANAID or not though.
I used to use an ISA auto 16/4, it worked fine. Though I NEVER let
those things go 'auto' - I ALWAYS force to 16Mbps. I just don't trust
the 'auto' thing, especially if you're using MAUs.
=======================================================================
Mike Stella Software / Systems Engineer
http://www.sector13.org/kazin Thirteen Technologies, LLC
=======================================================================
------------------------------
From: [EMAIL PROTECTED] (Phil Howard)
Subject: IP to process network interface?
Crossposted-To: comp.os.linux.development.system
Date: Wed, 24 Feb 1999 19:05:03 GMT
Is there a way to get an IP network interface into the kernel so
that the kernel can operate on it as any network interface, but
have the interface really connect via a process on that machine?
Any IP packets sent to that interface would be given to the process,
and any packets that process produces through the appropriate API to
this facility, would "arrive" into the IP stack in the kernel. We
can assume the process will have root permissions to do this up, for
obvious reasons.
Does this kind of thing exist? I've looked at some of the various
features of Linux, and none seem to do _this_ (as simple as the
concept seems to me). The only option I see as an alternative is
to use PPP via a pty, but I'd rather avoid that, if I can, for a
few reasons.
--
-- *-----------------------------* Phil Howard KA9WGN * --
-- | Inturnet, Inc. | Director of Internet Services | --
-- | Business Internet Solutions | eng at intur.net | --
-- *-----------------------------* phil at intur.net * --
------------------------------
From: "lunchmeat7" <[EMAIL PROTECTED]>
Subject: Re: Adding the chargen deamon
Date: Wed, 24 Feb 1999 14:04:07 -0500
if you
#cat /etc/inetd.conf | less
you can scoll up and down to see where it is, port 19 i believe and simply
check if its commented out with a #
if it is just edit it and uncomment it, and do a
ps aux | grep inetd
to find the process ID of inetd the internet daemon ( a number usually
around the low one-hundreds)
and then do a ...
kill -HUP (process id number)
to restart the process, it will read in the new inetd.conf file and start
over or you can reboot if you like.
i don't think chargen is a daemon, it might be part of the kernel itself.
Patrick Berge wrote in message ...
>I need the chargen deamon working for Linux to run a network maintenance
>tool I recently acquired. TCP/IP works fine I just need to use chargen. I
>have Red Hat Linux 5.0 with the 2.0.32 kernal and was not able to find the
>deamon in the Runlevel Manager? I am not new to Unix but I have never had
>to do system admin stuff except in AIX (I used some spiffy little SMIT
>tool), so I really don't know where to get this deamon or how to install
it.
>Any ideas?
>
>
------------------------------
From: [EMAIL PROTECTED]
Subject: Mapping a Linux drive in Win 95 via TCP/IP
Date: Fri, 26 Feb 1999 09:14:27 GMT
Hello there,
I've just joined this newgroup with the hope that you can answer the
one-and-only question I have (so far:).
I'm a newbie to Linux, coming from good 'ol DOS, Win 95/NT (please note:
_huge_ amounts of sarcasm there!). So please bear with me.
I have a Ethernet peer-to-peer network of Win 95 machines. Hooked up
to this network is one Linux machine (running Redhat 5.2). My wish is
to have each individual user to be able to access his (or her) directory
on the Linux machine _as if_ it were a Win 95 drive.
The Win 95 end of the bargin is the easy part (I hope). Hopefully I'll
just have to map the drive as usual. It's the Linux end of it that I'm
stuck with. I've tried no end of NFS stuff. I've even used a piece of
software from a place called Zannet where you put its own NFS server onto
the Linux machine. But the Zannet stuff crashes the Win 95 machine.
HELP! I've searched and searched and searched (etc.) to try and find
information about it (and information that I can understand).
If someone can give a little help I would be ever grateful. And you might
just be saving a poorely-funded school from having to fork out for a huge
amounts of money for an NT server.
TIA,
Oliver B. Tupman
ICT Technician, Alcester Grammar School.
http://webspace.dialnet.com/alcester_grammar/
============= Posted via Deja News, The Discussion Network ============
http://www.dejanews.com/ Search, Read, Discuss, or Start Your Own
------------------------------
From: "lunchmeat7" <[EMAIL PROTECTED]>
Subject: Re: Adding the chargen deamon
Date: Wed, 24 Feb 1999 14:10:19 -0500
to test to see if its there...
telnet localhost 19
and you should get a lot of characters across the screen.
------------------------------
** 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
******************************
