Linux-Networking Digest #56, Volume #12 Fri, 30 Jul 99 07:13:27 EDT
Contents:
PPP and Network Neighborhood (Kostis Mentzelos)
Re: Activate NIC at boot ("Kent Nilsen")
netatalk and the dreaded nbp_rgstr (Andy McFadden)
Re: NFS problem solved (mandrake)- additional info ("Kent Nilsen")
Re: OT: whats the best circuit sim. for Linux? (Einar S. =?iso-8859-1?Q?Ids=F8?=)
Re: OT: whats the best circuit sim. for Linux? (Einar S. =?iso-8859-1?Q?Ids=F8?=)
Ethertap device configuration - Writing RAW ethernet packets ("opteam4")
Re: Reverse Proxy & Load Balance + Redundancy ("S. Pidgorny")
Re: ppp problem - ISP does not respond to configure requests (W.G. Unruh)
Re: ifconfig - unbinding a IP from a NIC card? RH 5.2 (Josh Miller)
Re: Automatic reconnection using pppd (W.G. Unruh)
----------------------------------------------------------------------------
From: Kostis Mentzelos <[EMAIL PROTECTED]>
Subject: PPP and Network Neighborhood
Date: Fri, 30 Jul 1999 09:55:09 +0300
I have a local network with a unix server running samba and a lot of
win95 machines.
Of course I can see my server from any win95 Network Neighborhood.
Now I want to connect my home computer (win95) with the server at the
office.
So I configured the with PPP.
Everything woks fine with PPP connection.
when I am connected I can telnet, ftp, ping to my server.
But I can't see any one in my Network Neighborhood.
I thing this has to do with my smaba configuration.
Is it possible to connect from my home and access every computer at the
office from
Network Neighborhood?
Kostis Mentzelos
------------------------------
From: "Kent Nilsen" <[EMAIL PROTECTED]>
Subject: Re: Activate NIC at boot
Date: Fri, 30 Jul 1999 06:55:14 GMT
Morn igjen (even more Norwegian!)
I've been there and done that, no help... The driver is compiled into the
kernel, I have all the ifcfg-eth* files, and all three of them contain
ONBOOT=YES. The "Enabled" is there in Linuxconf. Weird stuff, it works fine
on other almost identical systems... Well, maybe I've got to recompile...
again, don't know how I could have done wrong in including the driver in the
kernel, but seems like I have to try it.
Thanks for the reply!
Kent R. Nilsen
Vidar Andresen <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]...
> In article <FwTn3.902$[EMAIL PROTECTED]>,
> "Kent Nilsen" <[EMAIL PROTECTED]> wrote:
> >Hello, I've got 3 3c905B NICs in a Mandrake 6.0 w/kernel 2.2.9-19MDK
> >(original kernel from distribution). My problem is that every time I
boot, I
> >have to start X, run netcfg and activate all NICs. I check the "Activate
at
> >boot", but it still won't do it automatically.
>
> Morn. (Thats norwegian)
>
> Linuxconf (if you do it that way) and set 1,2 and 3 to [*] Enabled.
>
>
> ��������������������Ĵ Adaptor 1 ���������������������Ŀ
> ڳ �
> �� [*] Enabled �
> �� Config mode (*) Manual ( ) Dhcp ( ) Bootp �
>
>
> It produce /etc/sysconfig/network-script/ifcfg-eth0
> ifcfg-eth1
> ifcfg-eth2
>
> With a setting of 'ONBOOT="yes"'
>
> if not? Check http://cesdis.gsfc.nasa.gov/linux/drivers/vortex.html
> for possible new driver, tweaks, /etc/conf.modules setting. And
> mailing list / web-archives.
>
> 'boot with pnp-os' in bios? cold boot or warm reboot?
>
> >So, I figured there is some obscure script that takes care of it, but
which,
> >and what is the proper syntax to get it to auto-activate?
>
> That would be to compile the driver it into the kernel and use lilo to
> activate it at boot.
>
> If you make a new kernel and _add_ it to lilo, that could be a way of
> testing it without disturbing the setup you already got.
>
> Mvh Vidar Andresen
------------------------------
From: [EMAIL PROTECTED] (Andy McFadden)
Subject: netatalk and the dreaded nbp_rgstr
Date: 30 Jul 1999 06:54:55 GMT
I've installed netatalk-1.4b2+asun2.1.3-4.i386.rpm under Red Hat 6.0 using
GnoRPM. No problems there. However, several seconds after the system
comes up, I get:
nbp_rgstr: Connection timed out
Can't register quake:Workstation@*
nbp_rgstr: Connection timed out
Can't register quake:netatalk@*
I found related Q&A on the netatalk FAQ-O-Matic, but neither of the two
suggestions (put an explicit "eth0" in /etc/atalk/atalkd.conf, and
uninstall/reinstall) had any effect.
I found one item on a netatalk message digest, but it was just a question
from 1.5 years ago from a guy whose machine crashed and then started
emitting these messages after the machine rebooted.
Has anybody figured out what this is all about? Internet searches on
"nbp_rgstr" turn up few hits (two hits in altavista, three hits from
hotbot, half a dozen articles on dejanews), but nobody seems to have an
answer.
The Ethernet card is a 3c59x "vortex"; ifconfig -a shows:
eth0 Link encap:Ethernet HWaddr 00:60:97:D8:CC:7D
inet addr:172.17.131.58 Bcast:172.17.131.63 Mask:255.255.255.248
EtherTalk Phase 2 addr:65280/156
UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
RX packets:3856 errors:0 dropped:0 overruns:0 frame:0
TX packets:3509 errors:0 dropped:0 overruns:0 carrier:0
collisions:0
lo Link encap:Local Loopback
inet addr:127.0.0.1 Mask:255.0.0.0
EtherTalk Phase 2 addr:0/0
UP LOOPBACK RUNNING MTU:3924 Metric:1
RX packets:125309 errors:0 dropped:0 overruns:0 frame:0
TX packets:125309 errors:0 dropped:0 overruns:0 carrier:0
collisions:0
Red Hat 6.0 appears to be Linux 2.2.5.
I believe the NBP registration is the machine talking to itself, which
makes me wonder if the lo config has more to do with it than the eth0. I
suppose it depends on how nbprgstr works.
Suggestions would be appreciated.
--
Send mail to [EMAIL PROTECTED] (Andy McFadden)
CD-Recordable FAQ - http://www.fadden.com/cdrfaq/ (a/k/a www.spies.com/~fadden)
Fight Internet Spam - http://spam.abuse.net/spam/ & news.admin.net-abuse.email
------------------------------
From: "Kent Nilsen" <[EMAIL PROTECTED]>
Subject: Re: NFS problem solved (mandrake)- additional info
Date: Fri, 30 Jul 1999 08:30:03 GMT
Kent Nilsen <[EMAIL PROTECTED]> wrote in message
news:j8Dn3.665$[EMAIL PROTECTED]...
> Hello (again)!
>
> Finally I got my NFS up and running, I've posted several messages and have
> gotten little response. The faqs and howtos are all for the old NFS
system.
> So, since other people are asking, and nobody seems to know much about
> knfsd, I post this in the hopes that it can help you:
>
> Problems: You get permission denied all the time when trying to mount a
SUN
> computer via the net, or maybe you manage to mount one NFS export and not
> the other. This might solve some other stuff too (lots of cool error
> messages in the nfs system).
I also had to ensure that the filesystem is mounted at boot by adding the
drive in /etc/fstab
Kent R. Nilsen
> Software: Mandrake 6.0 (and possibly RedHat 6.0, had similar problems
> there), both with KNFSD instead of old NFS.
------------------------------
From: Einar S. =?iso-8859-1?Q?Ids=F8?= <[EMAIL PROTECTED]>
Subject: Re: OT: whats the best circuit sim. for Linux?
Date: Fri, 30 Jul 1999 08:47:06 GMT
Some spice-clone would be my first choice. I haven't tried any
Linux-versions but a quick search found Berkeley Spice for Linux at
http://www.ee.mankato.msus.edu/~paco/spice.html
There also seems to be a tkspice ('simple GUI for Spice3') out there.
Text-based Spice can be very powerful, but it can be tricky to learn.
Good manuals are required. See if you can get spice with a gui first and
then maybe move on to text-based later.
//Einar
Brandon30X wrote:
> Any good circuit simulation programs will work,
> preferably with a nice interface.
> Thank You, sorry fo being off topic.
> ~Brandon
------------------------------
From: Einar S. =?iso-8859-1?Q?Ids=F8?= <[EMAIL PROTECTED]>
Subject: Re: OT: whats the best circuit sim. for Linux?
Date: Fri, 30 Jul 1999 09:08:39 GMT
This is where you want to go:
http://infopad.eecs.berkeley.edu/~icdesign/SPICE/
//Einar
Brandon30X wrote:
> Any good circuit simulation programs will work,
> preferably with a nice interface.
> Thank You, sorry fo being off topic.
> ~Brandon
------------------------------
From: "opteam4" <[EMAIL PROTECTED]>
Subject: Ethertap device configuration - Writing RAW ethernet packets
Date: Fri, 30 Jul 1999 11:28:26 +0200
Hi
I'm currently programming an ethernet bridg as a daemon.
I know there is an ethernet bridge solution in the kernel (and tried it, it
worked fine.)
But I need to nanlyzes the packet incoming and outgoing on each ethernet
segment.
I started to write a program using SOCK_PACKET , in PROMISCUOUS mode, but I
was only
able to read packets from eth0 and eth1 but not writing ther ethernet packet
back to eth0 and
eth1
I looked so at the tap* devices. I configured my kernel easily:
netlink_dev and ethertap support as modules
then inserting them with
#insmod netlink_dev
#insmod -x -o tap0 ethertap unit=0
#insmod -x -o tap1 ethertap unit=1
and activated my ethernet interface with
#ifconfig eth0 192.168.3.125 up promisc
#ifconfig eth1 192.168.3.126 up promisc
(routing through the default gw etc...)
#ifconfig tap0 192.168.3.127 up promisc
#ifconfig tap1 192.168.3.128 up promisc
then I add a route to tap0 and tap1 (to link eth0 and tap0, and eth1 and
tap1
In fact the aim is writing in /dev/tap0 is sending a raw ethernet packet to
eth0
and reading an ethernet packet in dev/tap0 is reading packet from eth0)
so I type
#route add 192.168.3.125 tap0
#route add 192.168.3.126 tap1
But thing doesn't work, my tap devices seem to be not connected to the
interface eth0 and eth1 as they
should...
could any one help me ?
thanx a lot !
============================================================================
================
LETEINTURIER Aurelien mail: [EMAIL PROTECTED]
Engineer at ANAGRAM NS,
OP'TEAM
Student at CNAM-IIE
FRANCE
------------------------------
From: "S. Pidgorny" <[EMAIL PROTECTED]>
Subject: Re: Reverse Proxy & Load Balance + Redundancy
Date: Fri, 30 Jul 1999 12:23:21 +0300
Crossposted-To:
comp.infosystems.www.servers.unix,microsoft.public.proxy,alt.unix.wizards
Such expensive gear. I'm pretty shure that commercial DDNS servers (also
monitoring host availability) could help much in this case.
Svyatoslav Pidgorny, MVP, MCSE
Tony Finch wrote in message
<[EMAIL PROTECTED]>...
>[EMAIL PROTECTED] wrote:
>>
>>I would like to load balance two apache servers running on two different
>>physical machines as well as make them redundant using Reverse Proxy
>>Server as load balancer and duplicating the content from one machine to
>>another one using a shared file system and scheduled duplications of
>>files. The content to be served is hihgly static.
>
>You'll probably be better off with a layer 4 switch rater than a
>reverse proxy, because it will be able to monitor the servers for
>availability and load. Reverse proxies are helpful for reducing the
>load on back end servers if the content is static (or if it is
>dynamically generated in such a way that the same URL leads to the
>same content), but they don't really improve availability. For
>back-end filesystem purposes a NetApp is good.
>
>Tony.
>--
>f.a.n.finch [EMAIL PROTECTED] [EMAIL PROTECTED] e pluribus unix
------------------------------
From: [EMAIL PROTECTED] (W.G. Unruh)
Subject: Re: ppp problem - ISP does not respond to configure requests
Date: 30 Jul 99 09:30:25 GMT
I suspect strongly that your ISP is using CHAP/PAP authentication and
that the login authentication is leading yo uup a blind alley.
Follow
axion.physics.ubc.ca/ppp-linux.html for step by step instructions for figuring out
what it is that your ISP wants.
"Amir J. Katz" <[EMAIL PROTECTED]> writes:
>Hi,
>I have a problem with my Linux system at home - it's a P-II 350, 64MB
>RAM with an external 28.8 modem, dual boot - Win98/Redhat Linux 6.0.
>When I am in Win98, I have no problem dialing up to my ISP and get
>connected.
>However, when in Linux, I use 'kppp' to dial in and it fails. It gets
>successfully past the 'connect' and 'user/password' authentication (both
>when I specified 'script-based' and 'PAP'), but when pppd starts, it
>times out, complaining that the configuration requests that it sends to
>the ISP side are not answered. I can see on the modem lights that it
>talks with the other side.
>I tried different servers at my ISP and the behavior is the same, so I
>assume that the problem is either on my side, or that all the ISP's
>server are configured wrong from Linux ppp's point of view.
>However, kppp does not provide any parameters that will affect the
>behavior of pppd in that aspect.
>Question is: How can I debug the ppp session so that I have some details
>when calling my ISP?
>Thanks.
>--------------2A4839C93792A6F1C5CCB6A5
>Content-Type: text/x-vcard; charset=us-ascii;
> name="amir_katz.vcf"
>Content-Transfer-Encoding: 7bit
>Content-Description: Card for Amir J. Katz
>Content-Disposition: attachment;
> filename="amir_katz.vcf"
>begin:vcard
>n:Katz;Amir J.
>tel;fax:+972-3-645-1100
>tel;work:+972-3-645-1145
>x-mozilla-html:FALSE
>url:http://www.bmc.com
>org:BMC Software, Inc.;IT Process Automation Business Unit
>version:2.1
>email;internet:[EMAIL PROTECTED]
>title:Unix, Perl and Linux Hacker
>adr;quoted-printable:;;P.O. Box 58168=0D=0A;Tel-Aviv;;61581;Israel
>fn:Unix Maven
>end:vcard
>--------------2A4839C93792A6F1C5CCB6A5--
------------------------------
From: Josh Miller <[EMAIL PROTECTED]>
Subject: Re: ifconfig - unbinding a IP from a NIC card? RH 5.2
Date: Fri, 30 Jul 1999 03:17:08 -0400
I'm not on a linux box right now, sgi's route command differ's a bit... so
I"m not sure the correct order, but man route has some great examples.
I think what you'll need is something like:
/sbin/route -del IP.To.Un.Bind
or maybe
/sbin/route -del IP.To.Un.Bind dev eth0:31
good luck
--
Josh I.
[EMAIL PROTECTED] wrote:
> In article <[EMAIL PROTECTED]>,
> Alex Harrington <[EMAIL PROTECTED]> wrote:
> > Not sure but maybe this will work??
> >
> > (I have split the ip into sections for ease of writing it as follows:
> > 192.168.1.1
> > a . b .c.d
> > so for 'a' read '192' etc..)
> >
> > /sbin/ifconfig eth0:d a.b.c.d
> > /sbin/route add -host a.b.c.d dev eth0:d
> >
> > replacing a,b,c and d with the correct IP address for your Linux box.
> >
> > This may work, it may not - I really don't know - just an intelligent
> > guess.
> >
> > NO GUARANTEES. You may be advised to wait for some more replies from
> > somebody who has more experience here than me.
> >
> > Alex -
> >
> > [EMAIL PROTECTED] wrote:
> > >
> > > Help!
> > > I accidently bound a workstation IP address to my Red Hat 5.2 server
> > > ethernet card and and now I get the message on the workstation that
> I
> > > have a IP address conflict.
> > >
> > > Here are the commands that I did on on my Red Hat 5.2 Primary
> Server:
> > >
> > > /sbin/ifconfig eth0:31 200.200.200.31
> > > /sbin/route add -host 200.200.200.31 dev eth0:31
> > >
> > > Please post the commands that will unbind this IP from the server so
> > > that I can reuse this IP for a workstation.
> > >
> > > Thank You,
> > > A new web administrator.
> > >
> > > Sent via Deja.com http://www.deja.com/
> > > Share what you know. Learn what you don't.
> >
>
> Sent via Deja.com http://www.deja.com/
> Share what you know. Learn what you don't.
------------------------------
From: [EMAIL PROTECTED] (W.G. Unruh)
Subject: Re: Automatic reconnection using pppd
Date: 30 Jul 99 09:41:38 GMT
Guillaume BRIALON <[EMAIL PROTECTED]> writes:
>Hi,
>I haven't found any pppd version implementing the -demand, -persist, and
>-idle options.
>These are very usefull for automatic reconnection.
>Does someone knows a pppd version enabling these options?
>Else, how do you enable automatic reconnection without these?
The later (> 2.3.5) versions are supposed to impliment the demand
(not -demand surely) option for Linux. However, diald is probably the prefereed
way of instituting demand dialling.
------------------------------
** 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
******************************
