Linux-Networking Digest #139, Volume #10 Sun, 7 Feb 99 15:13:39 EST
Contents:
NFS Install help needed. ("Bassman")
Re: Need help on network config (Frank Adcock)
[ff] The Freefire Bulletin #4 (1999-02-07) Newsletter (Bernd Eckenfels)
Re: 10base-T recommendation? (Christian Aasland)
Re: Netscape crashes on mailto: links (Fred Wilson Horch)
Re: Name resolution problem ("Carl R. Friend")
Re: 10/100 Ethernet SWITCH (to be used for Fast Ethernet LAN, and (Christian
Aasland)
Re: Netbios not on TCP/IP and Linux? ("Carl R. Friend")
Re: POP/IMAP setup ("Carl R. Friend")
How to set up a router (Sukanta Kumar Hazra)
Re: PPP, kernel 2.2.1 and ppp-compress-nn modules (Douglas E. Mitton)
Re: A better DyNIP? (Stefan Monnier)
Synoptic Lattis Hub 2813 (molesr)
Re: PPP problems under 2.2.1 (Christian Bienia)
Re: Time Synchronization (Tarcus)
Re: mail - "we do not relay" error (Tarcus)
----------------------------------------------------------------------------
From: "Bassman" <[EMAIL PROTECTED]>
Subject: NFS Install help needed.
Date: Sun, 7 Feb 1999 18:03:52 -0000
I have a small network partially up and running.
1 P2 333 dual boot Win95 and RedHat 5.2.
4 workstations. 486dx2 33MHz, various amounts of RAM, Slakware 3.5
2 have 202 Mb HDD and CDROM, 2 Have 2x202MB HDD.
All have 3com 3c509 NICS with PNP turned off ( all tested and work ok).
IDE Controller has only 1 port. Therefore, I can only put either the disks
or 1 HDD +cdrom on the PCs.
The 2 machines with only 1 HDD have been configured using the CDROM and work
perfectly. I have tested the network out and they are able to ping and
telnet each
other and the P2 machine.
I thought I would be able to install RH5.2 onto the dual HDD machines via
NFS installation. but get "permission denied" message.
I wondered if this was to do with RH5.2 ( or my setup) and so tried from one
of the 486's
with Slakware, same message.
It is obviously my error, what am I doing wrong?
Can anyone help?
Regards.
Bassman
------------------------------
From: Frank Adcock <[EMAIL PROTECTED]>
Subject: Re: Need help on network config
Date: Mon, 08 Feb 1999 02:21:44 +1000
This is a multi-part message in MIME format.
==============00CF10053152C25018B81786
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
What is your netstat -rn output, and your traceroute output ?
bbw wrote:
> I have Linux (SUSE 5.3) installed on 2 machines (which also has Win 98
> installed) and linked via a hub. The two machine can communicate if
> both boot as Win 98. If one of the machines is boot as Linux, they can
> ping each other (and 'arp -a' will show that the H/W address are
> correctly received). However, if both macines are boot as linux, they
> cannot ping each other (not reachable), and 'arp -a' will show that the
> H/W address of the remote machine is al zeros. Both macine can only
> ping themselves. I have checked with ifconfig, route, netstat etc.
> (also try setting the H/W address manually using arp), and cannot figure
> out what else to try to made the two machine see each other in Linux.
> Would appreciate if any body can help or suggest some thing to try.
> Thanks in advance.
==============00CF10053152C25018B81786
Content-Type: text/x-vcard; charset=us-ascii;
name="frank.vcf"
Content-Transfer-Encoding: 7bit
Content-Description: Card for Frank Adcock
Content-Disposition: attachment;
filename="frank.vcf"
begin:vcard
n:Adcock;Frank
x-mozilla-html:FALSE
url:http://www.bushlife.com.au
adr:;;;Scarborough;QLD;4020;Australia
version:2.1
email;internet:[EMAIL PROTECTED]
fn:Frank Adcock
end:vcard
==============00CF10053152C25018B81786==
------------------------------
From: Bernd Eckenfels <[EMAIL PROTECTED]>
Crossposted-To: comp.security.unix,comp.security.firewalls,de.comp.security
Subject: [ff] The Freefire Bulletin #4 (1999-02-07) Newsletter
Date: 7 Feb 1999 18:51:01 GMT
The Freefire Bulletin #4 (1999-02-07) Newsletter
The Freefire Project tries to help Developers in building Firewall and
IT-Security Solutions based on Free Tools. You can find additional
information about the Project at the Homepage
<URL:http://sites.inka.de/lina/freefire-l/index.en.html>
0 Executive Info (Ec)
=====================
1 High Availability (HA) with Open Source (Ec)
Mission critical applications
2 Linux-HA HOWTO (hm)
Harald Milz talks about the Project
3 fake (Ec)
A network superserver with additional features
4 Eddie & Erlang/OTP (Ec)
Server Farms and distributed network applications
5 FreeQualizer (Ec)
Not realy a free tool - Load Balancer for FreeBSD
6 Misc (Ec)
ONE-IP Project, GFS, Linux Virtual Server and the Linux NAT Page
7 Other Freefire News (Ec)
New Drawbridge 3.0, Juniper now Open Source
1 High Availabiltiy with Open Source
====================================
Open Source solutions - for firewalls or to secure important Servers -
getting more and more common. Companies have realized that external support
isnt always the better solution (compared to inhouse know how and Open
Source). As the tasks get bigger and bigger more vital parts of a company
depend on Open Source systems (By Definition a firewall has to control all
external network traffic). Therefore it is time to add support for HA to
Free Systems. The most important points for HA are reducing single points of
failure (SPOF) and fault tolerant software.
Therefore the main focus of this bulletin is on HA solutions.
<URL:http://www.inka.de/sites/lina/freefire-l/index.en.htm>
2 Linux-HA HOWTO/Project (hm)
==============================
In today's commercial Unix marketplace, High Availability (HA) is key to
selling server solutions. Virtually every Unix supplier has their own HA
software solution to provide customers with near-fault-tolerant server
systems at moderate prices. As a rule of thumb, redundancy is used to
prevent the overall IT system from having single points of failure, a method
that has been common in space flight and general aviation for decades. The
common objective is to mask unplanned outages from users in a manner to let
users continue to work quickly.
Unplanned outages can severely hamper your operations. Two 1995 studies by
Oracle Corp. and Datamation showed that average businesses lost between
80,000 and 350,000 USD per hour of unplanned outages. After the 1993 World
Trade Center bombing, 145 of 350 businesses which were located in the
building had to close down within a year because they had no redundant IT
structure.
That being said, and keeping in mind that Linux starts conquering commercial
data centers and business critical use, it is clear that Linux needs its
High Availability solution(s) as every other Unix.
<URL:http://sunsite.unc.edu/pub/Linux/ALPHA/linux-ha/High-Availability-HOWTO.html/>
3 fake (Ho)
============
Fake has been designed to switch in backup servers on a LAN. It has been
used to success to switch in backup Mail, Web and Proxy servers during
periods of both unscheduled and scheduled down time. Fake allows you to take
over the IP address of another machine in the LAN by bringing up an
additional interface and making use of ARP spoofing. The additional
interface can be either a physical interface or an IP alias.
<URL:http://linux.zipworld.com.au/fake/>
4 Eddiware & Erlang/OTP
========================
Eddie is a toolkit for bulding HA Server farms. It comes with 4 components:
IP Migration, Load-Balancing DNS, Intelligent HTTP Gateway and Content
Replication. Those Aplications are bult on Ericssons Erlang/OTP System. A
Open Source Programming Language/Runtime-Environment for distributed HA
Solutions. Erlang proofed its usefullness in serveral big telecommunication
Projects of Ericsson. Based on Erlang it should be trivial to develop
Application level Proxies (which instantly support distributed operation).
Thanks to the Erlang/OTP Crew for their helpfull responses. The Erlang/OTP
System is abvailable in binary for Linux, Solaris, Windows and FreeBSD but
should run on any vanilla Unix, too.
<URL:http://www.erlang.org/>
<URL:http://www.eddieware.org/>
<URL:http://www.ericsson.se:800/cslab/>
5 FreeQualizer (Ec)
===================
Additional High Availability Software for FreeBSD is for example FreeQualizer, a Free
(non-source) Load Balacer for 2 nodes.
<URL:http://www.coyotepoint.com/freequalizer.shtml>
6 Misc (Ec)
===========
The ONE-IP Project is research in progress based on an experimental *BSD
Implementation. Two different approaches are made to have multiple servers
respond to th same address. Looks like a promising idea. Especially the
Broadcast-Based Dispatching looks like a god solution to the SPOF problem
with current load-balancing server.
<URL:http://http://www.research.att.com/~ymwang/papers/HTML/ONE-IP/S.html>
A lot more usefull information on Load Balancing and Adress Rewriting can be
found on the Linux Address Translation Page from Linas Vepstas.
<URL:http://http://www.linas.org/linux/load.html>
Interesting work is going on in the GFS Project. Shared SCSI Filesystem is
a good step towards Linux Clusters for HA Solutions
<URL:http://gfs.lcse.umn.edu>
The LinuxDirector is a Load Balancing Tool included in the Linux Virtual
Server Package.
<URL:http://proxy.iinchina.net/~wensong/ippfvs/>
7 Other Freefire News
=====================
Russel Nepper announced Drawbridge 3.0. Its running on FreeBSD and
supporting Filtering of bridged Data.
<URL:http://drawbridge.tamu.edu>
Obtuse Systems Corporation has announced an Open Source Version of the Juniper Firewall
Toolkit. This is a major step towards a complete and free Firewall Solution
based on existing tools.
<URL:http://www.obtuse.com/open_source/>
Authors of the current issue
============================
Ec - Bernd 'eckes' Eckenfels <[EMAIL PROTECTED]>
hm - Harald Milz <[EMAIL PROTECTED]>
Ho - Horms <[EMAIL PROTECTED]>
(c) Copyright 1999 Bernd Eckenfels and others
Bernd Eckenfels <[EMAIL PROTECTED]>, GERMANY
------------------------------
From: Christian Aasland <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.hardware
Subject: Re: 10base-T recommendation?
Date: Sun, 07 Feb 1999 16:30:40 GMT
My experiences... at the bottom is the reason I will never buy cheap
net-cards again:
Got a Linux running as a router between 100TX, 10baseT (still working on
this) and cable modem... The cards break down as:
1. Cable-modem: 3Com 3c900 - PCI 10baseT. Works great! Uses the 3c59x
"Boomerang/Vortex" driver. $42
2. 100TX: 3Com 3c905b - PCI 10/100. $49. Works great.
3. 10baseT - 3c509b - ISA, $29. Can't get it to work in Linux.
The fastest machine in the house has a DECnet (Linksys) 10/100 PCI card
that has no bus-mastering or I/O offloading (like the 3c905c Parallel
Tasking cards have). This machine also has the highest ping time when
connecting to our server. The two 233mmx's running 3c905b's are in the
10-15ms range whereas the PII/350 with the Linksys is in the 33-40ms range.
I haven't tried swapping cards yet, but considering that all machines are
running Win98, voodoo and 128 megs ram, I'm pretty sure the cheap network
card is the culprit. Based on this I wouldn't reccommend buying cheap cards
to try and save a couple $$'s.
Just my $0.02 - take it for what it's worth!
Keith W Sheffield wrote:
> I'm going to try to network my home machines together. Initially, I'll
> just use a crossover cable, but I plan on getting a hub and using IP
> Masq at some point in the future. The machines I'm using will be
> running linux.
>
> Does anyone have any recommendations on a good but inexpensive 10baseT
> PCI ethernet card and any suggested web pages/tips that I should check
> out before I begin this project?
>
> thanks.
>
> --
> We were unanimous - in fact everyone was unanimous.
>
> Keith Sheffield [EMAIL PROTECTED]
--
Christian Aasland
[EMAIL PROTECTED]
ICQ#: 30268555
AIM:caasland
------------------------------
From: Fred Wilson Horch <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.help
Subject: Re: Netscape crashes on mailto: links
Date: Sun, 07 Feb 1999 11:37:11 -0500
I'm running RedHat 5.2 (2.0.36) and the glibc version of Netscape 4.5.
Once I open Messenger (I can open then close it), mailto: links work.
Otherwise, I get the same problem.
I assume the Messenger component initializes something. I found it so
annoying that I've configured Netscape to start up Messenger by default
(under Edit -> Preferences -> Appearance). Hope this helps.
--Fred
Ivo Naninck wrote:
> I see this too...
> However, I believe it doesn't happen when the messager-window is open, [...]
>
> > Pete wrote:
> > I'm running Netscape Communicator 4.5 on a Redhat 5.1 machine. Everytime I
> > click on a mailto: link, netscape crashes and dies. [...]
> > Does anybody understand why Netscape is choking on the mailto links and
> > know how to fix this problem?
------------------------------
From: "Carl R. Friend" <[EMAIL PROTECTED]>
Subject: Re: Name resolution problem
Date: Sun, 07 Feb 1999 13:35:23 -0500
Chris Joyce wrote:
>
> Woops, my message seems to have been truncated when I posted
> it.Here is the remainder of it:
> My /etc/resolv.conf:
> domain compuserve.com
> nameserver primary
^^^^^^^
> nameserver secondary
^^^^^^^^^
Your problem is that /etc/resolv.conf requires the _addresses_
of the nameservers, not their _names_ (chicken-and-egg problem).
Put the IP addresses for "primary" and "secondary" into
/etc/resolv.conf and all should be well.
--
______________________________________________________________________
| | |
| Carl Richard Friend (UNIX Sysadmin) | West Boylston |
| Minicomputer Collector / Enthusiast | Massachusetts, USA |
| mailto:[EMAIL PROTECTED] | |
| http://www.ultranet.com/~crfriend/museum | ICBM: N42:22 W71:47 |
|________________________________________________|_____________________|
------------------------------
From: Christian Aasland <[EMAIL PROTECTED]>
Crossposted-To:
comp.dcom.lans.ethernet,linux.redhat.misc,linux.samba,comp.os.linux.hardware
Subject: Re: 10/100 Ethernet SWITCH (to be used for Fast Ethernet LAN, and
Date: Sun, 07 Feb 1999 16:37:35 GMT
As stated, the switches are really expensive... not really sure a full switch
is better than a hub for small networks, as switches only prevent collisions
on busy nets.
Anyway, I found an IBM ISA 10/100TX adapter on www.ebay.com for $24. Works
great in anything but Linux, however, so stay away from it if that's what
you're planning on using it for.
Numbers wrote:
> I am looking to buy one of these for my 100-BaseT home LAN (which also has
> some 10-BaseT boxen too), and the net connection will be a 10-BaseT Cable
> Modem interface.
>
> I am only looking at spending < $150. What should I buy?
>
> p.s. - What's the overall most stable and fastest 100-BaseT ethernet
> adapter for the Linux 2.2 kernel? Also, is there such a thing as a ISA
> 10/100 Ethernet card?
--
Christian Aasland
[EMAIL PROTECTED]
ICQ#: 30268555
AIM:caasland
------------------------------
From: "Carl R. Friend" <[EMAIL PROTECTED]>
Subject: Re: Netbios not on TCP/IP and Linux?
Date: Sun, 07 Feb 1999 13:51:25 -0500
William Herrera wrote:
>
> No problem with sharing the Linux box drives of course. But the
> problem is in sharing the Windows95 box drives on Netbios over TCP/IP.
> I am told this is a security risk (and we are talking very
> confidential records here) with Windows that does not exist with
> netbios on netbieu due to the fact netbeui won't transfer info over
> into the TCP/IP based extra-nets.
Microsoft "networking" is, in and of itself, a security risk. Even
with "N"T, they still cling to the use of plain-text-equivalent pass-
words and obsolete protocols. Once you share anything sensitive from
a Win[9x|NT] box it's pretty much open for grabs.
On a Linux box (running Samba) shaing the sensitive data, you
could always set up firewalling filters (with ipfwadm or ipchains)
to allow only certain machines to connect to the sensitive data. In
that case, the kernel would block incoming requests before Samba ever
saw them. On top of that, Samba has its own control mechanisms. Any
firewall to the outside (Internet) should block both incoming and
outgoing NBT as there are several exploits which can lead to password
divulgence without the user even knowing it.
Of course, even with that, the data pass over the wire unencrypted
anyway so anybody with a sniffer (on your segment) could snarf a copy
of what passed between the server and the client box. But that's just
a built-in risk you'll have to look at and consider.
> Sigh. Wonde> why we have AppleTalk for Linux but not netbieu for
> Linux?
Perhaps because the general perception is that NetBIOS/BEUI/NBT
is so broken that it's not worth dealing with?
--
______________________________________________________________________
| | |
| Carl Richard Friend (UNIX Sysadmin) | West Boylston |
| Minicomputer Collector / Enthusiast | Massachusetts, USA |
| mailto:[EMAIL PROTECTED] | |
| http://www.ultranet.com/~crfriend/museum | ICBM: N42:22 W71:47 |
|________________________________________________|_____________________|
------------------------------
From: "Carl R. Friend" <[EMAIL PROTECTED]>
Subject: Re: POP/IMAP setup
Date: Sun, 07 Feb 1999 13:58:35 -0500
Sean McEwan wrote:
>
> I got PPP access to my linux box, now I am trying POP/IMAP services,
> and they half-work. I can retrieve mail using Netscape or Eudora, but
> cannot send mail.
It sounds like you don't have an SMTP (sendmail, usually) process
running on the server machine. POP only handles the retrieval of mail,
not the process of _sending_ it.
> From Eudora, I get 'Connection Refused (10061)'. From Netscape, I get
> connection refused. I don't have DNS working, so I am trying to mail
> to [EMAIL PROTECTED]
Yup. You don't have an SMTP process running on 192.168.1.2 that
the mail client needs to connect to; hence the "refused" messages. As
a last resort, if you're only using your machine to send/receive
Internet e-mail, you can point the "outgoing" mail at your ISP's
SMTP box. If you want full LAN functionality, you'll have to deploy
a mail-handler SMTP-compliant package on your LAN someplace (possibly
on 192.168.1.2).
--
______________________________________________________________________
| | |
| Carl Richard Friend (UNIX Sysadmin) | West Boylston |
| Minicomputer Collector / Enthusiast | Massachusetts, USA |
| mailto:[EMAIL PROTECTED] | |
| http://www.ultranet.com/~crfriend/museum | ICBM: N42:22 W71:47 |
|________________________________________________|_____________________|
------------------------------
From: Sukanta Kumar Hazra <[EMAIL PROTECTED]>
Subject: How to set up a router
Date: Mon, 8 Feb 1999 00:36:01 +0800
How can I set up my linux box running kernel Linux 2.2.0 to act as a
router. The situtaion is such
My linux box named "suk" has two NICs both of which are working fine. My
second computer named "suk2" is running NT4 . It has one NIC. Suk is
connected to the university LAN by eth0 and obtains ip by DHCP . suk works
perfectly and I can access the net and everything. I can telnet to suk
from suk2 and the personal network seems okay.
The ip of eth1 on suk is 192.0.0.1 and the ip of eth0 on the nt machine is
192.0.0.2 . I compiled ip_forwarding ,gatewaying , masquerading support in
the linux
2.2.0 kernel. Now all I want to do is to be able to have access to the
internet from my NT machine and browse webpages etc. There are no security
concerns etc here, so
can someone please suggest a simple method by which I can set up the
router.
I shall be extremely thankful, as I am just a beginner and even though I
read the manuals I wasn't able to make much sense out of them. If possible
can you kindly tell me the specific commands to use and other things, so
that I can set up the router easily.
Thanks a lot
Sukanta
======================================================================
+ "testing? What's that? If it compiles, it is good, +
+ if it boots up it is perfect." +
+ Linus Torvalds +
======================================================================
------------------------------
From: [EMAIL PROTECTED] (Douglas E. Mitton)
Subject: Re: PPP, kernel 2.2.1 and ppp-compress-nn modules
Reply-To: [EMAIL PROTECTED]
Date: Sun, 07 Feb 1999 19:05:08 GMT
Thank you! I found this to be valuable information as well!
I couldn't find any references to this in the ppp 2.3 or linux
sources.
Darrell Tangman <[EMAIL PROTECTED]> wrote:
>cybear <[EMAIL PROTECTED]> wrote:
>> I've gone blind searching for solution....
>
>Not the traditional cause for blindness, but never mind.
>
>> I upgraded from 2.0.36 (redhat 5.2) to 2.2.1 and upgraded to the latest
>> PPP, both are up and working. However, whenever I launch ppp I get the
>> following error message sent to /var/log/messages:
>
>> modprobe: can't locate module ppp-compress-21
>> modprobe: can't locate module ppp-compress-26
>> modprobe: can't locate module ppp-compress-24
>
>Try adding the following to /etc/conf.modules:
>
> alias ppp-compress-21 bsd_comp
> alias ppp-compress-24 ppp_deflate
> alias ppp-compress-26 ppp_deflate
>
>Without these ppp would still work fine, you just wouldn't get the
>advantage of the ppp compression schemes that may or not be available
>through your ISP in any case.
>
>--
>Darrell Tangman -- [EMAIL PROTECTED] -- Augusta, Georgia, USA
================================================
Doug Mitton - Brockville, Ontario, Canada
'City of the Thousand Islands'
Amateur Radio: VE3DMZ ([EMAIL PROTECTED])
EMail: [EMAIL PROTECTED]
http://www.cybertap.com/dmitton
User Group: http://signals.rmc.ca/klug
SPAM Reduction: Remove "x." from my domain.
================================================
------------------------------
From: Stefan Monnier <[EMAIL PROTECTED]>
Subject: Re: A better DyNIP?
Date: 07 Feb 1999 14:03:57 -0500
Miguel Cruz <[EMAIL PROTECTED]> writes:
> Yes, that reminds me of several others:
> www.ml.org
> www.ml.org
Last time I looked this one doesn't work anymore.
> www.ml.org
> Oh yes, and www.ml.org
Stefan
-
-
-
------------------------------
From: molesr <[EMAIL PROTECTED]>
Subject: Synoptic Lattis Hub 2813
Date: Sun, 07 Feb 1999 12:13:35 -0600
Reply-To: [EMAIL PROTECTED]
I bought a used Synoptics Lattis Hub 2813. But with no manual. I tried
to search thru the net with no success.
I tried Nortel and the only documentation they had on the 2813 was
release and upgrade notes. Does
anyone know of a place that has a owners or users manual on the 2813. I
am trying to setup the
TFTP IP address on it and I cannot get the monitor screen to come up.
Any documentation or help
will be appreciated.
Thank You
------------------------------
From: Christian Bienia <[EMAIL PROTECTED]>
Subject: Re: PPP problems under 2.2.1
Date: 7 Feb 1999 16:49:46 GMT
Clifford Kite wrote:
> I've been paying particular attention to PPP connection problems for many
> months now and I've only seen one post - a recent one - that said this
> worked for him. I would certainly be interested in learning how you
> came to _know_ that ppp-2.3.5 has such a bug. It seems very unlikely
> that a bug that would affect virtually everyone using ppp-2.3.5 could
> have remained undiscovered until now. I've used 2.3.5 without problems
> under kernels 2.0.36 and 2.1.131, although not extensively.
I've posted a detailed bug-report to the linux-ppp mailing list a few
weeks ago, but as you have requested my analysis, here you are ;-) :
I've tested pppd 2.3.5 with kernel 2.0.36, 2.2.0-pre9 and 2.2.1. pppd
2.2.0 was only tested with kernel 2.0.36 and earlier ones.
I started with SuSE Linux 6.0 that comes "ready for kernel 2.2" , a
pre-installed pppd 2.3.5 and kernel 2.0.36. The connection was
established properly: ifconfig and route showed that everything was
configured as it should, my external TA (a Zyxel Eltite 2864ID, TA is
the abbreveation of 'Terminal Adapter') showed me, that a connection was
established and that data was transmitted in both directions.
But for me, the pppd-connection seemd to be a kind of black hole: I
couldn't manage to get even one valid packet out of the Internet:
- ping [local_IP_assigned_by_PPP-server] worked perfectly
- but ping [remote_IP_assigned_by_PPP-server] didnt't
- any attempt to transmit data (dns, http, ftp, ...) failed completely
- tcpdump -i ppp0 'hung' and returned no output
Note: pppd seems to send the data properly (but I haven't tested that,
it's just an estimation, at least it does send something) because the
behaviour of the LEDs of my TA is "normal". For example if I ping a
remote host, a packet is sent (TXD blinks once) and after a minimum
delay a packet returns (RXD blinks once again).
I wasn't able to change this, so soon I tried to replace the pppd 2.3.5
binary with the pppd 2.2.0 binary (fortunately, I had a copy of the old
binary) and I only made this change. The ppp-connection worked
immediately as it should.
I then upgraded to kernel 2.2-pre9 and took the new binary again. I got
exactly the same behaviour. The same with kernel 2.2.1 that I am using
now. Of course, I couldn't test pppd 2.2.0 with the new kernels.
I then downloaded the sources of pppd 2.3.5. I compiled and installed
it, but it didn't change anything.
But I didn't only test different kernel/pppd/pppd-configuration
combinations: One of my friends is using SuSE Linux 6, too. He has the
same TA as I have (Zyxel Elite 2864ID, as mentioned), but a different
provider. I convinced him that he copied me his pppd-configuration
files, including chat-script and pap-secrets (we're rather good friends,
you know). I tried this other configuration. I could connect to his
provider using his account, but again, transmission of data was
impossible (same details as above).
I then gave up, because I have tried everything I could imagine and
stayed with pppd 2.2.0. Fortunately, I read a posting about a workaround
for this problem (replacing 'crtscts' with 'xonxoff'). I did so and now
I am surfing with pppd 2.3.5 and kernel 2.2.1.
Ok, here you have a summary of the conclusions I have made:
- it's not a configuration problem, as pppd 2.3.5 didn't complain
(neither in the logs nor on the console), the connection itself is
established properly (including route + ifconfig) and pppd 2.2.0 works
perfectly.
- the problem has nothing to do with my TA, as my friend uses the same
and it works fine with him and I always used 'crtscts' without any
problems.
- it has nothing to do with my provider, as it still didn't work when I
tried my firend's one with my machine.
--> The reason for my agony must be located somewhere in the new pppd
2.3.5 / the pppd 2.3.5 - specific kernel sources. It probably only
appears with very special configurations, as I am belonging to a
minority with my problem.
If anybody is interested in detailed debugging-logs or megs of
configuration-fiels, don't hestitate and mail me. I hope that this can
help someone to find out what exactly doesn't work. I would greatly
appreciate if somebody would inform me about a (real) solution or any
progress concerning this.
Greetings,
Chris
------------------------------
From: [EMAIL PROTECTED] (Tarcus)
Subject: Re: Time Synchronization
Date: 7 Feb 1999 19:02:46 GMT
In article <79khp1$[EMAIL PROTECTED]>,
autodata <[EMAIL PROTECTED]> writes:
> I have two time sync issues ...
>
> 1. How do I configure the Linux box (server on our LAN) to get the
> time from a time server via the Internet?
Use NTP.
> 2. How can I get the WinXX machines on the lan to sync to the Linux
> server?
Use NTP!
NTP comes with a daemon that you run all the time, this will server
out the time to other machines on the network, you then use an NTP
client for Windows (such as Tardis) to sync to the linux box. The
linux box then syncs to an external source via the "ntpdate -b
<hostnames>" command. You could run ntpdate from your ip-up script,
so every time the link comes up the time is synced.
The NTP homepage is;
http://www.eecis.udel.edu/~ntp/
--
>From the keyboard of Tarcus himself, running Linux in the UK.
-- There are no facts, only opinions --
------------------------------
From: [EMAIL PROTECTED] (Tarcus)
Subject: Re: mail - "we do not relay" error
Date: 7 Feb 1999 18:58:10 GMT
In article <[EMAIL PROTECTED]>,
[EMAIL PROTECTED] writes:
> What does "we do not relay" from a mail recipient mean?
It means that you've installed a secure version of sendmail but
haven't put your own network into the permitted lists. Look in the
docs for information on relaying. How to get around this depends
entirely on how you set things up and what your circumstances are, but
as long as you don't panic sendmail's pretty easy to set up --- on
about the tenth time through!
As someone else put it some other time in some other place, one of the
big advantages of sendmail is that it's so configurable, but one of
the big DISadvantages of sendmail is that it's so configurable!
At any rate, the relaying message appears when neither of the
recipient of a message or the *machine making the incoming connection*
are considered local, or are present in the access database. The
contents of the "MAIL From:" line aren't checked as it's easy to fake.
--
>From the keyboard of Tarcus himself, running Linux in the UK.
-- There are no facts, only opinions --
------------------------------
** 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
******************************
